Approximately 40% of Massachusetts households (1.1 million total) experienced food insecurity in 2025, up from 19% in 2019. Photo Credit: Greater Boston Food Bank.

The El Niño that 4Hunger.org has tracked since spring, it is now official, on September 21, the daily Niño 3.4 anomaly broke the record set in 2015–16, and by October 3 it had reached 3.28°C.² Forecast models put it on track to be the strongest El Niño on record, peaking near 4.0°C in December, the strongest El Niño ever recorded.³ El Niño winters tilt the odds toward rain across drought-stricken California, the Southwest and the Gulf Coast, but that rain does not arrive until at least December, and it peaks between January and March 2027.⁴

Even then, rain is not food. The winter wheat now going into dry ground will not be harvested until next summer, and the same El Niño threatens the rice, sugar, cocoa and coffee harvests abroad that stock American shelves.⁵

Americans cannot wait that long for relief, because they are facing four shocks at once. A two-year drought has produced the smallest winter wheat harvest in decades and shrunk the U.S. cattle herd to a multi-generational low. The war with Iran pushed diesel, the fuel on which every truck, tractor and food bank depends, to a record price and drove up gasoline and fertilizer with it. Tariffs are being passed through to grocery shelves. And the 2025 reconciliation law, the One Big Beautiful Bill Act (OBBBA), has begun cutting the Supplemental Nutrition Assistance Program (SNAP) and Medicaid, the two programs that stand between a price shock and an empty table. Each of these pressures was forecast, and each can now be measured.

The Squeeze, Measured

The OBBBA, signed on July 4, 2025, cut federal SNAP spending by almost $187 billion over ten years. The Congressional Budget Office (CBO) found that the law as a whole will leave the poorest tenth of households about $1,200 a year worse off, while the richest tenth gain about $13,600 a year, mainly through tax cuts.⁶ CBO also estimated that it would reduce SNAP participation by roughly 3 million people in an average month, most of them through expanded work requirements,⁷ which now reach adults up to age 64 and parents whose youngest child is 14 or older, and no longer exempt veterans, people experiencing homelessness or young adults leaving foster care.⁸

The real decline has been larger. SNAP participation fell from 41.68 million in June 2025 to 36.35 million in June 2026, a drop of 5.3 million people, or 12.8 percent, in a single year,⁹ running ahead of CBO’s estimates.¹⁰ In Massachusetts, the number of SNAP households fell 15 percent, and a class-action lawsuit alleges that the state’s SNAP assistance line, which automatically disconnected 74 to 81 percent of calls in the first half of 2026, is cutting off eligible people.¹¹ The Greater Boston Food Bank reports that a record 40 percent of the state’s households, about 1.1 million, experienced food insecurity in 2025.¹² In North Carolina, where SNAP rolls have fallen by 110,129 people, a Guilford County food nonprofit reports demand up 20 percent from last year.¹³ The OBBBA is shifting people in need off federal food aid and onto charity, which makes clear that how much hunger there is in America is a political choice.

The next phase began on October 1, when states became responsible for 75 percent of SNAP’s administrative costs, up from 50 percent. Starting October 1, 2027, states will also pay up to 15 percent of benefit costs, depending on their payment error rates.¹⁴ With federal Medicaid cuts landing on state budgets at the same time, the incentive is to tighten enrollment rather than expand it. Massachusetts, with a 12.5 percent error rate, would owe the maximum share, an estimated $213 million a year, yet its fiscal 2027 budget cuts caseworker funding by about $21 million.¹⁵

Health cuts compound the harm. KFF, a nonpartisan health policy research organization, puts the law’s federal Medicaid reductions at $911 billion through 2034, with 10 million more people uninsured; Medicaid work requirements are scheduled to begin on January 1, 2027.¹⁶ In Massachusetts, 99,000 to 202,000 people could lose MassHealth coverage by 2028,¹⁷ in a state that had the nation’s lowest uninsured rate, 2.8 percent, in 2024. These cuts come on top of a drop of roughly 8 million in Medicaid, CHIP and marketplace enrollment since early 2025.¹⁸ For a household on the margin, a medical bill and a grocery bill come out of the same budget.

A Population Without Slack

American households with the bottom 50% in income entered 2026 with almost no financial buffer. Sixty percent of the people in Feeding America’s survey do not feel financially stable,¹⁹ and only 30 percent of all Americans could pay a $1,000 emergency expense from savings.²⁰

Policymakers are now flying blind. On September 20, 2025, USDA terminated its annual Household Food Security report, calling it “redundant, costly, politicized, and extraneous.”²¹ The final edition found 47.9 million people food insecure in 2024.²² The Urban Institute found that 24.2 percent of adults reported household food insecurity in December 2025, although the two surveys are not directly comparable.²³ Just as the largest benefit cuts in SNAP’s history, cuts to Medicaid and subsidies for ACA health insurance take effect, the federal government has stopped publishing the report designed to show their consequences – how hungry are people in America.

Drought in the Meat Case and the Bread Aisle

Grocery inflation overall has been moderate: food-at-home prices rose 2.2 percent in the twelve months to August 2026.²⁴ The drought’s damage is concentrated in protein. USDA’s Economic Research Service (ERS) forecasts beef and veal prices to rise 9.4 percent in 2026, after rising 10.6 percent in the 12 months to June 2025.²⁵ Repeated droughts made feed expensive and withered pastures, forcing ranchers to sell breeding stock, and the beef cow herd is now the smallest since 1961.²⁶ Rebuilding it takes years.

Photo Credit: Thomas Barwick

Grain tells the same story. U.S. winter wheat production fell 27 percent this year, to 1.02 billion bushels, and harvested area fell to a record low.²⁷ Roughly a third of planted acres were never harvested, including about half in Oklahoma and seven in ten in Texas.²⁸ USDA projects that wheat stocks will fall 22 percent this marketing year and that the price farmers receive will rise from $5.06 to $6.40 a bushel.²⁹ The 2027 crop is now going into dry ground, where seed that sprouts after a light shower can die if dry weather returns.³⁰

War at the Pump: The Cost of Moving Food

The second shock is geopolitical. The U.S. and Israeli strikes on Iran that began on February 28, 2026, led to what the U.S. Energy Information Administration (EIA) called a “de facto closure of the Strait of Hormuz,” through which about 27 percent of globally traded oil and up to 30 percent of global fertilizer trade passed in 2024. Brent crude nearly doubled in the first quarter, from $61 to $118 a barrel.³¹ Diesel reached a record $6.53 a gallon on September 21 and, at $6.20 on October 5, was still 67 percent higher than a year earlier.³²

Food banks run on diesel trucks, and every extra dollar spent on fuel comes out of the money for food. Houston Food Bank’s diesel price rose 69 percent between January and April,³³ and food banks in Alabama and Virginia report fuel costs up 68 to 70 percent from last year.³⁴ “If I am spending more on fuel, I can’t buy as much food,” said Cheryl Schondek, the Greater Boston Food Bank’s chief operating officer.³⁵ The Food Bank of Iowa estimates that its higher diesel bill costs it the equivalent of more than 8,000 meals a month.³⁶

Households are paying at the pump as well. Regular gasoline averaged $4.354 a gallon on October 5, about 39 percent more than a year earlier.³² “Most typically, they’re having to pull it from the grocery budget,” Amy Breitmann, of Golden Harvest Food Bank in Augusta, Georgia, told NPR in May.³⁷

Gas station in Hudson, MA October 6, 2026, Photo Credit Mark W. Roberts

Fertilizer is where the war reaches the next harvest. By April, the World Bank reported, urea prices had climbed above $850 per metric ton, up 80 percent since February.³⁸ In late September, anhydrous ammonia was still 25 percent more expensive than a year earlier,³⁹ and ERS projects that farm fertilizer expenses will rise 15.3 percent this year and fuel expenses 28.8 percent.⁴⁰

Tariffs and the Shrinking Federal Pantry

The third pressure, tariffs, lands on a food supply already weakened by federal cuts. Since March 2025, USDA has cancelled the Local Food for Schools ($660 million) and Local Food Purchase Assistance ($420 million) programs, halted about $500 million in food deliveries to food banks⁴¹ and terminated the Regional Food Business Centers that helped small and mid-sized farms reach buyers such as food banks.⁴² Each cut removed both a market for farmers and food that stretched pantry budgets. Texas food banks have absorbed $88.4 million in federal funding cuts and 40 percent fewer federal commodities this year,⁴³ and the Greater Boston Food Bank lost nearly $2.3 million in emergency food resources to USDA reductions in 2025.⁴⁴

Food banks are now buying replacement food just as wholesale prices rise. In September, Campbell’s and Conagra announced price increases, citing tariffs alongside energy, fertilizer, diesel and crop costs.⁴⁵ The Yale Budget Lab estimates that the tariffs now in force will cost the average household about $1,100 a year.⁴⁶

The Farm Ledger

All four pressures converge on farmers themselves. Chapter 12 farm bankruptcy filings rose 46 percent in 2025, to 315,⁴⁷ and April 2026 produced the highest monthly total since February 2020.⁴⁸ Filings are on pace to rise again this year.⁴⁹ ERS forecasts net farm income down 5.5 percent after inflation and farm debt at a record $605.1 billion, with government payments supplying about 30 percent of net farm income.⁵⁰ Distress is rising fastest in the same Southern Plains and southeastern regions where the drought is deepest. A farm that fails this autumn plants nothing next spring.

Washington’s own decisions have cut both ways. The 2025 tariff war cut soybean growers off from China;⁵¹ purchases resumed after a truce in late October 2025, but a 10 percent Chinese retaliatory duty still limits sales.⁵² U.S. tariffs on imported farm inputs collected $958 million through October 2025.⁵³ With the Iran war layered on top, a Farm Bureau survey of more than 5,700 farmers in April found that 70 percent could not afford all the fertilizer they needed for their 2026 crop.⁵⁴

Grain barges on the Mississippi River. Photo credit: Getty Images

The government has paid farmers to absorb much of the damage, through higher reference prices,⁵⁵ $12 billion in Farmer Bridge Assistance⁵⁶ and a plan for new fertilizer capacity that will not come online for 12 to 18 months.⁵⁷ Most of that money flows to the largest operations. Only about 40 percent of farms grow crops eligible for reference-price payments, and the top 10 percent of those farmers have collected nearly three-quarters of all those payments.⁵⁸ The law raised the annual payment limit from $125,000 to $155,000,⁵⁹ and bridge payments grow with every acre planted.⁶⁰ Market concentration deepens the squeeze: four companies control more than 75 percent of the corn and soybean seed market.⁶¹

El Niño: Limited Relief on a Winter Schedule

Against these four pressures, El Niño offers one partial and delayed remedy. NOAA’s Climate Prediction Center (CPC) gives better than 90 percent odds of a very strong event through the winter.¹ Its September 17 outlook gives 60 to 70 percent odds of above-normal precipitation across southern California and the Southwest from December through April, and 70 to 80 percent over southern Georgia and northern Florida, but only 40 to 50 percent over the Southern Plains, where the drought is deepest.⁴ CPC expects the effects to “increase slowly at first before quickly ramping up during the core winter months.”⁶²

The forecast has not failed. It has simply not yet arrived. As of September 29, drought covered 59 percent of the Lower 48 and affected 111.6 million people in 42 states.⁶³ In Oklahoma, 99.8 percent of the state is in drought. Parts of Texas and Oklahoma would need 20 to 25 inches of rain over three months to end it,⁶⁴ and on drought-hardened soil, heavy downpours run off as flash floods instead of soaking in. What the wheat needs is steady rain, week after week, through the winter.⁶⁵

That deficit matters now because the 2027 wheat crop is being planted now. As of October 4, 36 percent of U.S. winter wheat was in the ground, against a five-year average of 46 percent. Oklahoma was 14 percent planted, and more than half its topsoil was short of moisture.⁶⁶ Seed sown into dry soil in October establishes poorly, no matter what rain falls in February.

Dust flies as a tractor prepares a dry field for wheat to be seeded. (Photo by Todd Johnson, OSU Agricultural Communications Services)

The Calendar

Set side by side, the policy calendar and the climate calendar run in parallel, not in sequence:

October 1, 2026. States begin paying 75 percent of SNAP administrative costs.¹⁴

October 2026. Winter wheat planting continues across the Southern Plains into topsoil that is predominantly short of moisture.⁶⁶ CPC’s next ENSO update is due October 8.

December 2026. CPC’s El Niño precipitation signal begins to emerge across the southern tier.⁴

January 1, 2027. Medicaid work requirements take effect.¹⁶

January–March 2027. The El Niño precipitation signal peaks across California, the Southwest and the Southeast.⁴

October 1, 2027. States begin paying up to 15 percent of SNAP benefit costs, depending on their payment error rates, for the first time in the program’s history.¹⁴

Throughout this period, no federal survey will measure American household food insecurity.²¹

Who Bears the Burden of These Four Price Shocks?

The burden of a price shock depends on the share of income spent on food. The wealthiest fifth of households spend less than 6 percent of their income on food; the poorest fifth spend 33 percent.⁶⁷ Those households are not evenly distributed. In 2024, 24.4 percent of Black households, 20.2 percent of Hispanic households and 30.9 percent of American Indian and Alaska Native households were food insecure, compared with 10.1 percent of white, non-Hispanic households.⁶⁸ In Massachusetts, 63 percent of Hispanic households and 51 percent of Black households were food insecure in 2025, and 78 percent of SNAP households received less than $300 a month in benefits.¹²

Four Shocks, Four Choices

None of these four shocks requires new science to answer. Each requires a policy decision by people who already have the data and the power to act.

The drought: protect next year’s harvest now. The latest extension of the 2018 Farm Bill expired on September 30, and Congress has not agreed on a replacement.⁶⁹ It should pass one in the lame-duck session that directs help to the small and midsized farms of the Southern Plains and the Southeast, not ever-larger checks for the largest operations. The Fair Seeds for Farmers Act, introduced by Sen. Elizabeth Warren and Rep. Jim McGovern of Massachusetts, belongs in that debate, too.⁶¹

The war: keep the trucks rolling and the fields fertilized. Emergency fuel assistance for food banks and restored USDA purchases for The Emergency Food Assistance Program cost far less now than after food banks have had to cut their delivery. Farmers need help with fertilizer this season, not in 12 to 18 months.⁵⁷

The tariffs: stop taxing the grocery cart. Food, and the fertilizer, chemicals and machinery farmers need to grow it, should be exempt from import tariffs, and USDA should restore the local food purchasing programs that bought from small farms and supplied food banks and schools.⁴¹

The cuts: stop shifting hunger onto states and charities. Congress should defer the October 2027 SNAP cost shift, pause the expanded work requirements while it examines why participation is falling faster than CBO projected,¹⁰ and delay the Medicaid work requirements. Massachusetts can and must fund the caseworkers, and phone lines its own SNAP system needs.¹⁵

Restore the measurement. Reinstating USDA’s food security report would cost little.²¹ A government that stops counting the hungry does not reduce their number; it only hides it.

El Niño’s rain is still coming, and it may yet rescue next year’s wheat. But no storm will restore a SNAP benefit, refill a food bank’s fuel tank, repeal a tariff or reopen a farm that has already gone under.

The drought is the only one of these four shocks that nature can end. The war, the tariffs and the cuts were set in motion by decisions in Washington, and decisions can be changed. Ask your members of Congress where they stand on each of these four choices before Election Day on November 3.

Endnotes

1. NOAA Climate Prediction Center, “ENSO: Diagnostic Discussion,” September 10, 2026.

2. Carbon Brief, “Analysis: ‘Super El Niño’ reaches ‘remarkable’ all-time record,” September 21, 2026; The Climate Brink, Climate Dashboard, October 3, 2026 data.

3. The Climate Brink, “The impacts of this year’s record-shattering El Niño,” September 28, 2026; see also note 2.

4. NOAA Climate Prediction Center, Prognostic Discussion for Long-Lead Seasonal Outlooks, September 17, 2026.

5. ConfectioneryNews, “Which food commodities are threatened most by ‘Godzilla’ El Niño?” September 16, 2026.

6. Congressional Research Service, Report R48552, August 15, 2025; Congressional Budget Office, distributional analysis of Public Law 119-21, August 11, 2025.

7. Congressional Budget Office, Estimated Effects of Public Law 119-21 on SNAP, August 11, 2025.

8. See note 6.

9. USDA Food and Nutrition Service, SNAP Data Tables, data as of September 11, 2026.

10. Center on Budget and Policy Priorities, “Congress Should Stanch SNAP Participation Losses, Which Exceed CBO’s Estimates,” June 30, 2026.

11. The Boston Globe, “Thousands in Massachusetts are losing food assistance because of disconnected SNAP calls, lawsuit says,” September 27, 2026; State House News Service via GBH, “Disconnected,” September 28, 2026.

12. Massachusetts DTA, Performance Scorecard, July 2026; Greater Boston Food Bank, Massachusetts Food Access Report, April 7, 2026.

13. North Carolina Health News, “SNAP changes push more North Carolina families to food pantries, food banks,” August 4, 2026.

14. See note 6.

15. CommonWealth Beacon, “New error rate highlights risks for Massachusetts SNAP program,” July 6, 2026.

16. KFF, Allocating CBO’s Estimates of Federal Medicaid Spending Reductions Across the States, July 23, 2025; Georgetown University Center for Children and Families, reconciliation law explainer, August 13, 2025.

17. The Boston Globe, “Over 200,000 people could lose MassHealth under coming federal changes, report says,” March 27, 2026.

18. FactCheck.org, “One Big Beautiful Bill Not (Yet) Driving ACA, Medicaid Declines,” July 29, 2026; U.S. Census Bureau, Health Insurance Coverage by State: 2023 and 2024, 2025.

19. Feeding America, Elevating Voices: 2026 Insights Report, September 29, 2026.

20. Bankrate, 2026 Annual Emergency Savings Report, February 4, 2026.

21. USDA, “USDA Terminates Redundant Food Insecurity Survey,” September 20, 2025.

22. USDA Economic Research Service, Household Food Security in the United States in 2024, as summarized by Center on Budget and Policy Priorities, December 30, 2025.

23. Urban Institute, Food Insecurity Remained High in 2025, As Safety Net Cuts Loom, March 18, 2026.

24. Bureau of Labor Statistics, Consumer Price Index, August 2026; USDA Economic Research Service, Food Price Outlook, September 25, 2026.

25. See note 24; Bureau of Labor Statistics, Consumer Price Index, June 2025.

26. USDA NASS, Cattle, January 30, 2026; USDA ERS, Livestock, Dairy, and Poultry Outlook.

27. USDA NASS, Small Grains 2026 Summary, September 30, 2026.

28. Calculated from note 27.

29. USDA, World Agricultural Supply and Demand Estimates, September 11, 2026.

30. Oklahoma State University Extension, “Basic rules for planting wheat this fall,” October 4, 2021.

31. U.S. Energy Information Administration, Today in Energy, April 7, 2026; IFPRI, “The Iran war’s impacts on global fertilizer markets and food production,” 2026.

32. U.S. Energy Information Administration, Gasoline and Diesel Fuel Update, October 6, 2026.

33. Houston Food Bank, “Effects of Rising Fuel Costs,” April 27, 2026.

34. FOX10, Mobile, Alabama, September 22, 2026; 12 On Your Side, Richmond, Virginia, September 10, 2026.

35. WROR, “Greater Boston Food Bank Defies Rising Costs, Ensures No Service Cuts,” May 11, 2026.

36. KCRG, “Rising fuel costs threaten Iowa’s food bank deliveries,” April 30, 2026.

37. NPR, “More people are going hungry now than at the height of the pandemic,” May 27, 2026.

38. World Bank, “Fertilizer prices surge as Strait of Hormuz disruptions tighten supplies,” May 14, 2026.

39. DTN, “Anhydrous Leads Fertilizer Prices Higher, Up 25% From a Year Ago,” September 30, 2026.

40. USDA Economic Research Service, September 2026 Farm Income Forecast, September 3, 2026.

41. CBS MoneyWatch, March 13, 2025; CNN, March 22, 2025.

42. Civil Eats, “USDA Cancels More Support for Regional Food Systems,” July 15, 2025.

43. The Texas Tribune, “Texas food banks struggle amid federal cuts, drop in donations,” September 3, 2026.

44. WBUR, “Greater Boston Food Bank faces federal funding cuts,” April 7, 2025.

45. Bloomberg, “From Campbell’s to Conagra, Food Industry Warns of Price Hikes,” September 11, 2026.

46. The Budget Lab at Yale, The State of U.S. Tariffs, August 24, 2026.

47. American Farm Bureau Federation, “Farm Bankruptcies Continued to Climb in 2025,” February 9, 2026.

48. Epiq AACER, bankruptcy filing data, May 6, 2026.

49. Brownfield Ag News, “Farm bankruptcies on pace to increase in 2026,” September 17, 2026.

50. See note 40; American Farm Bureau Federation, Market Intel, September 3, 2026.

51. Iowa Soybean Association, “How We Got Here: The 2025 U.S.–China Soy Saga,” November 4, 2025.

52. High Plains Journal, “U.S., China buy more time to iron out trade wrinkles,” October 2, 2026.

53. Farm Progress, “Trump’s tariffs generate $958M but cost farmers far more,” January 22, 2026.

54. Farm Policy News, “70% of Farmers Can’t Afford All Needed Fertilizer, Survey Shows,” April 2026.

55. Iowa State University CALT, “Reviewing the Agricultural Provisions in the One Big Beautiful Bill Act,” July 11, 2025.

56. USDA, Farmer Bridge Payments announcement, December 8, 2025.

57. Farm Policy News, “US Set to Use Tariff Funds to Address High Fertilizer Prices,” April 20, 2026.

58. Environmental Working Group, “For 41 years in a row, USDA sent taxpayer-funded payouts to roughly 9,000 farmers,” July 13, 2026.

59. American Enterprise Institute, Changes for Agriculture in the One Big Beautiful Bill Act, October 15, 2025.

60. American Farm Bureau Federation, “Farmer Bridge Assistance Program: Details on $11 Billion in Aid,” January 2, 2026.

61. Office of Sen. Elizabeth Warren, Fair Seeds for Farmers Act press release, July 23, 2026.

62. NOAA Climate Prediction Center, Prognostic Discussion for Long-Lead Seasonal Outlooks, August 20, 2026.

63. U.S. Drought Monitor, Current Conditions, data valid September 29, 2026.

64. U.S. Drought Monitor, Weekly Summary, October 1, 2026.

65. WFIT/NPR, “How the risk for flash flooding increases after a drought,” August 11, 2022.

66. USDA NASS, Crop Progress, October 5, 2026.

67. USDA Economic Research Service, Food Prices and Spending, April 30, 2026.

68. See note 22.

69. Farm Policy News, “Farm Bill Expires Again as Lawmakers Continue Negotiations,” October 2, 2026.

Approximately 40% of Massachusetts households (1.1 million total) experienced food insecurity in 2025, up from 19% in 2019. Photo Credit: Greater Boston Food Bank.

The El Niño that 4Hunger.org has tracked since spring, it is now official, on September 21, the daily Niño 3.4 anomaly broke the record set in 2015–16, and by October 3 it had reached 3.28°C.² Forecast models put it on track to be the strongest El Niño on record, peaking near 4.0°C in December, the strongest El Niño ever recorded.³ El Niño winters tilt the odds toward rain across drought-stricken California, the Southwest and the Gulf Coast, but that rain does not arrive until at least December, and it peaks between January and March 2027.⁴

Even then, rain is not food. The winter wheat now going into dry ground will not be harvested until next summer, and the same El Niño threatens the rice, sugar, cocoa and coffee harvests abroad that stock American shelves.⁵

Americans cannot wait that long for relief, because they are facing four shocks at once. A two-year drought has produced the smallest winter wheat harvest in decades and shrunk the U.S. cattle herd to a multi-generational low. The war with Iran pushed diesel, the fuel on which every truck, tractor and food bank depends, to a record price and drove up gasoline and fertilizer with it. Tariffs are being passed through to grocery shelves. And the 2025 reconciliation law, the One Big Beautiful Bill Act (OBBBA), has begun cutting the Supplemental Nutrition Assistance Program (SNAP) and Medicaid, the two programs that stand between a price shock and an empty table. Each of these pressures was forecast, and each can now be measured.

The Squeeze, Measured

The OBBBA, signed on July 4, 2025, cut federal SNAP spending by almost $187 billion over ten years. The Congressional Budget Office (CBO) found that the law as a whole will leave the poorest tenth of households about $1,200 a year worse off, while the richest tenth gain about $13,600 a year, mainly through tax cuts.⁶ CBO also estimated that it would reduce SNAP participation by roughly 3 million people in an average month, most of them through expanded work requirements,⁷ which now reach adults up to age 64 and parents whose youngest child is 14 or older, and no longer exempt veterans, people experiencing homelessness or young adults leaving foster care.⁸

The real decline has been larger. SNAP participation fell from 41.68 million in June 2025 to 36.35 million in June 2026, a drop of 5.3 million people, or 12.8 percent, in a single year,⁹ running ahead of CBO’s estimates.¹⁰ In Massachusetts, the number of SNAP households fell 15 percent, and a class-action lawsuit alleges that the state’s SNAP assistance line, which automatically disconnected 74 to 81 percent of calls in the first half of 2026, is cutting off eligible people.¹¹ The Greater Boston Food Bank reports that a record 40 percent of the state’s households, about 1.1 million, experienced food insecurity in 2025.¹² In North Carolina, where SNAP rolls have fallen by 110,129 people, a Guilford County food nonprofit reports demand up 20 percent from last year.¹³ The OBBBA is shifting people in need off federal food aid and onto charity, which makes clear that how much hunger there is in America is a political choice.

The next phase began on October 1, when states became responsible for 75 percent of SNAP’s administrative costs, up from 50 percent.Starting October 1, 2027, states will also pay up to 15 percent of benefit costs, depending on their payment error rates.¹⁴ With federal Medicaid cuts landing on state budgets at the same time, the incentive is to tighten enrollment rather than expand it. Massachusetts, with a 12.5 percent error rate, would owe the maximum share, an estimated $213 million a year, yet its fiscal 2027 budget cuts caseworker funding by about $21 million.¹⁵

Health cuts compound the harm. KFF, a nonpartisan health policy research organization, puts the law’s federal Medicaid reductions at $911 billion through 2034, with 10 million more people uninsured; Medicaid work requirements are scheduled to begin on January 1, 2027.¹⁶ In Massachusetts, 99,000 to 202,000 people could lose MassHealth coverage by 2028,¹⁷ in a state that had the nation’s lowest uninsured rate, 2.8 percent, in 2024. These cuts come on top of a drop of roughly 8 million in Medicaid, CHIP and marketplace enrollment since early 2025.¹⁸ For a household on the margin, a medical bill and a grocery bill come out of the same budget.

A Population Without Slack

American households with the bottom 50% in income entered 2026 with almost no financial buffer. Sixty percent of the people in Feeding America’s survey do not feel financially stable,¹⁹ and only 30 percent of all Americans could pay a $1,000 emergency expense from savings.²⁰

Policymakers are now flying blind. On September 20, 2025, USDA terminated its annual Household Food Security report, calling it “redundant, costly, politicized, and extraneous.”²¹ The final edition found 47.9 million people food insecure in 2024.²² The Urban Institute found that 24.2 percent of adults reported household food insecurity in December 2025, although the two surveys are not directly comparable.²³ Just as the largest benefit cuts in SNAP’s history, cuts to Medicaid and subsidies for ACA health insurance take effect, the federal government has stopped publishing the report designed to show their consequences – how hungry are people in America.

Drought in the Meat Case and the Bread Aisle

Grocery inflation overall has been moderate: food-at-home prices rose 2.2 percent in the twelve months to August 2026.²⁴ The drought’s damage is concentrated in protein. USDA’s Economic Research Service (ERS) forecasts beef and veal prices to rise 9.4 percent in 2026, after rising 10.6 percent in the 12 months to June 2025.²⁵ Repeated droughts made feed expensive and withered pastures, forcing ranchers to sell breeding stock, and the beef cow herd is now the smallest since 1961.²⁶ Rebuilding it takes years.

Photo Credit: Thomas Barwick

Grain tells the same story. U.S. winter wheat production fell 27 percent this year, to 1.02 billion bushels, and harvested area fell to a record low.²⁷Roughly a third of planted acres were never harvested, including about half in Oklahoma and seven in ten in Texas.²⁸ USDA projects that wheat stocks will fall 22 percent this marketing year and that the price farmers receive will rise from $5.06 to $6.40 a bushel.²⁹ The 2027 crop is now going into dry ground, where seed that sprouts after a light shower can die if dry weather returns.³⁰

War at the Pump: The Cost of Moving Food

The second shock is geopolitical. The U.S. and Israeli strikes on Iran that began on February 28, 2026, led to what the U.S. Energy Information Administration (EIA) called a “de facto closure of the Strait of Hormuz,” through which about 27 percent of globally traded oil and up to 30 percent of global fertilizer trade passed in 2024. Brent crude nearly doubled in the first quarter, from $61 to $118 a barrel.³¹ Diesel reached a record $6.53 a gallon on September 21 and, at $6.20 on October 5, was still 67 percent higher than a year earlier.³²

Food banks run on diesel trucks, and every extra dollar spent on fuel comes out of the money for food. Houston Food Bank’s diesel price rose 69 percent between January and April,³³ and food banks in Alabama and Virginia report fuel costs up 68 to 70 percent from last year.³⁴ “If I am spending more on fuel, I can’t buy as much food,” said Cheryl Schondek, the Greater Boston Food Bank’s chief operating officer.³⁵ The Food Bank of Iowa estimates that its higher diesel bill costs it the equivalent of more than 8,000 meals a month.³⁶

Households are paying at the pump as well. Regular gasoline averaged $4.354 a gallon on October 5, about 39 percent more than a year earlier.³²“Most typically, they’re having to pull it from the grocery budget,” Amy Breitmann, of Golden Harvest Food Bank in Augusta, Georgia, told NPR in May.³⁷

Gas station in Hudson, MA October 6, 2026, Photo Credit Mark W. Roberts

Fertilizer is where the war reaches the next harvest. By April, the World Bank reported, urea prices had climbed above $850 per metric ton, up 80 percent since February.³⁸ In late September, anhydrous ammonia was still 25 percent more expensive than a year earlier,³⁹ and ERS projects that farm fertilizer expenses will rise 15.3 percent this year and fuel expenses 28.8 percent.⁴⁰

Tariffs and the Shrinking Federal Pantry

The third pressure, tariffs, lands on a food supply already weakened by federal cuts. Since March 2025, USDA has cancelled the Local Food for Schools ($660 million) and Local Food Purchase Assistance ($420 million) programs, halted about $500 million in food deliveries to food banks⁴¹and terminated the Regional Food Business Centers that helped small and mid-sized farms reach buyers such as food banks.⁴² Each cut removed both a market for farmers and food that stretched pantry budgets. Texas food banks have absorbed $88.4 million in federal funding cuts and 40 percent fewer federal commodities this year,⁴³ and the Greater Boston Food Bank lost nearly $2.3 million in emergency food resources to USDA reductions in 2025.⁴⁴

Food banks are now buying replacement food just as wholesale prices rise. In September, Campbell’s and Conagra announced price increases, citing tariffs alongside energy, fertilizer, diesel and crop costs.⁴⁵ The Yale Budget Lab estimates that the tariffs now in force will cost the average household about $1,100 a year.⁴⁶

The Farm Ledger

All four pressures converge on farmers themselves. Chapter 12 farm bankruptcy filings rose 46 percent in 2025, to 315,⁴⁷ and April 2026 produced the highest monthly total since February 2020.⁴⁸ Filings are on pace to rise again this year.⁴⁹ ERS forecasts net farm income down 5.5 percent after inflation and farm debt at a record $605.1 billion, with government payments supplying about 30 percent of net farm income.⁵⁰Distress is rising fastest in the same Southern Plains and southeastern regions where the drought is deepest. A farm that fails this autumn plants nothing next spring.

Washington’s own decisions have cut both ways. The 2025 tariff war cut soybean growers off from China;⁵¹ purchases resumed after a truce in late October 2025, but a 10 percent Chinese retaliatory duty still limits sales.⁵² U.S. tariffs on imported farm inputs collected $958 million through October 2025.⁵³ With the Iran war layered on top, a Farm Bureau survey of more than 5,700 farmers in April found that 70 percent could not afford all the fertilizer they needed for their 2026 crop.⁵⁴

Grain barges on the Mississippi River. Photo credit: Getty Images

The government has paid farmers to absorb much of the damage, through higher reference prices,⁵⁵ $12 billion in Farmer Bridge Assistance⁵⁶ and a plan for new fertilizer capacity that will not come online for 12 to 18 months.⁵⁷ Most of that money flows to the largest operations. Only about 40 percent of farms grow crops eligible for reference-price payments, and the top 10 percent of those farmers have collected nearly three-quarters of all those payments.⁵⁸ The law raised the annual payment limit from $125,000 to $155,000,⁵⁹ and bridge payments grow with every acre planted.⁶⁰ Market concentration deepens the squeeze: four companies control more than 75 percent of the corn and soybean seed market.⁶¹

El Niño: Limited Relief on a Winter Schedule

Against these four pressures, El Niño offers one partial and delayed remedy. NOAA’s Climate Prediction Center (CPC) gives better than 90 percent odds of a very strong event through the winter.¹ Its September 17 outlook gives 60 to 70 percent odds of above-normal precipitation across southern California and the Southwest from December through April, and 70 to 80 percent over southern Georgia and northern Florida, but only 40 to 50 percent over the Southern Plains, where the drought is deepest.⁴ CPC expects the effects to “increase slowly at first before quickly ramping up during the core winter months.”⁶²

The forecast has not failed. It has simply not yet arrived. As of September 29, drought covered 59 percent of the Lower 48 and affected 111.6 million people in 42 states.⁶³ In Oklahoma, 99.8 percent of the state is in drought. Parts of Texas and Oklahoma would need 20 to 25 inches of rain over three months to end it,⁶⁴ and on drought-hardened soil, heavy downpours run off as flash floods instead of soaking in. What the wheat needs is steady rain, week after week, through the winter.⁶⁵

That deficit matters now because the 2027 wheat crop is being planted now. As of October 4, 36 percent of U.S. winter wheat was in the ground, against a five-year average of 46 percent. Oklahoma was 14 percent planted, and more than half its topsoil was short of moisture.⁶⁶ Seed sown into dry soil in October establishes poorly, no matter what rain falls in February.

Dust flies as a tractor prepares a dry field for wheat to be seeded. (Photo by Todd Johnson, OSU Agricultural Communications Services)

The Calendar

Set side by side, the policy calendar and the climate calendar run in parallel, not in sequence:

October 1, 2026. States begin paying 75 percent of SNAP administrative costs.¹⁴

October 2026. Winter wheat planting continues across the Southern Plains into topsoil that is predominantly short of moisture.⁶⁶ CPC’s next ENSO update is due October 8.

December 2026. CPC’s El Niño precipitation signal begins to emerge across the southern tier.⁴

January 1, 2027. Medicaid work requirements take effect.¹⁶

January–March 2027. The El Niño precipitation signal peaks across California, the Southwest and the Southeast.⁴

October 1, 2027. States begin paying up to 15 percent of SNAP benefit costs, depending on their payment error rates, for the first time in the program’s history.¹⁴

Throughout this period, no federal survey will measure American household food insecurity.²¹

Who Bears the Burden of These Four Price Shocks?

The burden of a price shock depends on the share of income spent on food. The wealthiest fifth of households spend less than 6 percent of their income on food; the poorest fifth spend 33 percent.⁶⁷ Those households are not evenly distributed. In 2024, 24.4 percent of Black households, 20.2 percent of Hispanic households and 30.9 percent of American Indian and Alaska Native households were food insecure, compared with 10.1 percent of white, non-Hispanic households.⁶⁸ In Massachusetts, 63 percent of Hispanic households and 51 percent of Black households were food insecure in 2025, and 78 percent of SNAP households received less than $300 a month in benefits.¹²

Four Shocks, Four Choices

None of these four shocks requires new science to answer. Each requires a policy decision by people who already have the data and the power to act.

The drought: protect next year’s harvest now. The latest extension of the 2018 Farm Bill expired on September 30, and Congress has not agreed on a replacement.⁶⁹ It should pass one in the lame-duck session that directs help to the small and midsized farms of the Southern Plains and the Southeast, not ever-larger checks for the largest operations. The Fair Seeds for Farmers Act, introduced by Sen. Elizabeth Warren and Rep. Jim McGovern of Massachusetts, belongs in that debate, too.⁶¹

The war: keep the trucks rolling and the fields fertilized. Emergency fuel assistance for food banks and restored USDA purchases for The Emergency Food Assistance Program cost far less now than after food banks have had to cut their delivery. Farmers need help with fertilizer this season, not in 12 to 18 months.⁵⁷

The tariffs: stop taxing the grocery cart. Food, and the fertilizer, chemicals and machinery farmers need to grow it, should be exempt from import tariffs, and USDA should restore the local food purchasing programs that bought from small farms and supplied food banks and schools.⁴¹

The cuts: stop shifting hunger onto states and charities. Congress should defer the October 2027 SNAP cost shift, pause the expanded work requirements while it examines why participation is falling faster than CBO projected,¹⁰ and delay the Medicaid work requirements. Massachusetts can and must fund the caseworkers, and phone lines its own SNAP system needs.¹⁵

Restore the measurement. Reinstating USDA’s food security report would cost little.²¹ A government that stops counting the hungry does not reduce their number; it only hides it.

El Niño’s rain is still coming, and it may yet rescue next year’s wheat. But no storm will restore a SNAP benefit, refill a food bank’s fuel tank, repeal a tariff or reopen a farm that has already gone under.

The drought is the only one of these four shocks that nature can end. The war, the tariffs and the cuts were set in motion by decisions in Washington, and decisions can be changed. Ask your members of Congress where they stand on each of these four choices before Election Day on November 3.

Endnotes

1. NOAA Climate Prediction Center, “ENSO: Diagnostic Discussion,” September 10, 2026.

2. Carbon Brief, “Analysis: ‘Super El Niño’ reaches ‘remarkable’ all-time record,” September 21, 2026; The Climate Brink, Climate Dashboard, October 3, 2026 data.

3. The Climate Brink, “The impacts of this year’s record-shattering El Niño,” September 28, 2026; see also note 2.

4. NOAA Climate Prediction Center, Prognostic Discussion for Long-Lead Seasonal Outlooks, September 17, 2026.

5. ConfectioneryNews, “Which food commodities are threatened most by ‘Godzilla’ El Niño?” September 16, 2026.

6. Congressional Research Service, Report R48552, August 15, 2025; Congressional Budget Office, distributional analysis of Public Law 119-21, August 11, 2025.

7. Congressional Budget Office, Estimated Effects of Public Law 119-21 on SNAP, August 11, 2025.

8. See note 6.

9. USDA Food and Nutrition Service, SNAP Data Tables, data as of September 11, 2026.

10. Center on Budget and Policy Priorities, “Congress Should Stanch SNAP Participation Losses, Which Exceed CBO’s Estimates,” June 30, 2026.

11. The Boston Globe, “Thousands in Massachusetts are losing food assistance because of disconnected SNAP calls, lawsuit says,”September 27, 2026; State House News Service via GBH, “Disconnected,” September 28, 2026.

12. Massachusetts DTA, Performance Scorecard, July 2026; Greater Boston Food Bank, Massachusetts Food Access Report, April 7, 2026.

13. North Carolina Health News, “SNAP changes push more North Carolina families to food pantries, food banks,” August 4, 2026.

14. See note 6.

15. CommonWealth Beacon, “New error rate highlights risks for Massachusetts SNAP program,” July 6, 2026.

16. KFF, Allocating CBO’s Estimates of Federal Medicaid Spending Reductions Across the States, July 23, 2025; Georgetown University Center for Children and Families, reconciliation law explainer, August 13, 2025.

17. The Boston Globe, “Over 200,000 people could lose MassHealth under coming federal changes, report says,” March 27, 2026.

18. FactCheck.org, “One Big Beautiful Bill Not (Yet) Driving ACA, Medicaid Declines,” July 29, 2026; U.S. Census Bureau, Health Insurance Coverage by State: 2023 and 2024, 2025.

19. Feeding America, Elevating Voices: 2026 Insights Report, September 29, 2026.

20. Bankrate, 2026 Annual Emergency Savings Report, February 4, 2026.

21. USDA, “USDA Terminates Redundant Food Insecurity Survey,” September 20, 2025.

22. USDA Economic Research Service, Household Food Security in the United States in 2024, as summarized by Center on Budget and Policy Priorities, December 30, 2025.

23. Urban Institute, Food Insecurity Remained High in 2025, As Safety Net Cuts Loom, March 18, 2026.

24. Bureau of Labor Statistics, Consumer Price Index, August 2026; USDA Economic Research Service, Food Price Outlook, September 25, 2026.

25. See note 24; Bureau of Labor Statistics, Consumer Price Index, June 2025.

26. USDA NASS, Cattle, January 30, 2026; USDA ERS, Livestock, Dairy, and Poultry Outlook.

27. USDA NASS, Small Grains 2026 Summary, September 30, 2026.

28. Calculated from note 27.

29. USDA, World Agricultural Supply and Demand Estimates, September 11, 2026.

30. Oklahoma State University Extension, “Basic rules for planting wheat this fall,” October 4, 2021.

31. U.S. Energy Information Administration, Today in Energy, April 7, 2026; IFPRI, “The Iran war’s impacts on global fertilizer markets and food production,” 2026.

32. U.S. Energy Information Administration, Gasoline and Diesel Fuel Update, October 6, 2026.

33. Houston Food Bank, “Effects of Rising Fuel Costs,” April 27, 2026.

34. FOX10, Mobile, Alabama, September 22, 2026; 12 On Your Side, Richmond, Virginia, September 10, 2026.

35. WROR, “Greater Boston Food Bank Defies Rising Costs, Ensures No Service Cuts,” May 11, 2026.

36. KCRG, “Rising fuel costs threaten Iowa’s food bank deliveries,” April 30, 2026.

37. NPR, “More people are going hungry now than at the height of the pandemic,” May 27, 2026.

38. World Bank, “Fertilizer prices surge as Strait of Hormuz disruptions tighten supplies,” May 14, 2026.

39. DTN, “Anhydrous Leads Fertilizer Prices Higher, Up 25% From a Year Ago,” September 30, 2026.

40. USDA Economic Research Service, September 2026 Farm Income Forecast, September 3, 2026.

41. CBS MoneyWatch, March 13, 2025; CNN, March 22, 2025.

42. Civil Eats, “USDA Cancels More Support for Regional Food Systems,” July 15, 2025.

43. The Texas Tribune, “Texas food banks struggle amid federal cuts, drop in donations,” September 3, 2026.

44. WBUR, “Greater Boston Food Bank faces federal funding cuts,” April 7, 2025.

45. Bloomberg, “From Campbell’s to Conagra, Food Industry Warns of Price Hikes,” September 11, 2026.

46. The Budget Lab at Yale, The State of U.S. Tariffs, August 24, 2026.

47. American Farm Bureau Federation, “Farm Bankruptcies Continued to Climb in 2025,” February 9, 2026.

48. Epiq AACER, bankruptcy filing data, May 6, 2026.

49. Brownfield Ag News, “Farm bankruptcies on pace to increase in 2026,” September 17, 2026.

50. See note 40; American Farm Bureau Federation, Market Intel, September 3, 2026.

51. Iowa Soybean Association, “How We Got Here: The 2025 U.S.–China Soy Saga,” November 4, 2025.

52. High Plains Journal, “U.S., China buy more time to iron out trade wrinkles,” October 2, 2026.

53. Farm Progress, “Trump’s tariffs generate $958M but cost farmers far more,” January 22, 2026.

54. Farm Policy News, “70% of Farmers Can’t Afford All Needed Fertilizer, Survey Shows,” April 2026.

55. Iowa State University CALT, “Reviewing the Agricultural Provisions in the One Big Beautiful Bill Act,” July 11, 2025.

56. USDA, Farmer Bridge Payments announcement, December 8, 2025.

57. Farm Policy News, “US Set to Use Tariff Funds to Address High Fertilizer Prices,” April 20, 2026.

58. Environmental Working Group, “For 41 years in a row, USDA sent taxpayer-funded payouts to roughly 9,000 farmers,” July 13, 2026.

59. American Enterprise Institute, Changes for Agriculture in the One Big Beautiful Bill Act, October 15, 2025.

60. American Farm Bureau Federation, “Farmer Bridge Assistance Program: Details on $11 Billion in Aid,” January 2, 2026.

61. Office of Sen. Elizabeth Warren, Fair Seeds for Farmers Act press release, July 23, 2026.

62. NOAA Climate Prediction Center, Prognostic Discussion for Long-Lead Seasonal Outlooks, August 20, 2026.

63. U.S. Drought Monitor, Current Conditions, data valid September 29, 2026.

64. U.S. Drought Monitor, Weekly Summary, October 1, 2026.

65. WFIT/NPR, “How the risk for flash flooding increases after a drought,” August 11, 2022.

66. USDA NASS, Crop Progress, October 5, 2026.

67. USDA Economic Research Service, Food Prices and Spending, April 30, 2026.

68. See note 22.

69. Farm Policy News, “Farm Bill Expires Again as Lawmakers Continue Negotiations,” October 2, 2026.

Photo credit: WFPA farmer shows a dried-up irrigated rice field fed by the Cimulu River tributary in Ciwasmandi, Sukajaya, Tasikmalaya, West Java, on Thursday, July 16, 2026. Antara Photo/Adeng Bustomi

It Begins Where It Is Named

Part three of this series told how Peru’s fishermen named El Niño, and what this one is already doing to their anchoveta (anchovy) catch.¹

The loss of the fishery is the gentler half of what a coastal El Niño does to Peru. The same warm water that empties the nets of anchovy loads the moisture in the atmosphere over one of the driest coastlines on earth, and the rain then arrives on ground that cannot absorb it because it is so dry. Peruvians have a word for what follows: “huaicos”, the torrents of mud and rock that come down the dry valleys and through the towns built in them. In 2017, a coastal El Niño killed around 150 people and displaced hundreds of thousands. This year’s rains have already killed at least 74, in floods and mudslides from December through March.²

So, the country that named the phenomenon lives with the full range of its impacts: a fishery that fails and a coast that floods. The name “El Niño” has traveled around the world, while the place that supplied it remains among the most exposed to what it describes.

This makes Peru the right place to ask the question this final article is about. The Pacific does what it has always done: it has formed an El Niño, this time in an ocean warmed by climate change. What varies, enormously, and what is still undecided, is how many people will die from this El Niño.

Two Droughts, One Ocean

In 1877, the monsoon failed over India. The Pacific was warm, the rains did not come, and somewhere between five and ten million people died in British India alone. Add the simultaneous famines in northern China and northeastern Brazil, driven by the same El Niño event, and the toll from that single El Niño runs into the tens of millions.³

In 1997 and 1998, the Pacific warmed again, to an intensity climatologists consider broadly comparable with the 1877 El Niño. The event was enormous. It caused roughly $36 billion in damage worldwide. However, it killed only about twenty-one thousand people.⁴

Set those two numbers beside each other and the conclusion is unavoidable. Between 1877 and 1997 the ocean did not change very much. What changed was everything around it. The world held far more people in 1997, but far fewer of them died in the 1997 El Niño, because of how governments and people responded.

Figure 1. The 1877–78 and 1997–98 El Niños side by side. Both were very strong El Niños; the number of people who died was not comparable. The 1997–98 toll is too small to register at the scale of 1877.

This is not an argument that water is harmless. El Niño sends the droughts that destroy crops and the floods that wash them away, and there is no arguing with either. Sudan’s farmers could not plant their sorghum in July because the rain did not come, and no decision by anyone anywhere could have grown a crop without water. The Nile fell lower than farmers in their sixties can remember, and cholera followed the falling river through Sudanese camps. In Uganda’s Karamoja region, the number of people in crisis or emergency hunger has risen by more than 40 percent in a few months. Ethiopia is watching unseasonal rain threaten its main harvest, and Somalia and Kenya brace for floods this fall. In Central America’s Dry Corridor, the first harvest of the year is already gone. The second is being planted into the face of this El Niño. In Peru, forecasters put the chance of a strong or extraordinary coastal El Niño this coming summer at more than 70 percent, and the warm water has already emptied the nets that supply a fifth of the world’s fishmeal.¹ India’s monsoon is heading for a deficient year, the Philippines warns it could lose as much as 700,000 tonnes of rice, salt water is forecast to push up the Mekong into Vietnam’s rice fields, and Papua New Guinea’s highlands face the drought and frost that affected 2.4 million people there in the last great El Niño. Peat smoke has already sent more than 170,000 Indonesians for treatment since August, 40,000 of them toddlers, and El Niño redraws the map of dengue and malaria. The water does the first damage, it does it directly, and it is not negotiable.

But between a failed harvest and a dead child there is a long space, and almost nothing in it is weather. Whether the household had reserves from last year. Whether it can afford to buy replacement food for crops that have been lost. Whether there is grain at an affordable price or food in the market at all, or whether the border has been closed to prevent exports. Whether anyone moved money or other aid before the lean season or after it. Whether there is a clinic left standing when the child stops eating.

That space is where the difference between 1877 and 1997 lies. The rains destroyed crops in both years. Only one of those harvest failures was allowed to become a famine.

“The temperature of the Pacific tells you where the rain will fail, and where crops will die. It tells you almost nothing about how many people will die. That number is set on land, by policy decisions made by governments around the world, most of them made before the drought and floods even arrive.”

The historian Mike Davis made this argument at length in the book Late Victorian Holocausts, and the mechanism he described is uncomfortably recognizable. India in 1877 was not short of grain. It was forced to export record quantities of grain it grew to Britain while its own population starved. The railways that could have carried food inland to the starving Indian peasants carried it to the ports for export instead. The theory given by the British at the time was that relief had to be rationed because generosity would corrupt the poor. The peasants had nothing to fall back on in the face of El Niño: British policy had pushed them into cash crops for export and into debt to pay the colonial land tax, leaving them no grain reserves of their own. The drought was real. The famine was administered by the British colonizers.⁵

That is the frame for everything in the previous five articles, and it is the reason this series has spent as much time on export bans, ration cuts and funding tables as on rainfall. Decisions already taken, and decisions being taken now, will set the death toll of this El Niño, too. So far there is no sign of the surge in humanitarian funding that 150 years of El Niño records say these countries around the world will need, particularly with this El Niño on course to be the strongest ever recorded.

We hope that this series will contribute to the calls for humanitarian aid to be funded now, before it turns into disaster relief.

This One Is Bigger. But That Will Not Change This Equation.

This El Niño is not merely comparable to 1997. On September 21st, the daily Niño 3.4 anomaly reached 3.11 degrees, breaking the record set in November 2015, two months before El Niños usually peak. The forecasts have climbed faster still: an era-adjusted peak near 3.0 degrees in ECMWF’s May forecast, then a 14-model median of 3.6 in July and 4.1 in September, when every one of 674 runs across fourteen forecasting systems exceeded the 2015–16 record. These are era-adjusted figures, measured against the climate of each event’s own time, and cannot be set beside the 2.6 NOAA lists for 2015–16 on its three-month index. NOAA gives a 75 percent chance this becomes the strongest event since 1950. No model has ever been tested at this intensity, but wherever the peak lands, it will be in territory no one has measured before.⁶

Heat alone has already killed more than 200,000 people since this El Niño began in June, and could kill 451,000 by February, according to the first projection of its death toll, published on September 23rd by the University of Chicago’s Climate Impact Lab. That number does not include a single death from hunger, flood, fire or disease. The last super El Niño shows what those can add, in 2015, smoke from Indonesia’s El Niño fires alone caused an estimated 100,000 excess deaths across Indonesia, Malaysia and Singapore in a single season. The same peat fires are burning again now.⁶

Figure 2. Peak era-adjusted Niño 3.4 anomaly for the two strongest El Niños on record, 1877–78 and 2015–16, against successive forecasts for 2026 and 2027. The whisker spans the middle 80 percent of the September ensemble. Models have never been verified at this intensity.

One quieter detail says more than any of those numbers. In February 2026, NOAA changed its ruler for measuring how big this El Niño will be. All official monitoring moved to a new “Relative Oceanic Niño Index”, which measures the central and eastern Pacific against the rest of the tropical ocean rather than against the fixed historical baseline. It had to, because the whole tropical ocean has warmed so much that the old fixed baseline had begun to make El Niños look stronger, and La Niñas weaker, than they really were.⁶

“If the temperature of the Pacific were the variable that decided the death toll, 2027 would be catastrophic beyond any precedent. The argument of this article is that it need not be, and that whether it will be is settled by things with no connection to the Pacific at all, above all, whether governments help the people that 150 years of El Niño records say will suffer from hunger.”

The Granary Is Also Dry

Start with the thing that turns a bad harvest into a survivable one: the ability to buy replacement food. Sudan imports about four fifths of its wheat. Egypt trades the title of the world’s largest wheat importer back and forth with Indonesia. The Philippines is covering a 700,000-tonne shortfall by importing. In every case the question is not whether the domestic harvest failed. It is what the replacement costs.

This year the replacement is short as well. As the previous article set out, drought has cut Europe’s wheat and maize, left the Rhine and the Danube too low to carry a full barge, and war has closed both ends of the Black Sea grain corridor.^{7, 8, 9}

So is the United States. USDA now expects an American wheat crop of 1.531 billion bushels, 23 percent below last year and the smallest in more than half a century, after drought on the Southern Plains left a third of all winter wheat acres unharvested. Fertilizer adds to the risk. The closure of the Strait of Hormuz in February put roughly a third of the world’s seaborne fertilizer trade at risk, and by April urea had climbed 80 percent. Urea fell back after shipping resumed in June, but attacks on ships resumed in July, and phosphate fertilizer remains tight. A farmer who pays more to plant grows less, and the shortfall lands in next year’s price.⁷

Figure 3. U.S. all-wheat production, 2021–2026, in billion bushels. The 2026 figure is USDA’s August forecast, unchanged in September. Source: USDA NASS.

None of that is El Niño. The European drought is attributed to climate change rather than the Pacific, and the Black Sea is a war. That is exactly the point. A household in Kassala cannot plant its sorghum because of an ocean cycle and then finds the replacement priced by a European heatwave and a Ukrainian port. It has not lost once. It has lost twice, and the second loss does not end when the rain returns.

Figure 4. What this El Niño has already done. Filled circles mark confirmed impacts, sized by the best available count of people affected; open circles mark impacts forecast next. Counts are not all directly comparable: in Sudan and the Sahel they include hunger driven by conflict, and Indonesia’s figure is people exposed to haze.

“The rain has shifted exactly where the forecasts said it would, and every shift has been predicted. The droughts came. The floods came. The fires came. The only thing that has not come is the humanitarian aid. The wealthy countries of the world still have time to show that they have not lost their humanity, but only if they fund proactive humanitarian aid now.”

Figure 5. The deadlines that are still open. For each region, the diamond marks the last point at which money buys prevention rather than relief; the bars show planting seasons, the flood, fire, salt and harvest-rain seasons, and the hunger peaks that follow. Indonesia’s deadline has already passed.

The Deadlines Are Already on the Calendar

There is no single deadline left as this El Niño rolls out around the world. There is a complex calendar of deadlines, set by planting seasons and by hungry seasons, when people’s reserves from previous harvests run out, on four continents. Each date on this calendar marks the last point at which money can still buy prevention to save crops from droughts and floods rather than provide disaster relief for the suffering, acute hunger and famine that arrive if these crops fail. The forecasts behind every one of those dates are already free, public and regularly updated, as Hunger Foretold described.

The first deadline falls next month. The short rains in southern Ethiopia, central and southern Somalia and northeastern Kenya begin in October, and the region’s forecasters have put a ninety percent probability on wetter-than-normal rain; FAO has already warned Somalia to expect serious flooding in the “Deyr” rainy season. Flood preparedness means moving people and herds to higher ground, clearing drainage, and stocking water purification against cholera, and all of it has to be in place before the rain falls, not after. The next deadline falls in Southern Africa.

A child wades through water in Uvira, eastern Democratic Republic of the Congo, after El Niño-induced torrential rains caused Lake Tanganyika to overflow in 2024. Photo: WFP/Benjamin Anguandia

Southern Africa plants its main maize crop of the year between October and December, as the yearly rains arrive. Zimbabwe, Zambia, Malawi and Mozambique get one growing season a year, and this year, the crops will be planted into the peak of an El Niño that has already broken the record and is forecast to become the strongest ever measured, arriving in a world warmed by climate change, which makes its impacts hit harder. Everything that determines whether the harvest in Southern Africa survives is decided in the weeks before planting: do drought-resistant seeds go into the ground, are drought-tolerant varieties available and affordable, do households have the cash to buy the fertilizers and fuel they need rather than having to eat their savings, and have the irrigation systems and village boreholes and wells been repaired. The region has been warned twice in a decade. In 2015 and 2016, the maize crop failed and the money for humanitarian aid arrived only afterward, when people were suffering across the entire subcontinent. In 2023 and 2024, it happened again. Both times the forecast was public, free and made a year in advance. Both times the funding arrived after the crops failed and mass hunger appeared, rather than before, when it could have prevented the worst of it.

Asia’s deadlines run in parallel. India sows its winter wheat from October into December. Vietnam plants its winter-spring rice in November, the export crop that the Mekong salinity forecast puts at risk across more than 350,000 hectares from November through April, so freshwater storage and salt-tolerant seed varieties must be in place before planting. The driest conditions across Southeast Asia were forecast months in advance to run from late September to February. Indonesia’s peat fires did not wait: they have been burning since August, and by September 21^st, Indonesia’s health ministry had recorded more than 170,000 cases of acute respiratory infection in the haze-hit provinces. This is what a missed deadline looks like. Peat fires are not ordinary fires. They burn below the surface of the ground; they are close to inextinguishable by rain alone. They produce the transboundary haze that in 1997 and again in 2015 turned into a regional health emergency, filling hospitals across Indonesia.

In swathes of Africa, Asia and Latin America, this year’s powerful El Niño is expected to unleash droughts, but also heavy rains and flooding. Photo: WFP/Gabriela Vivacqua

Early 2027 brings its own deadlines. Southern Africa’s lean season peaks from January to March, and in Madagascar 3.7 million people are expected to be in Crisis or worse through January. Ethiopia’s “belg” crop, grown during the short 2026 February-to-May rains, came through this year on mostly average rains. Its main “meher” crop, which supplies 90 to 95 percent of the country’s grain, is being harvested now, and the Famine Early Warning Systems Network (FEWS NET) is predicting unseasonal rains from October to January that could flatten this meher harvest, sprout it in the head and rot it in storage. Tarpaulins, drying equipment and storage support have to reach farmers before those rains do. The next “belg” crop is planted when the rains return in February, so drought-resistant seeds, fertilizers and fuel need to be in farmers’ hands by then.

In Central America’s Dry Corridor, which runs through Guatemala, El Salvador, Honduras and Nicaragua, families usually run out of stored grain around June, when the hungry months before the next harvest begin. This year they ran out of food reserves in February, because three years of failed or poor harvests, through the last El Niño in 2023 and 2024 and a hot, erratic 2025, had left them nothing in their reserves. Now they have already lost the first harvest, the “primera” corn crop, and the second, the “postrera,” is only now going into the ground, late and into the same dry conditions caused by El Niño that already killed the “primera” crop. If it fails too, next year’s hungry months are likely to begin early, and help has to arrive before they do.

The last deadlines on the calendar fall in the middle of 2027, in the places where this series began. In Sudan, if the late-planted 2026 sorghum harvest falls short, its grain will run out sooner than it should, bringing the next lean season forward from its usual start in May 2027 and starting it from a lower base of reserves. Across the Sahel, the lean season runs from June to August. In both Sudan and the Sahel, food, cash and other humanitarian relief have to be in place before the rains close the roads, which means they have to be bought and moved by early spring.

“Every date on this calendar gives a rough deadline before which money can still buy seed, fertilizer, water, and protective measures rather than disaster relief. It is before planting in each of these countries that humanitarian relief can prevent suffering, hunger and famine. After that period passes, disaster relief is more costly and is used in a scramble to reduce hunger, disease and death. The same amount of humanitarian aid helps fewer people and arrives late.”

The Appeal Nobody Can See

This brings us back to the question that has run through this series since its first article.

On June 18th, FAO (the Food and Agriculture Organization of the United Nations) and the World Food Programme launched the first joint anticipatory action appeal either agency has made. They asked for $202 million to protect 8.8 million people across 22 priority countries, running to March 2027. The agencies said they were already positioned to reach 1.2 million people and that a further $167 million would cover the remaining 7.6 million. They also said, in the launch material, that the systems, partnerships and operational plans needed to act were fully in place and coordinated for immediate action, and that what was needed was the financing.¹⁰

The WFP projects that this El Niño could push at least 49 million more people into acute hunger across 45 countries by the end of 2027, raising the number of people in those countries who cannot meet their daily food needs to 274 million. The appeal was meant to get ahead of El Niño and provide proactive relief that would protect 8.8 million vulnerable people. It bears repeating that no El Niño this strong has ever been recorded, so the actual impacts could be greater.

More than three months after this appeal was launched there is no published figure for how much has been raised, and this publication went looking. OCHA’s Financial Tracking Service records no 2026 flows to either agency tagged to El Niño or anticipatory action in any of the 22 countries. Only one government has announced a contribution earmarked to it: on July 3rd, Norway allocated NOK 30 million, about $3 million, split between FAO and WFP. Other money is moving through other doors: CERF has offered up to $100 million, with more than $20 million allocated for anticipatory action in six countries and $6 million for southern Madagascar, and on September 17th the Novo Nordisk Foundation gave WFP $7.75 million, with a pledge to jointly promote the appeal, but none of it is counted against the appeal, and together it comes to less than 20% of the $202 million requested. FAO’s own project page, which publishes the donor and the amount for every anticipatory action contribution it receives, has posted nothing since May 29th, three weeks before the appeal launched.¹¹ Time is getting short.

None of that is an oversight. FTS reports progress against coordinated plans, principally response plans and flash appeals, and a joint inter-agency appeal is not one of those, so it gets no funding bar. The flagship instrument for acting before a disaster sits outside the architecture built for watching humanitarian money move. It may well be better funded than it looks. There is no public way to find out, which is a finding in itself. ¹²

For scale, the entire previous El Niño cycle produced identifiable FAO anticipatory action funding of roughly $26 million, about half of it from Germany. The largest single line, $13.2 million, came from the Government of Somalia with the World Bank rather than from any traditional donor. This appeal seeks $202 million.¹³

Meanwhile the governments raising the alarm are the ones about to be hit. Uganda declared a drought in July and funded a six-month response across Karamoja from its own budget. Honduras placed eighty percent of its territory under alert. Brazil released 1.335 billion reais for El Niño - related actions. The Philippines is distributing drought-tolerant seed. These are low- and middle-income countries pre-staging what they can afford, against a forecast of a super El Niño.¹⁴

In Chad, WFP is prioritizing newly arrived people from Sudan for food assistance, for lack of resources. $167 million could help a lot.

Photo: WFP/Arete/Salamon Djek

What $167 Million Buys, and What It Costs to Wait

So let this article do the thing the previous five have carefully avoided, and ask for something.

The FAO/WFP appeal is failing as far as we can see. Not struggling, not underfunded in the way every humanitarian appeal is underfunded. The gap is $167 million, and it is the difference between reaching 1.2 million people and reaching 8.8 million. That is 7.6 million people whose protection currently depends on money that, apart from Norway’s roughly $3 million, nobody has given.

Figure 6. The $202 million appeal against what is publicly recorded as raised. The $35 million shown is the coverage the agencies stated they already had at launch, not a figure for funds received, because no such figure has been published. Searches of OCHA’s Financial Tracking Service, CERF allocations and FAO’s own contribution page on September 5, 2026 found nothing recorded against the appeal. The only earmarked contribution announced by late September was Norway’s NOK 30 million, about $3 million, which the figure shows as a thin gold band. The lower bar sets the appeal’s 8.8 million people against the 49 million more that WFP expects this El Niño to push into acute hunger by the end of 2027.

For scale, $167 million is roughly the price of two F-35 fighters. It is about a tenth of what the Panama Canal Authority is spending on one reservoir to make sure Panama Canal shipping continues uninterrupted. Using commonly cited figures, it is roughly three to four weeks of keeping one American carrier strike group at sea near the Strait of Hormuz. What the appeal is seeking comes to about $22 for each person who would benefit. This is what it costs to put drought-tolerant seed in the ground, repair a well or borehole, move cash to a household before it sells its goats or its farming tools rather than after, vaccinate herds against the diseases that follow floods, or buy animals from herders at a fair price before the pasture fails, rather than letting them be sold at a loss or die. FAO estimates that every dollar spent on anticipatory action saves up to seven in avoided losses.

The cost of not providing aid is measurable, because it was measured in Mali, in the central Sahel. In areas where food rations were cut for lack of funding, the number of people in crisis-level hunger rose 64 percent between 2023 and 2026. Where full rations were maintained, the number of people in crisis-level hunger fell 34 percent. Same region, same crisis, same years. That was ordinary food aid, delivered after the crisis had arrived. This current El Niño appeal is asking for money to fund anticipatory actions, the same money spent earlier, when it buys more and helps more people: seed instead of rations, working wells instead of water shortages, a vaccinated herd instead of dead goats and cattle.¹⁵

Figure 7. What happens when funding is cut: Mali, in the central Sahel, 2023 to 2026.

“This is not a request for generosity. It is a request that money already budgeted for famine relief in 2027 be spent in 2026 instead, when it buys seed rather than body bags, and when it costs a seventh as much.”

The timing is the whole argument, and it is now tight. Every deadline that passes turns the same dollar into imported grain bought at a lean-season price, delivered after the child is already acutely hungry and has stopped growing.

So, specifically. To governments around the world: fund the FAO/WFP anticipatory action appeal. The joint appeal needs $167 million, and it needs it before the first planting deadlines in October and November. To FAO and WFP: publish what you have raised for this appeal, weekly, whether or not the tracking architecture requires it, because an instrument that cannot show its own funding cannot make the case for more funding. To OCHA: give the FAO/WFP anticipatory action appeal a funding page, so that the next journalist who goes looking to see how much money has been raised finds a number instead of nothing.

None of that requires a new institution, a new fund, or a new theory of humanitarian response. The plans exist. Both agencies have said in writing that they are ready and coordinated for immediate action. The only missing input is the money, and there are only weeks left before the first of those deadlines pass.

Where to Watch All of This Yourself

• American drought, every Thursday: U.S. Drought Monitor and Drought.gov

• European drought, every ten days: European Drought Observatory

• The Pacific: NOAA Climate Prediction Center

• The crops: JRC MARS Bulletin and USDA WASDE

• The hunger: FEWS NET and WFP HungerMap Live

What the Maps Were For

This series began with a simple observation: the hunger arriving in 2027 is being forecast in public, on a schedule, by government scientists, in forecasts anyone can read for free. Five articles later, that has held up. NOAA said in May that a strong event was likely. It was. The models said the rains would fail in Sudan and in the Dry Corridor. They did.

The forecasting is not the problem. It is arguably the most successful piece of public scientific infrastructure in the world. When the hunger and famines arrive, and they will, they will be the best predicted crises in history. The question then will be who made the bad decisions, and why.

The problem is the gap between the forecast and the money needed for humanitarian aid, and that gap is not technical. It is not a modeling problem or a data problem or a capacity problem.

In 1877 the grain was there and the policies killed people. In 2026 the forecast is there, and it has been right so far about the regions it named for crop failures. The money is not being mobilized to get ready for this super El Niño. The question this El Niño will answer over the next fifteen months is whether a warning that arrives a year early is worth anything at all to a system that has rarely managed to mobilize the humanitarian aid needed to respond to an El Niño before the harvests fail and the hunger arrives.

That question does not have to be answered the way it was answered in 2015, or in 2023. Both ended in suffering, hunger and death. The question is being answered right now, and the window begins closing next month; most of it will be shut within 100 days. It is also being answered by whoever reads this and does nothing about it.

Call your Senators at: (202) 224-3121

Call your Representative at: (202) 224-3121

That number is the U.S. Capitol switchboard; ask the operator for each member’s office. To find out who your members are, go to senate.gov and house.gov. Call today, call tomorrow, the next day... Ask them to urge the State Department, which now manages most U.S. foreign aid, to fund the FAO–WFP Joint Anticipatory Action Appeal for El Niño. Ask them to promote even greater El Niño funding, because the World Food Programme projects that the 2026 - 2027 El Niño could push at least 49 million more people into acute hunger across 45 countries by the end of 2027, raising the number of people who cannot meet their daily food needs in those countries to 274 million. All of these people need aid!

If you live outside the United States, call your own elected representatives and demand the same.

“The maps were free. They were also right. What they have not been, so far, is funded: in three months, one donor government has put about $3 million toward the $167 million still needed by the appeal built to act on them.”

Endnotes

1. Peru’s multi-agency commission ENFEN activated an El Niño Costero alert on February 13, 2026 and expects coastal warming to persist into late 2026. PRODUCE set a total allowable catch of 1.9 million tonnes for the first North-Central anchoveta season of 2026, reported as the lowest in a decade and 36 percent below the 3 million tonnes set for the same season in 2025; operations were suspended from May 12th and extended on May 27th through June 10th owing to a high presence of juveniles in the catch. Fishmeal prices reported more than 75 percent above the previous year. Peru accounts for approximately 20 percent of global fishmeal and fish oil production. On the mechanism, warm surface water suppresses the coastal upwelling of cold, nutrient-rich water, collapsing the plankton on which the anchoveta feed.

2. On the coastal hazard: El Niño Costero events bring intense rainfall to the normally arid northern Peruvian coast, producing flash flooding and the debris flows known locally as huaicos. The 2017 coastal event is commonly reported as having killed approximately 150 people and displaced hundreds of thousands. The current coastal El Niño has already caused deadly flooding: Peru declared a state of emergency on February 25, 2026, and by March 20, 2026, Peru’s civil defense institute, INDECI, reported 74 deaths, 27,555 people left homeless and 196,483 otherwise affected by the rains since the start of the year (La República, March 20, 2026). The count covers the whole rainy season, including the highlands, not only the coast; it had stood at 68 in late February (Africanews, February 26, 2026). ENFEN’s update of July 17, 2026 rated the December 2026 to March 2027 season at 38 percent strong, 33 percent extraordinary, 23 percent moderate and 6 percent weak (reported by The Rio Times, August 24, 2026).

3. Mortality estimates for the 1876–78 famine in British India range from approximately 5.5 to 10.3 million, with wider ranges in some sources. Estimates for the total mortality of the 1876–79 El Niño famines across India, China and Brazil vary substantially between authorities and run into the tens of millions; these figures are contested and should be presented as ranges rather than single figures. Palaeoclimate reconstructions place the 1877–78 event among the strongest in the instrumental and proxy record.

4. Commonly cited figures for the 1997–98 El Niño: approximately $36 billion in global economic damage and roughly 21,000 deaths, the latter predominantly from flooding and storms rather than famine. Damage estimates vary with methodology.

5. Mike Davis, Late Victorian Holocausts: El Niño Famines and the Making of the Third World (Verso, 2001), on grain exports from India during the 1876–78 famine, the role of rail infrastructure in moving grain to ports, Famine Commission relief policy, and the effect of revenue demands and cash-crop conversion on peasant reserves. Davis’s mortality estimates and his causal argument have both been debated in the subsequent literature; the argument here rests on the mechanism rather than on any single mortality figure.

6. ECMWF seasonal ensemble, May 2026: median peak Niño 3.4 anomalies near +3°C by December with high-end members approaching +4°C. July 2026 multi-model forecast (Zeke Hausfather, The Climate Brink, July 13, 2026): median of individual-model peaks near +3.6°C era-adjusted across 667 ensemble members from fourteen seasonal prediction systems, with 91 percent of members exceeding the record. September 2026 multi-model forecast (Carbon Brief, Hausfather, September 21, 2026, updated September 23, 2026): fourteen seasonal forecast systems (NOAA CPC, Copernicus C3S, ECCC and JAMSTEC data), peak monthly anomaly around +4.1°C, 80 percent of the 674 individual runs between +3.4°C and +4.6°C, every run peaking above the 2015–16 record; August observations came in about 0.3°C below forecast. Observed: daily Niño 3.4 anomaly of +3.11°C on September 21, 2026, surpassing the previous daily record of +3.08°C set on November 18, 2015. All of these figures, and the historic values in Figure 2, are era-adjusted: single-month peaks measured against a centered 30-year climatology (ONI convention). They are not directly comparable with NOAA’s published three-month ONI values, on which the 2015–16 event peaked near +2.6°C and 1997–98 at +2.4°C. NOAA Climate Prediction Center, September 10, 2026: greater than 90 percent probability of a very strong event for Northern Hemisphere fall and winter, and a 75 percent chance that October to December reaches +2.5°C or more on the Relative Oceanic Niño Index, exceeding every event on record since 1950 (69 percent in the August 13, 2026 outlook). World Meteorological Organization, September 3, 2026: El Niño firmly established and forecast to strengthen to very strong before peaking toward the end of 2026. NOAA CPC ENSO Diagnostic Discussion, September 10, 2026: El Niño Advisory in effect; August values on the relative (RONI-convention) indices of +1.8°C in Niño 3.4, +2.5°C in Niño 3 and +3.4°C in Niño 1+2, with subsurface anomalies above +10°C at depth; relative values run lower than the era-adjusted figures above and are not directly comparable. IRI, September 2026 ENSO Quick Look: Niño 3.4 near +3.0°C in mid-September on NOAA OISST; all 22 plume models (13 dynamical, 9 statistical) forecasting a very strong El Niño (Niño 3.4 at or above +2.0°C) from September–November 2026 through December–February 2026–27, with a majority at or above +3.0°C; objective El Niño probabilities of 100 percent from September–November 2026 through February–April 2027. On convergence: forecaster Ben Noll, as reported by Meteored (theweather.com) in early September 2026, noted that ECMWF peak forecasts had stabilized near +4°C for November–December after months of increases; the multi-model median nevertheless rose from about +3.6°C in July to about +4.1°C in September. On the index change: CPC transitioned official monitoring to the Relative Oceanic Niño Index in February 2026, measuring eastern Pacific anomalies relative to the wider tropical ocean rather than a fixed baseline, to account for background warming; RONI and conventional Niño 3.4 values are not directly comparable and are quoted here as published. On heat deaths: Climate Impact Lab, “451,000 at Risk: Where Action Can Save Lives During the Super El Niño,” September 2026, projecting 451,000 additional heat-related deaths worldwide from June 2026 to February 2027 compared with the 1996–2025 average, more than 200,000 of them already by September; the projection covers heat only. On Indonesia’s 2015 fires: Koplitz and others, Environmental Research Letters, 2016, estimating about 100,300 excess deaths across Indonesia, Malaysia and Singapore from the 2015 fire-season haze.

7. European Commission Joint Research Centre, MARS Bulletin, Crop Monitoring in Europe, July 2026: EU soft wheat yield forecast cut to 5.88 tonnes per hectare, about 7 percent below 2025, with grain maize and sunflower down 6 to 7 percent. World Weather Attribution, July 2026: below-average rainfall from Portugal to southern Finland between April and June; France’s lowest maize harvest in 50 years. Coceral put EU maize production at 52.7 million tonnes, the lowest since 2007. The roughly one million hectares of maize often reported lost in Romania is a reduction in planted area since 2019, not a 2026 crop loss; Romania expects a larger 2026 harvest. WWA also found climate change made the drought approximately 80 times more likely in western Europe and 40 times in the east. United States: USDA NASS Crop Production, August 12, 2026, and WASDE, September 11, 2026: all-wheat production forecast at 1.531 billion bushels, down 23 percent from 1.985 billion in 2025, with 21.04 million of 31.26 million winter wheat acres expected to be harvested. Fertilizer: IFPRI, 2026, on the closure of the Strait of Hormuz from February 28, 2026 putting roughly one-third of global seaborne fertilizer trade at risk; World Bank, May 2026, urea above $850 per metric ton in April, up 80 percent since February and the highest since April 2022. Urea fell back to about $453 per metric ton by June after shipping through the strait resumed (WTO, July 10, 2026); IFPRI’s Food and Fertilizer Export Restrictions Tracker, August 2026, reported that attacks on shipping resumed in July, nitrogen prices eased and phosphate supplies remained tight.

8. Reported record low water levels on the Rhine; the Danube falling low enough that Hungary’s only nuclear power plant approached a full shutdown; the Po reported to have dried up; water use restrictions imposed across several French regions.

9. Ukraine’s agriculture ministry put agricultural exports for the 2026–27 marketing year at about 29.6 million tonnes against an earlier estimate of 64.4 million, with wheat down 53 percent to 8.3 million tonnes; a separate official figure for grain alone was revised to 38–40 million tonnes from about 43 million, and the two should not be conflated. By August 19, 2026, Russian and Ukrainian attacks on each other’s ports and shipping had shut more than 97 percent of the two countries’ combined grain export capacity in the Azov and Black Sea basin (Reuters, August 19, 2026).

10. FAO and WFP, Joint Anticipatory Action Appeal, launched June 18, 2026: $202 million to protect 8.8 million people across 22 priority countries, June 2026 to March 2027, described by both agencies as their first joint appeal of this kind; already positioned to reach 1.2 million, with a further $167 million required for the remaining 7.6 million. The launch material states that the systems, partnerships and operational plans needed to act are fully in place and coordinated for immediate action, and that what is needed is the financing. The 22 countries are Cameroon, Ethiopia, Kenya, Madagascar, Malawi, Mozambique, Nigeria, Somalia, South Sudan, Sudan, Uganda, Zimbabwe, Afghanistan, Pakistan, the Philippines, Timor-Leste, Colombia, El Salvador, Guatemala, Haiti, Honduras and Venezuela.

11. Searches conducted September 5, 2026 of OCHA Financial Tracking Service flows to FAO and WFP for 2026, CERF allocations for 2026, and the FAO anticipatory action project highlights page, which publishes donor, amount, project code and country for contributions to the SFERA Anticipatory Action Window. The most recent entry on that page at the time of searching was dated May 29, 2026. WFP reported on August 5, 2026 that since May it had triggered anticipatory action in six countries, providing over $14 million to support half a million people; only four of those six are among the 22 appeal countries and the period begins before the appeal launched, so the figure describes a wider portfolio rather than progress against the appeal. Funds announced outside the appeal: statement of the Emergency Relief Coordinator, July 13, 2026, offering up to $100 million from CERF with more than $20 million already allocated for anticipatory action in six countries; a CERF allocation of $6 million for Grand Sud, Madagascar, in early September 2026; and a Novo Nordisk Foundation contribution of $7.75 million to WFP announced September 17, 2026, described by WFP as the largest private-sector investment in its anticipatory action to date. Norad, July 3, 2026: Norway allocated NOK 30 million from its humanitarian reserve to support the joint appeal, half to FAO’s SFERA and half to WFP’s Anticipatory Action Trust Fund; about $3.0 million at early-July exchange rates and $3.2 million at late-September rates. This is the only government contribution earmarked to the appeal that we could find. WFP said on September 17, 2026 that it and the Novo Nordisk Foundation would jointly promote the appeal.

12. FTS reports funding progress against coordinated plans, principally humanitarian response plans and flash appeals. A joint inter-agency appeal of this type is not a coordinated plan and therefore receives no funding progress page.

13. Identifiable FAO anticipatory action funding across the 2023–24 El Niño cycle, from the FAO project highlights page: German contributions through the Federal Foreign Office and the SFERA Anticipatory Action Window of EUR 4.5 million and EUR 3 million for two global phases, plus Colombia $1 million, West Africa $3 million, Mozambique $500,000 and Tajikistan $132,104; Canada $1,222,000 for Somalia and $1 million for the Central American Dry Corridor; Norway $2,309,469 for Somalia; and $13,177,000 from the Government of Somalia with the World Bank. Several listed projects do not publish amounts, so the total is a floor rather than a complete figure.

14. Government of Uganda drought declaration, July 2026, with a six-month emergency response across all nine districts of Karamoja; Honduran government alert covering close to 80 percent of national territory; Brazilian federal release of R$1.335 billion against El Niño effects; Philippine Department of Agriculture distribution of drought-tolerant seed varieties.

15. Reporting via the International Food Policy Research Institute’s Food Security Portal, April 2026, drawing on World Food Programme data from the central Sahel: a 64 percent increase since 2023 in the population facing crisis-level hunger in areas where rations were reduced owing to funding shortfalls, against a 34 percent decline where full rations were maintained. The figures are for Mali: WFP, news release on West and Central Africa, January 2026, reporting a 64 percent rise in acute hunger (IPC 3+) since 2023 where rations were reduced and a 34 percent decrease where full rations were maintained. The same figures appear in UNRIC’s central Sahel briefing of June 2026 in looser wording.

Photo Credit: Doctors Without Borders

Nineteen African countries have already lost part or all of a harvest to this El Niño. The season that feeds most of Southern Africa has not yet been planted. It goes into the ground next month, and in August the region’s own meteorological services announced the forecast for these crops: below-normal rain across most of the subcontinent for the rest of 2026, with even less rain expected after January.

On June 18, the FAO and the World Food Programme made a joint appeal for $202 million to protect 8.8 million people from El Niño hunger before the harvests failed rather than after, based upon the arithmetic that a dollar spent early saves up to seven spent on humanitarian disaster relief afterward. Three months later, less than 20% of that sum has been found through any channel, and one of the largest contributions came from a Danish pharmaceutical company’s charitable foundation. The money requested will pay for: Cash transfers that reach households before they must start selling critical tools or breeding animals. Seeds that are drought-tolerant or flood-resistant. Livestock protection from drought and floods. Water harvesting and storage equipment. Flood defenses. Post-harvest storage. Dissemination of the El Niño forecast itself, with advice attached on what to do about it.

The crop failures in these nineteen countries were predicted before they even happened. That is what this series is about.

Photo Credit: WFP/Gabriela Vivacqua A farmer inspects a drought-stricken maize crop in Zambia.

Hunger Foretold showed that the El Niño hunger of 2027 is being forecast in public, on a schedule, by governments anyone can read for free. El Niño Has Landed described the rains that failed in Sudan this summer, the harvest that will fail in December, and the worst of the shortfall, which will not be felt until the second half of 2027. El Niño Crosses the Atlantic followed El Niño on the same ocean into the Americas, to the crops already lost there and those still at risk into 2027. El Niño Reaches Asia set out what is mostly still, for a few more weeks, a forecast across the half of the world where most of the planet’s rice is grown and showed the weakened monsoon and crops not yet planted at risk from this El Niño. This part returns to Africa, because Sudan did not fail alone, eighteen other countries failed with it, and because the reason these countries were so exposed was settled long before the Pacific warmed and this El Niño formed.

El Niño has already damaged crops in nineteen African countries: Burkina Faso, Cameroon, Chad, DR Congo, Djibouti, Eritrea, Ethiopia, Ghana, Guinea, Kenya, Mali, Mauritania, Niger, Nigeria, Senegal, Somalia, South Sudan, Sudan and Uganda. Egypt, whose own winter wheat had a good year, is exposed by a different route.

So where is the surge of humanitarian aid?

The appeal FAO and the World Food Programme launched on June 18 was the first joint anticipatory action appeal either agency has ever made, and it covers 22 countries.¹

The items the appeal would pay for divide by the calendar. Where a season has not been planted, the seeds and the water are the interventions, because they change what comes out of the ground: a shorter-maturing variety that finishes growing before the rain stops, a tank that holds rains that fall in November for the drought in January. Where the harvest is already gone, as it is in Sudan, nothing can be done for that crop at all, and the money does something less visible and no less important. It keeps a household from selling the goat, the cattle, the tools and eventually the land in order to eat this year. Selling off vital tools and livestock needed for farming is the transaction that turns one failed harvest into five failed years. In the 2023-24 El Niño, the two agencies reached more than three million people this way, months ahead of the El Niño’s peak.¹ There is little such help this year.

Three months after the appeal was launched, neither agency will say how much of the $202 million has been raised, and there is no entry against the appeal in the United Nations’ own funding tracker, because that tracker reports only against coordinated plans and an inter-agency appeal does not get a page. No government has publicly announced a contribution earmarked to the appeal itself. Money is moving, but through other doors and at a different scale. In July, the UN’s Emergency Relief Coordinator said he stood ready to disburse up to $100 million from the Central Emergency Response Fund to get ahead of this El Niño, and that more than $20 million had already been allocated for anticipatory action in six countries. In early September, that fund released a further $6 million for southern Madagascar. On September 17, the Novo Nordisk Foundation, a Danish philanthropy, put $7.75 million into WFP’s anticipatory work. The largest private contribution ever made to WFP’s anticipatory action, against an El Niño that has already tied the strongest ever measured, came from a pharmaceutical company’s charitable foundation.

Add every announced figure together, whether or not it is formally against the appeal, and the total is less than 20% of what FAO/WFP asked for. What has actually been delivered is smaller still: as of August 5, WFP’s anticipatory programs had reached half a million people with $14 million, against the 8.8 million people the appeal aims to protect. That is the “surge.”¹ Set against what is coming, it is not a surge at all. WFP’s own analysis, published in August, expects acute hunger across 45 already food-insecure countries to rise from about 225 million people to 274 million people by the end of 2027, an increase of roughly 22 percent, or at least 49 million additional people suffering from acute hunger or worse. What has been raised so far will reach a tiny fraction of the people in need.² Massive hunger and famine are coming, and not one major economy has announced a contribution to the appeal meant to prevent them.

The Season That Has Not Been Planted Yet

Begin with the part of Africa where nothing has been lost yet, because it is the only part where the outcome is still open to aid and relief. It is also the part the forecasters are most confident will be devastated by this El Niño. WFP’s own analysis, published in August, expects the largest effects of this El Niño event in parts of Central America and in Southern Africa, and the Southern African figure is the worse of the two. Southern Africa’s acutely food-insecure population is projected to rise by nearly seventy-five percent.Southern Africa has not appeared in the earlier parts of this series because its season has not yet begun. It begins with the planting that starts next month.³

The forecast is not one model’s output. It is a regional consensus, agreed at the thirty-third Southern Africa Regional Climate Outlook Forum in Swakopmund, Namibia, between August 25 and 28, by the national meteorological services of the region working with the European, American and British forecasting centers. It predicts below-normal rainfall across most of central and southern countries of the Southern African Development Community (SADC), an intergovernmental organization of sixteen member states that promotes economic integration, peace and security in the region. Below-normal rainfall is predicted from October to December, with even larger deficits from January to March 2027 across Angola, Namibia, Botswana, Zimbabwe, Zambia and parts of South Africa, with above-normal temperatures over nearly the whole region. The northern edge runs the other way, as it always does in these El Niño events: eastern DR Congo, most of Tanzania and the far north of Malawi and Zambia are expected to be wetter than usual, with flooding predicted to impact crops and livestock.³

The geography matters because of what grows in it. The dry half is rain-fed maize country, and maize is what Southern Africa eats. In the El Niño drought of 2015 and 2016, South Africa’s maize harvest fell to 7.8 million tonnes, the smallest crop since 2007, and the country became a net importer of the grain for the first time in eight years. The crop just harvested, grown in a favorable season before this El Niño reached the region, is a record 17.4 million tonnes. That is the baseline the coming season will be measured against, and the coming season is the one the forecast describes. A region that loses a large share of its principal crop does not only go hungry inside its own borders. It stops being the place its neighbors buy from. That is the mechanism the rest of this article describes: a shortfall in one exporting region becomes a higher price for every importer, and it reaches Sudan from as far away as Europe, where the Rhine, the great waterway that runs from the Swiss Alps to the North Sea, is running at record lows in the drought.³

No one has yet put a number on next year’s crop, and that is not an oversight. South Africa’s Crop Estimates Committee publishes its first indication, a survey of what farmers intend to plant, in late October. The preliminary planted area follows in January and the first production forecast in late February. Planting itself begins in the middle of October. The harvest this section is concerned with does not yet exist as a figure in any database, which is exactly why the decisions on proactive humanitarian aid that will determine how much of this crop is lost are still there to be made.

What does exist is the record of what El Niño has done to this crop before, and it is not a gentle record. In the 2023-24 season, when El Niño brought only a mild mid-summer drought, the maize harvest fell 22% to 12.9 million tonnes. The droughts of 2018-19 brought it to about 11.8 million tonnes. The 2014-15 and 2015-16 seasons averaged around 8.9 million tonnes. The clearest signal is in yields rather than totals: Agbiz’s figures show South African maize yielding 3.49 tonnes a hectare in a severe El Niño season, against more than 5.00 tonnes in La Niña years. The forecasters will not predict those numbers against the coming season until farmers have planted, which is the correct discipline. It does not stop a reader from noticing the range of the impact this El Niño is likely to have on these crops.³

Fortunately, South Africa is better placed for this El Niño than it was for the last one. The record harvest and the stocks carried over from 2025 leave roughly three million tonnes available for export in the 2026-27 marketing year, two wet seasons have left soil moisture and dam levels healthy, and FNB’s senior agricultural economist expects South Africa to need no significant maize imports in 2026 or 2027, with imports becoming a question only if El Niño runs into a second season. The danger in Southern Africa in 2027 is not mainly that South Africa fails. It is that Malawi, Zimbabwe, Zambia and Madagascar fail while the region’s supplier has a thinner surplus and more buyers bidding for it. Regional demand rises in the same season the regional surplus falls, and the countries that have just lost their own harvest are the ones who must go into that market to buy from a country that is already eating into its reserves.³

The human numbers are already on the record, before a seed is in the ground. FEWS NET (Famine Early Warning Systems Network) expects delayed and below-average rains to cut planting and seasonal farm labor in Zimbabwe, Malawi, Mozambique, Zambia and Lesotho, and deteriorating pasture to reduce livestock productivity in southern Angola. Malawi has 2.6 million people projected to face acute hunger in the 2026-27 lean season, with WFP preparing to assist 600,000 of them. Madagascar has 3.7 million people expected to be at Crisis or worse between October and January as this El Niño takes hold, and in early September the UN’s emergency fund released $6 million to get ahead of this hunger crisis in the south of the island. The African Development Bank has put the cost of this El Niño to affected African economies at up to $20 billion.⁴

This is the argument of this entire series at its simplest. In Sudan, as the rest of this article describes, the rains have already failed, and what is left to argue about is how fast the humanitarian aid response arrives and how much of the damage can still be prevented. In Southern Africa nothing has failed yet. The forecast is public, regional, consensus-based and a full month ahead of even planting the crops that are all at risk of failing. Seed choice, planting dates, water storage, pre-positioned relief, and the price at which a government offers to buy grain are all still open questions. Every one of them will be closed by December. If the hunger of 2027 in Southern Africa turns out badly, it will be possible to date the decisions that made it so, and they will have been taken in October and November of 2026, in public, with the forecast lying on the table and bureaucrats and politicians choosing not to act to prevent the hunger that they all know is coming.⁴ The UN General Assembly meets this week and all of the countries who could each provide the FAO/WFP funding will be in the room.

“The Southern African season has not even been planted. That is not a reason to wait. That is the entire opportunity.”

The rest of this article is about the places where that window has already shut. Every one of the nineteen countries that have lost part of a harvest to this event will have to buy grain to cover the gap. The price of that replacement grain is not set in Africa at all.

To Find the Drought, Start Looking at the Rivers in Europe and Africa

The European Rhine is at a record low. Through August, barges that normally carry grain and fuel through Germany were moving at about a fifth of their usual load. The Danube has dropped far enough that Hungary’s only nuclear power station came close to shutting down for want of cooling water, and far enough that Romania shut its sole working Danube-cooled nuclear reactor for the first time and sent naval engineers to detonate the riverbed to divert more water toward the intakes of the power plant. Copernicus, the European Union’s Earth observation program, recorded four of Europe’s great rivers at record lows in August: the Loire, the Po, the Rhine and the Danube. In eastern Croatia, the Drava and the Danube are at their lowest in years. None of the drought has ended. Germany’s federal hydrology institute says the rain that fell in late August bought only temporary relief and expects no real recovery before October.⁵

Rainfall ran below average across most of Europe between April and June, from Portugal in the southwest to southern Finland in the northeast, and the heatwaves followed. France is bringing in what its own growers expect to be a thirty percent fall in its maize crop, to 9.5 million tonnes, which would be the lowest maize harvest this century. The European Union has planted less than eight million hectares of maize for the first time since the century began, and the grain trade has cut the EU crop to its smallest since 2007. The European Commission’s crop monitors cut the soft wheat yield forecast to 5.88 tonnes a hectare in July, seven percent below last year, with maize and sunflower down six to seven percent, and the EC also warned that the heat had shortened grain-filling, so quality of these grains and seeds fell along with volume. Hot and dry conditions were forecast to persist. Romania, which has given up something like a million hectares of maize to successive dry years since 2019, got rain this season and expects a larger crop than last. The European Commission’s president called it the “summer of truth.” The EU has put the damage at about €180 billion, and the heat killed about 35,000 people in Europe alone, a bill for one summer roughly a thousand times the $202 million FAO and WFP asked for to protect 8.8 million people.⁶

The attribution is not in doubt. World Weather Attribution found that climate change made this drought roughly eighty times more likely in Western Europe and forty times more likely in Eastern Europe. That matches what this series reported in part one about El Niño itself: warming is making strong El Niños stronger, and this one has already tied the all-time record for Pacific warming, two months before El Niños usually peak. These climate impacts are arriving in a different hemisphere by a different route.⁷

This loss of grain production is not only a European story. It is the opening of an African one, because Europe and the Black Sea are where much of Africa buys wheat when its own harvests fail, and a shortfall in any major exporter raises the price for every buyer. Sudan imports roughly four-fifths of the wheat it eats. Egypt buys more wheat abroad than almost any country on earth, trading the title back and forth with Indonesia. Almost every nation in 4Hunger.org’s articles has lost part of a harvest this year, and every one of them will try to buy its way out of the shortfall on a market that has less grain than normal at the same time. A failed harvest is survivable if you can replace it. What is happening in 2026 is that the places people replace their food from are all failing together.⁷

Forty Years of Taking the Water Away

Before going country by country in Africa, it helps to say plainly what this region is, because the usual description of Africa is misleading. The conventional account runs: “here is a group of African countries with long-running wars and chronic hunger, and now a drought has arrived on top of them.” That framing makes the conflict the background, and climate change and this super El Niño the event that lands on top of the conditions already caused by civil strife. That framing gets the sequence backwards.

In much of this group of African countries, the wars are the result of loss of water due to climate change and overuse. Not entirely, and not simply, but enough that the line between a climate story and a conflict story stop being useful.

Take Lake Chad. In the 1960s, it covered roughly 9,700 square miles. It has since lost perhaps 90% of that surface area, and with the loss of the lake, lost the fishing and irrigated farming economy of a basin of countries that support tens of millions of people across Chad, Niger, Nigeria and Cameroon. What replaced that economy in the worst-affected areas was an insurgency recruiting from precisely the young men for whom the lake stopped feeding and providing a livelihood. The Lake Chad Basin is now one of the largest displacement crises in the world. The lake receded first.⁸

Or take Darfur in Sudan. The conflict that began there in 2003 is widely described, including by a former Secretary-General of the United Nations, as the first modern war with climate change among its root causes. Decades of declining rainfall pushed pastoralists south into cultivated land causing competition over water and grazing lands that turned violent. The violence was then organized and armed along ethnic lines. Sudan’s present catastrophe is the direct descendant of that war and of a loss of water driven in part by climate change. So, when this article reports that the rains failed in Gedaref, Sudan, this July, it is describing the third time in forty years that failing rain has helped redraw the lines of who lives where in Sudan.⁹

Take another fault line that is running through the Sahel right now. As the desert edge moves south, herders move with it, onto land that farmers have been cultivating for hundreds of years. That competition, over water rights and grazing corridors, both of which used to be sufficient for both ethnic groups, is among the drivers of violence across Nigeria’s Middle Belt, central Mali and Burkina Faso. FEWS NET (The Famine Early Warning Systems Network) counted 9,856 conflict events in the region between January and June, 41 percent of them in Nigeria. Those events get reported as security news. A good share of them are arguments about water and land that were settled peacefully when there was more of both water and useable land.¹⁰

This needs one qualification, because the research is contested. Many scholars argue that governance, marginalization and the politics of land tenure are the primary drivers, and that climate is an aggravator rather than a cause. They have good evidence. But notice what both accounts have in common. Lake Chad shrank partly through declining rainfall and partly through water drawn off upstream for irrigation and held behind dams. Herding corridors closed through land law as much as through the advancing desert. Rain-fed smallholders were moved into cash crops for export earnings and debt service. Whichever cause you weight most heavily, somebody decided how to respond to it, and for forty years the answer was to do very little while doing something was still cheap. Calling the result a long-running conflict makes it sound like terrain. The conflict was built by both climate change and governance errors.¹¹

“The usual telling is that El Niño has arrived on top of conflicts. In most of the countries in Africa, the more accurate telling is that the water left first due to climate change and overuse, the conflict then followed it, and El Niño is the third visitor to a house that has already been emptied twice.”

This is why the aid withdrawal described later in this article is not a separate subject. Remove the water over forty years. Let the farmland and the herding corridors go with it. Let the conflict that follows displace twenty million people. Cut the humanitarian funding that was barely allowing the population to barely survive in these countries. Then send an El Niño that has already tied the strongest on record into what is left. Africa is not a continent that happens to be gripped by civil conflict. It is a continent where the loss of water helped build the conflict, and withholding aid now penalizes the people who had the least say in climate change or any of it.

The fact that Sudan did not fail alone changes the shape of this story. Sudan sits inside a belt of countries that El Niño dries as a single unit: southern South Sudan, western and central Ethiopia, northern Uganda, Eritrea, Djibouti and Sudan itself. The belt’s rains all failed together in June and July, as the El Niño models predicted they would. Roughly 38 million people across the six IGAD (the Intergovernmental Authority on Development, a trade bloc and regional organization in East Africa and the Horn of Africa, with seven member states since Eritrea withdrew in December 2025) states of Djibouti, Ethiopia, Kenya, Somalia, South Sudan and Sudan, an overlapping but not identical group of countries, were already at IPC Crisis levels of hunger or worse before El Niño caused these rains to fail.¹² Crisis falling upon crisis.

Take the smaller members of that Sudanese belt first, because they will not make the headlines. In northern Uganda the rains came early in February, stopped, returned too weak to carry a crop and were gone by the end of May. Crop losses across Karamoja run between 50 and 70 percent, more than 600,000 people are short of food, and Kampala declared a drought in July. The same authorities now forecast near-normal to above-normal rain from September through December. In Uganda, months of sun on bare soil bake the surface into a crust, and then heavy rain on that crust runs off rather than soaking in, so Karamoja may lose one season to no rain and the next to too much.¹²

Djibouti imports almost everything it eats, so the failure of the regional rains reaches its households as a price increase rather than a harvest. Eritrea is the one country here that cannot be seen at all due to government policies. Satellites show the drought: ICPAC, the IGAD Climate Prediction and Applications Centre, a specialized climate center for the Greater Horn of Africa, records rainfall deficits from May through July with vegetation stress, and places most of Eritrea inside the belt of below-average rain. What cannot be seen is what it has done to people. Eritrea is not a FEWS NET country; it is absent from the IPC system, and the Global Report on Food Crises records simply that no data were available. Every other country in this article can be counted. Eritrea can only be observed from a satellite in orbit.¹²

South Sudan is the closest thing we have to a country already over the line into Emergency. FEWS NET (The Famine Early Warning Systems Network) expects Emergency or Phase 4 levels of hunger to remain widespread from October 2026 through January 2027. At the household level, an Emergency or Phase 4 Hunger designation means one of two things: 1) either the household already has large food consumption gaps showing up as very high acute malnutrition and excess mortality, people are visibly starving and dying; or 2) the household is avoiding starvation only by emergency measures such as selling their breeding animals, their tools, their land, pulling their children out of school, or taking on debt that they can’t afford.

Harvests in Greater Upper Nile are expected to be minimal at best. FEWS NET (The Famine Early Warning Systems Network) lists the causes in order, and the order is worth reading: 1) insecurity, 2) displacement, 3) below-average rainfall linked to El Niño during the main agricultural season, 4) flash flooding, and 5) renewed conflict in the dry season. Three of those five are human decisions. One arrived from the Pacific. The flooding is both climate and human decision: water that fell upstream in another country is released through a dam in Uganda and is leaving standing water on fields the people of South Sudan were driven onto because the fighting left them nowhere else to go.

Photo Credit: UNHCR/Tiksa Negeri

The flash flooding in that list is not a contradiction of the failed rains, and it needs explaining, because it is the clearest case in this article of damage arriving from somewhere else. The water is not local, and it is not new. The water causing the flooding today fell in the Lake Victoria basin, some six hundred miles south, during the exceptionally wet years that began in 2019, and that have been working their way north through Lakes Kyoga and Albert ever since, and which are periodically released through a dam in Uganda. The Sudd sits in a depression on impermeable clay, so once the water arrives it cannot drain or flow into a river; it leaves mainly through evaporation. By 2022, years of excess water had swollen the Sudd to roughly 63,000 square miles, about the size of Tunisia. People are being flooded off their fields in South Sudan by rain that fell six years ago, in another country.¹³

A risk of Famine is formally identified in four counties of South Sudan: Akobo, Nyirol, Nasir and Ulang. FEWS NET is explicit that Famine would occur there if conflict re-escalated and humanitarian access closed. That is not a remote scenario. It is a description of the past two years.¹⁴

Ethiopia has the largest population of any country in this belt, and El Niño has taken two harvests from it rather than one. The first was the “belg” harvest. The belg is Ethiopia’s shorter rainy season, running from February to May, and in belg-dependent areas it governs both crop planting and livestock water, setting the yields for barley, wheat and maize. Those rains came late and were followed by prolonged dry spells. Belg production came in 35 percent below average in West Hararghe and 31 percent below in East Hararghe, both in Oromia, and yields fell across the belg highlands of Amhara and Tigray. By July, headline inflation had reached 15.3 percent, its highest level in a year, in a country where the poorest households already buy most of what they eat. Higher prices, higher rates of hunger.¹⁵

The second was the “meher” harvest, Ethiopia’s main growing season, which accounts for more than 85 percent of the country’s crop production, and it matters more. The “kiremt” rains that are most important to a successful meher harvest arrived late as well, and then largely failed. FEWS NET’s August update finds rainfall deficits of more than 50 percent across much of the country and expects the June to September kiremt rains across most of Ethiopia to rank as the driest in the historical record. That shortfall is landing on households that had already lost the belg harvest.¹⁵

Rain did return in late July, but it did not rescue the season. FEWS NET’s August update finds rainfall still significantly below average and erratic, and expects the kiremt rains to stop early, by mid-September. It now expects the national meher harvest to be below average, possibly by as much as 30 percent, with little to no meher production at all in the lowlands of East and West Hararghe, central and eastern Tigray and northeastern Amhara, where maize has wilted, sorghum is severely stunted and some farmers have replanted twice. A late reprieve is not a full harvest, and a full harvest is what prevents acute hunger and famine.¹⁵

Then the sequence turns, in the way that should be the most alarming thing in this section. FEWS NET expects unseasonal rain across central and northeastern Ethiopia from October to December. It will bring some relief to pasture and water, but it will arrive exactly when what crop there is is being cut and dried. Rain at this time flattens the stalks so they cannot be cut, makes the grain germinate in the head before it is harvested, and puts the rest into storage too wet, so it molds. Sun is the only drying technology most Ethiopian smallholders have. At the same time the October to December rains over the southern and southeastern pastoral lowlands are forecast above average, bringing flash flooding to areas still recovering from the last drought. One country, one season, ruined twice by opposite weather: drought and flood in the same year.

The Rain That Failed Over Ethiopia Is Also a River

The rain that failed over the Ethiopian highlands this summer does not only fail as a harvest. It fails as a river. Roughly two-thirds of the water in the main Nile River comes off the Ethiopian highlands by way of the Blue Nile, and something close to eighty percent of the Nile River’s total flow originates in Ethiopia and the eastern basin. This is the one place in this article where the El Niño signal can be read as a number on a gauge rather than inferred from a field.¹⁶

The relationship is old and carefully measured. Work by Elfatih Eltahir and colleagues at MIT puts the El Niño-Southern Oscillation at about a quarter of the year-to-year variability in Nile flow and finds that eighty-three percent of El Niño events beginning between April and June have produced drought in the upper catchment of the Blue Nile. This El Niño began in late spring. That is not a coincidence noticed after the fact. It is the specific condition their work identified as the dangerous one, published years in advance, and it is why part one of this series could report that the failure of Sudan’s rains was forecast rather than unlucky. The same team has also shown that a warming climate should widen the swing in Nile flow from year to year by about half again, which is the Nile version of the finding this series reported in part one about El Niño itself.¹⁶

Downstream, the river now runs through a piece of infrastructure that did not exist during the last comparable El Niño. The Grand Ethiopian Renaissance Dam was inaugurated on September 9, 2025, and it can hold about 74 billion cubic meters of water. Part one of this series reported its reservoir standing about 20 feet lower in July 2026 than a year earlier, with parts of the Blue Nile basin receiving as little as half their normal rain. A dam is a buffer in a normal year and a question in a dry one, and the question is who absorbs the shortfall of rain. There is no agreement between Ethiopia, Sudan and Egypt that answers it. Egypt has spent fifteen years asking for a binding rule on releases during multi-year drought and has not got one. In August, Ethiopia’s water minister said his country needs nobody’s permission to develop its own rivers, and Egyptian ministers replied that Cairo would not permit further dams on the Nile.¹⁶

Egypt is the country at the end of the pipe, and its exposure is not the one this article has been describing everywhere else. Its own harvest was good. The 2026 wheat crop came in near ten million tonnes, about seven percent above the five-year average, and the state bought a record 4.72 million tonnes of it for the subsidized bread program. Egypt’s problem is that a good harvest is nowhere near enough. It eats roughly twice what it grows, buys something close to thirteen million tonnes abroad each year, and runs a bread subsidy that about seventy million people depend upon. Ninety-seven percent of its fresh water comes from the Nile, and water availability has fallen to about five hundred cubic meters per person per year, half the level at which the United Nations calls a country water scarce. On August 1, as the flood season opened, the Egyptian irrigation ministry declared a state of maximum emergency and put its Nile inflow committee on round-the-clock watch.¹⁷

So, Egypt meets this El Niño twice, from opposite directions, and neither arrival is a failed harvest of its own. First, less water is reaching it from an Ethiopian highland rainy season that failed nine hundred miles upstream, through a dam it does not control and cannot negotiate over. Second, the grain it must buy to cover the gap between what it grows and what it eats is priced on a world market that El Niño has tightened for reasons that have nothing to do with Egypt. A country can irrigate its way out of a drought, but it cannot irrigate its way out of both of those unrelated problems.¹⁷

“Sudan’s rain failed in July due to El Niño. Egypt will feel it in the flow of the Nile River, and it will feel El Niño again in the price of grains needed to make many loaves of bread, and in neither case will anything have gone wrong inside Egypt.”

Then there is the direction nobody thinks to look. Sudan’s drought has already caused a food crisis in Chad, because a million and a quarter people have crossed the border since the war in Sudan began and are eating Chadian harvests. The Cadre Harmonisé, the unified regional tool used to identify, analyze and classify acute food and nutrition insecurity in the Sahel and West Africa, put three million Chadians, almost fifteen percent of the population, at Crisis or Emergency levels of hunger for the lean season now ending, and two million Chadian children under five are expected to suffer acute malnutrition this year, nearly 484,000 of them severely. Severe acute hunger in childhood permanently alters brain development and metabolic health, leading to lifelong cognitive impairments, reduced physical capacity, and a significantly higher risk of chronic adult diseases like Type 2 diabetes and hypertension. A drought in one country becomes a food crisis and long-term health crisis in the next by way of the people who walked out of it.

Photo Credit: Doctors Without Borders

Across West and Central Africa nearly twenty million people are now forcibly displaced or stateless, more than fourteen million of them displaced within their own countries, driven by the wars in the Sahel, Sudan, the Lake Chad Basin and the DRC, and by climate shocks on top of those civil conflicts. The displacement crisis across West and Central Africa is a regional system collapse rather than an isolated national event. A failed harvest in Gedaref, Sudan, is therefore not one country’s emergency. It is arithmetic performed on host communities several countries away, who did not plant the crop and will not eat it either.¹⁸

“A drought does not stay where it falls, since much of Sudan’s population has already moved. Sudan’s failed sorghum will be eaten in advance, in eastern Chad.”

The Sahel is drying on the same signal due to El Niño. The Africa Climate Outlook expected El Niño to suppress the West African Monsoon through June to August, and that season has now run. It delivered rains unevenly, which is what the forecasters said it would do. FEWS NET (The Famine Early Warning Systems Network) records below-average rainfall observed across much of the Sahelian band, with moderate to severe deficits in western Senegal, southwestern Mauritania, central and southeastern Mali, central and southeastern Niger, northeastern Nigeria and much of Chad, and vegetation conditions below average in all of them. Crops in Mali, Niger and Chad are short of water. Temperatures are the amplifier: the models project above-normal heat across the region with confidence, and heat pulls moisture out of soil that is already short of it. The split of El Niño weather runs here too. While the Sahelian band dries, persistent heavy rain has put southeastern Nigeria under an elevated risk of flooding. FEWS NET’s (The Famine Early Warning Systems Network) regional read is more mixed than that list of deficits suggests: the agency records the shortfalls as concentrated in central Mali and eastern Chad, expects average to above-average rainfall across much of the Sahelian band over the season as a whole, and treats flooding rather than drought as the nearer risk in parts of Niger and Nigeria. Both pictures are true at once, and which one a household lives through depends on where it farms. In the Sahel, El Niño is a contributor rather than the cause of food insecurity.¹⁹

In West Africa, El Niño is arriving on the largest hunger crisis in this article. The Cadre Harmonisé analysis put 41.8 million people across West Africa and the Sahel in acute food insecurity at the end of 2025 and projected that 52.8 million people would be in acute hunger during the June to August lean season. Watch what happened to that projection as the season ran. By July, FAO and WFP had revised it up to 54.8 million people in acute food insecurity. The number expected to reach Emergency levels this year, more than three million, is more than double the 1.5 million of 2020. In Borno State in Nigeria, the updated Cadre Harmonisé projected about 10,300 people in Catastrophe during the lean season, the phase at which people starve. At the launch of the 2026 Sahel Humanitarian Overview in June, the UN reported 24 million people needing humanitarian assistance across Burkina Faso, Chad, Mali, Niger, northern Cameroon and northeast Nigeria. These are numbers larger than the eastern belt this article has spent much time describing, in a region most readers will not have seen mentioned in connection with El Niño at all.¹⁹ This hunger is due to a compounding mix of conflict, economic instability, climate shocks, and severe humanitarian funding shortages.

What moved that number? The rains were not the only thing that changed between January and July. This is the region where the World Food Programme went from assisting 1.3 million people in Nigeria during the 2025 lean season to having funding to help only 72,000 people by February 2026. That is ninety-four percent of an operation gone in a single year, in the country carrying the largest share of this region’s conflict. It is also where Mali ran the only controlled version of this experiment that anyone has. Rations were cut in some areas because the funding was cut and maintained in others. Where humanitarian aid was cut, crisis-level hunger rose 64 percent against 2023 levels; where full rations were maintained, crisis-level hunger fell 34 percent. Same country, same war, same year, and the only variable was whether humanitarian aid money was available so that the food arrived. Every other case for acting early in this article is a forecast or a cost ratio. This one is a record of what withdrawing the money has already done, and to whom. The revision from 52.8 million people affected to 54.8 million people affected is a forecast catching up with the withdrawal of aid as much as with the drought: in 2025 every one of the five largest donors cut its aid, all at once. The United States cut its aid by nearly 57 percent, alongside cuts by Germany, the United Kingdom, Japan and France.²⁰

The 41.8 million baseline that all of this is measured against is itself the product of the forty years described earlier in this article: the lake that receded, the land corridors that closed, the farmland that went to desert. El Niño did not build that number. It is only the most recent thing to be added to it.

The mechanism that turns a shortage into a famine is already at work here, and it is not a forecast. Cereal exports are currently banned in Chad, Niger, Mali, Burkina Faso, Cameroon, Ghana and Guinea, with Niger and Burkina Faso adding livestock and Burkina Faso adding cowpea. Most are standing measures, some dating to 2022, which is the point. This region has spent four years with its internal grain trade partly closed, so the mechanism that turned a shortage into a famine in 2008 is not something West Africa might resort to under pressure. It is the condition it is already in. The bans also do a second damage rarely counted. Since the grain cannot leave, staple prices have fallen to atypically low levels, which sounds like relief and is not: the farmers who grew it cannot earn from selling it. One policy, two harms.¹⁹

East of those countries living in drought, El Niño does the opposite. The regional climate forum that met in August put a ninety percent probability on wetter-than-normal October to December rains over southern Ethiopia, central and southern Somalia and northeastern Kenya, reinforced by a positive Indian Ocean Dipole, a climate pattern marked by an east-west temperature contrast across the tropical Indian Ocean, where the western side becomes unusually warm and the eastern side becomes unusually cold, making seasonal rainfall totals likely to pass about 16 inches in places in these parts of Africa. The FAO has already warned Somalia to expect serious flooding in the “Deyr” season. The rain will fall on pastoral areas still classified in Crisis, and the forecasters’ own warning list runs past drowned crops and drowned herds to washed-out roads and outbreaks of water-borne disease. They are still recovering from the drought of October 2020 to early 2023, when five consecutive rainy seasons failed across the region, the worst in forty years. What drove it was a rare triple-dip La Niña, which UNDRR rates the most severe such event in seventy years, alongside a persistently negative Indian Ocean Dipole. Both drivers held for three years without a break.²¹

One warm ocean, one region, drought in the west and flood in the east in the same weeks. That is exactly what was predicted, and exactly what El Niño has always done.

Figure 3. The belt fails as one unit, but not in one direction. From the Sahel through Sudan and South Sudan to western Ethiopia, El Niño suppressed the June to September rains. Across the same belt’s eastern edge it is forecast to deliver too much between October and December. The western drought does not end when the eastern flood begins; the two overlap in the countries between them.

Three times in this article a shortage in Africa has become a quarrel between governments rather than a problem inside one. Seven West African states have closed their borders to cereal exports. Water released from a dam in Uganda stands on fields in South Sudan, and nobody downstream has a say in when it comes. The flow of the Nile is set in Ethiopia and argued over in Cairo, with no agreement governing what happens in a dry year. El Niño does not create these disputes. It removes the surplus that made them unnecessary, and it removes it from every one of them, all at once.

The Other Critical Deadline

Southern Africa’s deadline is a planting date. The deadline for everywhere else in this article is quieter, and it has already started running. Food aid is not a transfer of money; it is a purchase, a shipment and a delivery, and the purchase cannot begin until the money is actually in hand. WFP’s own procurement rules say so plainly: buying may only start once the required funds are available. Conventional international purchase and shipping runs to three or four months, and longer when a commodity has to be sourced across an ocean. Where WFP has been able to position stock in advance, against money committed early, the same delivery has taken an average of nine days.²²

Put that against the calendar this article has described. Sudan’s sorghum harvest fails in December. The shortfall is felt from the middle of 2027. Money committed this autumn becomes food on a truck in time. Money committed in March 2027, when the joint appeal expires, becomes food in the summer, which is to say after the hunger it was meant to prevent has already started and disaster aid rather than proactive aid is needed to prevent starvation. The appeal does not fail on the day it closes unfunded. It fails about four months earlier, silently, on the day the last purchase that could have arrived in time was not made.²²

That is the whole argument of anticipatory action, and it is why the difference between one dollar and seven is not really about money. It is about when the dollar shows up. Every week that the $202 million goes unfunded is not a week of delay. It is a week subtracted from the far end, from the months in 2027 when the food would have been arriving. It is a week more of suffering by people in crisis levels of hunger.

Consider what all of that means for a household in Kassala or Kordofan, Sudan. Its own sorghum was lost when the rains failed in July. The wheat it would ordinarily buy instead comes mostly from Russia, through a Black Sea corridor where both ends are now closed at once, and it is priced on a world market that Europe’s poor harvest and low rivers have already tightened. Russian strikes have cut Ukraine’s expected farm exports roughly in half. Ukrainian strikes have taken more than ninety percent of Russia’s Azov and Black Sea export capacity offline, leaving Russia sitting on a grain harvest of about 140 million tonnes of grain that it cannot move. The prices have already responded to these circumstances. In August the FAO cereal index reached its highest price level since May 2024, with wheat fifteen percent above where it stood a year ago, and sorghum, the crop that has just failed in Gedaref, up again in the same month. A country that loses its harvest and then finds the price of the replacement has risen has not lost once. It has lost twice, and the second loss is the one that lasts, because the price increases do not end when the rains return.²³

“Nothing about the hunger in Africa began with the 2026-27 El Niño. Water has been leaving Africa for more than forty years. The grazing lands and the farmland went as the water disappeared. What changed this year is that the money holding hunger down across the continent was withdrawn. In 2025, official development assistance fell by a record 23%, humanitarian aid by 36%, and for the first time every one of the five largest donors cut aid at once. In the same months, the forecast for a super El Niño arrived. The response to this El Niño is the one part of this that can still be changed, and the funding can still be found. It will be cheapest if it arrives before the hunger disaster hits, rather than after the hunger and famine have begun.”

Endnotes

1. FAO and WFP, Joint Anticipatory Action Appeal, launched June 18, 2026: $202 million to protect 8.8 million people across 22 priority countries to March 2027; already positioned to reach 1.2 million, with $167 million required for the remainder. FTS reports against coordinated plans only, so an inter-agency appeal receives no funding page, and searches of OCHA FTS, CERF and FAO’s contribution page found nothing recorded against the appeal itself. Funds committed through other channels: statement of the Emergency Relief Coordinator, July 13, 2026, offering up to $100 million from CERF with more than $20 million already allocated for anticipatory action in six countries; a CERF allocation of $6 million for Grand Sud, Madagascar, in early September 2026; and a Novo Nordisk Foundation contribution of $7.75 million to WFP announced September 17, 2026, described by WFP as the largest private-sector investment in its anticipatory action to date and directable across all 22 appeal countries. WFP, August 5, 2026: anticipatory action triggered in six countries since May, over $14 million reaching half a million people; only four of the six are appeal countries and the period predates the launch. Delivery has since widened, with more than 170,000 people reached in Chad across seven provinces and more than 376,000 in the Central American Dry Corridor as of September 11, 2026. WFP names Austria, Germany, Ireland, the Republic of Korea, Norway and CERF among the donors to its anticipatory action work generally, rather than to this appeal. The appeal’s package of anticipatory actions, as listed by FAO and WFP on June 18, 2026: cash assistance, distribution of drought-tolerant and/or flood-resistant seeds and agricultural inputs, livestock protection measures, water harvesting and storage systems, flood protection infrastructure, post-harvest protection measures, agricultural advisories and the dissemination of early warning information, tailored to local context. FAO and WFP estimate that every dollar invested in early action can generate up to $7 in avoided losses and response costs. During the 2023-24 El Niño the two agencies supported more than three million people through anticipatory action, delivering assistance months before peak impacts. The “under a fifth” figure in the text is the sum of the CERF allocations to date, the Madagascar allocation and the Novo Nordisk contribution, against the $202 million requested.

2. World Food Programme, “World Food Programme ramps up preparedness and response plans as El Niño threatens to push tens of millions into acute hunger,” Rome, August 5, 2026. The analysis covered 45 countries already considered food insecure and where El Niño is expected to significantly influence rainfall, temperatures and the incidence of floods and droughts. Across those countries the acutely food-insecure population is projected to rise from approximately 225 million to 274 million by the end of 2027, an increase of about 22 percent, or at least 49 million additional people. The largest impacts are expected in parts of Central America and southern Africa, with significant effects also anticipated in eastern Africa and in South and Southeast Asia. Regional increases reported alongside the headline figure: approximately 83 percent in Central America and nearly 75 percent in southern Africa; more than 18 million additional people across East and Southern Africa; more than 16 million in Latin America and the Caribbean; 8.2 million in Asia and the Pacific; and nearly 6 million in West and Central Africa, a rise of 12 percent. WFP puts the food-security impact of the 2015-16 El Niño at 60 to 100 million people. El Niño is expected to peak between September and December 2026, with effects continuing through 2027 by way of disrupted planting and harvest cycles.

3. Southern African Regional Climate Outlook Forum (SARCOF-33), Swakopmund, Namibia, August 25-28, 2026, convened by SADC national meteorological and hydrological services with the SADC Climate Services Centre and input from ECMWF, NOAA and the UK Met Office: below-normal rainfall favored across most of central and southern SADC for October-December 2026, sharpening for January-March 2027 over Angola, Namibia, Botswana, Zimbabwe, Zambia and parts of South Africa; normal to above-normal rainfall over eastern DR Congo, most of Tanzania and the extreme northern parts of Malawi and Zambia; above-normal temperatures likely across most of the region from October 2026 to January 2027. On the projected increase in southern Africa specifically, see note 2: WFP’s August 5, 2026 analysis puts the rise in the region’s acutely food-insecure population at nearly 75 percent, against about 83 percent in Central America, with more than 18 million additional people across East and Southern Africa. South African maize: the 2016 harvest finished at 7.78 million tonnes, down 22 percent from 9.96 million tonnes in 2015 and the smallest crop since 2007, after the country recorded its lowest rainfall since records began in 1904; South Africa became a net maize importer for the first time since 2008. The Crop Estimates Committee’s seventh forecast, released August 26, 2026, put the 2025-26 crop at 17.4 million tonnes, 4.5 percent above the 16.65 million tonnes of the previous season and the largest on record. The 2025-26 crop was planted between October and December 2025 and harvested from April 2026, so it precedes this El Niño entirely; the season the SARCOF forecast covers is 2026-27, planted from October 2026 and harvested from April 2027. No production forecast for 2026-27 existed at the time of writing: the Crop Estimates Committee issues its intentions-to-plant survey in late October, its preliminary area estimate in January and its first production forecast in late February, with planting beginning in mid-October. Historical analogues from Agbiz: the 2023-24 season, affected by a mild mid-summer drought, saw the maize harvest fall 22 percent to 12.9 million tonnes; the 2018-19 droughts brought it to an average of about 11.8 million; the 2014-15 and 2015-16 seasons averaged around 8.9 million. Agbiz records maize yields of 3.49 tonnes per hectare in a severe El Niño season against more than 5.00 tonnes per hectare in La Niña years. On South Africa’s buffer: the 2025-26 crop plus carryover stocks leave about three million tonnes for export in the 2026-27 marketing year against annual domestic consumption of about 12 million tonnes; Paul Makube, senior agricultural economist at FNB Commercial, expects no significant South African maize imports in 2026 or 2027, with imports a question beyond 2027 only if El Niño persists for a second season, and notes that regional demand could rise if El Niño depresses harvests elsewhere in southern Africa.

4. FEWS NET Southern Africa: delayed onset, below-average rainfall and above-average temperatures expected across much of the region for the October 2026 to May 2027 season, reducing planting activity and seasonal labor demand in Zimbabwe, Malawi, Mozambique, Zambia and Lesotho, with deteriorating pasture reducing livestock productivity and milk availability in southern Angola and rainfall deficits compounding an early lean season in southern Madagascar; maize prices in Malawi significantly above average and expected to rise at least through January. WFP, September 11, 2026: Malawi 2.6 million people projected into acute hunger during the 2026-27 lean season with assistance prepared for 600,000 in the south of the country; Madagascar 3.7 million at Crisis or worse for the October-January lean season with 75,000 targeted under a government-led drought monitoring mechanism. CERF allocation of $6 million for Grand Sud, Madagascar, announced early September 2026. African Development Bank estimate of up to $20 billion in lost output across affected African countries, reported July 26, 2026.

5. Copernicus recorded the Loire, Po, Rhine and Danube at record lows in August 2026. The Rhine at Kaub fell to roughly nine inches, below the previous record of about 9.8 inches (25 centimeters) set in October 2018; German shippers reported cargoes at about 20 percent of normal loads. Hungary’s only nuclear power plant came close to a full shutdown for want of Danube cooling water; Romania shut its sole working Danube-cooled reactor at Cernavodă for the first time and carried out controlled underwater detonations of the riverbed to improve flow toward the intakes. The Drava and Danube in eastern Croatia at their lowest levels in years as of August 7, 2026. Water use restrictions were imposed across several French regions, with the grape harvest affected; England was declared in drought on July 29, 2026 after the driest July in nearly two hundred years across the south. As of mid-September the German Federal Institute of Hydrology reported that late-August rain brought only temporary relief, with no appreciable easing expected before early October, and the Kaub model gave a near-certain probability of the river remaining below the roughly 30-inch (77-centimeter) low-water reference through September 26. Commerzbank estimated that sustained critically low water would cost German GDP about 0.35 percentage points in the quarter.

6. European Commission Joint Research Centre, MARS Bulletin, July 2026: EU soft wheat yield cut to 5.88 t/ha, about 7 percent below 2025; grain maize 7.38 to 6.93 t/ha; maize and sunflower down 6 to 7 percent; total cereals about 1 percent below the five-year average. Heatwaves shortened grain-filling and brought the harvest forward, reducing quality as well as volume. On France: the maize growers’ association projected a 30 percent fall to 9.5 million tonnes, which would be the lowest this century, with further downward revision expected; Coceral cut EU maize production by nearly 8 percent to 52.7 million tonnes, the lowest since 2007; EU maize plantings fell below 8 million hectares for the first time this century. On Romania: the roughly one million hectares of maize lost is a reduction in planted area since 2019 under successive droughts, not a 2026 crop loss. Romania had ample rain in 2026 and its harvest was expected to rise, with one consultancy estimating 8.2 million tonnes against an official 7.75 million for 2025. On the cost of the summer: The Guardian, September 19, 2026, reporting Ursula von der Leyen’s State of the Union address of September 16, 2026 (“summer of truth”), the EU’s estimate of about €180 billion in damage and about 35,000 deaths in Europe. The comparison with the $202 million appeal holds at any plausible euro-dollar exchange rate.

7. World Weather Attribution: climate change made the 2026 European drought approximately 80 times more likely in Western Europe and 40 times more likely in Eastern Europe. See also the study reported in part one of this series on the intensification of recent strong El Niño events by anthropogenic warming. On the Black Sea: Ukraine’s agriculture ministry put agricultural exports for the 2026-27 marketing year at about 29.6 million tonnes against an earlier estimate of 64.4 million, with wheat down 53 percent to 8.3 million tonnes. A separate official figure covers grain alone and was revised to 38-40 million tonnes from about 43 million; the two measures should not be conflated. On the Russian side, more than 90 percent of Azov-Black Sea grain export capacity was taken offline following Ukrainian strikes and navigation restrictions by mid-August 2026, against a Russian harvest of about 140 million tonnes; analysts have estimated a combined global supply gap of 30 to 40 million tonnes. On the El Niño record: Carbon Brief, September 21, 2026, reporting the daily Niño 3.4 anomaly at 3.07°C on September 19, 2026, a statistical tie with the record of 3.08°C set in November 2015 and reached about two months earlier in the season; Zeke Hausfather of Berkeley Earth, quoted by AFP on September 21, 2026, confirming the tie on NOAA’s satellite-and-buoy record. Some forecasters using other datasets have described the record as already broken. On Sudan’s wheat: IFPRI’s assessment of Sudan’s wheat value chains, citing the World Bank (2020), puts imports at 80 to 85 percent of total wheat demand, with domestic production covering the remaining 15 to 20 percent.

8. Lake Chad is commonly reported as covering roughly 9,700 square miles in the 1960s and as having lost most of that since, with estimates of the reduction varying between roughly 90 and 95 percent; some recent satellite work finds partial recovery in the southern pool. Links between livelihood collapse around the lake and recruitment into armed groups are documented by UNDP and the International Crisis Group, though the weight of environmental against political factors is debated.

9. On Darfur: former UN Secretary-General Ban Ki-moon described the Darfur conflict as having ecological origins rooted in declining rainfall, in a Washington Post opinion piece in June 2007. The characterization of Darfur as a climate conflict has been contested by scholars who emphasize state policy, counter-insurgency strategy and the arming of militias as primary causes; see the subsequent literature for a fuller account. The argument made here is that declining rainfall contributed to the pressures that preceded the conflict, not that it caused it.

10. FEWS NET West Africa: 9,856 conflict events recorded between January and June 2026, 41 percent in Nigeria and 27 percent in Cameroon, with conflict identified as the principal driver of acute food insecurity in the region. On farmer-herder competition: southward movement of pastoral systems under desertification pressure is identified as a factor in violence across the Nigerian Middle Belt, central Mali and Burkina Faso in a substantial literature, alongside land tenure disputes, the collapse of customary dispute resolution mechanisms and the availability of weapons.

11. A large body of research argues climate is best understood as a threat multiplier acting through governance failure, marginalization and land tenure rather than as an independent cause of armed conflict, and several prominent climate-conflict claims have been criticized on methodological grounds. This article adopts the weaker position: that declining water availability over decades has increased vulnerability to conflict and food crisis, without asserting direct causation in any case. On Lake Chad, reduced rainfall, upstream abstraction from the Chari and Logone, and dam construction have all been identified as contributors; no proportion is asserted here.

12. European Commission JRC, March 2026, citing FAO: approximately 38 million people across Djibouti, Ethiopia, Kenya, Somalia, South Sudan and Sudan at IPC Phase 3 or worse. On northern Uganda: FEWS NET Uganda, June 2026 to January 2027, projecting Crisis across Karamoja on crop losses of 50 to 70 percent; IPC Karamoja, April to July 2026, about 474,000 people at Phase 3 or above; WFP, August 24, 2026, recording more than 600,000 short of food and 122,000 children under five needing treatment for acute malnutrition, with a drought declared in July across all nine districts. On Eritrea: ICPAC Climate Watch Advisory, August 11, 2026, recording rainfall deficits from May through July with soil moisture deficits and vegetation stress, and the FSNWG special report placing most of Eritrea within the below-average rainfall belt. Eritrea is not a FEWS NET country, is absent from the IPC system, and the Global Report on Food Crises records that no data were available. FAO removed Eritrea from its list of countries requiring external assistance owing to outdated information. Uganda Ministry of Water and Environment, September to December 2026 seasonal outlook: near-normal to above-normal rainfall across most of the country and near-normal to slightly above-normal in Karamoja, with El Niño and a developing positive Indian Ocean Dipole favoring enhanced rain.

13. Water (2025) on Sudd flooding: localized rainfall is not the primary driver in Jonglei and Unity States; upstream rainfall in the Lake Victoria basin is. The Sudd lies in a depression on impermeable clay, so floodwater recedes largely through evaporation. Lake Victoria outflow is regulated at the Nalubaale and Kiira dams at Jinja, Uganda, and University of Reading research notes Sudd flood levels vary with controlled releases. Lake Victoria rose more than 4.5 feet between September 2019 and May 2020, the largest increase since the 1960s, and remained near its historical maximum across four seasons; Mulangwa et al., EGUsphere preprint (2025), characterize 2019 to 2024 as multi-year propagation of excess storage rather than isolated rainfall events. Satellite analysis put the Sudd at roughly 63,000 square miles in 2022.

14. FEWS NET East Africa Food Security Outlook, June 2026 to January 2027: Emergency (IPC Phase 4) widespread in South Sudan October to January; minimal harvests in Greater Upper Nile; risk of Famine persisting in Akobo, Nyirol, Nasir and Ulang, with Famine assessed as the outcome should conflict re-escalate and access close.

15. FEWS NET Ethiopia Food Security Outlook Update, August 2026 to January 2027, based on information available as of August 27, 2026: June to September kiremt and karan/karma rains historically poor, with deficits exceeding 50 percent of average rainfall across much of the country, and likely to rank as the driest in the historical record across most of Ethiopia; rainfall still significantly below average and erratic despite moderate improvement since late July; kiremt rains likely to cease early, by early to mid-September; meher production, over 85 percent of national crop production, expected below average and possibly up to 30 percent below, with little to no meher production in lowland East and West Hararghe, central and eastern Tigray and northeastern Amhara; belg production 31 percent below average in East Hararghe and 35 percent below in West Hararghe; unseasonal rainfall likely across central and northeastern areas from October to December; slightly early, above-average October to December deyr/hageya rains with elevated flood risk in southern and southeastern pastoral areas; annual headline inflation 15.3 percent in July 2026, per the Ethiopian Statistical Service.

16. On Nile hydrology: the Blue Nile contributes roughly two thirds of main Nile discharge, and Ethiopia together with the eastern basin accounts for about 80 percent of total flow. Siam and Eltahir put ENSO’s share of interannual Nile flow variability at about 27 percent. Zaroug, Eltahir and colleagues (Hydrology and Earth System Sciences, 2014) found that 83 percent of El Niño events beginning in April-June produced droughts in the upper catchment of the Blue Nile, and that an El Niño followed by a La Niña carries a 67 percent chance of extreme flood. Eltahir and Siam (Nature Climate Change, 2017) project roughly a 50 percent increase in year-to-year flow variation under warming. GERD inaugurated September 9, 2025, reservoir capacity about 74 billion cubic meters; reservoir level and Blue Nile basin rainfall for July 2026 as reported in part one of this series. Ethiopian Water and Energy Minister Habtamu Itefa Gelata, August 21, 2026, on requiring no permission to develop Ethiopian water resources; Egyptian foreign and irrigation ministry statements of the same month.

17. Egypt: FAO’s June 2026 country brief forecast a 2026 wheat crop of about 10 million tonnes, close to 7 percent above the five-year average, with state procurement closing at a record 4.72 million tonnes and the 2026-27 import requirement put at 13.5 million tonnes. USDA has Egypt and Indonesia both near 13 million tonnes for 2025-26, close enough that the title of largest wheat importer passes between them. The subsidized bread program serves roughly 70 million people. Egypt’s Ministry of Water Resources and Irrigation, February 10, 2026, on per capita water availability of about 500 cubic meters a year against a UN scarcity threshold of 1,000; the ministry puts the Nile’s share of Egyptian water resources at 98 percent, other sources at 97 percent. The 2026-27 hydrological year opened on August 1, 2026 with a declared state of maximum emergency and round-the-clock monitoring by the Nile River Inflow Committee. The 1959 agreement allocates Egypt 55.5 and Sudan 18.5 billion cubic meters annually.

18. FEWS NET West Africa regional outlook, June 2026 to January 2027: Crisis! (IPC Phase 3!) in eastern Chad from June to January 2027, driven by Sudanese refugees and Chadian returnees. UNHCR: nearly twenty million forcibly displaced or stateless across West and Central Africa, more than fourteen million internally. Cadre Harmonisé via WFP: some three million Chadians at Crisis or Emergency for the June to August 2026 lean season. FEWS NET Chad, February 2026: 914,607 Sudanese refugees and 355,903 Chadian returnees since April 2023, with a new wave into Wadi Fira and Ennedi-Est after El Fasher. IPC Chad Acute Malnutrition, October 2025 to September 2026: 2 million children 6–59 months acutely malnourished, including nearly 484,000 severely.

19. Africa Climate Outlook via Welthungerhilfe, June 2026, on El Niño suppression of the West African Monsoon. FEWS NET West Africa Seasonal Monitor, June and July 30, 2026: below-average rainfall across many Sahelian areas including Mali, Niger and Chad, with moderate to severe deficits in western Senegal, southwestern Mauritania, central and southeastern Mali, central and southeastern Niger, northeastern Nigeria and much of Chad, and above-normal temperatures worsening moisture loss. NOAA CPC, August 20 to 26, 2026, on abnormal dryness in southeastern Mali and elevated flood risk in southeastern Nigeria. Cadre Harmonisé, November 2025 cycle: 41.8 million in acute food insecurity for October to December 2025, projected to 52.8 million for June to August 2026, revised to 54.8 million in the joint FAO-WFP briefing of July 9, 2026. UN OCHA, 2026 Sahel Humanitarian Overview, June 3, 2026: 24 million requiring assistance. Forecasts for August to October indicate above-normal rainfall in southwestern Niger and northwestern Nigeria. WFP, January 16, 2026 (reported by UN News), drawing on the Cadre Harmonisé: more than three million people projected at Emergency levels in 2026, more than double the 1.5 million of 2020. FAO Regional Office for Africa, January 26, 2026: more than 15,000 people at risk of Catastrophe (Phase 5) in the Dikwa, Kaga and Kalabalge localities of Borno State; the updated Cadre Harmonisé reported by FAO on July 6, 2026 put the Phase 5 population in Borno at 10,300.

20. World Food Programme, January 2026: 1.3 million people assisted in Nigeria during the 2025 lean season against funding sufficient to reach 72,000 in February 2026. On Mali: reporting via the International Food Policy Research Institute’s Food Security Portal, April 2026, drawing on WFP data from the central Sahel, recording a 64 percent increase since 2023 in the population facing crisis-level hunger in areas where rations were reduced owing to funding shortfalls, against a 34 percent decline where full rations were maintained. Note that neither FAO nor WFP has published a decomposition of the revision from 52.8 to 54.8 million between its climatic and financing components, and none is asserted here. The OECD reported that 2025 official development assistance fell 23.1 percent in real terms to $174.3 billion, the largest annual decline on record, with humanitarian assistance down 35.8 percent, core UN funding down 27.0 percent, and bilateral aid to sub-Saharan Africa down 26.3 percent. It was the first year in which all five of the largest donors reduced their aid simultaneously; United States assistance fell by nearly 57 percent, and its contributions to the UN by 87 percent. The OECD projects a further fall of between 5.8 and 6.9 percent in 2026, with bilateral aid to sub-Saharan Africa declining a further 11.6 percent. WFP also projected 13 million children in the region suffering malnutrition during 2026.

21. Greater Horn of Africa Climate Outlook Forum (GHACOF-74), August 17-18, 2026, convened by ICPAC with the national meteorological services of the region: a 90 percent likelihood of wetter-than-normal October-December 2026 rainfall over southern Ethiopia, central to southern Somalia and northeastern Kenya, and a greater than 90 percent chance of exceeding 400 millimeters (about 16 inches) in parts of central and southern Somalia, central Kenya, the Lake Victoria basin, Burundi, western Rwanda and western Tanzania, reinforced by a positive Indian Ocean Dipole. The accompanying Food Security and Nutrition Working Group report warns of risk to life, crops and infrastructure and of elevated water- and vector-borne disease outbreaks; FAO has warned separately of significant Deyr flooding in Somalia. FEWS NET East Africa Food Security Outlook, June 2026: Crisis (IPC Phase 3) persisting to at least January 2027 in southern and southeastern pastoral areas after below-average March to May rains and incomplete recovery from the 2020–23 drought. UNDRR, Horn of Africa Floods and Drought, 2020-2023: Forensic Analysis (2024): the 2020-2022 La Niña was the most severe in 70 years across Ethiopia, Kenya and Somalia, owing to its intensity and three-year duration.

22. On procurement lead times: WFP’s local food procurement policy documentation states that the process of purchasing food may only begin once the required funds are available, and identifies the arrival of funds during a country’s lean season as a recurring constraint. WFP’s 2020 update on food procurement reports conventional international lead times of several months, reaching four and a half months for rice once shipping delays were included, against an average of nine days for commodities drawn from stock forward-positioned through the Global Commodity Management Facility, which finances purchases in advance of confirmed donor funding. The Joint Anticipatory Action Appeal of June 18, 2026 runs to March 2027.

23. FAO Food Price Index, August 2026: the headline index at 133.3 points, up 1.9 percent on July and 2.5 percent year on year; the FAO Cereal Price Index at 116.3, its highest level since May 2024, with wheat up 2.6 percent on the month and 15.0 percent year on year, maize up 2.5 percent, sorghum up 3.9 percent and barley up 2.6 percent. FAO attributes the move to continued Black Sea export disruption, deteriorating production prospects in the European Union under prolonged heat and dryness, and input costs following the closure of the Strait of Hormuz. The September index, covering August, was the most recent available at the time of writing.

“Man’s inhumanity to man makes countless thousands mourn.” Robert Burns

Photo Credit: United Nations

One hundred and fifty years ago, the tropical Pacific warmed to what remains, by most reconstructions, the most extreme El Niño in the instrumental record. Rain that should have fallen as the monsoon on Asia fell into the ocean instead. Between 1876 and 1878 drought struck India, northern China, Brazil, Egypt, Morocco, Southern Africa and Australia all at once. Tree rings suggest Asia’s drought was the worst in eight hundred years. Somewhere between thirty and sixty million people died, roughly 3% of the human population, and in India the colonial government went on exporting record tonnages of wheat while the people of India starved.

Nobody in 1876 knew why the failed monsoon was happening. The word El Niño had not entered the scientific literature; the Southern Oscillation would not be described for another forty years. The rains simply stopped, and then people died, and the connection between Peru’s warm water and the Deccan’s failed harvest was invisible to everyone it killed.

That is the only thing about this year that is different. Everything else is the same physics. All of the physics and whether projections are reported widely, weekly and can be read by everyone.

What the hundred and fifty years of records now tell us

El Niño does not create or destroy rain. It moves it. Warm water in the eastern Pacific shifts the convection that drives global circulation, and the rain that falls reliably in one place falls somewhere else instead. The result is not one disaster but a pattern of them, and after a century and a half of observation the pattern is well enough mapped that the affected regions can be named in advance: drought through southern Africa, parts of the Horn, Indonesia, the Philippines, northeastern Brazil and Central America’s dry corridor; flood through eastern East Africa, coastal Peru and Ecuador, and parts of the Southern Cone.

On September 10, the United States Climate Prediction Center put the probability of a very strong El Niño developing through this autumn and winter above 90 percent, with a 75 percent chance of exceeding the strength of any event recorded since 1950. Sea surface temperature anomalies in the eastern equatorial Pacific already run an unprecedented more than three degrees Celsius above normal. In August, the Meteorological Services of Southern Africa agreed in regional consensus that rainfall will run below normal across most of the subcontinent from October, getting even worse from January. The Regional Forum for the Greater Horn put a ninety percent probability on the opposite for other parts of Africa: rain well above normal, with flooding across southern Ethiopia, central and southern Somalia and northeastern Kenya during the same months.

That is the 1876 pattern, repeated many times, now forecast, published, and free for any government to read.

The projections

The World Food Programme published its estimate in August. Across 45 countries that are already food insecure, and where El Niño will materially shift rainfall and temperature, the number of people in acute hunger is projected to rise from roughly 225 million to 274 million by the end of 2027. That is an increase of at least 49 million people, about 22%. The largest effects are expected in parts of Central America and southern Africa, with significant impacts in eastern Africa and South and Southeast Asia. The African Development Bank estimates the event could cost affected African economies up to $20 billion in lost output.

Nineteen African countries have already lost part of a harvest. Malawi has 2.6 million new people projected to suffer from acute hunger this lean season; Madagascar has 3.7 million expected at crisis levels or worse between October and January. Rice sowing has already been delayed across the Philippines, Cambodia, Thailand and Vietnam.

An appeal sized to what donors might give, not to the Hunger that is coming

On June 18 the Food and Agriculture Organization and the World Food Programme issued the first joint anticipatory action appeal either agency has ever made. They asked for $202 million to protect 8.8 million people across 22 countries, running to March 2027.

Read that appeal to help 8.8 million people against the 49 million people who are projected to suffer from El Niño induced acute hunger. The appeal covers eighteen percent of the expected additional caseload. The other eighty-two percent have no humanitarian aid pointed at them at all.

This was not an oversight. It was an appeal written by people who had watched the funding collapse of the preceding year. In 2025 official development assistance fell 23% in real terms, the largest annual decline on record; humanitarian assistance fell nearly 36%; bilateral aid to sub-Saharan Africa fell 26%; and for the first time every one of the five largest donors countries cut humanitarian aid simultaneously. The agencies asked for what they thought they might plausibly be given. They did not ask for what the forecast actually requires.

Even that has not been funded. Three months on, neither agency will say how much of the $202 million has been raised, and there is no entry against the appeal in the UN’s own financial tracking service, because that service reports against coordinated plans and an inter-agency appeal does not get a page. What has moved came through other doors: an offer of up to $100 million from the Central Emergency Response Fund with more than $20 million already allocated, $6 million released for southern Madagascar in early September, and $7.75 million committed on September 17 by the Novo Nordisk Foundation, the largest private-sector investment in anticipatory action ever recorded. A pooled fund and a pharmaceutical company’s charity. No donor government has publicly announced a funding commitment earmarked to the appeal or to El Niño relief.

What the whole problem would cost

Nobody has to estimate what this El Niño will cost, because the estimate exists. In July the Peterson Institute for International Economics published a costing of this event, built on a 2023 study in Science that attributed $4.1 trillion and $5.7 trillion in global income losses to the El Niños of 1982-83 and 1997-98. PIIE puts first-year losses at roughly $986 billion in 2027, about 0.8 percent of global GDP, and cumulative six-year losses at $17.6 trillion, and calls those assumptions conservative. The burden falls where it always does: $353 billion of the first-year loss lands on fifty-one countries, a hit of 3.6 percent of their combined GDP. And PIIE is explicit that while the first-year losses are largely baked in, the $6.2 trillion that follows is not. It represents crops not planted and clinics not stocked, all of it sensitive to whether help arrives in time.

Its recommendations are that development banks pre-position credit before the peak rather than after, and that humanitarian pre-positioning begin now, noting that WFP desperately needs capitalization after the United States cut its contribution by 54 percent in 2025. A Washington economics institute has arrived independently at the case the humanitarian agencies have been making since June.

So set the humanitarian appeal against the potential damage. The objection is always that the sums are unaffordable. They are not, and the agencies’ own figures settle it. The appeal protects 8.8 million people for $202 million, about $23 a head. WFP’s 2024 performance report puts the average cost of feeding someone already in crisis at $51 a year, or forty-five cents a day.

Protecting all 49 million people at risk of acute hunger in advance, at the appeal’s own unit cost, comes to roughly $1.1 billion. Feeding them afterward, at WFP’s own average, costs about $2.5 billion a year, and that is a floor, since emergency rations in hard-access countries run above the global average. For scale: WFP has received $3.7 billion this year from every donor in the world for every operation it runs, so the additional caseload alone would consume two-thirds of the WFP’s entire budget. Against the African Development Bank’s $20 billion, $1.1 billion of prevention is five percent of the damage on a single continent. Based upon the agencies’ own ratio of one dollar early to seven late, acting ahead of all 49 million stands against something near $8 billion in avoided losses.

Now hold those numbers against PIIE’s. The $202 million appeal is about one fiftieth of one percent of the losses this event is expected to inflict in 2027 alone. For every thousand dollars the world economy is about to lose, the two agencies have asked for twenty cents, and have not been given it. The $1.1 billion that would cover everyone forecast to move into acute hunger due to this El Niño is about one dollar and ten cents per thousand dollars of financial loss, roughly a tenth of one percent of the first-year damage.

Divide that where it belongs. The World Bank counts eighty-seven high-income countries, and not one of the twenty-two countries named in the appeal is among them. Spread across those eighty-seven high income countries, the $202 million already requested is $2.3 million each. The $1.1 billion that would cover everyone this El Niño event is forecast to push into acute hunger is about $13 million each, one mid-sized government contract, once. It is not a burden on anybody. It is a rounding error with a body count attached.

Why Does The Humanitarian Funding Have To Be Mobilized Starting This Month?

Southern Africa’s primary maize growing season is planted from October to December and harvested from April to June, straight into the forecast of the peak of El Niño. India’s “rabi” crops are sown October to December. In Brazil, soybean and maize planting opened this month in Mato Grosso, the country’s largest producing state, where El Niño is expected to bring heat and drought. Southeast Asia’s dry-season rice runs into the first quarter of 2027, and Central America’s “primera” maize crop is not sown until around May. So the decision window is October through December for most of the affected world, extending into the first half of 2027 for parts of Asia and the Americas, which is also, near enough, when the appeal expires.

Photo Credit: Oxfam South Africa

Within that window, the money has to arrive earlier still, because food aid is a purchase rather than a transfer. WFP’s own procurement rules state that buying may only begin once funds are actually in hand. Conventional international purchase and shipping runs three to four months. WFP also runs a fund that buys food before donors pay and stores it close to where it will be needed. When countries draw on those stocks, delivery has averaged nine days instead of months. Money committed this autumn becomes seed and food in time for the planting and the lean seasons that follow. Money committed in March arrives in the summer, after the hunger it was meant to prevent.

The appeal will not fail on the day it closes unfunded. It will fail months earlier, silently, on the day the last purchase that could have arrived in time is not made.

Photo Credit: ©KANPISUT - STOCK.ADOBE.COM

The Rice Belt Is Counting Already

Asia grows the largest grain crop in the world and the one where a shortfall stops being a regional tragedy and becomes a global price hike. Also, as of this month, the impacts of El Niño is no longer entirely a forecast.

India’s southwest monsoon has run short all season, and the shape of the shortfall matters more than the size. June received only 60 percent of the long-period average for rain. July rains then over-corrected to 103 percent, which rescued a great deal of rice plantings that had looked lost. By August 24, the cumulative rain deficit stood at 13 percent, and the season’s first three months finished 14 percent down. The India Meteorological Department now expects September to deliver less than 91 percent of normal rain. September carries only about a fifth of the season’s rain, so that would leave the monsoon at roughly 87 percent of its long-period average or lower: below the 90 percent line under which the IMD classes a season as deficient, and short of the department’s own May forecast. For the season to reach even 90 percent, September would need to deliver 107 percent of normal.¹

That recovery in July is worth dwelling on, because it is the difference between this article and the previous three. Kharif sowing was running 23 percent behind at the end of June. By the end of July, it was 3 percent behind, and by August 17 the gap was down to 1.5 percent. Live monitoring of water storage across 150 major reservoirs reached 64.3 percent of capacity by August 20, well up from the 44 percent at the start of the month. One month of rain undid most of two months of damage. That is what a system with irrigation and reserves can do, and it is precisely what Sudan and Guatemala do not have.²

But the averages hide the story. Rice acreage is still down 3.2 percent with an estimated harvest of 40.5 million hectares. The deficits that persist are in the eastern and northern rice belt, in the sugarcane areas of the west and south, and in the soybean regions, all of which are under moisture stress. The USDA’s Foreign Agricultural Service now expects Indian rice production to be down about 4.5 percent and maize down 9.3 percent. A crop assessment in late August found 32 percent of 446 monitored districts under watch, 15 percent of them for dry risk. Uneven rainfall over a subcontinent is not a smaller problem than uniform drought, it is a differently distributed one, and this one lands on the farmers with the least irrigation to fall back on. Roughly three in five hectares of Indian cropland have no assured access to water.³

The input costs for fuel and fertilizer have moved independently of the weather. Indian fertilizer companies have been importing urea at more than $950 a tonne, compared to $400 to $450 a tonne before the Strait of Hormuz closed in February. The cost of sugar went from 48 to 55 rupees between July 20 and August 20, and the government has permitted duty-free raw sugar imports to ensure that there is enough sugar to meet India’s needs. The phosphate and fertilizer squeeze this publication described in Hunger Foretold, is arriving in the largest agricultural economy in the world at the same moment as a weak monsoon.⁴

Figure 1. India’s monsoon rescued itself in July and then did not finish the job. June came in at 60 percent of the long-period average, July at 103, and the IMD expects September below 91. Sowing recovered from 23 percent behind last year to 1.5 percent. What the averages hide is where the shortfall sits: rice acreage is still down 3.2 percent, and the deficits that remain are in the eastern and northern rice belt.

The Crop That Has Not Been Planted Yet

The crop that hasn’t been planted is the “rabi” crop, sown from October into December and harvested in spring, and it is mostly wheat. Rabi does not run on monsoon rainfall. It runs on what the monsoon left behind: soil moisture at sowing, and water in the reservoirs to irrigate the rabi crop through the dry winter. Which means the reservoir figure quoted above is not really a kharif number at all. Storage at 64.3 percent of capacity in late August is the opening balance for the winter crop, and it is the number to watch through October as it draws down.

This matters because El Niño winters in northern India tend to be dry, and because wheat, not rice, is what India’s own food security system runs on. The Food Corporation of India’s buffer is a wheat and rice buffer, and wheat is the half that has been tight in recent years. A weak rabi arriving in March 2027, on top of a kharif already down 4.5 percent in rice, is the sequence that produces an Indian policy response rather than merely an Indian harvest problem.

The same is true across the region. Vietnam’s winter-spring rice is planted in November, Indonesia’s main planting begins as the rains arrive or fail, and Afghanistan’s winter wheat is sown from October. None of those crops exists yet. All of them are 2027’s food, and all are still a decision rather than a harvest.

What New Delhi Does Next Is the Whole Question

India controls roughly forty percent of the world’s traded rice. Which means the question for 2027 is not really whether Indian farmers lose yield. It is what the Indian government does about it, and how fast everyone downstream reacts to what India does.

There is a precedent and it is recent. In 2023, a merely moderate El Niño was enough to produce a ban on non-basmati exports within weeks, and rice prices reached fifteen-year highs. The ban was not a response to famine in India. It was a response to a forecast, and to the domestic politics of food inflation. Both of those conditions exist again now, with a considerably stronger event and a urea price more than double what it was in January.

“The mechanism that turns a shortage into a famine is not the weather. It is governments each acting reasonably to protect their own people, in an order that leaves the poorest countries last.”

Watch two things rather than the monsoon totals. The first is the Food Corporation of India’s buffer stock position against its quarterly norms, which are published monthly. The second is the Directorate General of Foreign Trade’s notification channel, which is where the 2023 ban appeared with no warning on July 20. Stockpiling is the observable step before a ban, and it is already happening elsewhere in the region.

The Importers Are Already Buying Rice

That part is no longer hypothetical. Malaysia and the Philippines have been increasing rice purchases to build national reserves ahead of the dry season, and that precautionary demand has moved the market on its own. Thai and Vietnamese benchmark grades have climbed above $400 a tonne, with Thai prices rising $40 to $50 in a fortnight at one point this summer. The Bangkok export price reached $490 a tonne at the end of June, up 27.5 percent since the start of the year.⁵

This is how 2007 and 2008 began. Not with a harvest failure large enough to starve anyone, but with importers buying early, exporters noticing, and each government reasonably deciding to secure its own supply first. World rice production that year was not short. The price tripled anyway.

Thailand, the second-largest exporter, has had its El Niño warning already. The country received roughly sixty percent less rain than normal between January and April, and the Office of the National Water Resources has said it could face a drought crisis more serious than 2015. Its 2025-26 paddy crop came in at 35.44 million tonnes, with the main crop down slightly and the off-season crop down 1.6 percent, on dry spells and water limits outside the irrigation zones.⁶

The Delta That Feeds the Importers

This brings us to the country missing from almost every rice story written about this El Niño, including, until now, this one. Vietnam.

Vietnam is where the Philippines buys. Of Philippine rice imports so far this year, 2.59 million tonnes came from Vietnam, compared to 378,404 tonnes from Thailand and 286,052 tonnes from Myanmar. Vietnam is not a minor supplier to the region’s most exposed importer. Vietnam is the supply.

Vietnam has published its own forecast, which almost nobody outside the country has read. The Department of Hydrometeorology, under the Ministry of Agriculture and Environment, expects El Niño to persist with a probability above 95 percent, reaching moderate to strong intensity with a 60 to 65 percent chance of becoming very strong by late 2026 or early 2027. Note in passing that Hanoi’s own agency is more cautious than the WMO on that last point, which is worth remembering the next time a single forecast is quoted as settled.

What the Vietnamese forecast says next is the part that matters for everyone downstream. Southern Vietnam’s rainy season is expected to end in late October or early November, several weeks earlier than usual, particularly across the Mekong Delta. From roughly November through April or May of 2027, the Delta is forecast to face severe drought and saltwater intrusion. During high tides in early 2027, the four grams per liter salinity boundary, the point past which rice will not grow, is projected to push 90 to 125 kilometers up the Vam Co river, 85 to 90 kilometers along the Cua Tieu, and 70 to 80 kilometers along the Ham Luong. That would directly affect more than 350,000 hectares of winter-spring rice, along with fruit-growing areas, aquaculture, and the domestic water supply for millions of people in coastal communities.⁷

The winter-spring crop is the export crop. It is also the crop that has not been planted. Vietnam’s Department of Crop Production has already instructed provinces in the Mekong Delta and the southeast to plan the 2026-27 winter-spring rice season around water availability, tidal surge and salinity risk, which is to say around a forecast rather than around a harvest. The ministry has published two planning scenarios, one high-risk and one very high-risk, and has distributed them to provincial authorities as preparedness tools.

So, the sentence that follows in the next section of this article, that the Philippines is buying on a market where the largest exporters are short, is not a rhetorical flourish. It is a description of a forecast that the supplying government of Vietnam has already published and already begun planning against.

There is one more thing in the Vietnamese record, and it is the most encouraging fact in this article. During the 2015-16 event, reduced Mekong flows combined with severe saltwater intrusion caused widespread disruption to domestic water, agriculture and aquaculture across the delta. In 2023-24, salinity intrusion was again severe. The losses were considerably lower. The Vietnamese assessment attributes the difference to improved forecasting and earlier response measures. Same river, same ocean cycle, comparable salinity, far less damage, because somebody acted on the forecast in time. That is the entire argument of this series, run as a natural experiment, in the Delta that feeds both Vietnam and the Philippines.⁸

Figure 2. Vietnam’s Mekong Delta salinity forecast for early 2027: the four grams per liter boundary projected 90 to 125 km inland on the Vam Co, 85 to 90 km on the Cua Tieu, 70 to 80 km on the Ham Luong, affecting more than 350,000 hectares of winter-spring rice. The rainy season is expected to end several weeks early, in late October or early November.

The Philippines Has Two Weeks of Rice

If you want to know where in Asia this becomes a hunger story rather than a price story, it is the Philippines.

More than half of Philippine rice land is rainfed, and the irrigation that serves the rest depends on rainfall too. The historical record shows El Niño cutting Philippine rice yields by five to ten percent, and as much as twelve in severe episodes. In April, national authorities warned that fifteen provinces were at risk of severe El Niño-induced drought. The agriculture department expects this event to cost about 700,000 tonnes of “palay”, which is unmilled rice or rice grains still in their husks before processing, putting the projected harvest at between 18.6 and 18.8 million tonnes; some analysts see it falling to 17 or 18 million.⁹

The Philippines is covering its own rice gap by importing, at a pace that has already passed the whole of last year’s total imports, most of it from Vietnam. The United States Department of Agriculture projects 5.1 million tonnes of imports this year against the Philippine department’s own estimate of only 3.6 to 3.8 million. The two numbers rest on different harvests. USDA’s follows from what it forecasts the Philippines will grow. The Philippines Department’s estimate follows from what the Philippines is planning to grow, a record 20.3 million tonnes of palay. But the same Department has already conceded the smaller harvest is likely as set forth above.The import plan therefore assumes a harvest the Philippines own Department has stopped predicting, and the gap is roughly 1.3 million tonnes of additional rice imports that nobody in the Philippines has yet ordered.¹⁰

Behind the import estimates sits the number that should worry anyone reading this. The National Food Authority of the Philippines holds buffer stocks sufficient for only somewhere between seven and fifteen days of national consumption, compared to the thirty-day strategic reserve that is widely treated as the minimum a country should maintain. The Rice Tariffication Law restricts the authority’s capacity to expand those stocks. So, the Philippines enters the driest months of a very strong El Niño with only about two weeks of rice, buying from a Delta that expects salt in its irrigation canals by January, on a market where Thailand, its second supplier, is short, and where India, the largest exporter of all, has banned exports before. That is the whole margin.¹¹

Figure 3. The Philippines enters the driest months of a very strong El Niño holding buffer stocks of roughly one to two weeks against a thirty-day benchmark, importing on a market where the two largest exporters are short and the largest of all has banned exports before.

The Maritime Continent, Where You Can See the El Niño Switch Thrown

There is one region 4Hunger.org has not visited before, and it is the one where the El Niño mechanism actually lives.

In a normal year, the rising branch of the “Walker circulation”, the east–west atmospheric loop that sits over the tropical Pacific, sits over the islands between the Indian and Pacific Oceans: Indonesia, Malaysia, the Philippines, Papua New Guinea. El Niño shifts that rising branch east into the open Pacific, and the Maritime Continent dries. Every other regional impact in these four articles, from Sudan’s failed sorghum to the reservoirs filling in California, is downstream of that single displacement. It is the switch, and it is being thrown now.

Indonesia is already burning. Burned area reached 32,637 hectares by February, twenty times the same period in 2025, and that was before the dry season had properly begun. Satellite thermal detections over mapped peatlands, a proxy for fire activity rather than a count of fires, have totaled 23,546 since January, rising from 5,114 in February to 12,942 in March. Many are inside oil palm and timber concessions.

A positive Indian Ocean Dipole is expected later this year. The dipole is the Indian Ocean’s own version of the same seesaw as El Niño, and its cool pole sits directly off Indonesia, suppressing the convection that brings rain. The Maritime Continent is therefore dried twice over: once by the Walker circulation shifting east, and again by an ocean cooling beneath it. That combination is the signature of the worst fire years on record.¹²

Peat fires are not ordinary fires. They burn below the surface of the ground; they are close to inextinguishable by rain alone. They produce the transboundary haze that in 1997 and again in 2015 turned into a regional health emergency across Indonesia, Malaysia and Singapore. The 2015-16 event cost an estimated $5 billion in regional losses. In 1997 and 1998, harvests fell by up to 40 percent, with coffee down around 40 percent and palm oil and tea around 30. Roughly 500 people died in Irian Jaya of food shortage and disease.¹³

Indonesia is Also the 4^th Most Populous Country on Earth and Will Need to Import Rice

The fires and the palm oil are how Indonesia usually enters a story like this one, as a source of smoke and a line in a commodity index. It is also the fourth most populous country on earth, and 280 million people there have to eat.

The United Nations’ Asia-Pacific assessment of August 28 named Indonesia as one of the clearest large-area drought concerns in the region. Forecasts show widespread probabilities of below-normal rainfall through September to November, which is precisely the window in which the dry season is meant to end. The main rainy and planting season is meant to begin then too. If El Niño prolongs the dryness or delays the onset of the rains, what suffers is water availability, land preparation for rice crops, and the planting of the country’s main rice crop. Localized crop failures have already been reported across Java, West and East Nusa Tenggara and parts of Sumatra, including damage to rice.¹⁴

The same assessment contains the sentence this entire series exists to underline. Vegetation observations do not yet show widespread severe deterioration, which means there is still valuable lead time to monitor soil moisture, reservoirs, irrigation and crop conditions, and to prepare before impacts intensify. That is a United Nations agency saying, in the flat prose of a situation report, that the damage has not happened yet and the interval to prevent widespread damage is open right now.

Indonesia is better placed than most to use this warning and this time. The government projected rice output of 19.3 million tonnes in the first half of this year, holds reserves of around 5.3 million tonnes, and has allocated roughly $3.8 billion to food security measures. That is a state with the fiscal room to act on a forecast. Whether it does, and whether the reserve goes to price stabilization in Java or to the eastern islands where the crop failures are already being reported, is a policy question to be made in Jakarta, and not a meteorological one.

The Palm Oil Lag

Indonesia and Malaysia together produce about 85 percent of the world’s palm oil, which makes this a commodity story as well as a fire one. The Indonesian producers’ association projects that an extreme El Niño could cut national crude palm oil output by as much as five million tonnes, close to 10 percent. But the timing is the point. Prices rise before output falls. Traders are buying on the forecast that the palm oil crop will be short of what is normal, so palm oil already costs more, while the trees themselves are producing normally. The drop in output comes later, starting in the first half of 2027.

That is the same lag this series has described in Sudanese grain stores, in soil phosphate, in the Nile behind Aswan and in the Sudd in South Sudan. Palm is the swing supply in the global vegetable oil market, so a palm shortfall increases demand for soybean oil, which is the mechanism by which a dry season on Sumatra becomes a rise in cooking oil prices in Lagos, Nigeria and Karachi, Pakistan. Malaysia’s own rice production, which is a smaller matter globally but not to Malaysians, is expected to also fall significantly under extreme El Niño conditions. This is part of why Kuala Lumpur has been buying rice, too.

Papua New Guinea sits in the same region and deserves separate mention, because it faces a failure mode nothing else in these four articles has. Its highlands are overwhelmingly home to subsistence farmers, with almost no market buffer, and a strong El Niño brings not only drought but highland frost, which kills the sweet potato gardens the people in the highlands live on. In 2015 and 2016, the government estimated 2.4 million people were affected by these frosts, 1.3 million of them severely, with up to 400,000 in acute food shortage. The World Bank describes Papua New Guinea and Solomon Islands as facing a double catastrophe this year, of Middle East price shocks and El Niño drought together.¹⁶

The Countries That Are Not In the Rice Trade

Myanmar has appeared once in this article, as a supplier of 286,052 tonnes of rice to the Philippines. That is the wrong fact about Myanmar. Approximately 12.4 million people there were facing crisis-level food insecurity in May, a figure expected to climb to 20 million or more by mid-year, due to rising food prices and conflict even before El Niño arrives. Myanmar sits inside the FAO’s mapping of agricultural drought risk across South and Southeast Asia, alongside Pakistan, India, Thailand, Cambodia, Vietnam, the Philippines, Indonesia and Timor-Leste. It has the largest hunger caseload of any country in this article and the least capacity to respond to what is coming.¹⁷

Pakistan’s “kharif” rice has come in mixed due to delayed monsoon rains, and reduced wheat acreage has created localized pressure. It has also adopted its first national anticipatory action strategy, built on the 2025 Khairpur floods, where acting on a forecast worked. Bangladesh brought in a good “boro” harvest, but boro is the irrigated dry-season crop, and, therefore, the one insulated from a weak monsoon; the rain-fed “aman” crop, harvested in November and December, has yet to be counted.

Cambodia, Laos and Timor-Leste appear in FAO’s drought-risk mapping and almost nowhere else. There is no published El Niño crop assessment or food security analysis for any of them. This is worth noticing rather than passing over: a country with no crop assessment is not a country with no problem, it is usually a country without a monitoring system. Cambodia and Laos are also two of the four most hydropower-dependent economies in the region, which puts them among the most exposed to loss of power due to falling reservoirs and among the least likely to produce data showing it.¹⁸

El Niño does not reduce harvests everywhere. Across Afghanistan, where 14 million people face severe food insecurity after four consecutive drought years, the El Niño is expected to drive above-average precipitation countrywide from October 2026 into January 2027. That should enable above-average and timely winter wheat planting, and improve snowpack, contingent on temperatures staying average or below. For the first time in five years, the forecast for Afghan farmers is good news, and it is El Niño delivering it.¹⁹

“El Niño is not a random disaster that happens to the world. It is a redistribution of rains that causes drought in some countries and increased rains and floods in others in a pattern that has recurred over 150 years.”

There is also a flood half to this story. The same event that dries the Maritime Continent is expected to bring above-average rain to parts of South and Central Asia, and the International Rescue Committee names Bangladesh, Pakistan and Afghanistan among those countries most at risk of El Niño flooding. Vietnam expects fewer storms in the East Sea but stronger ones. Fewer and worse is harder to plan for than simply more.

The Reservoir Has Three Claimants

Every reservoir in this article appears twice: once as irrigation, and once as something else. That is not a rhetorical device. It is an accounting problem that several governments are about to resolve in public, and the way they resolve it will decide things that have nothing to do with farming.

Hydropower supplies roughly one sixth of Southeast Asia’s electricity, and the systems that lean on it hardest are Cambodia, Laos, Myanmar and Vietnam. Three of these countries have already been discussed in this article as having the least capacity to absorb a reduction in rainfall. During the moderate El Niño of 2023, regional hydropower output fell by 21 terawatt hours against the previous year, and it did so despite 500 megawatts of new capacity coming online. Adding generators does not help when there is no water to turn them.

This year the decline is already measurable. Hydropower generation across key Asian markets fell by about 13 average gigawatts year on year in June, with India and Vietnam accounting for more than eighty percent of the drop. Power systems across the region are burning more coal and liquefied natural gas to hold their grids together. The Philippines precedent from the last event is starker still: available hydroelectric capacity fell to 725.5 megawatts against a dependable 3,472, a decline of eighty percent.²⁰

The timing is what matters here. Analysts expect the acute impacts in early 2027, when depleted reservoirs coincide with the dry-season peak in demand for cooling. That is the same window as the salinity in the Mekong Delta, the same window as India’s rabi harvest, and the year China’s own meteorological agency has named as the one El Niño will hit it hardest. Whatever else 2027 turns out to be, it is the year several kinds of reserve run out at once.

The obvious consequence is emissions: a dash for coal and gas that turns a temporary shortfall into long-lived fossil infrastructure. The less obvious one is why this series cares. Fuel also is an agricultural input. The same governments buying emergency LNG to run air conditioners are the ones whose fertilizer already costs more than double what it did in January because the Strait of Hormuz closed. Bangladesh and Pakistan, both acutely price-sensitive LNG importers, have already curtailed electricity consumption through fuel conservation measures. A government that spends its foreign exchange keeping the lights on in February has less of it to buy urea in March, and less again to buy rice in May.

So when Thailand and Indonesia decide what their reservoirs are for, they are not choosing between farmers and cities. They are choosing between this year’s electricity and next year’s harvest, out of the same water.

Figure 4. The reservoir empties, the grid buys replacement fuel, and the cost arrives twice. The first bill is counted: coal and gas displace hydropower, and a temporary shortfall hardens into long-lived fossil infrastructure. The second is not. Foreign exchange spent on emergency fuel in February is not available for urea in March, and the smaller planting that follows is recorded, if at all, as a bad harvest rather than as a consequence of the drought.

China Is Forecasting Its Own 2027

China grows more rice than any other country, imports enough to move the world price when it chooses to and has published one of the most specific national El Niño forecasts anywhere. China is also the only government in these four articles to predict that 2027 will be its own worst year in writing.

Start with what Beijing thinks the ocean is doing, because the Chinese agencies did their own analysis with their own instruments. The National Climate Center reports the sea surface temperature index in the El Niño monitoring region at 2.09 degrees in July and averaging 2.59 degrees in August, both the highest readings for those months since records began in 1951. Chen Lijuan, the center’s chief forecaster, says the event is developing unusually quickly and expects it to peak around November or December, potentially as the strongest El Niño on record. In July, the Ministry of Natural Resources’ National Marine Environment Forecasting Center had already warned that warming in the central and eastern equatorial Pacific was running ahead of the 1997-98 event at the equivalent stage. Three Chinese agencies, monitoring separately, made the same prediction as the WMO and NOAA.²¹

What they forecast next is the opposite of almost everything else in this article. Asia’s El Niño story so far has been a drought story. China’s is a flood story, and it arrives in phases, because ocean heat takes months to work its way through atmospheric circulation. Through this autumn, rainfall is expected above average along and south of the lower Yangtze and across South China, with parts of the southwest at risk of extreme localized rain, while temperatures in the southwest and south run significantly above normal and heat waves are likely. Through the winter, as the event matures, most of the country should be warmer than usual, though forecasters warn that temperatures could still swing sharply as cold air masses push south. Taken on its own, none of that is a catastrophe.

Then comes the sentence that should have been in this article from the beginning. China’s climate in the summer of 2027, the National Climate Center says, will be more severely affected by El Niño than in 2026. The lag is the whole point: the western Pacific subtropical high is expected to strengthen and shift south, dragging moisture into central and eastern China months after the ocean has begun to cool. The middle and lower reaches of the Yangtze are forecast for particularly heavy precipitation and a heightened risk of widespread, severe flooding. Major grain-producing areas in the Northeast and North China are told to prepare for above-average rainfall and floods that could damage crop production. Northwest China and Xinjiang go the other way, into drought driven by unusually high temperatures and limited rain. Authorities have been warned to watch for crop diseases and pests encouraged by warm, humid conditions.

Figure 5. China’s El Niño arrives in phases and does its worst damage after the ocean has peaked. Autumn 2026: above-average rain along and south of the lower Yangtze and across South China, with heat waves in the south and southwest. Winter 2026-27: warmer than normal, with sharp swings as cold air pushes south. Summer 2027: the lag effect at full strength, heavy rain and flood risk on the middle and lower Yangtze and across the northeastern grain belt, drought in the northwest and Xinjiang.

The precedent the Chinese agencies keep returning to is not 2015. It is 1998, the year after the last comparable event, when abnormal rainfall produced floods across multiple river basins, the worst of them on the Yangtze. More than 4,000 people were killed and 200 million affected. The National Marine Environment Forecasting Center expects abnormally high precipitation in the Yangtze basin again in the summer of 2027, with severe pressure on flood control.²²

Every other government in these articles is being asked to act on a forecast of this season. China has published one for eighteen months out, named the basins and the crops where El Niño impacts will be severe, and told its provinces to prepare. Nobody in Beijing will be able to say afterwards that the warning came too late.

What the El Niño Predictions for Asia Add Up To

This article has gone country by country, which understates the total. Here is the total as the institutions have published it.

FAO’s analysis of 41 years of satellite imagery finds that a comparable El Niño produced losses of around 15 million tonnes of rice across Southeast Asia. The World Bank has warned that if the event develops fully, rice yields could fall by a fifth to a half across the hardest-hit parts of South Asia, where rice underpins food security for hundreds of millions of people. Those are the numbers that belong at the top of any Asian El Niño story, and they are conspicuously absent from most of them.²³

One omission belongs on a watch list rather than in these totals, and it is disease. El Niño reshapes the epidemiology of dengue, malaria and heat illness, and it does the same for crops and livestock: drought-stressed maize carries aflatoxin, so a harvest can exist and still not be safe to eat. It deserves its own article, and 4Hunger.org intends to give it one.

There have been near-record rice harvests in Cambodia, China, Indonesia, Thailand and Vietnam which have built a genuine cushion, and Chinese food prices have been falling on adequate supply. But those are the harvests already in, gathered before this El Niño peaked, and the same assessment warned that surging input costs threaten the next planting cycle. The cushion is real, and it is a cushion against one bad season rather than two.

The Window That Is Still Open

In June, the FAO and the World Food Programme launched a $202 million appeal to protect 8.8 million people across 22 high-risk countries, many of them in this region, before the harvests failed rather than after. As part two of this series recorded, neither agency will say how much of it has been raised, and the only published delivery figure covers six countries and half a million people.

The ocean is running ahead of the forecasts. The Multivariate ENSO Index, which folds sea surface temperature, air pressure, winds and cloud into a single figure and so measures whether the atmosphere has actually coupled to the ocean, stood at 2.41 for June and July. That is the highest June to July value since records began in 1979. Fifteen models now put the Niño 3.4 anomaly at or above 3 degrees for October to December, beyond the top of the scale forecasters normally use, and the mid-August assessment found little sign of any transition away from El Niño before the spring of 2027.²⁴

The forecast has hardened even more. On September 3, the World Meteorological Organization said El Niño was firmly established in the central and eastern equatorial Pacific and would strengthen to very strong before peaking at the end of the year, with a probability approaching certainty that it persists through December to February. It also forecasts a positive Indian Ocean Dipole, the dipole mode index averaging 0.9 degrees over September to November. That is the combination that dries the Maritime Continent hardest, and it is now the expectation rather than only a risk. The next report is the ASEAN Meteorological Centre’s October to December outlook, due in early October.25

What Is Fixed and What Is Not

The El Niño cannot be stopped. The heat is already in the ocean. NOAA measured subsurface anomalies of ten degrees at depth in July, and nothing any government does between now and February will change the temperature of the Pacific. Anyone suggesting otherwise is selling something.

The rain, though, is genuinely uncertain, and this article contains its own proof. India’s monsoon looked like a catastrophe at the end of June and something close to a near-miss by the end of July, and the recovery was large enough to pull sowing back to within two percent of last year. That recovery was not simply luck. The same Indian Ocean Dipole that dries Indonesia warms the Arabian Sea and tends to strengthen the Indian monsoon, so the two teleconnections are statistical linkages by which an ocean anomaly in one basin shifts rainfall in another. The ocean warming in one place changes the weather in another. But these connections only make things likelier, not certain, and so it could make things worse, but sometimes they break the good way, too. They could pull on India in opposite directions, and the season turns on which of them prevails. Afghanistan is about to find that out in the opposite direction from everyone else. Planning on that is not a strategy. Assuming the outcome is already fixed is just as wrong.

Indonesian peat is the clearest case. Once it catches fire, it burns below the surface and rain barely touches it. Canal blocking and rewetting work only before ignition, and the worst fire months are ahead. That is a thing which is cheap in prevent in September and impossible to put out in December.

Photo Credit: Niken D. Sitoningrum, Mongabay

Vietnam’s winter-spring planting decisions in the Mekong Delta are being taken now, against a salinity forecast the government has already published, and the 2023-24 season proved what taking them early is worth. Philippine dry-season planting decisions have not all been taken, and the drought-tolerant seed the agriculture department is distributing now sets the yield in March. Indonesia has reserves of 5.3 million tonnes and a choice about where they go. India’s rabi wheat goes in the ground from October on reservoir water that is being allocated this month. Beijing will decide what its reserves will be used for before the Yangtze rises in 2027, but not after. Thailand and Indonesia are about to decide whether their reservoirs serve irrigation, hydropower or drinking water; Panama made that call a year early, which is why it still has a working canal.

And the largest lever of all costs nothing. If India does not ban rice exports, that is worth more to global food security than every anticipatory action budget in this article combined. It is not a question of money or capacity. It is a political choice, and it is still unmade.

So, the honest position is this. El Niño is fixed. The rainfall is uncertain. The response is open, and the record is not encouraging. Southern Africa was warned of a bad harvest a year ahead in 2015 and again in 2023, but the money still arrived after the harvest failed both times. Preventable is a claim that works in something like physics. It has never yet been a reliable predictor of human behavior.

Figure 6. Three things get called preventable and only one of them is. The event is settled physics; the rainfall is genuinely uncertain; the response is open, and it is the response that decides how many people go hungry.

The rains that failed in Sudan failed in July. The primera in Guatemala is gone. Asia is the one place in these four articles where the political decisions that determine the outcome of global hunger are still, for a few more weeks, in front of the people who have to make them.

“This is the part of the map where the warning has arrived before the damage. It will not stay that way for long, and the record of the last five events is that the money needed to prevent hunger arrives after the harvest fails, every time. Let’s hope that the 2026/2027 El Niño is different.”

Endnotes

1.1. India Meteorological Department: cumulative rainfall to August 24, 2026 at 564 mm, 13 percent below the long period average; June at 60 percent of LPA and July at 103 percent; a 14 percent deficit across the first three months of the season. IMD Director General Mrutyunjay Mohapatra, September 2026: monsoon precipitation expected at less than 91 percent of the September long-term average of 167.9 mm. IMD’s second long-range forecast, issued in May 2026, put the full season at 90 percent of LPA plus or minus 4 percent, and ICRA had projected a figure near 90 percent if August and September came in at 93 percent; with June to August finishing 14 percent below normal, that scenario no longer holds. The seasonal figure in the text is computed from IMD’s 1971-2020 long-period average of 868.6 mm, of which June to August account for 700.7 mm (about 81 percent) and September 167.9 mm (about 19 percent): 86 percent of 700.7 mm plus 91 percent of 167.9 mm gives 755 mm, or 87 percent of LPA. IMD classifies a season below 90 percent of LPA as deficient, 90 to 95 percent as below normal.

2. ICRA mid-monsoon update, August 2026: kharif sowing contraction narrowing from about 23 percent year-on-year at June 25 to about 3 percent at July 31. ICICI Bank report: kharif sowing at 105.7 million hectares as of August 17, 1.5 percent below the 107.3 million a year earlier; live storage across 150 reservoirs at 117.3 billion cubic meters as of August 20, 64.3 percent of capacity, against 44.39 percent across 166 reservoirs on August 1 per USDA FAS.

3. ICICI Bank: rice acreage down 3.2 percent year-on-year at 40.5 million hectares as of August 17. USDA Foreign Agricultural Service, via Bloomberg, September 1, 2026: weak rainfall may cut Indian rice production by about 4.5 percent and maize by 9.3 percent. Agriwatch crop condition assessment, August 21, 2026: of 446 districts monitored, about 32 percent under watch, 15 percent for dry risk and 5 percent showing very poor crop condition. Roughly 60 percent of Indian cropped area is without assured irrigation.

4. Indian fertilizer importers reported paying more than $950 per tonne for urea against $400 to $450 before the closure of the Strait of Hormuz in February 2026. Domestic sugar prices rising from 48 to 55 rupees between July 20 and August 20, 2026, with the government permitting duty-free import of raw sugar. See Hunger Foretold (this publication, 2026) on the phosphate and fertilizer squeeze.

5. Reported precautionary stockpiling by Malaysia and the Philippines ahead of the dry season, with Thai and Vietnamese benchmark grades rising above $400 per tonne. Foreign Policy, August 5, 2026: Bangkok export price at $490 per tonne at the end of June, up 27.5 percent since the start of 2026. On the 2007-08 precedent, world rice production did not fall materially; prices roughly tripled on export restrictions and precautionary buying.

6. Thailand received roughly 60 percent less rain than normal between January and April 2026; Thailand’s Office of the National Water Resources warned of a drought crisis potentially more serious than 2015. Office of Agricultural Economics: 2025/26 Thai rice output at 35.44 million tonnes of paddy, comprising 26.99 million tonnes main-crop (down 0.32 percent) and 8.45 million tonnes off-season (down 1.63 percent).

7. Viet Nam Department of Hydrometeorology, Ministry of Agriculture and Environment, reported July 2026: El Niño persisting with probability above 95 percent, moderate to strong intensity, 60 to 65 percent chance of very strong by late 2026 or early 2027. Mekong Delta forecast for early 2027: the 4 g/l salinity boundary extending 90 to 125 km along the Vam Co, 85 to 90 km along the Cua Tieu and 70 to 80 km along the Ham Luong during high-tide periods, directly affecting more than 350,000 hectares of winter-spring rice as well as fruit-growing areas, aquaculture and domestic water supplies. Southern Regional Hydro-Meteorological Centre: southern rainy season expected to end in late October or early November, several weeks early, with seawater potentially reaching 80 to 100 km inland along Mekong Delta rivers and canals. Department of Crop Production and Plant Protection directive of June 2026 instructing Mekong Delta and southeastern provinces to plan the 2026-27 winter-spring rice crop around water resources, tidal surge and salinity risk. Two planning scenarios, high-risk and very high-risk, issued to provincial authorities.

8. Vietnamese national assessment comparing the 2015-16 and 2023-24 events: reduced Mekong flows and severe saltwater intrusion in 2015-16 caused widespread disruption to domestic water supply, agriculture and aquaculture; salinity intrusion in 2023-24 remained severe but losses were considerably lower, attributed to improved forecasting and earlier response measures. Officials note that the scale of impact depends on El Niño intensity, rainfall distribution, reservoir operations and transboundary Mekong flows as well as on the effectiveness of response.

9. PAGASA historical assessments: El Niño events reduce Philippine rice yields by 5 to 10 percent, rising to 12 percent in severe episodes. World Bank Food and Nutrition Security Update, May 2026: Philippine authorities warned in April that 15 provinces were at risk of severe El Niño-induced drought. Briones and Tolin (2024) on more than half of Philippine rice land being rainfed. Philippine Department of Agriculture, September 1, 2026: El Niño expected to cut palay production by about 700,000 tonnes, with the harvest between 18.6 and 18.8 million tonnes; analyst projections of 17 to 18 million tonnes reported by BusinessMirror.

10. Philippine rice imports for 2026 to date already exceeding the full 2025 total, with Vietnam the leading supplier at 2.59 million tonnes, followed by Thailand at 378,404 tonnes and Myanmar at 286,052 tonnes. USDA projects Philippine rice imports at 5.1 million tonnes against the Department of Agriculture’s estimate of 3.6 to 3.8 million tonnes, which assumes record palay output of 20.3 million tonnes. On the internal inconsistency: the same department separately expects El Niño to reduce palay production by about 700,000 tonnes. A caution: the 20.3 million figure reaches this article through analyst reporting rather than a primary departmental release, and should be checked against the Department of Agriculture and the Philippine Statistics Authority directly before being relied upon.

11. National Food Authority buffer stocks equivalent to seven to fifteen days of national consumption against a thirty-day strategic reserve benchmark; the Rice Tariffication Law restricts the authority’s capacity to expand buffer stocks.

12. Pantau Gambut and Indonesian government data via NASA: burned area of 32,637 hectares by February 2026, approximately twenty times the same period in 2025; 23,546 peatland hotspots recorded since January, rising from 5,114 in February to 12,942 in March, many within oil palm and timber concessions. BMKG confirmed an El Niño index of 1.59 on August 4, 2026 and identified East and West Nusa Tenggara, Java, Bali and South Papua as areas at risk. Forecasts indicate the most anomalous dry conditions across Southeast Asia running from late September 2026 through February 2027.

13. On the 2015-16 event: an estimated $5 billion in regional economic losses from drought and fires. On 1997-98: harvests reduced by up to 40 percent, with coffee down about 40 percent and tea and palm oil about 30 percent below normal, and roughly 500 deaths recorded in Irian Jaya from food shortage and disease; months of transboundary haze led to the ASEAN Zero Burning Policy of 1999 and the 2003 Agreement on Transboundary Haze Pollution.

14. United Nations Asia and the Pacific snapshot of El Niño and the Middle East crisis, as of August 28, 2026: Indonesia identified as one of the clearest large-area drought concerns, with widespread probabilities of below-normal rainfall through September to November overlapping the end of the dry season and the transition into the main rainy and planting season, with potential effects on water availability, land preparation and planting of key crops, particularly rice; vegetation observations not yet showing widespread severe deterioration, leaving lead time to monitor and prepare; localized crop failures reported across Java, West and East Nusa Tenggara and parts of Sumatra, including damage to rice production. World Bank Food and Nutrition Security Update, June 2026: Indonesia projecting rice output of 19.3 million tonnes from January to June, with 5.3 million tonnes in reserve, and approximately $3.8 billion allocated to food security measures.

15. Indonesia and Malaysia together account for approximately 85 percent of global palm oil production. Gapki, the Indonesian Palm Oil Association, projects that an extreme El Niño could reduce Indonesian crude palm oil output by up to five million tonnes, with analysts expecting prices to firm ahead of production declines and the yield impact arriving from the first half of 2027. Malaysian production fell approximately 20 percent during the 2016 event. World Bank, May 2026: extreme El Niño conditions may significantly reduce Malaysia’s rice production.

16. Papua New Guinea government estimates for the 2015-16 event: 2.4 million people affected, 1.3 million severely, with up to 400,000 in acute food shortage; highland frost accompanying drought damaged sweet potato gardens. World Bank, May 2026, describing Papua New Guinea and Solomon Islands as facing a double catastrophe of Middle East price shocks and El Niño drought, compounded in PNG by Cyclone Maila.

17. World Bank Food and Nutrition Security Update, May 29, 2026: approximately 12.4 million people in Myanmar facing crisis-level food insecurity (IPC Phase 3 or above) in May, expected to climb to 20 million or more by mid-year, driven by rising food prices. FAO Agricultural Stress Index System analysis, drawing on 41 years of satellite imagery, maps agricultural drought risk across South and Southeast Asia from Pakistan and India through Myanmar, Thailand, Cambodia and Viet Nam to the Philippines, Indonesia and Timor-Leste.

18. European Commission Joint Research Centre, September 3, 2026: Pakistan facing mixed kharif rice conditions due to monsoon delays; Bangladesh and Sri Lanka with good prospects from abundant rainfall; Afghanistan expecting positive spring and summer crop outputs despite 14 million people facing severe food insecurity. World Bank: Pakistani cereal production above average, with dry conditions and reduced wheat acreage creating localized pressure; for Bangladesh, the concern is not nationwide shortage but the convergence of heatwaves, reduced rainfall, salinity intrusion and localized floods with higher production costs. WFP, September 11, 2026, on Pakistan’s adoption of its first national anticipatory action strategy following the 2025 Khairpur floods.

19. FEWS NET Afghanistan Food Security Outlook Update, August 2026, based on information available as of August 31: the ongoing El Niño event expected to drive above-average precipitation countrywide between October 2026 and January 2027, enabling above-average and timely winter wheat planting following four consecutive years of drought between 2021 and 2024, and improving snowpack and snow water volumes, contingent on temperatures remaining average to below average. Second season rice and maize planting complete across eastern and northeastern producing areas at near-average levels.

20. Hydropower contributes roughly one sixth of Southeast Asia’s electricity output, with Cambodia, Laos, Myanmar and Viet Nam operating the most hydropower-reliant power systems; regional hydropower generation fell by 21 terawatt hours in 2023 against 2022 despite 500 MW of new capacity. S&P Global Commodity Insights, reported July 2026: hydropower generation across key Asian markets down about 13 average gigawatts year on year in June 2026, with India and Viet Nam accounting for more than 80 percent of the decline, forcing greater reliance on coal and LNG; analysts expect the acute impact in early 2027 as depleted reservoirs coincide with dry-season cooling demand. International Energy Agency Electricity Mid-Year Update 2026: El ño typically reduces hydropower and wind generation in Southeast Asia and Latin America, increasing reliance on coal and gas; Bangladesh and Pakistan, price-sensitive LNG importers, have curtailed electricity consumption through fuel conservation measures. Philippine Department of Energy on the 2024 event: available hydroelectric capacity of 725.5 MW against 3,472 MW dependable capacity, an 80 percent decline.

21. China National Climate Center, September 2026: sea surface temperature index in the El Niño monitoring region at 2.09°C in July 2026 and averaging 2.59°C in August, both the highest for the corresponding periods since records began in 1951; chief forecaster Chen Lijuan on the event developing unusually quickly and peaking around November or December, potentially the strongest on record. Ministry of Natural Resources, National Marine Environment Forecasting Center, July 2026: warming in the central and eastern equatorial Pacific already exceeding that observed at the same stage of the 1997-98 event, with a strong or super El Niño expected by autumn and winter. Phased outlook via the National Climate Center and China Meteorological Administration: above-average autumn rainfall along and south of the lower Yangtze and in South China, with particularly high precipitation in parts of Southwest China and significantly above-average temperatures in the southwest and south; a warmer-than-average winter nationally with sharp fluctuations as cold air pushes south. Chinese state media are the primary route for these figures; they are quoted as published.

22. China National Climate Center on the lag effect: China’s climate in summer 2027 expected to be more significantly affected by El Niño than in 2026, with the western Pacific subtropical high strengthening and shifting south; above-average rainfall across much of central and eastern China, particularly heavy precipitation in the middle and lower Yangtze, elevated risk of severe flooding in multiple basins, and major grain-producing areas in Northeast and North China advised to prepare for above-average rainfall and flooding affecting crop production, while Northwest China and Xinjiang face elevated drought risk from high temperatures and limited rainfall; warning on crop diseases and pests encouraged by warm, humid conditions. On 1998: abnormal rainfall and floods across multiple Chinese river basins, most severely the Yangtze, with more than 4,000 killed and 200 million people affected, per Xinhua. National Marine Environment Forecasting Center expects abnormally high precipitation in the Yangtze basin in summer 2027.

23. FAO Agricultural Stress Index System analysis: a comparable El Niño produced losses of around 15 million tonnes of rice across Southeast Asia. World Bank warning that rice yields could fall by a fifth to a half across the hardest-hit parts of South Asia and East Africa. On the counterweight: World Bank Food and Nutrition Security Update, May 2026, recording near-record rice harvests across Cambodia, China, Indonesia, Thailand and Viet Nam as a potential supply cushion while warning that surging input costs threaten the next planting cycle; Chinese food prices down 1.6 percent year on year in April 2026 on adequate supply. International Rescue Committee, July 2026, naming Bangladesh, Pakistan and Afghanistan among countries most at risk from El Niño flooding. On Central Asia: European Commission Joint Research Centre, September 3, 2026, recording finalized winter cereal harvests with excellent yields. On health: World Health Organization guidance on elevated risk of vector-borne disease and heat-related illness in South Asia during El Niño events.

24. NOAA PSL: MEI.v2 at +2.41 for June to July 2026, the highest June to July value in a record beginning in 1979. IRI mid-August objective outlook: weekly Niño 3.4 at +2.7°C centered on August 12, July monthly value +2.03°C, and 15 models forecasting October to December Niño 3.4 anomalies at or above +3.0°C, beyond the highest defined intensity category. NOAA July anomalies: Niño 3.4 +1.4°C on the relative index, Niño 3 +1.7°C, Niño 1+2 +2.9°C, with subsurface anomalies reaching +10°C at depth. On the February 2026 change of official index, see part five. The July figure above is on the relative index while the IRI values are conventional Niño 3.4; the two are not directly comparable and are quoted here as published.

25. World Meteorological Organization, September 3, 2026: El Niño firmly established and forecast to strengthen to very strong before peaking toward the end of 2026, with Global Producing Centres indicating a near-100 percent probability of persistence through December 2026 to February 2027, and a positive Indian Ocean Dipole with the Dipole Mode Index averaging +0.9°C during September to November. NOAA Climate Prediction Center maintained its El Niño Advisory on August 13, 2026, with a greater than 90 percent probability of a very strong event and a 69 percent chance that October to December exceeds every event on record since 1950. FAO and WFP Joint Anticipatory Action Appeal, launched June 18, 2026, seeking $202 million to protect 8.8 million people across 22 high-risk countries; see part two of this series on its funding status. The ASEAN Meteorological Centre October to December outlook and associated ASEANCOF bulletin are expected in early October 2026.

Photo Credit: WFP/ Stephanie Savariaud

Where the Name “El Niño” Came From

In the nineteenth century the fishing crews working out of the northern Peruvian ports noticed that every few years, around Christmas, the cold water they depended on gave way to something warm pushing down from the north. The anchoveta went with it. The nets came up empty and stayed empty for a season. They called the current El Niño de Navidad, the Christ Child, for the time of year it arrived. The warm water also pushes the surviving shoals deep and south, so what the fleet does find is undersized. That is why Peru suspended the season in May: too much of the catch was juveniles.

They were not naming a global phenomenon. They were naming a thing they had to plan around, in the way Central American farmers named the canícula after the dog days of Sirius. Working words, given by people who needed to know when the weather would turn against them.

The science came from a famine.

In 1877 the monsoon failed over India and millions died. The government asked its new meteorological department to forecast the rains. In 1904 the department went to Gilbert Walker, a mathematician appointed because predicting a monsoon was a statistical problem nobody had solved. Walker spent twenty years correlating air pressure from stations around the world, looking for anything that would tell him a year ahead whether India would eat. He found a seesaw in pressure across the Pacific and called it the Southern Oscillation.¹

It took until the 1960s for Jacob Bjerknes to show that Walker’s atmospheric seesaw and the fishermen’s warm current were one thing: a single coupled system of ocean and air, and the largest source of year-to-year climate variation on earth.

The forecasts this series keeps quoting exist because of a famine and were built by a man trying to see the next one coming. That is worth holding onto as this article crosses the Americas, from the Guatemalan smallholder who has lost the first of two harvests, to the reservoirs filling in California, to the Peruvian fishery the whole thing is named after, which is failing again.

A System Built Around a Second Chance

Central America’s Dry Corridor is a belt of dry tropical forest running down the Pacific side of the isthmus, from southern Mexico to Panama, in the rain shadow of the mountain chain along its spine. Moisture comes off the Caribbean, drops on the eastern slopes and arrives on the Pacific side already spent. The result is a strip of semi-arid land in an otherwise humid region, farmed almost entirely by smallholders growing maize and beans for themselves to eat. It is built around a second chance. The primera goes in with the first rains in May and June and is harvested in August and September; the postrera follows, sown from September and harvested in November and December. Between them sits the “canícula”, the midsummer dry spell that pauses the rains for a few weeks in July and August. A household that loses the primera can still eat off the postrera. That is the design.

It is also what makes 2026 different, because both halves of this system are failing at once.

The primera is already gone in the worst-affected areas. Between June 20th and July 20th, much of Guatemala’s Dry Corridor recorded only five to eight days with any rainfall at all, and the maize went into stress and began wilting. FEWS NET attributes the damage directly to El Niño conditions and an extended canícula, and places the poorest households of the Dry Corridor in Crisis through September, along with targeted areas of Alta Verapaz in the north and the Western Highlands along the Mexican border. Neither of those lies inside the corridor. Alta Verapaz is cloud-forest country, normally among the wetter parts of Guatemala; the Highlands are volcanic, above two thousand meters, and largely indigenous. That the damage has reached both means it is landing on households already food-insecure for reasons that have nothing to do with rainfall.

In Honduras, cumulative rainfall deficits of around forty-five millimeters were recorded across parts of Yoro, El Paraíso and Olancho in May and June, and roughly 8,800 hectares of maize and 3,100 hectares of red beans have been written off. El Salvador’s small and medium producers expect the first harvest of basic grains to come in around twenty percent down.²

The postrera, the second chance, is being sown now into the same dry conditions. FEWS NET expects widespread delays and reduced planted area across Honduras and El Salvador, because the primera harvest ran late, because soil moisture during the July and August planting window was inadequate, and because fertilizer prices have made the inputs unaffordable. There is another contributing factors, the 2026 hurricane season is forecast to be below average, which is itself an El Niño effect. In most years tropical moisture from tropical storms and even hurricanes is what refills the corridor between seasons. This year the thing that usually rescues the postrera has been suppressed by the same ocean that killed the primera. ³

Figure 1. The Dry Corridor plants twice a year so that one failure is survivable. In 2026 the primera was sown late and lost, and the postrera is going in delayed and on reduced area. The tropical moisture that normally refills the corridor between seasons has been suppressed by the same El Niño. Season windows are typical; exact dates vary by altitude and country.

“A two-season system exists so that one failure is survivable. Central America is losing both halves of it in the same year, to the same ocean.”

Meanwhile the lean season arrived early. In Guatemala’s Dry Corridor it began in February rather than the usual June, because households came into the year with no staple grain reserves after consecutive poor harvests. FEWS NET describes an early and prolonged lean season, with the poorest households already selling productive assets, the tools and animals that produce next year’s food, to buy this month’s food. Between two and two and a half million Guatemalans were projected to need food assistance between June and August. ⁴

The prices tell the rest. White maize in Honduras stood almost 31 percent above last year in June, and more than 35 percent above the five-year average, in a country whose poorest households were already buying rather than growing most of what they eat. Transport costs are running some 15 percent above last year. The Honduran government has placed close to eighty percent of the country under alert, and both El Salvador and Panama have declared formal El Niño alerts. ⁵

There is a precedent for what follows, and it is recent enough that some readers will remember the news coverage without ever having connected it to the weather. The 2015 and 2016 El Niño left about 3.5 million people across Guatemala, Honduras, El Salvador and Nicaragua needing humanitarian assistance. It also contributed to one of the largest northward migration waves in the region’s modern history, and the largest share of it came out of the Western Highlands, the same volcanic departments FEWS NET has just placed in Crisis. ⁶

That is worth stating plainly, because it is the point at which this article stops being about somewhere else. A family that loses the primera, then loses the postrera, then sells the animals, has run out of things to sell. What it does next is leave. The United States has spent a decade arguing about immigration and comparatively little time on taking protective actions before the El Nino that we have known is coming for months.

The Corridor Is Seven Countries, Not Three

It is worth being precise about how much country the failed rain covers, because the Dry Corridor is usually written about as though it meant three places. It runs from southern Mexico through Guatemala, El Salvador, Honduras and Nicaragua into Costa Rica, with dry zones reaching Panama. Nicaragua is the hardest of them to see, and that is itself the finding: it sits squarely in the corridor, it was one of the four countries behind the 3.5 million refugees figure in 2015 and 2016, and there is less independent monitoring in Nicaragua than in its neighbors. When the impacts of this El Niño are counted up, Nicaragua is the country most likely to be undercounted.

Panama, Where the Damage Becomes a Freight Rate

Panama is being hit in an entirely different way, and it is the one that will reach a North American checkout counter fastest. To see why, it helps to know that the Panama Canal does not run on sea water.

Unlike Suez, which is a ditch at sea level, Panama lifts ships about eighty-five feet over the Isthmus of Panama and lowers them down the far side. The lifting is done by filling and draining lock chambers from Gatún Lake, a rainwater-fed reservoir that also supplies drinking water to much of Panama’s population. Every ship that crosses the canal uses roughly fifty-two million gallons of that fresh water, and most of it runs out to the sea and is gone. The canal is the largest freshwater consumer in world transport. ⁷

When it does not rain, the lake falls, and the canal has two levers. It can reduce the draft of ships passing through the canal, or it can reduce the number of ships.

Draft is how deep a loaded ship sits in the water, and it is set by how much cargo is aboard. A vessel told it may draw no more than forty-eight feet instead of fifty is a vessel that has to leave cargo behind. Every half foot is containers left on a dock, or grain that did not sail.

Rainfall across the canal watershed has run 34 percent below the historical average from May to August, and the inflows that actually refill the lake are 44 percent below. Gatún stood at 84.1 feet on 21 August. The authority has cut the maximum draft in steps through the year: 49.5 feet from 3 July, 49 on 24 July, 48.5 on 15 August, 48 from 2 September and 47.5 from 1 October, against a design maximum of 50. ⁸

For most of the year the canal insisted the draft cuts would not affect the number of ships. That changed on 20 August. Under Advisory A-29-2026 daily transits fall from thirty-six to thirty-four for booking dates from 4 September, and to thirty-two from 15 September. The second lever has been pulled, and it is the one that did the damage last time. The market has already repriced around it: on 25 August a South Korean gas carrier paid $5.3 million at auction for a single priority slot, reported as the highest bid ever recorded, and slots have been averaging around $1.1 million.

In the 2023 and 2024 El Niño, the lake sat just above seventy-nine feet from June to October. Draft of ships passing through the canal was reduced to forty-four feet. Daily transits dropped from the usual thirty-six to as few as eighteen, and the canal handled 9,936 vessels that financial year against 12,638 the year before, a fall of twenty-nine percent. Ships waited weeks, paid at auction to jump the queue, went round South America or shifted to the Suez Canal. The canal carries about five percent of world shipping, and roughly seventy percent of what moves through it is coming to or going from the United States. A rainfall deficit in one Central American watershed surfaces later as a freight rate, and then as a grocery bill, by a route that has nothing to do with maize.⁹

The canal was one agency that read the reports and went into this El Niño prepared in a way it was not in 2023: water-saving measures at the locks were taken from late 2025, hydroelectric generation at Gatún suspended to prioritize storage, and a 2026 dry season that was among the wettest since 1950 used to build reserves. Full lakes at the start are not nothing. A repeat of 2024 is a risk, not a schedule.¹⁰

Figure 2. What happened the last time the canal’s watershed ran dry. Daily transits fell from around 36 to 18 between November 2023 and February 2024, with maximum draft cut from 50 feet to 44. Restrictions announced this June take effect from July. The canal entered this event with reserves built during an unusually wet dry season, so a repeat is a risk rather than a schedule.

The Caribbean and the Northern Rim

Beyond the isthmus the same failed rains forecast reaches further than the corridor. FAO puts the probability of below-normal rainfall across the wider Caribbean region at seventy percent, with the highest agricultural drought risk along the Dry Corridor, in Colombia and Venezuela, and across Cuba, the Dominican Republic and Haiti, where roughly six million people are already projected at Crisis or worse. In 2015 and 2016 Haitian harvests in some areas fell by as much as seventy percent and food insecurity doubled within months. Cuba enters this event having effectively run out of diesel dur to U.S. sanctions. The mirror image again holds: Bolivia, Ecuador and Peru are forecast above-average rains.¹¹

South America, Where El Niño Reaches into the Ocean

Off the coast of Peru, El Niño does something it does nowhere else. It stops the deep ocean waters from coming up.

In a normal year, cold water rich in nutrients rises along that coast from the deep, feeds the plankton, and supports the anchovy: the largest single-species fishery on earth. Warm surface water suppresses that upwelling. The plankton fail, and the anchovy disperse, go deep or move south. Peru’s monitoring commission activated a coastal El Niño alert in February and expects the warming to persist into late 2026. The first fishing season of the year opened with a quota of 1.9 million tonnes, the lowest in a decade and 36 percent below last year, and even that was suspended in May and again into June because too much of the catch was juveniles. Fishmeal prices are running more than 75 percent above a year ago. ¹²

Peru supplies roughly a fifth of the world’s fishmeal and fish oil, which is why this belongs in an article about hunger rather than one about fishing. Fishmeal is protein feed. It goes into salmon farms in Norway and Chile and shrimp ponds across Southeast Asia, and when it becomes scarce buyers move to soybean meal instead, which tightens that market too. The 1997 and 1998 El Niño cut Peruvian fishmeal production to around 815,000 tonnes. The 2023 season was cancelled outright, costing more than $1.4 billion in lost exports and around half a percent of Peru’s national income. ¹²

Inland, the pattern runs the way it does everywhere else in this article. Northern Brazil, the Amazon basin, Colombia and Venezuela can expect reduced rainfall, higher temperatures, more fire and lower river flow which damages the Amazon forest and its river. In countries that generate much of their electricity from water is a power problem as well as a farming one. ¹³

Southeastern South America usually does well out of an El Niño. Argentina, Uruguay, Paraguay and southern Brazil tend to get more rain, and that can lift maize and soybean output in some of the largest exporting countries in the world. The same ocean that is emptying Peruvian nets may be filling Argentine grain carts. That comes with a qualification, and not a small one. El Niño is simultaneously Argentine agriculture’s best friend and its worst enemy, and which one depends on strength. Moderate events water the crop; strong events drown it. The same 2015 and 2016 season that produced Argentina’s second-largest soybean harvest on record, 59.1 million tonnes, also damaged more than three million hectares through waterlogging as the Paraná and Uruguay rivers rose. This event is forecast very strong. ¹⁴ Whether the world is better or worse fed for this El Niño is not one number. It is a ledger with entries on both sides, and the entries land on different people.

Brazil is not uniformly better off either. The south gains, but Matopiba and the Center-West run drier, and Eduardo Martins of GAAS warns that up to a fifth of national soybean output could be lost if drought settles there. Eduardo Assad of Embrapa is more cautious, putting it nearer ten percent, in line with what 2015 and 2024 actually cost. Even the worse figure would leave 145 to 150 million tonnes and Brazil the largest exporter. October rain decides how big the harvest will be. ¹⁴ But soybeans are the wrong crop to worry about. ¹⁵

Brazil grows roughly three quarters of its maize as safrinha, the second crop, planted from January into the same fields the soybeans have just left. So the risk to Brazilian maize runs through the soybean harvest rather than around it. Late rains mean late planting, late planting means a late harvest, and a late harvest pushes the safrinha past the safe window into late February and March, to mature through the drying end of the season. The USDA’s Brasília post sets out that sequence explicitly. A record soybean crop that arrives late is not good news for maize. It is the mechanism by which the maize is lost. ¹⁶

Two other crops are also exposed, one of them by the rain rather than the lack of it. Brazilian wheat grows in the south, on the wet side of this event, and rain at harvest means sprouting in the head and grain downgraded to animal feed rather than for human consumption; production is already forecast down 22.5 percent to 6.1 million tonnes, pushing imports to a record. Coffee and citrus, flower in September and October in the belt where the outlook shows the risk of drought, which makes the next eight weeks a flowering watch, and a time of worry.¹⁷

North America Gets the Other Half of the Deal

Now cross the Rio Grande, and the story inverts.

A strong El Niño splits North American winter weather along a line. The southern tier gets an active Pacific storm track; the northern tier gets a mild, dry one. For the United States that is on balance welcome. NOAA has an El Niño advisory in effect with 97 percent odds the event persists into early spring 2027. The southern track is expected to bring rain to California, the Southwest and the southern Plains, and an exceptional Sierra Nevada snowpack that works as a natural reservoir into summer. Areas of the Southwest that have struggled with persistent drought stand to benefit. The Atlantic hurricane season is suppressed, because El Niño raises the wind shear that tears developing storms apart; in the eastern Pacific it lowers that shear instead, so storms there become more frequent and stronger. The moisture that fed the Grand Canyon flood was drawn inland from a Pacific tropical storm. ¹⁸

That is the sentence this series has not yet had occasion to write. In the one part of the world best equipped to absorb a bad year, this El Niño is mostly a good one. Despite that, two days before this article was written, it killed people.

On 29 August a flash flood surged down Bright Angel Creek at the bottom of the Grand Canyon. Two people died, more than sixty were evacuated by air, and the Transcanyon Waterline that supplies the park was destroyed, leaving it on limited reserves. Between half an inch and an inch of rain fell in half an hour onto steep, arid ground. Most of it, in the words of the National Weather Service meteorologist who described it, ran off instantly. The Dragon Bravo burn scar left by last year’s North Rim fire channeled the rest.

The forces behind it were identified quickly: the Arizona monsoon, a marine heatwave, a heat dome pulling moisture inland from a Pacific tropical storm, and this El Niño. Atmospheric moisture over the United States has run five to fifteen percent above normal this summer, and the analysis traced that to record warm Pacific water. Rain lands on ground too hard and too dry to take it, runs off instead of soaking in, and does its damage on the way. ¹⁹

The rain the American Southwest needs and the rain that killed two people in a canyon are the same rain.

Figure 3. The same event, four different deals. The southern tier of North America gains rain and snowpack; the northern tier pays for it in 2027; Mexico, Central America and the Caribbean lose a harvest. Seasonal outlooks are probabilistic rather than certain.

“The same ocean that is taking a second harvest from a Guatemalan smallholder is filling reservoirs in California. That is not a coincidence in the weather. It is the whole argument of this series in a single season.”

The costs North America does carry are deferred and regional. The Pacific Northwest and the northern Rockies sit on the dry side of the split, and below-normal mountain snowpack in the Cascades and the Columbia basin is a high-confidence outcome. That is a 2027 water and hydropower problem rather than a 2026 one, and it lands on the same wheat country that has already had a hard run. Snowpack is a reservoir that fills in winter and releases slowly through the dry months. A warm winter does not drain it. It causes the snow to fall as rain and stops the snowpack from forming. ²⁰

Mexico is the exception on this side of the border: if the event persists into spring 2027, reduced summer rainfall is expected across its maize belt. And the Midwest has a hazard unrelated to drought, in the March to April transition, where a rapid warm-up followed by a late freeze, the 2016 pattern, threatens early-planted crops and fruit trees. ²¹

So Where Does American Hunger Come From?

Not from a failed American harvest. It comes from two other places.

The first is price. American shelves are supplied by a global market, and that market is being squeezed at both ends of this article. Central American maize and beans, Asian rice, and the fertilizer that grows all of it are all under pressure at once. A household that buys most of what it eats is exposed to that regardless of how good the Sierra snowpack is.

The second is policy, and it is the one that can still be changed. The 2025 reconciliation law cut SNAP by roughly $187 billion over ten years, extended work requirements to adults aged 55 to 64 and to parents of children aged 14 and over, and for the first time in the program’s sixty-year history required states to share its costs. More than four million people are estimated to have lost benefits between July 2025 and April 2026. The price wave from this El Niño reaches American shops in 2027, into a support system that has just been cut. ²²

Figure 4. The American sequence. The weather arrives on the helpful side of the split, and the exposure is a price reaching shelves in 2027 into a support program cut in 2025. Benefit-loss estimates come from an organization that opposed the legislation; the Congressional Budget Office’s figure, covering work requirements alone, is lower.

The Part That Already Worked

One more thing belongs here, because it is the only piece of good news in this article and it is load bearing.

During the 2023 and 2024 El Niño, anticipatory action was implemented across nine countries in this region. Cash transfers, agricultural inputs, rehabilitated water systems, irrigation, animal health brigades. It reached more than a hundred thousand people in some 250 rural communities. In several Central American countries, it produced increases of up to forty percent in maize and bean production. ²³

That is not relief distributed after a failure. That is more food grown, in the same soil, in the same bad year, because somebody moved before the rain did not come. The corridor has already run the experiment. The postrera is being planted now.

“The question in the Dry Corridor is not whether early action works. It has been measured, in these countries, in the last El Niño. The question is whether humanitarian aid and crop support funding gets to these smallholder farmers before December.”

Endnotes

1. On the naming: Peruvian fishermen used El Niño de Navidad for the warm current appearing off northern Peru around Christmas. On the science: the India Meteorological Department was founded in 1875; after the 1877–78 famine the government asked Henry Blanford to attempt seasonal monsoon forecasts, the first regular one issued 4 June 1886. Sir Gilbert Walker, Director General 1904–1924, chosen for his mathematics, introduced correlation and regression to long-range forecasting and identified the Southern Oscillation. Jacob Bjerknes established in the 1960s that it and the eastern Pacific warming are one coupled system.

2. FEWS NET Guatemala, August 2026: Dry Corridor and targeted areas of Alta Verapaz and the Western Highlands in Crisis (IPC Phase 3) through September; five to eight days of rainfall between 20 June and 20 July; damage attributed to El Niño and the canícula. FEWS NET Honduras, July 2026, citing DICTA-SAG: 8,772 ha of maize and 3,080 ha of red beans lost; deficits of about 45 mm in Yoro, El Paraíso and Olancho. Salvadoran small and medium producers association: expected 20 percent loss in the first basic-grain harvest.

3. FEWS NET Honduras and El Salvador, April to July 2026: widespread postrera delays and reduced planted area from primera delays, inadequate July–August soil moisture and high fertilizer prices; the 2026 hurricane season expected below average, consistent with El Niño.

4. FEWS NET Guatemala: lean season onset February 2026 against a typical June to August peak; 2.0 to 2.4 million people projected to need food assistance June to August; poorest households selling productive assets.

5. FEWS NET Honduras, July 2026: white maize 30.8 percent above the previous year and 35.4 percent above the five-year average in June; transport costs nearly 15 percent above last year. Honduran alert covering about 80 percent of national territory; El Niño alerts declared by El Salvador and Panama, August 2026.

6. About 3.5 million people across Guatemala, Honduras, El Salvador and Nicaragua required humanitarian assistance after the 2015–16 event, which is widely identified as a contributing factor in the subsequent rise in northward migration, with the Western Highlands the principal origin area. Migration is multi-causal; drought is one driver alongside violence and poverty.

7. Gatún Lake is a rainwater-fed reservoir about 85 feet above sea level, supplying the locks and serving as a source of potable water for Panama. The canal authority puts consumption at roughly 52 million gallons per transit, much of it discharged to the ocean. The 2016 Neopanamax locks recycle about 60 percent of each chamber via water-saving basins, but the chambers are larger, so the net saving is around 7 percent per transit.

8. Panama Canal Authority advisories, 2026: maximum authorized Neopanamax draft reduced from the design maximum of 50 feet to 49.5 feet effective 3 July, 49.0 on 24 July and 48.5 on 15 August. Advisory A-29-2026 of 20 August postponed the 48.0-foot reduction from 26 August to 2 September and the 47.5-foot reduction from 3 September to 1 October, while capping daily transits at 34 vessels for booking dates from 4 September and 32 from 15 September, with Neopanamax slots reduced to nine and Panamax slots to 25, falling to 23 from 15 September. The canal averaged 34.03 transits a day in July and about 35 through June. Cumulative watershed rainfall 34 percent below the historical average for May to August, inflows 44 percent below; Gatún Lake at 84.1 feet on 21 August. On auctions: SK Gas reported paying $5.3 million on 25 August 2026 for a priority slot for the gas carrier G. Spirit, reported by Bloomberg as the highest bid recorded; slots averaging about $1.1 million. Advisories are revised at short notice and should be checked against pancanal.com before publication.

9. 2023–24 precedent: Gatún just above 79 feet June to October 2023; draft cut to 44 feet; daily transits from 36 to as few as 18 by February 2024; reported FY2024 transits 9,936 against 12,638 in FY2023, down 29 percent. Waits of up to three weeks, auction premiums, and diversion via Suez, the Cape or North American rail. The canal handles roughly 5 to 6 percent of world maritime trade, about 70 percent of it US-bound or US-origin. Verify transit figures against ACP monthly statistics.

10. Panama Canal Authority, May 2026: water-saving measures at the locks from late 2025, hydroelectric generation at Gatún suspended to prioritize storage, and reserves built during a 2026 dry season among the wettest since 1950.

11. FAO El Niño agricultural risk assessment, June 2026: about 70 percent probability of below-normal rainfall regionally, risk highest along the Dry Corridor, in Colombia and Venezuela, and in Cuba, the Dominican Republic and Haiti, where harvests fell by up to 70 percent in places during 2015–16. World Bank: roughly 6 million Haitians at Crisis or worse through mid-2026. European Commission JRC, June 2026, on dryness threatening maize and rice from Colombia through Central America and the Caribbean, above-average rain for Bolivia, Ecuador and Peru, and diesel depletion in Cuba.

12. ENFEN activated an El Niño Costero alert on 13 February 2026, expecting coastal warming into late 2026. PRODUCE set a first-season quota of 1.9 million tonnes, the lowest in a decade and 36 percent below 2025; operations suspended from 12 May and extended to 10 June over juveniles. Fishmeal prices more than 75 percent above the previous year. Peru is roughly 20 percent of global fishmeal and fish oil. Production fell to about 815,000 tonnes in 1998; the cancelled 2023 season cost over $1.4 billion in exports and an estimated 0.5 percent of GDP. FAO: the Peru Current produces 12 to 20 percent of world fish landings.

13. El Niño is associated with reduced rainfall, higher temperatures, fire risk and lower river flow across northern Brazil, the Amazon basin, Colombia and Venezuela, with consequences for hydroelectric generation; and with increased rainfall across southern Brazil, Uruguay, Paraguay and northern Argentina, which may lift maize and soybean output. A Peruvian fishmeal shortfall also raises demand for soybean meal. In the eastern Pacific El Niño lowers wind shear, raising cyclone frequency and intensity while suppressing Atlantic activity. Added autumn and winter moisture across the southern Plains improves prospects for hard red winter wheat.

14. El Niño brings above-normal October to March rainfall to central and northern Argentina, the core Pampas provinces typically at 120 to 160 percent of normal. In moderate events this lifts yields; in strong events it floods. The 2015–16 event damaged more than three million hectares through waterlogging as the Paraná and Uruguay rivers rose, in the same season that produced 59.1 million tonnes of soybeans, the second largest on record, with maize yields about 7 percent above the ten-year average.

15. Eduardo Martins, GAAS, warning up to 20 percent of Brazilian soybean output could be lost if drought settles over Matopiba and the Center-West; Eduardo Assad, Embrapa and FGV Agro, putting it nearer 10 percent in line with 2015 and 2024. A 20 percent reduction would still leave about 145 to 150 million tonnes. 2026/27 projections: StoneX 183.1, Safras & Mercado 180.1, USDA 186, S&P Global 177 million tonnes. October rainfall is the decisive variable. Several figures are from regional business press; CONAB, the Rosario and Buenos Aires exchanges and USDA WASDE are primary.

16. Safrinha maize is planted January to March after the soybean harvest and is roughly three quarters of Brazilian maize production. USDA FAS Brasília, Grain and Feed Update 2026: El Niño could delay the soybean harvest, and later reaping would compromise the optimal safrinha window. The principal supply-bearish scenario is delayed central-Brazil rains, excessive southern rainfall and a late soybean harvest pushing the safrinha into deteriorating weather; percentage seeded before late February is the indicator. CPTEC/INPE: below-normal rainfall in parts of the North, Northeast and central Brazil, above-normal in the South.

17. USDA FAS: Brazilian wheat production forecast down 22.5 percent to 6.1 million tonnes for MY2026/27, driving record import demand, on reduced area, lower investment and rising input costs. FAO recorded reduced Paraná and Rio Grande do Sul wheat plantings after low-quality harvests in 1997–98. Coffee and citrus flower in September and October; persistent dryness then damages flower development and cherry formation, though a short dry spell followed by timely rain can assist.

18. NOAA Climate Prediction Center El Niño advisory: 97 percent odds the event persists into early spring 2027; the 13 August 2026 update raised the probability of a very strong event for October to December to 95 percent. Regional effects: an active southern storm track bringing precipitation to California, the Southwest and southern Plains; an exceptional Sierra Nevada snowpack; drought relief across the Southwest and southern Rockies; suppressed Atlantic hurricane activity.

19. National Park Service and Associated Press, 29 to 31 August 2026: two deaths, more than sixty air evacuations from Bright Angel Creek, and destruction of the Transcanyon Waterline. NWS Flagstaff on half an inch to an inch of rain in half an hour on steep terrain, most of which ran off instantly, with the Dragon Bravo burn scar a likely major contributor. Ben Noll, Washington Post, 31 August 2026, attributing the event to the Arizona monsoon, a marine heatwave, an inland heat dome drawing Pacific moisture and El Niño, with atmospheric moisture over the US 5 to 15 percent above normal. Single-storm attribution is not possible; El Niño is one factor among several.

20. Strong El Niño events typically bring milder, drier winters to the Pacific Northwest, northern Plains, Great Lakes and Ohio Valley, with below-normal Cascade snowpack and reduced Columbia basin runoff described as a high-confidence outcome for 2027 water supply and hydropower.

21. Reduced summer rainfall is expected across Mexican maize areas if the event persists into spring 2027. The March to April transition carries a risk of rapid warm-up followed by late freeze, the 2016 pattern, threatening early-planted crops and fruit trees.

22. The 2025 reconciliation law reduced SNAP by roughly $187 billion over ten years, extended work requirements to adults aged 55 to 64 and parents of children aged 14 and over, and for the first time required states to share program costs. The estimate of more than four million people losing benefits between July 2025 and April 2026 comes from Center on Budget and Policy Priorities tracking; CBPP opposed the legislation, and the Congressional Budget Office projects 2.4 million in an average month for work requirements alone.

23. FAO and WFP, Latin America and the Caribbean briefing, May 2026: during 2023–24 anticipatory actions ran in nine countries, reaching more than 100,000 people in 250 rural communities through cash transfers, agricultural inputs, water system rehabilitation, irrigation and animal health brigades; in some Central American countries these produced increases of up to 40 percent in maize and bean production.

Five of these countries, together with Ghana and Guinea, have already closed their borders to cereal exports. Roughly 38 million people across Djibouti, Ethiopia, Kenya, Somalia, South Sudan and Sudan were at IPC Crisis levels of hunger or worse before this season’s crops were even at risk. None of it is a surprise, all of this was predicted and the worst of the hunger caused by these failed rains is still more than a year away.

The First Part of El Niño Hunger Has Already Stopped Being a Forecast; Without Increased International Response It Will Become a Humanitarian Disaster.

Where is the Surge of Humanitarian Aid?

Last week 4Hunger.org published Hunger Foretold, which revealed one thing in detail: that the hunger arriving in 2027 is being forecast right now, in public, by government scientists, on a schedule anyone can read for free.

We also revealed the three pressures that are converging in the next eighteen months: 1) European and American droughts, one driven less by missing rain than by an atmosphere supercharged by climate change to pull water out of soil faster than rain can replace it, the other traditional, but shifting regions into a different grain exporting region; 2) the strongest El Niño ever forecast, is still building in the Pacific; and 3) a phosphate fertilizer shortage created by China and Russia’s export policies and a lack of sulfur on the global market reducing United States fertilizer production rather than by a lack of phosphate on the global market, the bill for not applying phosphate fertilizer this year will not arrive until the soil reserve of phosphate runs out, creating lasting hunger two to four years from now. Hunger Foretold set out when each of these global pressures would impact the food reaching tables around the world, argued that the lag between a failed harvest and a hungry household is a space in which humanitarian actions can still be mounted, and concluded that whatever acute hunger and famines that follow will be the result of policy decisions made by politicians in countries where the hunger lands and also in governments thousands of miles away from the hunger. ¹

Hunger Foretold was about a forecast. This article is about El Niño landing, destroying crops and causing hunger, a description of the damage already caused as of September 2, 2026, and the forecast of what is yet to come in the next 18 months.

The rains that feed Sudan’s sorghum did not come in June and July, and as a result, the Nile River fell lower than farmers in their sixties can remember seeing it in August. The U.S. Climate Prediction Center reclassified southern South Sudan and swathes of northern, eastern and central Ethiopia into drought. Guatemala’s Dry Corridor went into Crisis. Honduras counted thousands of lost hectares of maize and beans. Not one of those outcomes was a surprise. All of them were predicted by the sources detailed in Hunger Foretold. Every one of the regions that have been hit by decreased rain and drought sat in the El Niño predictions months before the rains did not come. There is a difference between a projection and a harvest that has already failed, and in six countries that line has now been crossed. ²

“The models said the rain would fail in Sudan, in South Sudan, in Ethiopia, in the Dry Corridor of Central America this June, July and August. The rain has failed in Sudan, in South Sudan, in Ethiopia, and in the Dry Corridor of Central America just as predicted. What follows is not a prediction. It is the next page of El Niño already in progress and the human and environmental suffering it is causing.”

This article asks a different question from the last one. Not whether the hunger is coming, but where it will land, in what order, and what happened the last five times it landed in the same places. The El Niño map is not new. It has been drawn and redrawn multiple times since 1877, and the striking thing about it is not its complexity but its repetitiveness. The same regions fail. The same crops fail in them. The same governments are surprised. What changes from one event to the next is how many people die, and that, it turns out, has almost nothing to do with the temperature of the water in the Pacific Ocean but actions of governments around the world.

Figure 1. What the models said, and what the world has since recorded. Solid bars are outcomes already on the register: rains that failed, droughts declared, hectares counted. Dashed outlines are windows still ahead. Every bar here, solid or dashed, was drawn from forecasts issued before the season began. Teleconnections are probabilistic rather than certain, which is precisely why the top half of this chart matters.

The Same Map, Every Time

Strong El Niños do not damage the world’s agriculture at random. They damage it in a pattern documented across a hundred and fifty years of observation, and the pattern is repetitive to the point of monotony. The rain fails in the Maritime Continent (the vast tropical region in Southeast Asia and the Indo-Pacific made up of thousands of islands, peninsulas, and shallow seas), southern Africa, Central America and northeastern Brazil. It arrives differently in different parts of the world, delivering torrents of rain over the Horn of Africa and coastal Peru, where it drowns crops rather than merely watering them. India’s monsoon underperforms. Australia’s spring turns hot and dry across the wheat belt exactly during its vulnerable grain-fill window. ³

The reason this matters is not meteorological elegance. It is that the maps have been alerting governments a year in advance, which regions, which countries, which counties will be in trouble, and the map has repeatedly been right, so that being surprised by the hunger of 2027 will require a deliberate effort not to read the maps and reports of the past year or to pretend not to understand their predictions.

The pattern is old. What has changed is how hard it hits. A study published in Science this month, led by Julia Cole at the University of Michigan, finds that the strong El Niños of recent decades have themselves been made more intense by the warming caused by burning fossil fuels. The relationship runs both ways: El Niño delivers a surge of global heat, and that accumulated heat in turn makes the next El Niño stronger. Andrew Dessler of Texas A&M, who was not involved in the work, called it an important validation of what the models had predicted, and said a substantial share of the damage from the coming El Niño event should be attributed to the fossil-fuel economy. This matters for how the rest of this article should be read. The 1877 map and the 2026 map are the same map. The El Niños have become stronger. ⁴

The Nile River Has Not Risen

Sudan’s rainfed belt, Gedaref, Kassala, Sennar, the Blue Nile, works to a calendar with no slack in it. Sorghum and sesame go in with the rains in June and July and come out from November. The Gezira scheme, one of the largest irrigation systems ever built, draws on the Blue Nile, not dependent on direct rainfall. Both harvests depend on the same rains arriving at the same time, which is the vulnerability that matters, because in Sudan in an El Niño year, the rainfed crops and the irrigated ones do not hedge each other. They fail together.

This year they did. The rains did not come in June or July. The reservoir behind Ethiopia’s Grand Renaissance Dam stood six meters lower in July 2026 than in the same month last year, and parts of the Blue Nile basin received as little as half their normal rainfall. In Gedaref, where most of the country’s sorghum and sesame grows, fields that are ordinarily green by August were still brown last month, and farmers gathered behind their imams to pray. The agriculture ministry belatedly has advised farmers to switch to drought-resistant varieties and to delay planting. This advice is itself an admission that they ignored the El Niño reports, since neither of these options is available to a farmer who has already sown their crop. ⁵

Photo Credit: WFP/Francis Thawani

Those farmers who can still plant say it costs four times what it did last year, which is where this story joins the others in Hunger Foretold. The closure of the Strait of Hormuz raised the price of fertilizer both in Iowa and in Gedaref by the same mechanism, and the Sudanese farmer has no crop insurance, no disaster programme and no balance sheet to absorb it. A farmer reported that a farm his family has held for a century, recovered from the Rapid Support Forces only last year, is being planted at four times the cost into a season the climate and El Niño prediction models condemned in spring of 2026.

The El Niño predictions did specifically predict the joint failure of these two crops. Elfatih Eltahir at MIT has spent three decades documenting the link between El Niño and the flow of the Nile, and his work shows that El Niños beginning in late spring, as this one did, carry a high probability of drought across southeastern Sudan and northwestern Ethiopia. This was not an unlucky season. It was a forecast season of failed crops, and the forecast is correct. ⁶

Now comes the part that will fool people and be extra cruel. FEWS NET expects Sudanese food security to improve between October and January, in parts of Gedaref, Kassala, Al-Jazirah and Sennar, as the harvest, such as it is, comes in. Hunger conditions ease. World attention leaves. Then FEWS NET predicts, almost in passing, that because the harvest is below average, the reduced household grain stocks will not actually be felt until later in 2027. The worst months of this drought will impact the people of Sudan more than a year after the rains failed to fall. This means that in Sudan there is a window this winter when everything will look as though it is getting better, but it is not. and this is exactly when international humanitarian aid could be mobilized to prevent the worst of the hunger and famine when the disappointedly small crop is eaten and there is no more food for the people of Sudan to eat. ⁷

What has changed is the calendar. Sudan’s lean season runs from May to September and peaks in August and September, which is to say now. The sorghum and millet harvest comes in through November and December. So, the months ahead are, in any year at all, with either a good or catastrophic harvest, the least hungry part of the Sudanese year is between October and January, when grain in the house from a family’s own fields, or bought at market with a grain new supply arriving, prices falling back from their lean-season peak, and the largest block of paid agricultural work in the annual cycle. Sorghum, millet and wheat flour prices rose between 5 and 13 percent in a single month from June to July, and stand 45 to 60 percent above where they were in February. Those prices will come down at harvest. The fall in prices will be read as recovery. ⁷

Photo credit: FAO/Mohamed Ahmed

But it is not recovery. It is the annual cycle briefly outrunning the hunger and it is measured against September rather than against a normal year. Because the harvest is small, the grain it puts in the store runs out sooner than it should, which brings the next lean season forward and starts it from a lower base, in households that have already sold whatever they had left. The improvement is real, it is temporary, and it is the reason the window of abundant food closes.

“The rain failed in Sudan in June and July 2026. The immediate hunger it causes arrives in the second half of 2027. Everything in between is the time in which humanitarian aid could be mobilized, but the harvest arriving in that window makes it the season in which it will look to outsiders as no longer necessary, but acute hunger and maybe famine is just around the corner.”

Figure 2. Why the worst month is more than a year after the rain. Sudan sows sorghum and sesame with the June rains and harvests from November, so a season that fails in July produces a harvest that fails in December. FEWS NET expects conditions to ease as that harvest arrives, and the shortfall to be felt only in the second half of 2027. The window in the middle is the one in which the emergency looks as though it is receding.

All of these El Niño impacts land on a country in the fourth year of a war that has killed more than two hundred thousand people and left nearly 19.5 million people, two out of every five Sudanese, at IPC Crisis levels of hunger or worse. That figure breaks down into 14 million in Crisis, more than 5 million in Emergency and nearly 135,000 in Catastrophe hunger, the household phase at which people starve. A risk of famine is identified in fourteen areas across North Darfur, South Darfur and South Kordofan, under a reasonable worst-case scenario of intensified conflict and further restrictions on humanitarian access. Famine itself was confirmed in El Fasher and Kadugli in September 2025. Those towns no longer meet the technical thresholds, which reflects how many people died or fled rather than any improvement. An estimated 825,000 children under five are expected to suffer severe acute malnutrition this year, seven percent more than in 2025 and 25 percent more than before the war, and close to nine million people were displaced inside Sudan by the end of March. National figures hide what is happening in particular places. Surveys in North Darfur in late 2025 found global acute malnutrition running at 52 percent in Um Baru and 34 percent in Kernoi. Thirty percent is the threshold at which malnutrition alone meets the famine standard. Across Darfur three quarters of surveys found rates above 15 percent, the level the World Health Organization calls critical.

Sudan’s food supply has two halves, and both are now exposed. It buys about four fifths of its wheat abroad, which leaves that half at the mercy of world prices and the shipping lanes. Sorghum is the half it grows to feed itself, and sorghum is what failed this summer. ⁸

A falling river does something else that a hunger article can forget to say. When the Nile drops, when people are pushed into camps, and when four years of war have taken the clinics with them, the water people drink gets worse at exactly the moment they are least able to survive it. Sudan declared a cholera outbreak in late June, in West Kordofan, on the line dividing the army’s territory from the paramilitaries. It is the third wave in three years, and it began ten weeks after the last one was declared over. By 10 July the World Health Organization had recorded more than 1,330 confirmed cases and 114 deaths, mainly in the Darfurs and the Kordofans, and said the true death toll was likely far higher. The Sudan Doctors Network reports over 200,000 people around Bara, among them 20,000 children, facing cholera and measles with no medicine to treat either. Those are the figures from before the rains. The WHO warned in the same briefing that Sudan’s rainy season was about to surge, and that cholera cases balloon when it does, because the water gets dirtier and the roads close at the same time. The previous wave ran from July 2024 to January 2026, reached all eighteen states, infected 124,418 people and killed 3,573. The agency has received about a third of the money it needs for Sudan this year. ⁹

Hunger and cholera are not two problems. A malnourished child dies of diarrheal disease at many times the rate of a fed one, and cholera kills fastest where people are already weak, crowded and drinking whatever water is left. The drought does not have to reach famine thresholds to kill. It only has to arrive where the water has gone bad and is contaminated with disease. ⁹

“The forecast said the rain would fail in Sudan. The rain failed. The same forecasts say the worst hunger arrives in the second half of 2027. There is no reason to think it will be wrong the second time. There is time to prepare an adequate humanitarian aid response.”

The drought does not stop in Africa, Part two: “El Niño Crosses the Atlantic” follows the impacts that have already happened in North and South America and shows how this El Niño has also impacted the Dry Corridor from Mexico to Panama, broader South America and how it will mostly benefit North America. Following that we will continue globe trotting examining how this El Niño is following the exact pattern of all El Niños since 1877. Stay tuned.

Endnotes

1. Hunger Foretold: The Coming 2027 Hunger and the Free Public Maps and Reports That Are Predicting It (this publication, 2026).

2. Agence France-Presse, Gadaref and Al-Jazira, 22 August 2026. Grand Ethiopian Renaissance Dam reservoir six meters below July 2025 (University of Tübingen); Blue Nile basin rainfall as low as half of normal (Climate Hazards Center, UC Santa Barbara). NOAA Climate Prediction Center, Global Weather Hazards Summary, 20–26 August 2026. FEWS NET Guatemala (August 2026), Honduras (July 2026, citing DICTA-SAG) and El Salvador (April 2026).

3. Ropelewski, C.F. & Halpert, M.S. (1987), Monthly Weather Review 115, 1606–1626. Iizumi, T. et al. (2014), Nature Communications 5, 3712.

4. Cole, J. et al., Science, August 2026, on the intensification of recent strong El Niño events by anthropogenic warming; reported by the Guardian, 27 August 2026, including comment from Andrew Dessler, Texas A&M. The study addresses recent decades rather than the 2026–27 event individually.

5. Agence France-Presse, Gadaref and Al-Jazira, 22 August 2026, including agriculture ministry guidance on drought-resistant varieties and delayed planting.

6. Elfatih A. B. Eltahir, Massachusetts Institute of Technology, on El Niño and reduced Nile flow, and on drought probability across southeastern Sudan and northwestern Ethiopia when an event begins in late spring.

7. FEWS NET Sudan, projecting improvement to Stressed (IPC Phase 2) in parts of Gedaref, Kassala, Al-Jazirah and Sennar between October 2026 and January 2027, with the reduction in household stocks not felt until later in 2027. FEWS NET seasonal calendar: rains June–September, lean season May–September, millet and sorghum harvest November–December. FEWS NET Sudan, August 2026, on the August–September lean-season peak and sorghum, millet and wheat flour prices 45 to 60 percent above February.

8. IPC Sudan, February–May 2026 with projections to January 2027, and the joint FAO/WFP/UNICEF release of 15 May 2026: 19.5 million at IPC Phase 3 or above, comprising 14 million in Phase 3, over 5 million in Phase 4 and nearly 135,000 in Phase 5; risk of Famine in fourteen areas of North Darfur, South Darfur and South Kordofan. 825,000 children under five expected to suffer severe acute malnutrition in 2026; close to nine million displaced within Sudan as of end-March. Famine confirmed in El Fasher and Kadugli in September 2025, when 21.2 million were at Phase 3 or above; FEWS NET subsequently recorded thresholds no longer met, reflecting mortality and displacement rather than improvement. SMART surveys, November–December 2025, via FEWS NET Sudan, June 2026: global acute malnutrition of 52 percent in Um Baru and 34 percent in Kernoi, North Darfur, against a famine threshold of 30 percent. Note that the June–September 2026 and October 2026–January 2027 projections cover only 56 localities, around 8.3 million people, against 195 for February–May.

9. World Health Organization briefing, Geneva, 10 July 2026: more than 1,330 confirmed cases and 114 deaths, mainly in the Darfurs and Kordofans, with the true toll likely higher. Outbreak declared in West Kordofan in late June 2026, ten weeks after the previous outbreak was declared over on 3 March; that wave, from July 2024, reached all 18 states, infecting 124,418 and killing 3,573. Sudan Doctors Network on 200,000 people around Bara. WHO reported roughly one third of its required Sudan funding received.

Figure 1: Photo credit: World Peace Foundation, toa555/stock.adobe.com

This is what the expert warnings say, when each El Niño impact comes due, and what can still be done to minimize the hunger and other impacts of the 2026-7 El Niño.

A Different Kind of Disaster

Most disasters arrive without permission. An earthquake gives no notice. A flash flood gives minutes of warning. A hurricane gives days.

The hunger now forming in the world’s food system is not like that. Its coming has been predictable for months, it will take another year for the 2026-7 El Niño to fully arrive, and the instruments watching it are free for all to read. It is a food crisis on a schedule where prevention is possible.

“The cruelest thing about the hunger from the 2026-7 El Niño now forecast is that IT IS FORECAST. Nobody in 2027 will be able to say the maps were unavailable or ‘How could I have known?’ The reports are free, public, and updated weekly. Humanitarian actions still can be taken to respond to the coming hunger, but will we act?”

That is the question this article asks. Everything else said in this article is evidence for why and how we should act, and the last section tells exactly what we still can do to anticipate and prevent the coming hunger.

The Warning System for Coming Hunger Already Exists

Start with what is being published right now, this week and every week, by people paid by taxpayers to publish it.

Figure 2. The core public monitoring systems for global food security. None of them requires a subscription, a login, or a security clearance.

The U.S. Drought Monitor releases a map every Thursday morning showing where American drought sits and, crucially, a change map showing where drought has worsened over the previous week. Europe’s Drought Observatory does the same every ten days. NOAA monitors the temperature of the tropical Pacific continuously and publishes a monthly forecast of how strong this El Niño is going to be. Crop analysts in Washington and Brussels publish yield forecasts for every major growing region, every month. FEWS NET publishes which districts of which countries will run short of food, and roughly when.

None of this is obscure. All of it is written in plain language. Any journalist, aid official, member of Congress or member of the European Parliament can read the whole system in an afternoon.

This means the honest way to describe the hunger that is coming is not as a surprise, and not as a tragedy. It is a food crisis on a schedule.

But here is a warning: this system is not permanent. In September 2025 the U.S. Department of Agriculture announced it would stop publishing its annual Household Food Security report, the thirty-year record of how many Americans cannot reliably afford to eat. ¹ Ending the count does not end the hunger. It ends the ability to see it. Everything this article demonstrates about the value of public data applies with equal force to why we need to keep the U.S. Household Food Security Report as a public resource.

Why Hunger Arrives Late

Before we discuss the schedule itself, one idea has to be clear, because without it the rest looks like alarmism.

The single most misunderstood thing about food crises is timing. People expect a drought to produce hunger the way a fire produces smoke; immediately, visibly, in the same place and time.

It does not work that way, for a simple reason: the harvest a drought destroys is usually not the one you, people of the Global South or the rest of the world are currently eating.

A dry summer damages a crop that gets harvested in autumn, sold or stored over the winter, fed to livestock or put on the market in the spring, or sold to people through the spring and summer and shows up as the price of grain, meat and milk or grain a year later. A farmer who cannot afford fertilizer this season plants a weaker crop next season. A field left unsown in September is a shortfall the following August. Every link in that chain adds months to when hunger will arrive.

The gap between the weather and the hunger is not evidence that the warnings were wrong. It is the normal delay of a food system, and it is exactly the space of time during which something could be or could have been done. It is why there is time, between now and when the worst of El Niño hits, in which humanitarian assistance can be mobilized.

That delay is the reason for hope in this story. It is also the reason for urgency. The two are both caused by the same facts and knowledge.

Three Pressures, One Year

Three separate things are pressing on the world’s food supply at once this year. They are usually reported as unrelated stories, which is why the picture often looks confusing. Taken in order, they are straightforward and combine to form a clear picture.

Each of these three is set out in full in the following sections, with the sources. I highly suggest that you read the details of all three because the droughts, El Niño and the Fertilizer crisis are going to cause incredible suffering and none of them are what you think. What follows first is a summary of what the three of them together are already doing and are going to do to people around the world:

The Calendar

With the three pressures in view, the sequence becomes readable. Here is the schedule, as the public data currently describes it, it is beginning to happen as predicted and we have hard evidence that El Niño’s impacts on rains and crops have already started to occur.

June to August 2026 - already happened. The rains that feed Sudan’s sorghum and sesame did not arrive, and the Nile River fell with the lack of rain. The lack of rain will impact both rainfed and irrigated crops in Sudan and the region. This is the one row in the predicted schedule of El Niño’s rearrangement of rainfall that can already be checked against the world. It happened as forecast.

September 2026 - now. Farmers across northern Europe have about four weeks to plant this autumn’s oilseed crop. The window is set by daylight and by pests, not by frost, and it does not move. Planting needs rain at sowing and rain again each week for the four weeks thereafter. If this rain does not come, a crop that was meant to be a record becomes a shortfall, and that shortfall belongs to 2027, not to this year. Almost nobody is watching this. It may be the most consequential four weeks in the whole sequence. ² Oilseed matters more than its acreage suggests, because one crop feeds two markets. Crushed rapeseed yields roughly 40 percent oil and 60 percent meal. The oil goes into cooking oil and biodiesel, the meal is what European dairy herds and pig and poultry farms are fed.

A shortfall raises the supermarket oil prices and the feed bill at the same time. Vegetable oils also trade as a single pool. Palm, soy, rapeseed and sunflower make up 88 percent of all the cooking oils that the world eats. Commercial buyer substitute freely, so a European failure to produce rapeseed does not stay in Europe. The increase on global demand will arrive just as El Niño impacts next year’s palm oil production. Analysts estimate Indonesian and Malaysian output of palm oil could fall by some 4.5 million tonnes based on the historic pattern of where past El Niños have caused damage. The market has already banked the expected record European crop, causing rapeseed oil prices to ease over the summer on expectations of larger supplies in 2026/27. That assumption is what these four weeks decide. If the harvest is smaller than expected, vegetable oils tighten worldwide, and that matters more than the acreage suggests. Vegetable oils supply about a tenth of the world’s daily calories, second only to cereals, and they are the cheapest dense calorie a poor household can buy, it is the item a wealthy shopper substitutes with something else and the one that a poor one simply goes without.

There is a lever here. Around 31 million tonnes of vegetable oil is burned in the EU as biodiesel each year, the calories from this oil are enough to feed more than 320 million people. A large share of Europe’s rapeseed oil goes to biofuel rather than food. If the crop comes in short, Europe faces a choice: relax the EU Biofuel Mandate and release those calories for food, or import and outbid poorer buyers from around the world to drive up food prices and make EU biofuels. This is the article’s argument in miniature. ² Whether a bad rapeseed harvest results in hunger depends on a decision by the EU.

October 2026 to February 2027 - the feed bills. Grain harvested this autumn becomes animal feed this winter. Feed is the largest single cost in producing meat, milk and eggs. This is the first stage that ordinary shoppers in wealthy countries actually notice, and they notice it at the meat counter rather than in the bread aisle.

December 2026 to February 2027 - the ocean peaks. The El Niño reaches its forecast maximum. Southern Africa plants its single annual maize crop in December, directly into the forecast peak of when El Niño is expected to cause droughts.

January to March 2027 - the lean season. Southern Africa’s hungry months. This is the first point in the whole sequence where hunger stops being a projection and becomes a measurement, hungry mouth by hungry mouth. By then, every decision that could have softened the hunger will already have been made or not made. We do not have much time.

Spring to autumn 2027 - the price wave in rich countries. European food inflation is expected to peak near 3 percent, roughly a percentage point of which is attributable to this year’s weather. ³ In the United States, the same global prices arrive on top of drought and input-cost inflation, and tariffs already in the system.

Summer 2027 - a new American drought. The snowpack that fails to accumulate in the Northwest this winter becomes the water that is not there next summer. This is a region that in a good year grows and exports a great deal of wheat but will have a reduced harvest due to the expected drought.

Late 2027 - Sudan’s shortfall arrives. The harvest now failing will be eaten first and missed later. FEWS NET expects conditions to ease around the November harvest, and the reduced food stocks due to the El Niño will be felt in the second half of 2027.

2028 - the slow damage. The phosphate skipped in 2026 starts to show up as a lower yields and smaller crops, in the countries least able to absorb a reduced harvest.

Where These Three Crop Drivers Actually Land

One distinction matters more than any number in this article.

In Europe and America, the crops lost due to drought are called inflation. In Nairobi, Khartoum and Port-au-Prince, the same crop loss is measured in missed meals and increasing hunger.

A European household spends roughly a tenth of its income on food. A three percent food-price rise there is an irritation, a slightly worse year, an argument about supermarket margins. In countries where food takes up to forty percent of household income, an identical price move is a family dropping from two meals a day to one.

That is why the hunger from this convergence of factors will not land where the droughts happened. Europe will eat. America will eat. Both will complain about grocery bills, and both complaints will be legitimate, particularly for the families in those countries who were already struggling. Nearly one in four American adults reported living in a food-insecure household in 2025, and more than one in four of working age. ⁴ Massachusetts, counted by household rather than by adult, is worse still: a record 40 percent, roughly 1.1 million households, reported food insecurity in 2025, more than double the 2019 rate. ⁵

But the acute hunger will land in Sudan, which imports roughly four fifths of its wheat and already has more than nineteen million people short of food. It will land in southern Africa, where a single failed maize season has no backup. It will land in Haiti, Somalia, Chad and South Sudan, which sit at the top of the humanitarian warning lists precisely because they have no food cushion left. ⁶

In Sudan it is no longer a forecast. The rains that should have come in June and July did not come, and the Nile River has fallen due to the missed rains. The reservoir behind Ethiopia’s Grand Renaissance Dam stood six meters lower this July than last, and parts of the Blue Nile basin have had as little as half their normal rainfall. In Gedaref, where most of the country’s sorghum and sesame is grown, fields that are ordinarily green by August were still brown this month, and farmers gathered behind their imams to pray for rain. The agriculture ministry is telling farmers to switch to drought-resistant varieties and to delay planting. Those who can still plant say it costs four times what it did last year, because the same closure of the Strait of Hormuz that raised fertilizer prices in Iowa raised them in Gedaref. On a farm his family has held for a century in Al-Jazira, which he recovered from the Rapid Support Forces only last year, a farmer told reporters that he is sixty-three and has never seen the Nile River this low in August. ⁶

The Nile River running low and the reservoir depleted were foreseeable too. Elfatih Eltahir at MIT has spent thirty years documenting the link between El Niño and the flow of the Nile, and his work shows that when an El Niño event begins in late spring, as this one did, the probability of drought across southeastern Sudan and northwestern Ethiopia is high. The warning for Sudan existed before the rains failed. Nor is Sudan alone in having crossed from projection into measurement.

In the week of August 20, 2026, the U.S. Climate Prediction Center upgraded southern South Sudan and parts of northern, eastern and central Ethiopia to drought after ninety days of severe rainfall deficits. Recorded dryness due to lack of rain is worsening across Central America and Venezuela. In the Dry Corridor of Guatemala, Honduras and El Salvador that dryness has already become hunger. Much of this region saw only five to eight days of rain between late June and late July. Honduras has lost close to nine thousand hectares of maize and three thousand hectares of beans. The FEWS NET now places the poorest households of Guatemala’s Dry Corridor in Crisis through September. Southern Africa is the exception that proves the schedule: its rains are not due until October, so it is still, for now, a projection. ⁶

The World Food Programme estimates that a total of 49 million people will be pushed into acute hunger by this El Niño. They would be added to the 266 million people almost half of which are children, across 47 countries who were already in acute hunger in 2025. That is a figure that has roughly doubled in a decade, and which includes 39 million in Emergency conditions and 1.4 million in Catastrophe (famine), the phase in which people starve. The WFP has anticipatory plans running in more than fifty countries. It published that estimate this month, which is to say that the number of people this event will make hungry is itself already a matter of public record. ⁶

Whether more people enter famine is not really a forecasting question either. The IPC already projects Sudan’s Catastrophe (Famine) population rising from roughly 135,000 in the spring to 200,000 across fifteen areas by the September lean season, and a risk of Famine is formally identified in fourteen areas of Darfur and South Kordofan through September, persisting in thirteen into January. Four other places also carry that risk this year: Sudan, South Sudan, Gaza and Somalia. Every one of those analyses says the outcome turns on conflict intensity and humanitarian access. On decisions, in other words, and not on rainfall. ⁶

“Again, none of those countries closed a shipping strait, warmed the Pacific, or dried out the Loire valley, nor did they make the decisions to cut the El Niño preparation networks.”

What We Already Know How To Do

Here is the part of this story that should be reported far more often than it is: we know how to prevent El Niño from turning into a humanitarian crisis. Not in theory. In practice, with methods that have been tested and costed. Unfortunately, the systems needed to do this have been dismantled since 2025 by cuts to U.S., European and Japanese foreign aid.

El Niño is the most predictable large-scale climate shock on Earth. It has been monitored for 150 years and the way that rains will be redistributed is well known. That predictability was turned into machinery after the 2015–16 event, which the UN itself judged to have been responded to too late. The decade since was spent building trigger systems: creating agreements that release money and supplies automatically when a forecast crosses a threshold, rather than waiting for a famine to be declared. In 2023–24 those systems reached three million people ahead of impact. ⁷ They work. They were simply too small in the past, but now a wrecking ball is being taken to them. They need to be rebuilt and quickly.

Because this is a schedule of changing rains and predictable hunger crises rather than a surprise, the responses can be unusually concrete and focused. It is a policy choice what we do and whether we prevent the impending crisis.

Fund the response before the failure. Cash, seed and pre-positioned food released on a forecast, rather than on a famine declaration, cost a fraction of late emergency aid and save lives that late aid cannot. The UN’s food agencies have asked for $202 million to protect nearly nine million people across 22 countries, explicitly ahead of impact of the coming super El Niño. The U.N. agencies say they are already positioned to act for 1.2 million of the people at risk, and that an additional $167 million would extend coverage to the remaining 7.6 million people. ⁸ Southern Africa plants in December. That money and aid has to move before then, into a humanitarian system whose largest donor, the U.S., cut its funding from roughly $14 billion to $3.7 billion in a single year, 2025. We need to restore those cuts.

Keep the grain moving. The lesson of the 2007–08 food crisis is that export bans spread from country to country, and that hoarding, not harvest failure, is what turns a shortage into a famine. Governments built shared systems to see real global stocks precisely so they could resist that panic. Using them, and pressing exporters early and publicly against blanket bans, will do more for the world’s poor in 2027 than any single shipment of aid.

Protect the households in America and Europe at the end of the chain. The price wave reaches American and European shops in 2027, on top of inflation, fuel and fertilizer price increases, and tariffs already in the pipeline. Cutting food assistance to America’s poor and working classes in the same window, kicking poor and working families off SNAP, is not a neutral budget decision, it is going to create suffering across America. The 2025 reconciliation law cut SNAP by roughly $187 billion over ten years, extended work requirements to adults aged 55 to 64 and to parents of children aged 14 and over, and for the first time in the program’s sixty-year history required states to share its costs. ⁹ That is a choice about who absorbs the cost of a Pacific Ocean weather and climate event they had no part in causing.

“Not one of these three actions requires a scientific breakthrough, a new institution, or a treaty. Each requires a decision, taken this autumn, by people who already have the information they need to prevent the impending crisis.”

The Point is: Famines and Massive Acute Hunger Aren’t Predetermined. They Still Can Be Prevented If You Act.

There is a version of the next eighteen months in which the warnings were missing, the science was contested, and the failures were genuinely unforeseeable. That version would be a tragedy, and no one would be to blame.

That is not the version we are in.

We are in the version where the maps are published every Thursday, the ocean is measured every day, the crop forecasts arrive every month, and every one of them is free. The hunger of 2027 will be the most thoroughly forecast hunger in all of human history.

Which means that every hunger crisis this El Niño produces will not be an accident of weather. It will be a decision, made in the autumn of 2026, by people and politicians who had the maps and the reports in front of them and should have known the consequences.

But a decision is not a fate determined. The same forecast that makes this hunger foreseeable is what makes it preventable, and that is the more important half of the sentence. The seed can be moved before the planting. The money can be released on the forecast rather than on the funeral. The grain can be kept moving. Every one of those things has been done before, and every one of them worked.

You can be one of the people who makes the politicians and policymakers look and listen, and who makes sure they know what to do to prevent these impending disasters.

Call your elected officials, state and federal, this week, next week and the week after that.

Ask simple questions:

1) have you mobilized the money needed for the 2027 El Niño lean seasons and hunger?

2) will it move before December when southern Africa plants?

Send them a copy of this article so you can be sure that they know that they have to act.

Only action by you and all of the people around you will prevent foreseeable famines from killing tens of thousands of innocent men, women and children in countries around the world far from the causes of climate change and the hunger that is accompanying this El Niño.

1. The Droughts That Are Already Here

Europe has just come through a summer in which half the continent was in some level of drought, much of Europe in severe drought, and the Rhine, the Loire, the Po and the Danube all fell to record lows. Across the Atlantic, the American drought peaked in March and April at the highest level for that time of year in the whole modern record, then eased somewhat while shifting to impact crops in the Plains.

Figure 3: The important thing about the European half is not how dry Europe was. It is WHY there is drought in Europe.

The drought that hit Europe this summer did not happen because the rain failed. Rainfall across western Europe was close to normal. What changed is that a hotter atmosphere caused by climate change pulls far more water out of the soil than it used to, so the same amount of rain no longer produces the same amount of water for crops to use and to create streams and rivers. Scientists examining the event found that the drying that caused Europe’s drought has become roughly eighty times more likely than in the slightly cooler climate without climate change. ¹⁰

Put plainly: the weather and rain did not have to get much worse for the crop losses to get much more severe. The available water moved. That is why this is a climate story and not just a bad summer weather story, and it is why the same pattern is likely to repeat and get worse as the world continues to warm.

Neither Drought Ends This Year

It would be reasonable to assume that a wet El Niño autumn where El Niño increases rain closes the drought chapter in parts of Europe and America. It does not, and the reasons differ on each continent.

In Europe, the thing driving the drought is not going away, because it is not the weather, it is not reduced rain. Rainfall varies year to year. The thirst of a warming atmosphere only goes in one direction. The same scientists who studied this summer, project that with a further 1.4°C of warming, drying conditions from an atmosphere demanding more evaporation will become roughly ten times more likely again in western Europe, and the odds of a soil-moisture deficit like this year’s will double. ¹¹

The most counterintuitive result happens in eastern Europe. The models there project more rain, not less in the coming years, enough that a year as dry as this last one would become only about a third as likely. Yet drought risk still rises. Warmer air pulls water out of soil and plants faster than the additional rainfall that is expected can replace it, so the ground dries out even as the skies get wetter. This is expected to continue and get worse as the world warms. More rain, more drought.

Figure 4. The water budget behind the paradox. Rain is only the income side of the ledger. When a warmer atmosphere raises the withdrawal faster than the deposit grows, the balance in the soil falls even in a wetter year. ¹² ¹³

“Climate change creates warmer air and increased evaporative demand, pulling water out of soil and plants faster than the expected additional rainfall can replace that water, so the ground dries out even as the skies and atmosphere get wetter. More rain, more drought.”

The reason that matters is a question of buffers. More rain sounds like reprieve, and where there is irrigation, it partly is. But eastern European farming is overwhelmingly rainfed. No canals, no allocations, no reservoirs sit between the atmosphere’s demand and the crop, so when evaporative demand outruns the rain, it comes straight out of the soil this year’s crop is rooted in. The same physics operates across the American West, this is not a European affliction, but there it meets infrastructure first, reservoirs, aquifers, rivers that have been dived for decades. As a result, a soil-moisture problem becomes a water-rights problem, which buys years of argument and some genuine protection. Eastern Europe has no such intermediary. The statistic arrives as a diminished harvest; in the same season the drought is measured.

This is not one study’s quirk. Research across Europe has found that the meteorological droughts themselves are getting shorter, while high evaporative demand stretches the total damage out longer, so that warm-season droughts and “flash droughts” are steadily replacing other kinds of drought. ¹² European river basins now lose 20 to 40 percent more water than their rainfall shortfalls alone would predict, because so much more water evaporates on the way into and down the rivers. ¹³

On the good side, long-term well records across Portugal, Spain, France and Italy show that water levels in wells are broadly stable, with one in five rising, showing better management has helped several semi-arid regions recover. ¹⁴ Europe’s groundwater is not uniformly collapsing. Europe’s soils and growing seasons are where the damage from increased evaporation is concentrated, and where crops are being decimated. Where people have managed water deliberately, the water has responded. That is worth holding onto and expanding.

In America, the drought will not end so much as relocate. That is a statement about geography, not about severity. Roughly two fifths of the country was in drought this August, and Lake Mead fell to a record low on the sixth of August 2026. ¹⁵ The same thirsty atmosphere is at work in America as in Europe; what moves is where this evaporative demand lands. The El Niño now building is forecast to push storms into the southern and southwestern United States and starve the northern tier, with below-normal precipitation and increased evaporative demand over the Northwest and north-central states set against a wetter southern tier. ¹⁶ For the Southern Plains, after eight years of drought, that is real relief.

But look at where the dryness is being sent. In Idaho, Oregon and Washington, a warmer El Niño winter means less precipitation overall and more of it falling as rain instead of snow, which means less snowpack, less summer runoff, and higher drought and wildfire risk the following summer. ¹⁷ Snowpack is a buffer of exactly the kind eastern Europe has never had: a reservoir that fills in winter and lets go slowly through the dry months. A warm winter does not drain it. It stops the snowpack from forming. That is a 2027 drought being manufactured this winter, aimed at the spring wheat country and the Pacific Northwest grain export belt, which are already among the driest parts of the country. Federal drought scientists caution that one wet winter will not refill the Southwest’s depleted reservoirs and aquifers regardless of how wet the El Niño makes it. ¹⁸

So, the honest summary is not that these droughts end in 2027. It is that the atmosphere gets thirstier on both continents and droughts continue, and what differs is only what stands in the way. Europe’s fields have nothing. America has snowpack and reservoirs, and this winter is forecast to take some of both, moving the shortfall north onto different farmers. Neither continent can expect a drought-free year 2027.

2. The El Niño in the Pacific Ocean That Is Still Coming

The second pressure has not arrived yet, and it is the largest.

A very strong El Niño is now forecast for this winter, with better than nine-in-ten odds, and a real chance of being the strongest event measured since records began in 1950. ¹⁹ El Niño is not a storm. It is a large-scale rearrangement of where rain falls on Earth, and it moves rain away from croplands in southern Africa, Southeast Asia and Central America, in a pattern that has repeated for a century and a half.

Figure 5. Pacific sea-surface temperature anomalies. NASA Worldview.

It deserves saying that this rearrangement is not all bad news. El Niño may actually help some of the places currently in drought. Forecasts suggest Western and Southern Europe turn wetter this autumn, and that the American Southwest and Southern Plains get an unusually wet winter. After years of drought in both regions, that would be a genuine mercy, and some farmers will end up owing this El Niño a debt.

But three things are true at the same time. Rain in October does not restore a crop harvested in September. The relief is forecast for the wrong regions on the wrong schedule: wetter in the Mediterranean, drier over exactly the northern European grain belt now deciding what to plant. Heavy rain arriving on hillsides burned bare this summer produces floods, not recovery.

The relief is real, regional, and late. It arrives for some farmers after the crop it might have saved is already in the trash, and it arrives in the same season the tropics, which already have had too little rain to grow their crops, have to plant their single corn crop of the year. The bottom line is that the 2026-7 El Niño is going to be severe and cause crop losses where there are the smallest buffers to prevent hunger and even starvation. We know where this is going to happen, where crops are going to fail, where people will be hungry.

3. The Fertilizer Squeeze

The third pressure is quieter and slower. Fertilizer costs spiked earlier this year after the closure of the Strait of Hormuz. Nitrogen prices have since fallen back substantially, and that part of the crisis has genuinely eased. ²⁰

However, the phosphate shortage which was also caused by the closure of the Strait of Hormuz hasn’t eased. The closure lifted the price of both nitrogen and phosphate fertilizers, but only the price of nitrogen (urea fertilizer) has come back down, because the two fertilizers are produced differently.

Nitrogen is manufactured: urea is made from natural gas, so when the energy shock passed the price of urea followed it down, and new capacity to manufacture urea can be built anywhere the gas is cheap.

However, phosphate is mined, in only a handful of countries that hold most of the world’s phosphate-rich rock, and two of them have stopped selling phosphate. China, the largest exporter, suspended phosphate exports in December to protect its own spring planting and has extended that suspension at least through August; traders do not expect phosphate shipments from China to resume quickly. Russia, another of the five countries that dominate the trade, has capped total fertilizer exports at 20 million tonnes from June to November, including some 7 million tonnes of the compound fertilizers that carry phosphate. Between them they have taken a large share of globally traded phosphate supply off the market, and there is nobody positioned to replace this shortage. Morocco and Saudi Arabia are already running flat out, and no significant new capacity is expected before 2027. ²¹

Then there is the acid. Turning rock into phosphate fertilizer takes sulfuric acid, and sulfur is not mined to order. It is stripped out of crude oil and sour gas during processing. This means how much of sulfur exists depends on how much oil and natural gas the world is refining, and where. The Gulf is traditionally one of the largest sources of sulfur production, which is why the same war that closed the Strait of Hormuz also tightened the world supply of sulfur, since these countries have not been processing typical amounts of oil and sour gas since exports are restricted. China also has restricted exports of sulfuric acid alongside the restriction on the export of phosphate fertilizer itself.

Ammonia and urea supplies on the global market have rebounded since the spring, as shipping fears receded and China loosened a separate restriction on the export of urea. The global sulfur supply has not similarly rebounded. The price of sulfur went from $155 a tonne to $400 a tonne between late February and early May 2025. A plant with its own phosphate rich rock and its own manufacturing facility still cannot make phosphate fertilizer without sulfuric acid. That is not hypothetical. In May, the company Mosaic which mines most of America’s phosphate rock and makes most of its phosphate fertilizer, halved output at its plants in Bartow, Florida and Faustina, Louisiana and scaled back further production in Brazil, explicitly to avoid buying sulfur at the high prices being asked due to the closure of the Strait of Hormuz. By July it was reducing two more American plants as its sulfur inventories fell and prices remained high. The largest phosphate producer in America is switching capacity off in the middle of a phosphate fertilizer shortage, because the sulfuric acid needed to produce it is too expensive for Mosaic to make any profit on the production of phosphate fertilizer. Washington has acknowledged the squeeze in the plainest terms available to it. On 29 June, 2026, the President signed a proclamation declaring an emergency over threats to the availability of phosphate fertilizer, and suspended anti-dumping and countervailing duties on Moroccan phosphate for up to eight months, citing conflicts in fertilizer-producing regions and trade actions by major producing countries. ²¹

So, the price of phosphate fertilizer stays high, and farmers around the world put less of it on their fields. ²¹ Crops feed on the reserve of phosphate built up in the soil over past seasons, not on what was spread this spring. As a result, a farmer who skips an application of phosphate fertilizer this year because of the high prices sees no penalty at this year’s harvest, because the crop draws down its nutrient savings from the soil instead. The penalty for not applying phosphate because it is too expensive this year arrives in the reduced crop production in the next two to four years. The reduced productivity of the fields arrives first for the farmers whose small crops and limited resources forced them to skip applying phosphate in the first place. ²¹ The failure to be able to afford applying phosphate causes hunger for years to come after the phosphate fertilizer is not applied to fields.

Figure 6. Why the phosphate bill arrives late. The crop draws on a reserve accumulated over past seasons, so the harvest immediately after a skipped application looks normal. The reserve, not the yield, absorbs the loss first. ²¹

Three pressures, then: a drought whose cause is strengthening as climate change becomes more severe, a Pacific Ocean pattern that peaks this winter, and a nutrient input squeeze that plays out slowly over years. Now is the time to take preventative action and for you to reach out to your federal and state lawmakers and push them to prevent a lean rain year from turning into a humanitarian disaster.

Where to Watch All of This Yourself

• American drought, every Thursday: U.S. Drought Monitor and Drought.gov

• European drought, every ten days: European Drought Observatory

• The Pacific: NOAA Climate Prediction Center

• The crops: JRC MARS Bulletin and USDA WASDE

• The hunger: FEWS NET and WFP HungerMap Live

Sources

1. U.S. Department of Agriculture, Economic Research Service, announcement of September 2025 discontinuing the annual Household Food Security in the United Statesseries after the 2024 report (ERR-358, published 30 December 2025).

2. European winter crop establishment analysis, August 2026: a record 6.4 million hectare EU rapeseed planting intention against a drilling window closing in mid-September. On the wider oilseed complex: IFPRI puts vegetable oils at roughly 10 percent of daily caloric food supply, about 300 kcal per person per day, second only to cereals, and notes they are an essential cooking item particularly for poor consumers unable to shift to butter or animal fats; the same analysis puts the 31 million tonnes of vegetable oil converted to biodiesel globally at a calorie equivalent of feeding more than 320 million people a year. FAO projected that 44 of every 100 additional calories in developing countries to 2030 would come from vegetable oils. Palm, soybean, rapeseed and sunflower account for about 88 percent of global edible oil consumption, with palm alone some 46 percent of traded volume and Indonesia and Malaysia supplying 88 percent of that; Marex estimated Indonesian and Malaysian output could fall by about 4.53 million tonnes on the pattern of past strong El Niño years. FAO’s vegetable oil price index reached a four-year high in July 2026 even as rapeseed and sunflower oil quotations eased on expectations of larger 2026/27 supplies.

3. Oxford Economics, July 2026, estimating a weather contribution of up to one percentage point to European food inflation in 2027.

4. Urban Institute, Food Insecurity Remained High in 2025, As Safety Net Cuts Loom, March 2026, reporting the December 2025 round of the Well-Being and Basic Needs Survey: 24.2 percent of adults aged 18 and over, and 27.7 percent of working-age adults, reported household food insecurity in the previous twelve months. More than 10,000 adults participated. Note that this is not the same measure as the discontinued USDA series: the WBNS uses the six-item short form of the Household Food Security Survey Module and reports the share of adults, where USDA used the eighteen-item module and reported the share of households, last putting it at 13.7 percent, or 47.9 million people, in 2024. The two figures are not interchangeable, and the Urban Institute presents its survey partly as a substitute for the federal data that is no longer collected.

5. The Greater Boston Food Bank and Mass General Brigham, Massachusetts Food Access Report: Hunger on the Rise, 7 April 2026: 40 percent of Massachusetts households in 2025, up from 37 percent in 2024 and 19 percent in 2019, with very low food security rising from 6 percent to 25 percent over the same period. Note that GBFB’s statewide survey and USDA’s Current Population Survey use different instruments and are not directly comparable.

6. World Food Programme on Sudan’s wheat import dependence; Joint Research Centre INFORM Warning country risk scores, 15 June 2026. On the 2026 season: Agence France-Presse, reporting from Gadaref and Al-Jazira, 22 August 2026, including the agriculture ministry’s guidance on drought-resistant varieties and delayed planting, and farmers’ accounts of planting costs quadrupling. Reservoir level at the Grand Ethiopian Renaissance Dam six metres below July 2025, per University of Tübingen data; Blue Nile basin rainfall as low as half of normal this summer, per the Climate Hazards Center at UC Santa Barbara, with below-average rainfall forecast for the remainder of the season. FAO describes the season as unfolding under heightened climate risk driven by El Niño conditions. On the El Niño–Nile relationship: Elfatih Eltahir, Massachusetts Institute of Technology. On observed conditions elsewhere: NOAA Climate Prediction Center, Global Weather Hazards Summary, 20–26 August 2026. On Central America: FEWS NET Guatemala Key Message Update, August 2026, classifying the poorest households of the Dry Corridor and targeted areas of Alta Verapaz and the Western Highlands in Crisis (IPC Phase 3) through September, after much of the Dry Corridor recorded only five to eight days with rainfall between 20 June and 20 July. FEWS NET Honduras Key Message Update, July 2026, citing the Secretariat of Agriculture and Livestock (DICTA-SAG): approximately 12,532 manzanas (8,772 hectares) of maize and more than 4,400 manzanas (3,080 hectares) of red beans lost to adverse weather as of July, with delayed rainfall onset postponing planting into July in Copán and other western departments. FEWS NET El Salvador, April 2026, expecting Crisis outcomes to expand through September in the Eastern Dry Corridor among smallholders entering 2026 with limited reserves. On the global total: World Food Programme, August 2026, estimating that 49 million additional people will be pushed into acute hunger by the strengthening event, with Central America and southern Africa expected to bear the brunt and significant impacts also forecast in eastern Africa and South and Southeast Asia; WFP reports anticipatory plans active in more than fifty countries and expects the event to peak between September and December 2026, with effects running through 2027 via subsequent harvests. On the existing caseload: Global Report on Food Crises 2026, published 24 April 2026 by the Global Network Against Food Crises, estimating 266 million people in 47 countries and territories at IPC/CH Phase 3 or above in 2025, or 22.9 percent of the analysed population, up from 22.3 percent in 2024. More than 39 million were in Phase 4 (Emergency) and 1.4 million in Phase 5 (Catastrophe) across Haiti, Mali, the Gaza Strip, South Sudan, Sudan and Yemen, a ninefold rise in the Phase 5 population since 2016. Famine was confirmed in parts of Gaza and Sudan. The 2024 comparison covered 53 countries and 296 million people, so the year-on-year fall in the absolute number reflects fewer countries analysed rather than improvement; the report notes the lowest data availability in a decade, attributed to funding cuts and restricted access. On famine risk: IPC, Sudan Acute Food Insecurity Situation for February–May 2026 with projections to January 2027, recording nearly 135,000 in Phase 5 (Catastrophe) and projecting a rise to 200,000 across fifteen areas in June–September, with a risk of Famine identified in fourteen areas of North Darfur, South Darfur and South Kordofan under a reasonable worst-case scenario of intensified conflict and further restrictions on humanitarian access. FAO and WFP, Hunger Hotspots 2026, June to November outlook: a risk of Famine persists or has been identified in Sudan, South Sudan, the Gaza Strip and Somalia. IPC South Sudan projection update, April–July 2026: 7.8 million at Phase 3 or above, 73,000 in Catastrophe, with risk-of-Famine analysis conducted for Luakpiny/Nasir, Ulang, Nyirol and Akobo. Note that in February 2026 FEWS NET moved Kadugli and Dilling from Famine Possible to Emergency with populations in Catastrophe, stating explicitly that this reflected technical thresholds no longer being met rather than substantive improvement, the reasons including the scale of deaths that likely occurred under siege and large-scale displacement out of the towns.

7. FAO and WFP, on anticipatory action reaching approximately three million people during 2023–24, with coverage remaining well below identified needs.

8. FAO and WFP Joint Anticipatory Action Appeal, launched 18 June 2026 and running to March 2027, the agencies’ first joint anticipatory action appeal: $202 million to protect up to 8.8 million people across 22 priority countries. The agencies describe themselves as already positioned to act for 1.2 million people, with a further $167 million required to reach the remaining 7.6 million. Note that the appeal documents do not state whether the roughly $35 million difference represents funds already committed or existing operational capacity, and no public figure for contributions received against the appeal could be located.

9. Congressional Budget Office and Food Research & Action Center analyses of the 2025 reconciliation law’s SNAP provisions.

10. World Weather Attribution, Increasingly hot Europe faces more severe droughts and growing challenges for water and land management, 23 July 2026. The comparison is to a climate 1.4°C cooler.

11. World Weather Attribution, 23 July 2026. Projections for a further 1.4°C of warming: intense evaporative conditions around ten times more likely in the western region, and the likelihood of a comparable soil moisture deficit doubling. In the eastern region the same models project rising six-month precipitation, with totals as low as 2026’s becoming roughly a third as likely, while warming still raises overall drought risk.

12. The rise of compound warm-season droughts in Europe, Science Advances. The study finds the duration of meteorological droughts declining while high atmospheric evaporative demand extends total event duration.

13. Massari et al., Evaporation enhancement drives the European water-budget deficit during multi-year droughts, Hydrology and Earth System Sciences, 2022: runoff deficits of −20% to −40% beyond what precipitation deficits predict, with annual evaporation rising 11–33% during droughts.

14. Multi-decadal groundwater observations reveal surprisingly stable levels in southwestern Europe, Communications Earth & Environment, 2024: of 12,398 wells across Portugal, Spain, France and Italy (1960–2020), 20% rising, 68% stable, 12% declining.

15. U.S. Drought Monitor and NIDIS, 4 August 2026: 40.7 percent of the United States and Puerto Rico, and 48.5 percent of the lower 48 states, in drought. USDA Weekly Weather and Crop Bulletin, 11 August 2026, on Lake Mead’s record-low surface elevation of 6 August 2026.

16. NOAA Climate Prediction Center long-lead seasonal discussion, 2026: El Niño teleconnections ramping up through autumn 2026 and winter 2026–27, with below-normal precipitation over the Northwest and north-central states set against above-normal precipitation over the southern tier.

17. USDA Northwest Climate Hub, El Niño in the Northwest: What Can We Expect?, on reduced snowpack, decreased runoff and summer water availability, and increased drought and wildfire risk.

18. National Integrated Drought Information System, on the insufficiency of a single wet winter to refill drought-depleted Southwestern reservoirs.

19. NOAA Climate Prediction Center ENSO update, 13 August 2026: a greater than 90 percent chance of a very strong event this autumn and winter, and a 69 percent chance of an event exceeding any since 1950.

20. DTN Retail Fertilizer Trends, weekly survey of US agricultural retailers, March to August 2026. Urea averaged $674 per ton in the second week of March, peaked at $865 in the first week of May, and stood at $678 in the second full week of August. DAP averaged $851, $914 and $917 on the same dates. Only weeks reporting both nutrients are plotted.

21. World Bank Commodity Markets Outlook, 2026; CoBank Knowledge Exchange, August 2026, citing Argus projections of a 13 percent fall in global phosphate consumption this year. On the U.S. emergency: Proclamation 11038, Declaration of Emergency and Authorization for Temporary Duty Free Importation of Phosphate Fertilizer from Morocco, 29 June 2026, invoking section 318 of the Tariff Act of 1930 and declaring an emergency with respect to threats to the availability of sufficient supplies of fertilizers to meet expected agricultural demand; White House fact sheet and USDA press release of the same date. The proclamation notes that farmers apply more than half their annual phosphate between autumn and early spring, and that domestic production cannot support demand. In February 2026 the administration had already invoked the Defense Production Act over domestic production of elemental phosphorus. On the supply side: China’s National Development and Reform Commission met producers on 11 December and suspended phosphate fertilizer exports in principle until August 2026 to secure domestic supply for spring planting, withdrawing an estimated 7 to 9 million tonnes, or roughly 30 percent of globally traded volume. Analysts noted that China’s sixty-day inspection and quarantine period could delay cargoes until October even once the suspension lapses. The World Bank notes that the Hormuz route carries a large share of global sulphur and ammonia shipments, both required to manufacture DAP fertilizer, or diammonium phosphate, and that Morocco’s OCP accelerated maintenance at its phosphate plants in response. The two inputs have since diverged: Middle East ammonia fell from about $680 to $560 per tonne across July 2026 and the July Tampa contract settled at $665, roughly $50 above pre-conflict levels, while sulphur remained exceptionally elevated and continued to carry cost through the phosphate chain. Russia’s restriction is separate and concurrent: a government decree set an export quota of 20 million tonnes of mineral fertilizer for 1 June to 30 November 2026, of which more than 8.7 million tonnes is nitrogen and upwards of 7 million tonnes compound fertilizer, extending quota arrangements in place since 2021 (Interfax; Tridge). Trade reporting describes licensing and quota regimes in both China and Russia as limiting spot availability of DAP and MAP fertilizer, Monoammonium Phosphate. Analysts note that Morocco and Saudi Arabia are already at capacity, with new phosphate production not expected before 2027–28. China has also restricted exports of sulphuric acid, the reagent used to convert phosphate rock into phosphoric acid. On Mosaic: the company announced curtailments on 11 May 2026, halving rates at Bartow and Faustina and scaling back in Brazil, and extended reductions to its Riverview and Uncle Sam plants in July as sulphur inventories fell (Argus Media; World Fertilizer; Glacier FarmMedia). Chief executive Bruce Bodine described the move as a temporary way to avoid buying incremental sulphur at prevailing prices. On the first-quarter call management put marginal sulphur near $1,200 a tonne against realised costs nearer $540, making some production uneconomic; the phosphate segment recorded a $48 million operating loss on a $280 million rise in raw material costs. Mosaic accounts for roughly 64 percent of US phosphate rock mined and about 80 percent of domestic phosphate fertilizer manufactured. Between 2018 and 2022, China, Morocco, the United States and Russia together accounted for close to 80 percent of world ammoniated phosphate production (IFPRI).

Aug 13, 2026

I am delighted to be joined by Tom Harris from Climate Uncovered Substack for this article. Tom is a climate advocate who is interested in letting the data do the talking and bringing complex scientific research papers to a wider audience. I highly recommend his Substack and hope you subscribe for more of his highly insightful and relevant content.

Physics sets the shock. Policy sets how many people go hungry. Slowly though food insecurity, chronically through lack of calories, and acutely through the starvation that kills. Every number and figure we cover below carries a source, and a confidence level wherever the source states one, because the impacts on global and domestic hunger in a 2°C world due to climate change does not need to be exaggerated, the truth is scary enough.

The 2ºC Dinner Table is Already Being Set

In 2025, for the first time in the history of the world’s hunger monitor, famine was confirmed in two countries in the same year. Parts of Gaza. Parts of Sudan, with four more countries on the brink of famine. Across six countries, 1.4 million people ended 2025 in the category the international humanitarian classification calls Catastrophe, the level at which starvation and death are no longer projected but counted body by body. That figure is nine times higher than it was a decade ago.

All of that happened at about 1.4°C of warming, in a world that still had a functioning humanitarian system.

Those figures measure acute hunger, the crisis level tracked by the Integrated Food Security Phase Classification, when a shock strips households of the last buffers between survival and starvation and famines are declared. This is different from chronic hunger, the persistent shortage of calories and nutrients that the UN Food and Agriculture Organization (FAO) estimates affect roughly 645 million people worldwide. It is also different from food insecurity, the broadest category, which captures people who have had to skip meals or go without food for a day or two. Because separate systems measure these three categories, they can move in different directions: chronic hunger can decline in the same year acute hunger rises. This article tracks all three and does not let the more reassuring number obscure the more dangerous one.

Figure 1. Global hunger at 2°C

The threshold this article discusses is what happens to hunger when the world crosses to 2°C above the historic (1850-1900) baseline, and the fact that this is no longer a problem of the 2050s. With the rate of warming now measured rather than modelled, 2°C could arrive as early as 2036. That is inside the working career of every official who will decide how to prepare for it, inside the term of most mortgages held by the people reading this, and before the high school graduation of a child about to start their education this autumn.

So, the question is not whether a hotter world makes more people hungry. That was answered in Somalia in 2022, where a drought that human warming made roughly a hundred times more likely killed an estimated 43,000 people, about half of them children under five. The question now is how many more, where, and whether anybody with the power to change the answer has the courage and the empathy to do something about it in the short time that remains.

The published answers are neither reassuring nor agreed. Estimates of how many additional people will go chronically hungry in a 2ºC world run from roughly one million to roughly 183 million.¹² That is not a scientific inaccuracy or a rounding disagreement. It is a spread of nearly two orders of magnitude, due to the single most important fact in this article. The choice of how many people will go hungry depends on the policy choices our leaders make right now.

A range that wide usually means the underlying science is immature. Here it means something else, and that difference matters more than any single number in this article. The models producing the low end and the high end broadly agree on the physics. They agree on how much heat a maize plant can tolerate during pollination, where rainfall belts are shifting, and what a hotter atmosphere does to evaporative demand. What they disagree about is what we and our leaders will do about it: whether trade stays open when exporters take fright of a food shortage, whether smallholders can access credit and improved seed, whether safety nets hold through the price spikes that are coming, and whether adaptation finance arrives before the shock or only in the wreckage, hunger and famine after it. Numbers of this size are easy to read as fate. They are not. For once the climate change story is not about ever-increasing suffering, at least it needn’t be.

2ºC sets the size of the shock. Policy choices decide how much of the shock becomes hunger.

These projections need a baseline. The FAO’s 2026 edition of The State of Food Security and Nutrition in the World puts the number of people facing chronic hunger at approximately 645 million in 2025.³ This article uses that as a level but does not treat the reported direction of where hunger is going as established.⁴

The Horizon: When 2ºC Arrives

The last IPCC assessment, AR6 published in 2023, working from evidence published up to 2019, indicated that under a moderate emissions pathway (SSP2-4.5) the 2°C threshold would not be breached for several decades, most likely in the 2050s. That estimate has aged badly, and it has aged badly for reasons of scientific measurements rather than politics.

The most compelling new evidence concerns the Earth’s Energy Imbalance (EEI): the difference between the solar energy arriving at the top of the atmosphere and the energy radiating back to space. It is the cleanest single indicator of whether the planet is accumulating heat, and how fast. NASA’s CERES satellite record now shows an approximately fourfold increase in the rate at which the Earth system is accumulating energy since 2003, and independent analysis finds the imbalance has more than doubled in recent decades.⁵ Observed accumulation is running above the projection for RCP8.5, the high-emissions scenario in the IPCC latest report, even though human emissions are following a lower path than the RCP8.5 pathway assumed. Warming is actually occurring faster than anticipated. Impacts are arriving decades earlier than hoped.

A large part of the explanation is a better understanding of cloud feedbacks. Earlier assessments could not determine whether clouds would amplify warming or restrain it; by AR6 the sign was clear, and it was amplification, though the uncertainty band remained wide. Work by James Hansen and colleagues indicates the feedback is strongly positive and is the primary driver of the rise in EEI: the planet is dimming, losing cloud cover while the clouds that remains have grown less reflective. On this analysis, the climate’s sensitivity to greenhouse gasses sits at the high end of the IPCC’s likely range.⁶

Since the EEI is a rate of accumulation, rather than a measure of the stock already accumulated, a rising EEI means that the energy available to warm the planet is itself accelerating. That acceleration is now visible in every downstream measure: surface air temperature, sea surface temperature, ice melt, sea level rise, ocean heat content, atmospheric relative humidity and the intensity of extreme weather events.

A 2026 analysis by Foster and Rahmstorf⁷ found that global temperature is accelerating even after accounting for natural variability, including the El Niño–Southern Oscillation. Their analysis places the crossing of the 1.5°C threshold between 2026 and 2028. Extrapolating the emerging warming rate puts the crossing of the 2°C threshold as early as 2036. That is approximately two decades earlier than the IPCC’s last estimate, and that temperature increase is now close to unavoidable.

Figure 2. Two estimates of when 2ºC arrives, and what changed between them.

The consequence is that everything downstream of the changes in climate arrive faster. But plant breeding programs take years to move a trait from the trial plot to the farmer’s field. Irrigation schemes, grain reserves and cold chains take even longer to finance and construct. Trade agreements take longer still. On the timeline the physical evidence now supports for increasing global temperatures, 2°C arrives with very little warning and very little room for us to prepare.

The Calories That Quietly Disappear

The most useful single number in the agricultural literature comes from Hultgren and colleagues (2025), an eight-year study of six staple crops across 12,658 regions, capturing two-thirds of global crop calories. For every degree of global warming, the world loses approximately 120 kilocalories per person per day, about 4.4% of daily consumption. Critically, that estimate is net of adaptation: it already assumes farmers change cultivars, shift planting dates and adjust management in response to the climate they experience, despite the fact that farmers notoriously resist changes to their farming practices.⁸ In the West this temperature change and reduction in crop production mostly arrives as higher grocery prices. In the Global South it arrives as food insecurity and acute hunger.

Figure 3. Global staple-calorie losses, and how much of them adaptation actually recovers.

The timeline matters as much as the magnitude. The same study, written before the energy-imbalance work by James Hansen and colleagues described above, puts the global calorie loss at roughly 7.4% by 2050 and 11.2% by 2100 under moderate emissions once adaptation is included, and 24% by 2100 on a high-emissions pathway. The 24% figure is the one the media picks up; the mid-century figure is the one that should be of immediate concern, especially as its timing is on the optimistic side. It sits inside the planning horizon of every finance ministry, development bank and grain trader now operating, and is far harder to dismiss as a future scenario.

Adaptation’s contribution can be stated precisely: adaptation and income growth together offset 23% of global losses in 2050, rising to about a third by 2100. That is real mitigation. But it is also a ceiling, and even with the adaptation fully implemented, it leaves the majority of the loss standing in a model already generous about how quickly temperatures will rise and farmers will adjust. The offset is also wildly uneven. In Africa it reduces projected end-of-century losses from 16.0 to 11.6%; in North America, from 21.0 to 20.8%, which is within rounding of nothing at all. The regions with the most agronomic capital have the least room left to improve, because they already farm close to the optimum.

The distributional result inside the same study is worth stating, because it cuts against the intuition that climate damage is a poor-country problem. Projected end-of-century crop yield losses average 41% in the wealthiest regions and 28% in the lowest-income regions. The steepest damage falls at both extremes of the agricultural economy: on modern breadbaskets currently enjoying the world’s best growing conditions, and on subsistence communities depending on small harvests of cassava and millet.

Behind the smooth averages sits a distinctly un-smooth implementation mechanism. Tropical staples such as maize, sorghum and rice are already grown close to their ceiling temperatures. Yield does not decline gently as the thermometer rises; it holds, and then collapses, because the damage is concentrated in a narrow reproductive window. Agronomists have measured the critical thresholds for seed set at flowering with unsettling precision: roughly 37.2°C for rice, 37.9°C for maize and just 27.3°C for wheat.⁹ Above that threshold during the flowering period leads to sterilized pollen and failed fertilization. Grain filling is also reduced, so even a handful of hot days can destroy a whole crop. In this context, averages hide the real hunger smallholders in the Global South and smaller farmers in the Global North will experience.

The nutrients that go missing too

There is a second loss inside the first, and it does not appear in tonnage of food harvested. Elevated atmospheric CO₂ makes staple plants grow more and nourish less. Wheat grown under the concentrations of CO₂ expected in the atmosphere later this century loses approximately 9.3% of its zinc, 6.3% of its protein and 5.1% of its iron. Up to 35% of the global population depend on these crops for the bulk of those nutrients.

Translated into people rather than percentages, the effect is huge. Smith and Myers (2018) estimate that elevated CO₂ could push an additional 175 million people into zinc deficiency and 122 million into protein deficiency, while 1.4 billion women of childbearing age and young children, precisely the groups for whom iron status matters most would lose more than 4% of their dietary iron, by 2050. The burden of these nutrient losses concentrates in some particular countries rather than being spread in India alone, the same analysis finds roughly 50 million people newly zinc-deficient, 38 million newly protein-deficient, and 502 million women and children in the iron-vulnerable group.

Figure 4. Nutrient dilution under elevated CO₂, and the populations it reaches.

The IPCC’s Sixth Assessment reaches compatible conclusions and assigns them high confidence: protein content in rice, wheat, barley and potatoes falls by 6–14%, placing an additional 150 million people at risk of protein deficiency; zinc dilution adds 150–220 million to the zinc-deficient population and worsens deficiency in more than a billion people who are already deficient.¹⁰

A harvest can arrive exactly on target and still carry less nourishment than the same harvest did twenty years ago. The one number everybody watches is the weight of grain harvested, but weight of grain does not say how nutritious it is.

The outcomes show up in clinics, rather than on commodity exchanges, as stunted growth, impaired immune function, complications in pregnancy, higher child mortality. Hidden hunger is a term of art, but the hiding is literal. The metric everyone watches is the weight of grain, and the weight of grain is the one thing that does not change, how nutritious the grain is, on the other hand, does.

A One-Percent Grain Shortfall Becomes a Six-Percent Price Shock

Between a lost calorie and a hungry person sits a market, and markets for staple foods are peculiar. Demand is highly inelastic: people do not respond to a bad harvest by eating less bread. Supply is equally unresponsive within a season, because when one crop fails another cannot be planted for immediate replacement. The entire adjustment therefore falls on price.

The elasticities estimated by Roberts and Schlenker give the standard result.¹¹ Their framework identifies demand using current-period yield shocks and supply using past shocks working through inventories and finds commodity demand to be extremely inelastic. A shortfall of roughly 1% in global caloric availability (a shortfall of corn, rice, wheat, sorghum or cassava) therefore produces a price movement several times larger, on the order of 6%; 2% crop reduction becomes roughly 12% price increase, 5% shortfall of grains becomes a 30% increase in price, and a 10% crop reduction becomes a 60% price increase⁸ (Some UK farmers are reporting a 20% drop in wheat yield this year due to the unprecedented 2026 drought and heat). Realized responses depend on stocks-to-use ratios, trade policy, export restrictions and speculative positioning, all of which can push the multiplier higher. This is not a market failure. It is what a well-functioning market does when neither supply nor demand can adjust. The impacts are real though.

Until recently this was a modelled claim. It is now an observed one. Kotz and colleagues (2025) document 16 cases across 18 countries between 2022 and 2024 in which extreme heat, drought or rainfall drove measurable food-price spikes, many following weather beyond all historical precedent prior to 2020.¹² Olive oil across the European Union rose by around 50%. Japanese rice rose by around 48%. U.S. vegetable producer prices rose about 80% after drought across California and Arizona. Cocoa rose roughly 280% after a February 2024 West African heatwave that attribution analysis found had been made approximately 4°C hotter by climate change.

Figure 5. Documented price responses to recent climate extremes, and the mechanism that produces them. EU olive oil rose around 50%, and US vegetable producer prices around 80% after drought in California and Arizona.

The persistence of the increase matters as much as the spike. Earlier work by Kotz and co-authors, a collaboration between the Potsdam Institute and the European Central Bank, found that higher temperatures and related crop shocks raise food and headline inflation persistently for roughly 12 months after the event.¹³ A single hot season is not a single bad month at the checkout. It is a year of elevated prices working through processing, retail and wage bargaining and landing on those least able to pay.

Then comes the political amplifier. Faced with domestic price pressure, exporting governments often restrict exports. Each restriction is locally rational and globally destructive: it removes supply from the world market at precisely the moment the world market is short, prompting the next exporter to do the same. The food price crises of 2007–08 and 2010–11 both followed this pattern, and both did more damage through policy cascade than through the underlying weather events.

For a household in a wealthy country spending a tenth of its income on food, a 6% rise in staple crop prices is an irritation. For a household spending half or more of its income on food, which describes a large share of the population in the regions covered below, it is the difference between three meals and two, or two meals and one a day.

When the Breadbaskets Fail Together

World grain trade rests on a small number of exporting regions: the US Midwest, the Southern Cone of South America, the Black Sea and eastern China, with Australia and India significant in wheat and rice. Five regional breadbaskets account for roughly 60% of global grain production. The system’s implicit insurance policy has always been that these regions do not fail in the same year: their weather has historically been close to uncorrelated, so a drought in one is covered by a surplus in another.

The system’s insurance policy has always been that the breadbaskets do not fail in the same year. Climate change is quietly cancelling this policy.

Work by Gaupp and colleagues shows that warming erodes exactly this assumption.¹⁴ Amplified Rossby wave patterns and hemispheric heat extremes couple the growing seasons of distant regions, converting independent risks into correlated ones.¹⁵ The probability of simultaneous failure does not rise gently with temperature; it ratchets up. The return period for simultaneous climate risk across all five maize breadbaskets falls from roughly every 16 years historically to under every 2 years in a world at 2°C.

Table 1. The ratchet: how correlated failure risk changes with warming Rows are expressed in comparable units so they can be read against one another. Gaupp’s return periods are for all five breadbaskets simultaneously exceeding their climate thresholds; the 1.5°C intermediate values are 21 years for wheat, 9 for soybean and under 3 for maize, which is why this is a ratchet rather than a slope. Tigchelaar’s figure rises to 86% at 4°C. McKinsey’s figures refer to the decade centered on 2030 under a high-emissions scenario.

Tigchelaar and colleagues put the same risk in the units a trader watches. The top four maize exporters are the United States, Brazil, Argentina and Ukraine, and between them they account for 87% of global maize exports. The probability that all four suffer simultaneous production losses greater than 10% in a given year is presently close to zero; at 2°C it rises to 7%, and at 4°C to 86%. At 2°C the projected decline across those four amounts to some 53 million tonnes, equivalent to 43% of current global maize export volume.¹⁶

The McKinsey Global Institute’s 2020 analysis puts the same risk in the units a planner uses. The probability of a greater than 15% global grain production shock roughly doubles by 2030, from about 1-in-100 to about 1-in-50 in any given year, giving an 18% chance of occurrence between now and 2035. A smaller but still severe shock of a 10% loss moves from a 6% to an 11% annual probability, and from a 46 to a 69% chance of occurring at least once in the next decade.¹⁷ That is the decade in which 2°C plausibly arrives.

A 2024 study by Zhang and colleagues puts numbers on the physical mechanism. At approximately 2°C, the likelihood that the world’s breadbaskets concurrently expose more than half their cropland to at least five days of extreme heat during the reproductive window in a typical year, not an exceptional one, rises from virtually zero in the 1984–2013 baseline to 43% for maize, 33% for rice and soybean, and 27% for wheat.¹⁸

The Limits of What Adaptation Can Do to Protect Crops

The World Bank’s Turn Down the Heat assessment contains a threshold that is easy to skim past: below roughly 2°C, adaptation offsets a substantial share of agricultural losses; above it, it becomes rapidly harder and covers only a shrinking fraction of the damage caused by the increasing temperature. 2°C is where the returns bend downward. The same report finds that under 2°C, South Asian crop production may fall by at least 12%, requiring more than twice the imports to meet per-capita demand, with childhood stunting rising by around 35% by 2050.¹⁹ The IPCC projects soft adaptation limits for multiple staple crops in many growing areas at 2°C, particularly in the tropics.

The Hultgren figure quantifies this from the other direction: adaptation and income growth together offset 23% of global losses at mid-century. Call that the best available estimate of what a farming system doing its rational best can recover. Note what ‘adaptation’ means inside these models. It means switching cultivar, shifting a planting date, drilling a well, installing drip irrigation. Every one of these changes requires capital, credit, information, secure land tenure and, frequently, a water right. Smallholders who lack the reserves do not adapt the way the model predicts. At 2°C their adaptation increasingly means land abandonment and migration, which the models score as adaptation and the households experience as catastrophic loss.

A result that points the other way

Honesty requires including one finding that cuts the other way. Zhao and colleagues (2025) analyzed 92,096 hybrid-trial observations across the US Corn Belt between 2000 and 2020 and found that modern breeding delivered both yield gains and improved drought resistance; the trade-off breeders had long assumed would bind progress did not. Projected forward, continued genetic improvement of this kind could cut drought-related yield losses by approximately 17.8% by 2100 relative to older hybrids.²⁰

The caveats are substantial and belong in the same breath. This is one crop, in one region, against one stressor. It addresses drought during grain filling, not extreme heat during pollination, which is the mechanism that does most of the damage to crops. It assumes the historical rate of developing genetic gains continues, and that farmers can afford the improved hybrids, fertilizers and other inputs that go with the new hybrids each season. It does not overturn Hultgren, whose estimates are already net of adaptation, and it does nothing for maize, sorghum or millet farmers in the Sahel or rest of the Global South.

A Map of Who Pays First

The additional hunger caused by a 2°C world does not distribute itself evenly, and the concentrations of damage and hunger are severe. South Asia carries roughly 33 million additional people at risk of hunger. Sub-Saharan Africa carries roughly 15 million additional hungry, plus approximately 43 million pushed below the poverty line. Across Sub-Saharan Africa, South Asia and Southeast Asia together sits around 80% of total exposure to additional hunger.

Figure 6. Where the additional exposure lands, and the populations already carrying a nutritional deficit.

The reasons compound rather than merely accumulate. These regions have the highest dependence on rain-fed agriculture and the thinnest irrigation coverage. They spend the largest share of household income on food, so the price channel of crop loss to price increase described above hits them hardest. Their staples are already grown nearest to their thermal ceilings, so a given increment of warming does more damage. Their safety nets are thinnest and their fiscal space to expand them smallest. Their populations are also the youngest, which means the nutritional deficits fall during the developmental window in which they may do permanent harm.

An independent analysis by Oxfam approaches the same map from the opposite end. Rather than ranking by modelled exposure, it ranks by revealed damage: the ten countries with the most extreme-weather-linked UN humanitarian appeals since 2000, where climate was assessed as a major contributor. They are Somalia, Haiti, Djibouti, Kenya, Niger, Afghanistan, Guatemala, Madagascar, Burkina Faso and Zimbabwe. Across those ten, acute hunger rose from 21.3 million people in 2016 to 47.5 million in 2021, a 123% increase in six years, with 18 million at emergency levels of hunger or worse.²¹

The same analysis supplies the arithmetic of responsibility that the modelling literature tends to leave out. Those ten countries account for roughly 0.13% of cumulative global carbon emissions. The G20 accounts for around 77%. The distribution of the damage is close to the inverse of the distribution of the cause, and no adaptation figure anywhere in this article alters that ratio. So far, the G20 are not fully meeting funding pledges to pay for adaptation in low-income countries hit hardest by climate change.

That is where the nutrition findings and the geography findings meet. The populations most exposed to calorie loss are, to a close approximation, the same populations carrying the nutrient losses described earlier. The same households absorb both shocks.

A third layer then sits on top: heat attacks food after harvest as well as before it. Higher temperatures accelerate spoilage and raise the risk of mycotoxin contamination, above all aflatoxin, a carcinogenic mold toxin in maize and groundnuts that worsens under exactly the combination of heat and drought stress a 2°C world delivers. Food-borne pathogen risks rise wherever cold chains are absent. Calories can be grown, harvested and delivered, and still not be safely eaten. Utilization is the pillar of food security that receives the least attention and is among the most temperature sensitive.

The Broad Base: Food Insecurity Now and at 2ºC

Next to chronic and acute hunger sits the third measure, and it is the broadest. The FAO’s Food Insecurity Experience Scale asks people directly whether, in the past year, they have had to eat less, skip meals, or go without food entirely for lack of money. By that measure roughly 2.3 billion people, about 28% of humanity, were moderately or severely food insecure in 2024, several hundred million of them severely food insecure, meaning they had at times run out of food or gone whole days without eating. The figure has barely moved since 2020 and sits roughly 680 million higher than in 2015. In sub-Saharan Africa it exceeds 58%, more than one person in two do not know where their next meal is coming from.²²

Food insecurity measures access, not calories or crisis, and that is what makes it the base of the other two measures of hunger: chronic hunger and acute hunger. It captures the enormous population living one shock away from starvation: not necessarily undernourished on the calorie accounting, not yet inside the IPC’s crisis phases, but unable to count on enough safe and nutritious food from one day to the next. Roughly 2.6 billion people already cannot afford a healthy diet at all, before any further warming arrives.

2°C carries no published headcount for this measure, for the same reason it carries none for acute hunger: food insecurity is observed in surveys, not projected by crop models. But its transmission belt has already been traced. Access is governed by price, and a 1% shortfall becomes a price rise several times larger, with a single hot season lifting food inflation for a year afterwards. A durable rise in staple prices does not simply tighten budgets; it moves people down the scale, and from severe insecurity into the acute pipeline the next section describes. A world that lets food insecurity spread is quietly filling the reservoir from which the next decade’s famines will be drawn.

The Sharp End: Will Acute Hunger Get Worse?

Everything modelled so far is chronic hunger: the slow arithmetic of calories per person per day, nutrient density, prices and purchasing power. That is what the crop literature is built to produce, and it is the measure behind the one million to 183 million range. It is not the measure that produces famines. Acute crisis tracks shocks rather than averages: what happens when a harvest fails, a currency collapses, a road closes or a war starts, and a household with no reserves has nothing between it and hunger or even famine.

Figure 7. The acute hunger picture in 2025: how severe, and what drove it.

How much of this is already climate related?

The honest answer is: a large and rapidly growing share of acute hunger is caused by climate change, but not the largest share. In 2025 conflict and insecurity were the primary driver for 147.4 million people across 19 countries, that is 56% of the total. Weather extremes were the primary driver for 87.5 million people across 16 countries in 2025, about a third of the total acutely hungry people in the world.²³

The trend inside that second number is the one to watch, especially in the context of 2°C of warming. As a primary driver of acute food insecurity, weather extremes accounted for 15.7 million people in 2020 and 87.5 million in 2025, a fivefold rise over a period in which global mean temperature moved by a fraction of a degree. Much of it traces to the 2023–24 El Niño, which devastated southern Africa, but this only temporarily rose temperatures to 1.6°C. A near-certain super El Niño for 2026–27 is already in the FAO and WFP hotspot outlook, which warns of deterioration across 13 countries and territories, with northeast Nigeria and Somalia joining Sudan, South Sudan, Yemen and Palestine at the highest level of concern.²⁴

For the causal link itself, the closest thing to a controlled experiment is the Horn of Africa. Between late 2020 and 2022, five consecutive rainy seasons failed across southern Ethiopia, Somalia and eastern Kenya, the worst drought in four decades. Attribution analysis found the agricultural drought was made roughly 100 times more likely by human-caused global warming, and that an event of that severity would not have occurred in a climate 1.2°C cooler.²⁵ More than 20 million people were pushed into Crisis or worse, and in Somalia an estimated 43,000 excess deaths occurred in 2022 alone, about half of them children under five.²⁶

A child stunted in a famine year is stunted permanently. There is no later harvest that gives those years of nutrition back. Stunting causes a lifetime of physical, metabolic, and cognitive challenges.

What 2°C does to the shock machinery

No published study puts a number on how many people will be in acute hunger or famine in a 2°C world. The IPC classifies observed conditions rather than projecting them, and the humanitarian and conflict variables that dominate it are not forecastable on a multi-decade horizon. Anyone offering a precise figure is extrapolating past the evidence. What can be done instead is to ask, driver by driver, whether the machinery that generates acute crises runs more often at 2°C. On every input for which projections exist, it does, meaning it is much more likely that acute hunger will increase in a 2°C world.

· Drought exposure rises. The IPCC’s assessment finds an additional 350 million urban residents exposed to water scarcity from severe drought at 1.5°C, rising to 410.7 million at 2°C; across all populations the additional exposure roughly doubles between the two thresholds. Drying is amplified in the tropics, which is where the protracted crises already are.²⁷

· Extreme heat exposure rises far faster. Around 700 million people, approximately 9% of the world, would face extreme heatwaves at least once every twenty years at 1.5°C. At 2°C the figure is more than 2 billion, or 28% of the world.²⁸ Heat does not only cut yields; it destroys the casual agricultural labor that landless households in crisis contexts depend on for the income to buy food.

· Shocks become more correlated and more frequent. The breadbasket evidence above is the machinery of acute crisis stated in physical units. A correlated shock is worse for acute hunger than for chronic hunger, because the mechanism that normally rescues a failed region, imports from one that did not fail, is exactly what a correlated shock removes.

· The price channel is an acute-hunger channel. The twelve-month persistence of heat-driven food price inflation matters far more to a household in Phase 2 hunger than to one in a wealthy country. Phase 2 acute hunger is where the people who fall into Phase 3 severe acute hunger come from, and a year of elevated prices is how they get there.

· Recovery capacity is not being rebuilt. The same 33 countries have appeared in every edition of the crisis report since 2016, and more than 80% of the severely food-insecure now live in protracted crisis contexts. These are places where a bad harvest is no longer a setback but a cumulative loss, because the resilience that used to absorb it has gone. Each new shock lands on the residue of the last.

The estimate, and its uncertainty

Setting those pieces together supports a directional judgement rather than a point estimate, and it should be labelled as such: this is inference from the drivers, not a projection anyone has published.

Acute food insecurity will almost certainly be worse at 2°C than it is now, and the mechanism will be frequency in addition to the then level of the acute hunger crises. The expectation is not that the world settles at a permanently higher number, but that the peaks come closer together. A bad year arrives before the previous one has been absorbed. Set the El Niño impacts beside a fivefold rise in weather-driven acute hunger over five years and a fraction of a degree, and a doubling of the weather-attributable component by mid-century reads as conservative. Whether the total number of people suffering from acute hunger doubles depends overwhelmingly on conflict, which nobody can forecast and which can be expected to worsen where there is less food. Conflict, climate and the cost of food are the three factors driving rates of hunger around the world.

The number that deserves the closest watch is not the headline count but the Catastrophe population. It has risen ninefold since 2016 to 1.4 million people in 2025, and famine was confirmed in two places in the same year in 2025 for the first time. Catastrophe is where a shock meets the absence of a response, which makes it the most sensitive indicator of both trends at once: hazards increasing, humanitarian capacity contracting. The 2026 crisis report expects no improvement, reporting the fall in humanitarian food-sector funding at around 39% against 2024, with development assistance down at least 15%. In the past, peoples of the world rallied to provide food to people in famines, (remember Live Aid?), and the fact that this did not happen to any great extent in 2025 does not bode well for the hungry people when 2°C arrives.

All of this points at the conclusion the chronic analysis reached, arriving from the other direction. The increase in acute hunger in a 2°C world is real, already growing, and already measurable in mortality rates - people are dying. But the difference between a bad decade and a catastrophic one still runs through whether anyone is funded to respond, whether trade stays open when the shock lands, and whether countries in permanent crisis are given any means of rebuilding the buffer that used to absorb the next crisis. WFP puts the return on taking anticipatory action at three to seven dollars saved for every dollar spent. 2°C decides how often the shock arrives. It does not decide how many people it kills.

What It Means in America, Britain and Australia

Crops do not fail gradually. The relationship between temperature and yield established by Schlenker and Roberts is sharply asymmetric and piecewise: below a crop-specific threshold, around 29°C for maize, 30°C for soybean, 32°C for cotton, additional warmth is mildly beneficial; above it, damage accumulates several times faster.²⁹ Warming does not drive a smooth curve. It moves more of the growing season onto the wrong side of a hard biological edge.

Figure 8. The temperature threshold above which US crop losses accelerate, and what crossing it costs.

The modelled consequences for the United States are stark: reductions of 30 to 46% in maize, soy and cotton yields by 2100 under slower warming, and 63 to 82% under unmitigated warming. Hultgren’s global analysis projects maize losses of up to 40% for mid-latitude grain belts even after farmers adapt, and separate work by Liu and Basso projects net US soybean yields will be 26 to 57% lower by 2050.³⁰ The climate is already moving underneath the system: roughly half the country shifted a half-zone warmer between the 2012 and 2023 revisions of the USDA Plant Hardiness Zone Map.³¹

American hunger from climate change nevertheless arrives through a different door than in Malawi or Bangladesh. The United States is a net exporter with deep reserves and a highly capitalized farm sector; supermarket shelves will not go empty. What changes are the prices on the shelf and, upstream, crop insurance costs, farm incomes, land values and the fiscal exposure of federal support programs. Low-water events on the Mississippi already interrupt grain movement in drought years. There is also a trade dimension the Hultgren authors state bluntly: as northern latitudes gain and mid-latitudes lose, agricultural advantage is going to transfer to Canada, Russia and China.

The households that feel these price rises are the ones already spending a high share of their income on food, and they feel it in the same year that public budgets are absorbing disaster and insurance costs. It is the price elasticity described earlier, operating inside a wealthy country: a modest availability shock, a multiplied price response, and the impact concentrated on the poor and working classes.

Britain’s exposure is structurally different, because Britain is a net importer of food. It is the case where the correlated-failure mechanism described earlier lands directly on a wealthy population. Two of the three worst UK wheat harvests on record have occurred this decade, in 2020 and 2024, both caused by relentless rainfall and wheat self-sufficiency fell from 96% in 2023 to 79% in 2024, with imports the following autumn running at double the five-year average. Across 2020 to 2024 the cumulative shortfall exceeded seven million tonnes. Then 2025 failed from the opposite direction, in one of the hottest and driest springs and summers on record. 2026 looks like being even worse. A country short of wheat in a wet year and short again in successive dry ones is not having bad luck. It is experiencing climate variance; both can be bad and expected to get worse as 2°C is reached.³²

The second half of that exposure is that Britain’s suppliers sit in the regions this article has already identified as losing when the temperature rises to 2°C. More than 90% of imported vegetables come from the Mediterranean, with Spain being the largest single supplier. Spain has been in a historic drought. Headline self-sufficiency across all farming sectors has fallen from 86 to 78%. Domestic shortfall and supplier shortfall are not independent events; they are two expressions of the same warming, which is the correlation problem operating at the scale of one country’s shopping basket.³³ Europe is also the fastest warming continent meaning the effects of warming arrive here first.

Australia sits between the two and supplies the longest-running natural experiment. Its wheat yields stopped rising in 1990. Hochman and colleagues simulated 50 representative sites across the grain zone and found that water-limited yield potential fell 27% between 1990 and 2015, driven by declining rainfall and rising maximum temperatures, with elevated CO₂ preventing a further 4% loss. National yields nonetheless held flat, because technology closed the gap between what farmers achieved and what the climate permitted, lifting relative yields from 39 to 55% of potential.³⁴

There is a qualification here that cuts both ways. The Zhao breeding result is a Corn Belt study, and if any agricultural region on earth has the pipeline, the capital and the agronomic data to bend its own curve, it is this one, and that is the strongest reason to think the American numbers may land nearer the optimistic end of their range. The Australian record is what taking a bet on agriculture looks like after twenty-five years: it paid, and it paid by spending the yield gap it was closing, which is why Hochman’s authors end by asking whether the gains can continue. Neither case is a reason to expect the same in the Sahel or the rest of the Global South.

The Window That Is Still Open

If most of the distance between one million and 183 million is policy, then the policies that should be implemented need exploring. Four fronts carry most of the leverage.

First, keep the grain moving

This is the largest single lever in the literature and the cheapest to implement. The difference between a world in which staple crop trade stays open during a crop price shock and one in which exporters close their borders is a swing of approximately 53 million people at risk of hunger. It requires no new technology, no infrastructure and no transfer of funds. It requires only that governments refrain from doing the thing they are most tempted to do at the worst possible moment. The temptation is strongest when the stakes are highest and price shocks and grain availability are at their worst. The mechanism by which countries agree not to close their borders has to be built in advance: pre-commitments against export bans, exemptions for humanitarian purchases, and regional reserves. Such an agreement takes negotiation and time. The process to agree to open borders must start now, before the crisis really starts to gain momentum.

Second, breed for nutrition and break the monoculture

Biofortification now has a hard target rather than a general aspiration, and the nutrient figures above state exactly what it is up against. Zinc-biofortified wheat and rice, high-iron beans and pearl millet, and quality-protein maize are existing technologies with existing delivery channels. What they lack is scale.

Alongside biofortification sits diversification away from over-reliance on four dominant monocultures. Millets, sorghum, cassava and legumes carry higher thermal tolerance and, in several cases, better micronutrient density than the crops they were displaced by. Rebuilding seed systems, agronomic knowledge and consumer demand for them is slow work, which is an argument for starting now rather than for postponing, given we may have only a decade to work with.

Third, fund the response before the failure, not after it

The compressed timeline established at the start of this article is the whole argument here. If 2ºC plausibly arrives in the mid-2030s rather than the 2050s, then adaptation finance disbursed in the 2040s is finance disbursed after the event. Irrigation, on-farm and regional storage, decentralized solar-powered cold chains, processing hubs close to production, extension services and seed systems all take the better part of a decade to plan, fund and build. Front-loading is not a preference about spending profiles. It is arithmetic about lead times.

Fourth, build safety nets that trigger on a forecast, not on a famine

The FAO’s 2026 affordability data provided this warning: approximately 2.6–2.7 billion people, 31.3 to 32.5% of the world, already cannot afford a healthy diet before any 2ºC shock arrives.³⁵ A 6% staple price move applied to a household already priced out of adequate nutrition does not produce belt-tightening. It produces substitution into cheaper, less nutritious calories, which compounds the dilution effect described earlier.

The design feature that matters is automaticity. Safety nets that expand by discretionary political decision expand late, after the harm is done and after the politics have hardened. Safety nets indexed to observable triggers, such as food price indices, rainfall deficits, satellite-observed vegetation stress which release resources when certain thresholds are crossed. The technology for this exists and such systems are in operation in several countries. Expansion of these systems is a question of coverage and financing and getting countries to allow these inventions. Linking safety net response to climate signals such as Earth’s Energy Imbalance may even provide an outlook away from natural variability in weather.

Every famine that follows a forecast drought is a policy choice made a year earlier, by people who had the forecast and failed to act.

The direction of travel

Everything above is written as though these four fronts were open. But we do not live in an ideal rational world. Three of them are currently closing, not through climate stress but by political choice.

2025 brought the largest single-year contraction in development assistance ever recorded. Aid from OECD donors fell 23.1% in real terms to $174.3 billion dollars, back to roughly where it stood in 2015 when the 2030 Agenda was adopted. The United States drove three-quarters of the fall, cutting its own aid by 56.9%, the largest reduction by any provider in any year on record, with Germany, France, the United Kingdom and Japan all cutting foreign aid as well. Humanitarian assistance fell 35.8%, and core contributions to the UN system fell 27%. A further 5.8% fall is projected for 2026.³⁶

The incidence is not evenly spread, and the geography should look familiar. The OECD projects bilateral aid to the least developed countries falling by 13 to 25%, and to sub-Saharan Africa by 16 to 28%. That is the same map as the 80% of climate exposure described earlier, and as the 60% of the 2030 chronic-hunger projection that sits in Africa.

The operational consequences are already measurable. In Uganda, donor shortfalls forced the suspension of food assistance for a million refugees and ration cuts of up to 80%, alongside a rise in global acute malnutrition among refugees from 5.5% at the end of 2024 to 7.7% in late 2025.³⁷ Peer-reviewed modelling of the US cuts alone projects more than 14 million additional deaths by 2030 across 133 countries, including 4.5 million children under five.³⁸ None of those cuts were forced by anything other than a political choice.

None of this makes the four fronts wrong. It makes three of them counterfactual. Front-loaded adaptation finance and automatically triggered safety nets are the levers that set the multiplier between one million and 183 million, and both are being dismantled in precisely the years before 2ºC arrives, which raises the standing of the one lever that costs nothing. Keeping staple trade open is the only front that does not require money that is no longer there, and on the accelerated timeline it may be the only one that can still be built in time.

Conclusions - The 2ºC Warning

Food is not merely an economic commodity. It is the baseline condition on which every other form of order rests. Governments survive inflation in most things; they survive it in bread less often. The historical record of price spikes in staple grains is also a record of unrest, of conflict, of migration, of state capacity failing under a load it was not designed to carry. When the harvest fails in several places at once, the consequences do not stay agricultural for long.

The instinct in wealthy countries is to treat this as somebody else’s emergency, buffered by distance and purchasing power. That instinct misreads the system twice over. A globally integrated food network transmits scarcity through price, and price reaches everywhere the network reaches. And the modelling does not even grant wealthy agriculture the protection it assumes: the steepest projected losses fall on today’s best growing regions, because they have the least headroom left. Wealth buys position in the queue. It does not buy exemption from the queue.

Which returns the argument to where it began. One million to 183 million. The physics that sets the shock is now largely determined; the acceleration in the energy imbalance has taken the choice about whether 2°C arrives, and much of the choice about when, out of our hands. What remains entirely in our hands is the multiplier: whether borders stay open, whether safety nets hold during food price shocks, whether the money arrives before the harvest fails rather than after.

The 2°C dinner table is being set. What ends up on it, and who is seated at the table, is still, for now, ours to decide. It will not be for much longer.

Action you can take

The choices remaining are narrowing as 2°C approaches faster than anyone anticipated and are therefore far more consequential. They are being made now, in budget lines and trade rules and appropriations votes, by people who will not do anything about it unless somebody demands that they do. So, ask them.

The aid contraction described above was not a natural disaster; it was a set of votes and decisions made by elected officials. Votes can be taken again to reverse those decisions. The trade commitments that keep grain moving in a crisis are negotiated in advance or not at all. The safety nets that decide whether a price spike becomes hunger are written into domestic law by legislators who answer to constituents.

· Call your representatives, at all levels, and ask what they intend to do about food security before the shock arrives rather than after it.

· Ask what they are doing to restore humanitarian and development funding. Ask whether they will commit, publicly and now, against export bans in the next crisis.

· In the US, demand that USAID and SNAP funding be restored and other actions are taken in anticipation of the super El Niño we are about to experience. We have 150 years of data; we know where crises are going to occur. We should be stocking food and other humanitarian supplies before these crises happen.

Thanks for reading Climate Uncovered! Subscribe for free to receive new posts and support my work.

If you found this article interesting or useful, please tick the ❤️ button as it helps the algorithm point more people in this direction and gives us feedback on what subjects are most useful. Thank you!

Please also visit and subscribe to Mark’s 4Hunger.org Substack!

A note on the evidence used

The figures in this article come from three classes of evidence, and they do not deserve equal weight. Observed and attributed is the strongest: the price cases documented by Kotz and colleagues, the USDA hardiness-zone revision, the UK harvest record and the rise in the Earth’s energy imbalance are all measurements of events that happened. Assessed with stated confidence covers the IPCC’s nutrient-dilution findings, which carry an explicit high confidence label. Modelled projections, which include Hultgren, Smith and Myers, Gaupp, Tigchelaar, Zhang, Zhao, Hochman and the McKinsey analysis, are conditional on assumptions about emissions, adaptation speed, trade and income growth, and the conditions are the argument.

This is why the headline range is so wide. Studies at the low end assume open trade, effective adaptation and continued income growth; studies at the high end assume the opposite. Neither is wrong. They answer different questions, and reporting the range honestly is more useful than selecting a point estimate that flatters a position. Three things would move the numbers toward the optimistic end: sustained genetic gains of the kind Zhao documents, a durable regime holding staple trade open through a synchronized shock, and biofortification reaching scale faster than dilution accumulates. Three would move it the other way: a synchronized breadbasket failure arriving before adaptation finance does, a cascade of export restrictions, and the contraction in assistance persisting through the 2030s.

END NOTES:

1. Foster, G., & Rahmstorf, S. (2026). Global warming has accelerated significantly. Geophysical Research Letters, 53, e2025GL118804. https://doi.org/10.1029/2025GL118804. The 2036 date is an extrapolation of the emerging warming rate, not a coupled-model projection.

2. Rosegrant, M. W., Sulser, T. B., Dunston, S., et al. (2024). Food and nutrition security under changing climate and socioeconomic conditions. Global Food Security. IFPRI IMPACT model. ‘At risk of hunger’ is IMPACT’s analogue of the prevalence of undernourishment and is not identically defined, so comparison with FAO’s figure is indicative rather than exact.

3. The one-million-to-183-million range. Low end: a 2026 multi-model assessment puts the direct climate effect at 9.6 million additional people at risk by 2050 under business as usual, with lower estimates in the same family. Mid-range: IFPRI (2022), Global Food Policy Report, Ch. 14, roughly 70 million by 2050. High end: IPCC AR6 WGII, which assesses food prices up to a third higher by 2050, placing an additional 183 million people in low-income households at risk of chronic hunger. The width is a statement about model structure, chiefly the weight given to price transmission, trade and adaptation, rather than about the physics.

4. FAO, IFAD, UNICEF, WFP and WHO. The State of Food Security and Nutrition in the World. Rome: FAO, 2025 and 2026 editions. Source for the 645 million figure for 2025, the 2030 projection, and the cost-and-affordability series. The 18-country data gap is noted in the same editions.

5. 4Hunger.org, MILLIONS OF PEOPLE ARE STARVING:The major media headlines says global hunger is falling. Third year in a row. The number from the 2026 Global Report on Food Crises they are quoting describes last year’s rate of hunger, rests on the thinnest data in a decade (eighteen countries and territories aren’t even included…

Read more: 12 days ago · 7 likes · 3 comments · Mark Roberts

6. Mauritsen, T., Tsushima, Y., Meyssignac, B., Loeb, N. G., Hakuba, M., Pilewskie, P., et al. (2025). Earth’s energy imbalance more than doubled in recent decades. AGU Advances, 6, e2024AV001636. https://doi.org/10.1029/2024AV001636. Rate-of-accumulation figures derive from the NASA CERES satellite record.

7. Hansen, J. E., Kharecha, P., Sato, M., Tselioudis, G., Kelly, J., Bauer, S. E., … Pokela, A. (2025). Global Warming Has Accelerated: Are the United Nations and the Public Well-Informed? Environment: Science and Policy for Sustainable Development, 67(1), 6–44. https://doi.org/10.1080/00139157.2025.2434494

8. Hultgren, A., Carleton, T., Delgado, M., Gergel, D. R., Greenstone, M., Houser, T., et al. (2025). Impacts of climate change on global agriculture accounting for adaptation. Nature, 642, 644–652. Six staple crops, 12,658 regions; all loss figures quoted here are net of adaptation.

9. Smith, M. R., & Myers, S. S. (2018). Impact of anthropogenic CO₂ emissions on global human nutrition. Nature Climate Change, 8, 834–839. Includes the India-specific figures cited in the text.

10. IPCC (2022). Bezner Kerr, R., Hasegawa, T., Lasco, R., et al. Food, Fibre and Other Ecosystem Products. In Climate Change 2022: Impacts, Adaptation and Vulnerability, WGII contribution to AR6. Cambridge University Press. Protein, zinc and iron findings carry high confidence.

11. Roberts, M. J., & Schlenker, W. (2013). Identifying Supply and Demand Elasticities of Agricultural Commodities: Implications for the US Ethanol Mandate. American Economic Review, 103(6), 2265–2295. https://doi.org/10.1257/aer.103.6.2265

12. Kotz, M., Donat, M. G., Lancaster, T., Parker, M., Smith, P., Taylor, A., & Vetter, S. H. (2025). Climate extremes, food price spikes, and their wider societal risks. Environmental Research Letters, 20(8), 081001. https://doi.org/10.1088/1748-9326/ade45f

13. Kotz, M., Kuik, F., Lis, E., & Nickel, C. (2024). Global warming and heat extremes to enhance inflationary pressures. Communications Earth & Environment, 5, 116. Source for the ~12-month persistence and the 2022 European summer estimate.

14. Gaupp, F., Hall, J., Mitchell, D., & Dadson, S. (2019). Increasing risks of multiple breadbasket failure under 1.5 and 2°C global warming. Agricultural Systems, 175, 34–45.

15. Kornhuber, K., Coumou, D., Vogel, E., Lesk, C., Donges, J. F., Lehmann, J., & Horton, R. M. (2020). Amplified Rossby waves enhance risk of concurrent heatwaves in major breadbasket regions. Nature Climate Change, 10, 48–53.

16. Tigchelaar, M., Battisti, D. S., Naylor, R. L., & Ray, D. K. (2018). Future warming increases probability of globally synchronized maize production shocks. Proceedings of the National Academy of Sciences, 115(26), 6644–6649. https://doi.org/10.1073/pnas.1718031115

17. McKinsey Global Institute (2020). Will the world’s breadbaskets become less reliable? A case study accompanying Climate risk and response: Physical hazards and socioeconomic impacts (January 2020). Source also for the stock-to-use and price-doubling estimates quoted in the text.

18. Zhang, H., Yu, X., Qiao, F., & Li, C. (2024). Increasing concurrent exposure of global breadbaskets to reproductive heat extremes. Environmental Research Letters, 19, 094030. Probabilities are drawn from a high-emissions pathway at the point where it passes through roughly 2°C and 3°C.

19. World Bank (2013). Turn Down the Heat: Climate Extremes, Regional Impacts, and the Case for Resilience. Prepared for the World Bank by the Potsdam Institute for Climate Impact Research and Climate Analytics. Source for the South Asia production, import and stunting figures.

20. Zhao, H., Tack, J. B., Kluitenberg, G. J., Kirkham, M. B., Sassenrath, G. F., Zhang, L., et al. (2025). Concurrent improvements in maize yield and drought resistance through breeding advances in the U.S. Corn Belt. Nature Communications, 16. https://doi.org/10.1038/s41467-025-64454-3. 92,096 hybrid-trial observations, 2000–2020.

21. Oxfam (2022). Hunger in a Heating World. Oxford. Ten-country hotspot selection, the 21.3-to-47.5-million rise between 2016 and 2021, and the emissions shares (≈0.13% for the ten countries; ≈77% for the G20).

22. FAO. Prevalence of moderate or severe food insecurity, FIES, SDG Indicator 2.1.2. About 2.3 billion people (28.0%) moderately or severely food insecure in 2024, up from roughly 1.6 billion (21.4%) in 2015; sub-Saharan Africa above 58%. The ~2.6 billion unable to afford a healthy diet is from the same series; see note 33.

23. Driver breakdown and the 2020–2025 weather-extremes trend: Global Report on Food Crises 2026, see note 36.

24. FAO and WFP (2026). Hunger Hotspots: FAO–WFP early warnings on acute food insecurity, June to November 2026 outlook. Rome. Thirteen hotspots; highest-concern list as cited.

25. World Weather Attribution (2023). Human-induced climate change increased drought severity in the Horn of Africa. The agricultural drought of 2020–22 assessed as roughly 100 times more likely under present warming.

26. Excess-mortality estimate for Somalia in 2022 as reported by the World Health Organization and partners; approximately 43,000 deaths, around half of them children under five. Regional caseload figures from UNHCR and FEWS NET.

27. IPCC AR6 WGII, Chapter 6 (Cities, Settlements and Key Infrastructure), citing Liu, W., Sun, F., Lim, W. H., Zhang, J., Wang, H., Shiogama, H., & Zhang, Y. (2018). Global drought and severe drought-affected populations in 1.5 and 2°C warmer worlds. Earth System Dynamics, 9, 267–283. https://doi.org/10.5194/esd-9-267-2018

28. Dosio, A., Mentaschi, L., Fischer, E. M., & Wyser, K. (2018). Extreme heat waves under 1.5°C and 2°C global warming. Environmental Research Letters, 13, 054006. Exposure defined as at least one extreme heatwave in a twenty-year period.

29. Schlenker, W., & Roberts, M. J. (2009). Nonlinear temperature effects indicate severe damages to U.S. crop yields under climate change. PNAS, 106(37), 15594–15598. Source for the piecewise temperature–yield relationship, the crop thresholds and the 30–46 / 63–82% loss ranges.

30. Mid-latitude maize figure: Hultgren et al. (2025), note 5. Soybean projection: Liu, S., & Basso, B. (2021). Scientific Reports, 11, 11085.

31. US Department of Agriculture, Agricultural Research Service, Plant Hardiness Zone Map, 2023 revision, compared with the 2012 revision.

32. US Department of Agriculture, Agricultural Research Service, Plant Hardiness Zone Map, 2023 revision, compared with the 2012 revision.

33. Energy and Climate Intelligence Unit (2024). Climate and Food: Home and Away. Source for the imported brassica share, the Spanish drought exposure, and the fall in headline self-sufficiency across all farming sectors from 86 to 78% by volume.

34. Hochman, Z., Gobbett, D. L., & Horan, H. (2017). Climate trends account for stalled wheat yields in Australia since 1990. Global Change Biology, 23, 2071–2081. Fifty representative sites across the Australian grain zone; water-limited yield potential down 27% over 1990–2015, relative yields up from 39 to 55% of potential.

35. FAO et al., The State of Food Security and Nutrition in the World 2026, cost and affordability of a healthy diet. See note 1.

36. OECD (2026). Preliminary official development assistance levels in 2025. Paris, 9 April 2026. Total DAC ODA of USD 174.3 billion, down 23.1% in real terms; US ODA down 56.9%; humanitarian ODA down 35.8%; core UN contributions down 27%.

37. World Food Programme, 2026 Global Outlook and associated operational reporting: USD 13 billion required for 110 million people in 2026 against USD 6.5 billion raised in 2025. Uganda ration and malnutrition figures from Refugees International (2026), drawing on UNHCR data.

38. Cavalcanti, D. M., de Oliveira Ferreira de Sales, L., da Silva, A. F., et al. (2025). Evaluating the impact of two decades of USAID interventions and projecting the effects of defunding on mortality up to 2030. The Lancet, 406, 283–294. More than 14 million additional deaths by 2030 across 133 countries (95% UI 8.5–19.7 million).

How seventy percent of the world’s great aquifers came to be in decline, why Lake Mead hit its all-time record low three days ago, and what happens in Somalia, in Iran, in the Dry Corridor of Guatemala, and in the Imperial Valley of California when the water that grows a family’s food runs out from under them.

By USDA photo by Bob Nichols - https://www.usda.gov/media/blog/2014/02/11/southern-plains-climate-hub-seeks-address-three-huge-problems, Public Domain, https://commons.wikimedia.org/w/index.php?curid=89060994

The Ledger Nobody Kept

At some point between midnight and one in the morning on Friday, August 7, 2026, the surface of Lake Mead fell to 1,040.46 feet above sea level. It was the lowest the reservoir has stood since it was filled behind Hoover Dam in the 1930s, breaking the previous record set on July 28, 2022 by a margin that Reclamation puts at four one-hundredths of a foot and some wire reports put at a tenth. Either way it is a matter of inches, which is the sort of margin that makes a milestone easy to dismiss.¹ By Saturday morning, Lake Mead held only 7.0 million acre-feet against its capacity to hold 26.1 million acre-feet: That is only 26.8 percent full. Lake Powell, upstream, stood at 22.8 percent full. Federal forecasters expect Mead to set a new low record in nearly every month from here through 2028. Those projections do not yet account for the mandatory federal cuts now expected to be imposed on the states by the Interior Department.²

On July 31, the Interior Department published the framework that will govern the Colorado River for the next decade: up to three million acre-feet of mandatory reductions must be made in water use, reductions falling entirely on Arizona, California and Nevada. These cuts will be imposed by the Secretary of the Interior after the seven basin states failed to produce an agreement of their own.³ What those reductions do to the food supply is the subject of a later section. What they mean here is simpler. The decline of the largest reservoir in the United States has stopped being only a hydrological question and become an allocation question, about which farms lose water, and about who decides.

This is a matter of Western water politics and of food. It is also a matter of who gets to stay where they were born and who will be forced to move due to no water.

Six months earlier, in January, the United Nations University’s water institute published an assessment of the world’s water reserves that put the American Southwest in a context most Americans never see. Its title is Global Water Bankruptcy, and the accounting was blunt: seventy percent of the world’s major aquifers are in long-term decline. A broader test than the rate thresholds used in the peer-reviewed work discussed below, and a lower bar, but a bar that seven aquifers in ten now fail. More than forty percent of the water used for irrigation on Earth now comes out of aquifers that are being steadily drained. Three billion people live in places where total water storage, both surface and subsurface together is falling, and more than half of global food production is concentrated in exactly those places. Some 170 million hectares of irrigated cropland, an area larger than France, Spain, Germany and Italy combined, sits under high or very high water stress.⁴

That is the ledger. It has been running a deficit for decades, and almost no country keeps an honest publicly available record of aquifer water levels.

Figure 1. One shortage, four separate failures, one destination. Unlike a drought, aquifer depletion does not end. It works through four independent channels at once, each of which alone can take land out of production, and it arrives at the same place a season or two later. The compensation step is what separates the American case from the Somali one: where farmers are paid to fallow, the displacement does not occur.

4Hunger.org has spent this year documenting three shocks to the world’s food supply in sequence: the fertilizer supply collapse that followed the closure of the Strait of Hormuz in February (see Fertilizer Famine); the two-year American drought that drove the U.S. cattle herd to its smallest size since 1951 and the wheat crop to a 22 percent year-over-year collapse (see Parched Ground); and the record El Niño now building in the Pacific (see Scorched Harvest)⁵. Each of those is an event. Each has a start date, a peak, and a plausible end.

This article is about the thing underneath all of them, which is not an event at all. It is a slow, compounding, almost entirely invisible drawdown of the water the world’s food supply is built upon, and there is no foreseeable end to the draining of these water resources, because there is no end to the needs for water that are overusing these water reserves. An El Niño passes. A strait reopens. An aquifer that has been pumped past its recharge rate for sixty years does not recover in a good year, or in a good decade, or in some cases there is nothing at all that will enable the aquifer to recharge and rebound.

“A drought takes a harvest. A depleted aquifer takes the farm, and then the whole town, and then the option of ever farming in the region again.”

A Map of What Is Already Gone

In 2024, a team led by Scott Jasechko published in Nature what remains the most comprehensive picture of global groundwater ever assembled: roughly 170,000 monitoring wells across 1,693 aquifer systems in more than forty countries, covering about three-quarters of the world’s groundwater withdrawals. Thirty-six percent of those aquifer systems are declining faster than 0.1 meters a year. Twelve percent are falling faster than half a meter a year. In thirty percent of them, the rate of decline in 2000–2022 was faster than it had been in 1980–2000, an acceleration of water withdrawal more than twice as common as chance would produce.⁶

Figure 2. The state of the world’s groundwater. Of 1,693 aquifer systems monitored across more than forty countries, roughly a third are falling faster than 0.1 metres a year, and in thirty percent the decline has accelerated since 2000. The sixteen percent that are recovering did so after deliberate intervention, not on their own. Shares from Jasechko et al., Nature, 2024.

The geography of the decline identified maps almost perfectly onto the geography of the world’s grain.

In the United States, the High Plains Aquifer, the Ogallala aquifer, underlies 111.8 million acres across eight states and supplies roughly thirty percent of all groundwater used for irrigation in the country.⁷ The USGS puts the area-weighted decline from predevelopment to 2019 at 16.5 feet, which sounds survivable until the number is disaggregated by state. Texas has lost 44.1 feet. Kansas has lost 27.3. Nebraska has lost four-tenths of a foot.⁸ That single table explains why there is no unified Ogallala aquifer politics: the states with the water have no reason to accept limits, and the states without it are past the point where limits will help. In the Texas High Plains Water District, average saturated thickness in the Ogallala is now 51 feet.⁹ A 2020 study projected that twenty-four percent of currently irrigated High Plains land will be unable to support irrigation by 2100. In a detail that earlier work missed was that most of the acres that lose irrigation will fall back to dryland farming with crops that can grow on rainfall alone, with food production at roughly a third of that of irrigated yields, Unfortunately, what the study also found was that not all of the farmland will be able to support even dryland farming. On thirteen percent of the transitioning land the soil is too thin to raise a rainfed crop at all, and it will go back to grazing. Irrigation, on that ground, was never improving the soil. It was substituting for good soil and allowing crops to grow there that would ever have on the poor soil alone.¹⁰

Figure 3. One aquifer, eight political realities. Water-level change in the High Plains aquifer from predevelopment to 2019. Texas has lost more than a hundred times what Nebraska has, and South Dakota has gained. This distribution is why there is no unified Ogallala politics: the states with water have no reason to accept limits, and the states without it are past the point where limits would help. Figures from USGS Scientific Investigations Report 2023-5143, Table 2.

In India, the Central Ground Water Board’s 2025 assessment found 730 of the country’s 6,762 groundwater assessment units classified as over-exploited. Punjab, the state that fed the Green Revolution, is now extracting 26.27 billion cubic meters a year when the sustainable removal from the aquifer is only 16.80 billion cubic meters, an extraction of 156 percent of the replenishable removal. In 111 of Punjab’s 153 blocks, more water leaves the ground each year than enters it. In Sangrur district the figure is a removal of 310 percent of the replenishable removal.¹¹ Both of these signal aquifers being overused and which will be depleted over time.

In Iran, where agriculture consumes roughly ninety percent of all available water, groundwater extraction has caused land subsidence across 31,400 square kilometers. Almost two percent of the country’s land surface is sinking faster than a centimeter a year, with more than a hundred sites exceeding that rate and one area near Rafsanjan dropping 34 centimeters annually. Subsidence is what happens when unconsolidated aquifer systems containing significant layers of fine-grained, compressible materials like clay and silt, aquitards, lose the pore water that helps carry the weight of everything above them: the grains pack together, the layers thin, the ground surface drops, and the space that once held water closes and does not reopen.¹² Seventy-seven percent of that fast subsidence correlates with where Iran’s most productive agriculture occurs. Most of it is, in the authors of the report’s framing, “irrecoverable”: inelastic compaction accounts for at least sixty percent of the deformation, which means the pore space of the aquifer which holds the water collapses and much of the aquifer’s capacity to hold water again is permanently lost.¹³

In Afghanistan, the Kabul aquifer has dropped 25 to 30 meters in a decade, with extraction exceeding recharge by 44 million cubic meters a year. Nearly half the city’s wells are already dry.¹⁴

There is one genuine counter-example. On the North China Plain, where water tables were falling one to two meters a year, By 2008 the plain had pumped away roughly 60 cubic kilometers of groundwater more than the rain put back: about 48 million acre-feet, or two and a half years of everything the entire Colorado River basin consumes.

China then went at the problem from both ends at once. It brought new water in, and it forced less water out. The new water came through the South-to-North Water Diversion, a canal system built to carry 9.5 cubic kilometers a year north from a reservoir on the Han River. In 2023 it actually delivered roughly 4.3 million acre-feet, and about fourteen percent of everything the region used that year. The reduction came from enforced limits on pumping, which cut what farms, factories and cities drew out of the ground by around 12 cubic kilometers a year. That is close to 10 million acre-feet: a single year’s saving roughly half the size of the entire Colorado River’s annual consumption.

Together they produced something no one had documented at this scale: the water tables rose about 0.7 meters a year across 130,000 square kilometers, an area the size of Louisiana, between 2020 and 2024, By 2024 the water in the aquifer stood higher than it had in 2005.¹⁵ Aquifer depletion, in other words, is not a law of physics. It is a policy outcome. It can be reversed by policy, at enormous cost, where a state decides to as long as the soils have not collapsed.

The Ground Gives Way, But It Does Not Shake

One question follows naturally from all of this, and it is worth answering plainly, because the answer is usually gotten backwards: does draining an aquifer cause earthquakes?

Very largely, no, and the physics runs opposite to the intuition. Pumping water out of rock lowers the pressure in its pores, and lower pore pressure presses the two sides of a fault together more firmly rather than loosening them. Whatever destabilizing effect exists arrives indirectly, as the drained rock contracts and transfers stress to its neighbors, and as the crust flexes upward once the weight of the water is gone.¹⁶

Scale settles it. Groundwater withdrawal changes crustal stress by something on the order of one to fifteen kilopascals. An earthquake releases a thousand to ten thousand. Research published this year found groundwater cycles in northern California associated with as much as a ten percent modulation in the rate of small earthquakes, and noted, in the same breath, that ocean and earth tides of equivalent strength produce no such effect at all.¹⁷ In the global database of human-induced earthquakes, groundwater extraction accounts for under one percent of the 1,377 recorded cases; hydraulic fracturing, mining and reservoir impoundment account for most of the rest.¹⁸ The most heavily monitored test of the mechanism is the Groningen gas field in the Netherlands, where pore pressure fell by more than ten times what the most exhausted aquifer experiences, and the largest earthquake in fifty years was magnitude 3.6.

None of this describes injection, which is a different process with a different record. Pumping fluid into the ground raises pore pressure and unclamps faults, and it took Oklahoma from one or two magnitude-3 earthquakes a year before 2009 to 888 in 2015, including a magnitude 5.8. The two are routinely confused, and they are opposites.¹⁹

What depletion does to the ground is not seismic, and it is not in dispute. The ground sinks. More than eighty percent of documented land subsidence in the United States is a consequence of groundwater use, across more than seventeen thousand square miles in forty-five states, and Arizona has mapped a hundred and sixty-nine miles of earth fissures opening above emptied basins. These are not earthquakes. They are the slow and permanent collapse of the space the water used to occupy, the more consequential hazard, and the one from which there is no recovery.

Figure 4. What depletion looks like underground. Pumping lowers the water table decade by decade; when it falls below a well’s screen the well fails, and when it falls below a clay layer the clay compacts, the ground surface drops, and the pore space that held the water closes permanently. Salt concentrates at the surface as the water that remains gets more brackish. Schematic, drawn to the mechanisms described in this article; vertical scale is exaggerated.

Three Ways a Water Shortage Kills a Harvest

One: not enough water at the one moment that matters

A crop does not need water evenly through the duration of its growth. It desperately needs abundant water at two or three specific points, and a deficit that arrives during flowering or during grain fill does damage that later rains cannot undo. Wheat builds its pollen about a week before it flowers, in a stage botanists call the young microspore, and this is the thirstiest moment in the plant’s life. Developing pollen grains cannot feed themselves; they are supplied with sugar from the leaves. A plant short of water closes its pores to stop losing moisture, which also shuts down photosynthesis, which starves the anthers, the small sacs at the tip of the stamen where pollen is made. The pollen still forms. It simply forms sterile.

What follows is a failure that conceals itself. The plant does not die. It has water enough to stay green, put up a head, and flower on schedule. But nothing is fertilized, no grain sets, and the crop stands in the field looking like a crop until the combine goes through and finds husks. Rain the following week changes nothing. The pollen died during the dry days, and a plant cannot make it twice.

Grain filling is the second window, and it fails in a different currency. Once a kernel has been fertilized, the plant spends roughly a month pumping starch into it, drawing on two accounts: the sugar its leaves are making now, and the sugar it banked earlier in the stem. Drought closes both. The pores shut, photosynthesis falls away, and the plant, reading the shortage as the end of its season, begins shutting down early, drying its leaves and halting the deposit before the kernel is full. Nothing is missing from the head this time. Everything in it is simply too small. And because a dry spell usually arrives with heat, the two stresses stack: high temperatures shorten the filling period on their own, even where there is water enough. The farmer learns the result at the elevator, where shriveled grain and low-test weight are docked once on the scale and again on the price. Between them, these two windows account for most of what a dry year actually costs. One sets how many grains a field will carry. The other sets what each of them weighs. Rain in between is welcome. Rain after either is too late.²⁰

In a controlled Chinese trial on two winter wheat cultivars, drought imposed at that single stage cut viable grain number on the main stem by 51.8 percent in the drought-tolerant variety and 79.5 percent in the sensitive one.²¹ The field still looked like a wheat field. But, it was not one because the plants held no wheat.

This is why “flash droughts,” the rapid-onset soil moisture collapse driven by heat and wind rather than by a long rainfall deficit, has become an important new term of art. Global projections show flash drought frequency rising through the century, and the cropland-specific numbers are the ones that matter here: under a high-emissions pathway, the annual risk of flash drought on North American cropland rises from 32 percent in 2015 to 49 percent by 2100; European cropland from 32 percent to 53 percent; African cropland from 35 percent to 47 percent.²²

The best current modeling of what this does to the food supply is a study published in Nature Communications this year. Run through 2050 under a middle-of-the-road emissions scenario, drought reduces globally averaged production of maize, rice, soybean and wheat by somewhere between roughly 0.2 percent and just under 2 percent. That is a small number, but the story is in the distribution. Maximum country-level production losses exceed 10 percent in 62 countries and exceed 20 percent in 24 countries, with single-crop maxima reaching 76 percent for maize, 67 percent for rice and 64 percent for wheat.²³ A global food system can absorb a two percent average. It cannot absorb a 76 percent maize failure in a particular country that eats its own maize and has no foreign exchange to import someone else’s.

Figure 5. Two windows, two currencies. A water deficit does the most damage at two specific moments, and it damages the crop differently at each. Stress at flowering sterilizes pollen and removes grains that never form; stress during grain fill leaves the grains in place but light. Rain between the two windows helps; rain after either does not. The curves are illustrative of the two sensitivity windows, anchored to measured grain-number losses of 51.8 and 79.5 percent in controlled winter-wheat trials.

Two: the wrong kind of water

A. Salty

When fresh water runs short, farmers do not stop irrigating. They irrigate with what is left, brackish groundwater, drainage water, recycled wastewater, and in hot climates the evaporation concentrates the salts that come with it. The result is a slow poisoning that is invisible until it is permanent and kills the crop.

The mechanism is worth understanding, because it explains why the damage is so hard to undo.

A root does not pump. It has no muscle for drawing water out of soil. It relies on osmosis: water moves on its own from where dissolved salts are dilute toward where they are concentrated, and because the sap inside a root is saltier than the water in ordinary soil, water flows inward without the plant spending anything to move it. That gradient is the entire engine.

Salt in the soil weakens that gradient, and enough of it reverses the direction. When the water around the roots becomes as salty as the sap inside them, water stops entering the plant altogether. When it becomes saltier still, water begins moving the other way, and the soil draws moisture back out of the root. The soil is visibly damp. The plant is in drought. It dies of thirst standing in wet ground.²⁴

What does get in causes a second kind of harm. Sodium and chloride accumulate in the leaves and scorch them, and they crowd out the potassium and nitrate the plant needs, so a salt-stressed crop is starved as well as parched.

And here is the trap. Every irrigation adds a little salt, and evaporation removes pure water while leaving that salt behind, so concentration climbs season after season. The only way to reverse it is to flood the field with enough clean water to carry the salt down below the root zone. The cure for salinity is water in abundance, which is precisely what a farmer who has resorted to brackish water does not have.²⁵

In December 2024, the FAO published the first global assessment of salt-affected soils in fifty years. 1,381 million hectares, 10.7 percent of the Earth’s land area, are salt-affected, with roughly another billion hectares at risk. Ten percent of irrigated cropland and ten percent of rainfed cropland are already affected. Under current warming trajectories the affected share of the global land surface could reach between 24 to 32 percent by the end of the century. In the worst-hit regions, salinity stress costs up to seventy percent of yield in crops like rice and beans. Ten countries, Afghanistan, Argentina, Australia, China, Iran, Kazakhstan, Russia, Sudan, the United States and Uzbekistan, hold seventy percent of the world’s salt-affected soil.²⁶ The most-cited estimate of the annual global cost of salt-induced degradation in irrigated areas is $27.3 billion in lost crop production. That figure is more than a decade old, counts nothing but forgone yield, and has never been replaced; FAO’s 2024 global assessment, the first in fifty years, did not attempt one. What has been measured since is the loss in food rather than in money. The World Bank puts the annual production lost to water salinity at 124 trillion kilocalories, enough to feed more than 170 million people every day. That the best current accounting of salinization is denominated in calories rather than dollars is itself a finding about how seriously the problem has been costed.²⁷

B. The Water That Carries Something Else

Salinity is the slow failure, and it is at least visible in the yield. There is a second failure that never appears in production statistics at all, because it does not damage the crop. It damages the people who grow it and the people who eat it.

When a city runs short of fresh water, the one supply that remains reliably available to the farms on its edge is the water the city has already used. Roughly 29 million hectares of the world’s irrigated land, about a tenth of the total, is watered from sources carrying a substantial load of urban wastewater, and about 95 percent of that is untreated or inadequately treated. Formal, engineered reuse of properly treated effluent covers perhaps 1.5 million hectares. The ratio of unsafe to safe is around thirty to one.²⁸

The reason is cost, and the pattern is exactly what you would guess. Of the 267.5 billion cubic meters of municipal wastewater the world produced in 2023, 45 percent was discharged with no treatment whatsoever. High-income countries treat 91.5 percent of theirs. Low-income countries treat 5.8 percent.²⁹

What that water carries is not hypothetical. In Kumasi, Ghana, farmers irrigating with wastewater carried Ascaris infection were 15.8 percent against 6.0 percent in a matched control group, about four times the odds.³⁰ In Santiago, Chile, where 13,500 hectares of salad crops were irrigated with raw sewage, typhoid ran at 210 cases per 100,000 at its peak, and when cholera arrived in 1991, 68 percent of the first 41 cases had eaten raw wastewater-irrigated salad. Chile banned the practice. That is the encouraging half of the story: this is a solved problem wherever anyone decides to follow the solution, do not water food crops with sewage.³¹

The residue is more permanent than the pathogens. In Mexico’s Mezquital Valley, irrigated with Mexico City’s untreated sewage for roughly a century, soil chromium now runs 7.7 to 8.6 times background and cadmium 3.0 to 4.7 times. Antibiotic-resistance genes in those soils are two orders of magnitude more abundant than in comparable fields.³²

Treated wastewater reuse is not a hazard; it is one of the best adaptations available for making due with the limited water resources that we have. Israel reuses about 87 percent of its treated effluent, supplying roughly half its agricultural water. Even the untreated kind carries real value, the nutrients in it are worth $179 to $451 per hectare per year, are taken up more efficiently than synthetic fertilizer, and after forty years leave soil carbon half again higher. When Mexico built a treatment plant that stripped those nutrients out, Mezquital farmers protested, and on the economics they were not wrong. The problem is not that poor farmers use wastewater. It is that they are made to use it raw and untreated.³³

Figure 5. Who treats their sewage before it reaches a field. Share of municipal wastewater receiving any treatment, 2023. The farms most likely to be irrigating with untreated waste are in the countries least able to treat it, which is why the hazard falls almost entirely on the poor. Figures from Qadir, Drechsel and Jones, Discover Water, 2025.

Three: not enough water to move the harvest

Even a crop that survives the water risk in the field has to get to a port. In the United States it usually gets there by water, and for four consecutive autumns it has struggled to make this trip.

The USDA’s modal share analysis finds, is that barges carried 44 percent of U.S. grain exports in 2022, 48 percent of all corn, 48 percent of all soybeans, and 32 percent of all wheat.³⁴ That is still close to half of the corn and soybeans this country sells to the world, moving on a river whose depth is now a recurring emergency.

The Mississippi at Memphis set a modern record low of 10.81 feet below normal in October 2022, then broke it at 11.52 feet below normal in October 2023. During the 2023 episode, loading drafts between Cairo, Illinois and the Gulf were cut by an average of 24 percent, tow sizes by 17 to 38 percent, and barge freight rates exceeded 900 percent of tariff, more than nine times the normal baseline rate due to a severe shortage of available barge capacity.³⁵ Low water returned in 2024 and again in 2025, when the Ohio River was contributing just eight percent of Lower Mississippi flow against a typical fifty percent due to a massive flash drought, and southbound grain shipments on the river fell 79 percent in little more than a month with corn movements down 72 percent.³⁶

What a low river actually does to a farmer is measurable to the penny. Peer-reviewed work on the 2022 event found that with the Mississippi gauge at −5 feet, soybean basis fell 58 cents a bushel for farms within five miles of the river, 29 cents at ten miles, and 12 cents at twenty-five.³⁷ The grain is fine. The farmer simply gets less for it, because the only affordable way out of the country has narrowed.

Figure 6. What a river gauge does to a farmer’s cheque. With the Mississippi five feet below normal, the discount on soybeans deepens the closer a farm sits to the river, because the farms nearest the water are the ones with no alternative to it. The grain is unaffected; only its price is. Measured for the 2022 low-water episode by Mitchell and Biram, Applied Economic Perspectives and Policy, 2025.

As of the week ending August 1, 2026, before harvest, before the seasonal low-water window, barge rates at St. Louis were running at 638 percent of tariff, up 39 percent from a year ago. Cairo-to-Memphis rates were up 47 percent.³⁸ The river has not yet made news this year. The price of using it already has.

The Water in the American Basket

Nearly half the contiguous United States, 48.5 percent, was in moderate drought or worse as of August 4, up from 29.9 percent a year earlier.³⁹ The commodity-specific numbers are worse than the map: 50 percent of the winter wheat crop, 58 percent of spring wheat, 57 percent of alfalfa hay, and 49 percent of the national cattle herd are in drought.⁴⁰ USDA’s July estimate put the 2026 wheat crop at 1.536 billion bushels, down 22.6 percent year over year on 5.1 million fewer harvested acres.⁴¹ Corn conditions stand twelve points below last year.⁴² The January cattle inventory of 86.2 million head was the smallest in seventy-five years.⁴³

Against that, the Colorado River. The basin supports 40 million people, seven states and thirty tribal nations, and its natural flow has fallen so far below the volume the 1922 Compact divided that the arithmetic no longer works: inflows averaged 12.9 million acre-feet a year between 2000 and 2024 compared to the baseline of 18 million acre-foot.⁴⁴ Agriculture takes 52 percent of all consumption in the basin and 74 percent of direct human consumption; alfalfa and grass hay alone account for 46 percent of all direct water consumption, and cattle-feed crops for 62 percent of agricultural water.⁴⁵ The Colorado River, in other words, is mostly used to feed livestock.

Figure 7. What the Colorado River actually grows. Feed for cattle takes 62 percent of the river’s direct human use, and a single crop, alfalfa and grass hay, takes 46 percent, nearly twice what every city, factory and business in the basin uses combined. Shares from Richter et al., Communications Earth and Environment, 2024; percentages are of direct human consumption and exclude reservoir evaporation and riparian use.

The seven basin states were given until February 14, 2026 to agree on how to operate the river after the 2007 interim guidelines expired. They failed. The decision therefore passed to the Interior Secretary, and on July 31 the Bureau of Reclamation published its final plan.

It is not a fixed rulebook. Rather than setting operating rules for a decade, it sets a procedure that is rerun every two years between 2027 and 2036, adjusting to how much water is actually in the system. Two numbers do most of the work. The first is how much water Glen Canyon Dam will release from Lake Powell into the lower river each year, which can now be set anywhere between 5.0 and 12.0 million acre-feet depending on how full Powell is in the autumn. The second is how much the states below it must give up: shortages of up to 3.0 million acre-feet.

Where that shortage falls is the fault line. All of it falls on the Lower Basin, Arizona, California and Nevada, and all of it is mandatory. The four Upper Basin states are asked for up to 200,000 acre-feet, and their contribution is voluntary.⁴⁶

The plan is candid about how the water will be found: by paying farmers and irrigation districts to leave fields dry. Combined Lower Basin reductions could eventually reach 3.6 million acre-feet a year: roughly forty percent of what those three states are entitled to draw from the river.⁴⁷

Within the Lower Basin’s share, roughly sixty percent falls on Arizona, thirty-five percent on California and four percent on Nevada.⁴⁸ And the authority behind it is worth being precise about, because these cuts are not the product of a negotiation. The Secretary of the Interior has served as water master of the Lower Colorado since the Boulder Canyon Project Act of 1928 and the Supreme Court’s decree in Arizona v. California, and may set operating rules without the states’ consent when the states cannot agree among themselves. Running underneath this is an older and more mechanical process that is already in force: each August, Reclamation’s 24-Month Study projects Lake Mead’s January elevation and thereby sets the shortage tier that automatically reduces deliveries to Arizona, Nevada and Mexico the following year.⁴⁹ No Record of Decision has yet been signed, which means the post-2026 framework remains a preferred alternative rather than law. Arizona’s water agency has called it “unacceptable,” warning that the reductions “would devastate Arizona’s water users and its economy,” while the Upper Basin governors said they were “encouraged.” Litigation is the widely expected next step.

Two facts about what that water grows. Yuma, Arizona supplies 80 to 90 percent of America’s leafy greens in the winter months. The Imperial Valley, which holds the single largest entitlement on the river, roughly 97 percent of it going to agriculture: produces about two-thirds of the vegetables Americans eat between November and March.⁵⁰ Imperial County’s unemployment rate is around 19 percent, and roughly one job in six there is in agriculture.⁵¹

California’s own groundwater reckoning is running on a parallel track. Two San Joaquin Valley subbasins are under state probation, with pumpers facing $300 per well and $20 per acre-foot; the Public Policy Institute of California projects that between 500,000 and 900,000 acres of valley farmland will come out of production by 2040 depending on how well the state manages the transition, with the worst case costing roughly 50,000 jobs and $4.5 billion in regional GDP.⁵² And the physical consequence of a century of overdraft is already eating the infrastructure built to fix it: the San Joaquin Valley sank an average of nearly an inch a year between 2006 and 2022, a 33-mile stretch of the Friant-Kern Canal lost up to sixty percent of its carrying capacity to subsidence, and after $325 million and eight years spent rebuilding ten miles of it, the ground beneath the new concrete has already dropped another foot: taking back 250 cubic feet per second of the capacity the project restored.⁵³

None of this stays in the West. USDA now forecasts beef and veal prices up 10.7 percent in 2026 and fresh vegetables up 6.8 percent, against 3.1 percent for food overall.⁵⁴ Those are precisely the two categories most exposed to water: beef through a seventy-five-year-low herd with half of it standing in drought, and fresh vegetables through a river that hit its all-time record low last Friday. A collapse in egg prices is currently masking both in the headline index.

And this arrives at American households with less protection than they had a year ago. The 2024 Household Food Security report, published in December 2025, found 13.7 percent of U.S. households, nearly 48 million people, in food-insecure homes, including more than 14 million children.⁵⁵ It was the last such report the government will publish; USDA announced in September 2025 that it was discontinuing a thirty-year annual series, and the Census supplement that fed it was cancelled the same month.⁵⁶ Independent surveys already put the number far higher: the Urban Institute found that nearly one in four American adults, 24.2 percent, lived in a food-insecure household in 2025, and more than one in four working-age adults.⁵⁷ Both figures are defensible; they measure the same condition with different survey machinery, and they disagree by a factor of nearly two. The instrument that would have settled which is closer to the truth no longer exists. Meanwhile the SNAP provisions of the 2025 reconciliation law begin to bite: on October 1, 2026, seven weeks from now, states must absorb 75 percent of SNAP administrative costs, with a share of benefit costs following in 2027, against roughly $200 billion in program reductions over a decade.⁵⁸

“The federal government stopped measuring hunger in the same year it enacted the largest cut to food assistance in the program’s history. You cannot be held accountable for a number that no longer exists.”

The People Who Move, and the People Who Cannot

Water scarcity does not usually produce a photograph. It produces a decision, made at a kitchen table, one household at a time, and the person who makes it is almost never recorded as a climate migrant⁵⁹.

The strongest global evidence is a 2024 study in Nature Climate Change that examined 107,840 migration flows between subnational regions across 72 countries from 1960 to 2016 and found that drought and aridity significantly increase internal migration, most strongly in arid zones, and most strongly of all in rural, agriculture-dependent areas. Within countries, people move from poorer regions to wealthier ones.⁶⁰ The mechanism is not that water disappears. It is that a livelihood does.

For Central America, the link runs through a specific and well-documented channel. A 2023 study in Scientific Reports matched 323,579 people apprehended in family units at the U.S.--Mexico border between 2012 and 2018 against growing-season moisture in the 54 departments of El Salvador, Guatemala and Honduras they came from. Departments that had recently experienced drier-than-average growing seasons sent 70.7 percent more emigrants to the United States. The authors are careful, and so are we: across model specifications, the effect ranged from 8.9 percent to 257 percent, and apprehension data cannot confirm that the households who lost crops are the same people who arrived at the border.⁶¹ But the direction is not seriously in dispute, and the corroborating evidence is strong: when coffee leaf rust destroyed an average of 71 percent of production for affected smallholders in eastern Guatemala, household migration roughly doubled, including in communities with no prior migration tradition.⁶²

In Somalia, the accounting is more direct than almost anywhere. Nearly 62,000 people were displaced by drought between the start of 2026 and April, and three of every four new displacements in Somalia are now drought-related, a 22 percent year-over-year increase, with a further 125,000 projected by mid-year if conditions held.⁶³ The southern Deyr cereal harvest came in 83 percent below its thirty-year average. Some 6.5 million Somalis are projected in crisis-level food insecurity or worse by March, including two million in emergency conditions and 1.84 million acutely malnourished children.⁶⁴

In Iraq, the migration tracking network has recorded 28,379 families, 170,274 people, displaced by drought between 2016 and 2024 across twelve central and southern governorates.⁶⁵ In Iran, the vice president for rural development said in January that of the country’s 69,000 villages, 31,000 now stand deserted, and that a nation which was 70 percent rural in 1976 has inverted entirely.⁶⁶ That figure is cumulative over five decades and water is one cause among several, but it arrives alongside reservoir declines exceeding 70 percent at two of Tehran’s principal sources, and a president who has said publicly that parts of the capital may have to be evacuated and the seat of government relocated.⁶⁷

The most-quoted number in this field is the World Bank’s projection of 216 million internal climate migrants by 2050 across six world regions, with water scarcity explicitly modeled as a driver.⁶⁸ It is worth knowing exactly what that number is: it is the high end of a 44-to-216-million range, produced under a pessimistic scenario combining high emissions with unequal development, published in 2021 and not updated since. The European Commission’s own research service now cautions that migration forecasts of this kind “are indicative at best” and recommends assessing populations’ exposure to climate extremes rather than predicting migrant counts.⁶⁹

What This Publication Cannot Tell You

Advocacy journalism earns its authority by being the first to name its own weak points. Four of them here are load-bearing.

First: drought displacement is systematically undercounted, and we cannot tell you by how much. The Internal Displacement Monitoring Centre, the world’s authoritative counter, only began recording drought as a displacement trigger in 2017. Its 2026 report notes that fewer drought movements were recorded in the Horn of Africa and southern Africa last year, “but that was partly the result of reduced reporting,” and that Iraq’s drought displacement figure fell to a third of the prior year’s “in part the result of less reporting.”⁷⁰ Sudden-onset disasters generate evacuation orders and camp registrations. Drought generates a family that sells its goats, then its tools, then leaves, and gets recorded, if at all, as an economic migrant. A causal analysis of Somalia from 2016 to 2023 found that market food prices, not rainfall, were the dominant proximate driver of movement, which is precisely why the connection to water gets lost in the record.⁷¹

Second: drought is not the leading driver of global hunger, and we will not claim it is. The Global Report on Food Crises for 2026 found 266 million people in 47 countries at crisis-level acute food insecurity or worse. Conflict was the primary driver for 147.4 million of them; extreme weather for 87.5 million, and the weather-driven figure fell year over year.⁷² Every famine the IPC has classified in this decade, Sudan in 2024 and 2025, Gaza in 2025, has been driven by war, not by rain. Water scarcity is a multiplier and a slow strangler. It is not the main killer, and a movement that overstates it will be correctly discredited.

Third: the most famous case in this literature does not hold up as it is usually told. The claim that Syria’s 2006–2010 drought displaced 1.5 million people who then triggered a civil war is the standard example in climate-security writing. The drought was real and was the worst in the instrumental record. But the 1.5 million figure was traced by later researchers to a single official’s remark in a humanitarian news report, most likely describing people affected by rather than displaced by drought; UN and Syrian government estimates put actual drought-related migration at 40,000 to 60,000 families, four to twelve percent of Syria’s urban growth over the period.⁷³ Satellite analysis subsequently showed Syrian croplands recovering quickly after 2009, with only 0.5 percent of previously cultivated land permanently abandoned.⁷⁴ The original authors have since revised their language from “displaced” to “affected” and describe climate as one contributing factor among many.⁷⁵ We cite the drought. We do not cite the war.

Fourth, and most uncomfortable: the poorest people do not migrate. A study of 115 countries across four decades found that in middle-income countries, warming increased both urbanization and emigration, but in poor countries, higher temperatures reduced the probability of migrating, both to cities and abroad, consistent with severe liquidity constraints.⁷⁶ The relationship between climate stress and movement is an inverted U. Above a wealth threshold, a shock mobilizes people. Below it, a shock immobilizes them. The people in the very worst trouble are the ones you will never see at a border, and a systematic review of dryland mobility research found that of 183 empirical studies, only ten addressed involuntary immobility at all.⁷⁷

This has a mirror image inside the United States that ought to be said plainly. American irrigators facing Colorado River cuts are being paid to fallow: hundreds of dollars an acre-foot, funded federally.⁷⁸ There is no evidence of population collapse in Colorado River farming counties, and we will not invent it. What the American case actually demonstrates is that a rich country can buy its way out of the displacement that a poor country has to live through. That is a more damning fact than a depopulation statistic would have been, and it is true.

“The question is never whether a society can survive losing its water. It is whether that society can afford to be paid for losing it.”

The Window That Is Still Open

Groundwater is the slowest-moving crisis in the food system, which means it is also the one where action taken this year still changes the outcome. Four things are required.

First, treat water accounting as food security infrastructure. The most recent aquifer-wide USGS assessment of the Ogallala reports water levels as of 2019. Anyone citing “the latest federal data” on the aquifer that irrigates the American grain belt is citing measurements seven years old. Meanwhile USDA has terminated the thirty-year household food security series.⁷⁹ A country that intends to manage a resource has to measure it: fund the USGS groundwater monitoring network, restore the food security survey, and require the states drawing on interstate aquifers to publish annual, well-level data.

Second, pay for demand reduction before the well fails, not after. The foundational Kansas projection found that cutting groundwater use twenty percent today would extend the region’s peak agricultural production into the 2070s and yield more cumulative output beyond 2070 than continuing at current rates.⁸⁰ Kansas’s Local Enhanced Management Areas have already demonstrated the mechanism at small scale: one district cut applied water per acre by 34 percent primarily by reducing water intensity rather than idling acres, and the Greater GMD4 LEMA is in renewal hearings right now, with the next order due for the 2028–2032 term.⁸¹ Conservation is not a sacrifice of production. On the arithmetic, it is the only path that preserves production.

Third, make anticipatory financing the default for slow-onset water crises. Somalia’s 2026 displacement was forecast months before it happened; the Deyr harvest failure was visible in satellite data before the harvest. Anticipatory action: cash, seed, drought-tolerant inputs and pre-positioned food released on forecast triggers rather than on famine declarations: costs a fraction of late emergency response. It requires a humanitarian budget that exists. The U.S. contribution has fallen from roughly $14 billion to $3.7 billion in a single year, and no trigger system survives the removal of the money behind it.⁸²

Fourth, protect the domestic floor while the price wave arrives. Beef at nearly eleven percent and fresh vegetables at nearly seven percent are water prices, arriving in American grocery stores through the Colorado River and the drought map. The SNAP cost-shifts that begin on October 1 will transfer part of that cost onto state budgets and, from 2027, onto the households least able to carry it.⁸³ A Farm Bill built for volatility: insurance and disaster programs that reach diversified operations and pay in the season the loss occurs, not two years later: is the other half of the same obligation.

How to Watch It Come

Every figure in this article came from a public dataset. If you want to watch the water yourself:

· The American drought, weekly: the U.S. Drought Monitor updates every Thursday; NIDIS’s drought.gov carries current conditions and the Mississippi River dashboard.

· The Colorado River, daily: Reclamation publishes Lower Colorado reservoir levels every day and the 24-Month Study monthly; the post-2026 process page holds the Final EIS.

· The aquifers: USGS groundwater monitoring; the Kansas Geological Survey annual water-level measurements; India’s Central Ground Water Board assessments.

· The crops and the freight: USDA’s Crop Progress, the monthly WASDE, the weekly Grain Transportation Report, and the GEOGLAM Crop Monitor.

· The hunger and the displacement: FEWS NET, IPC, WFP’s HungerMap Live, and IDMC’s Global Internal Displacement Database.

A Ledger That Comes Due

Water is the only input in agriculture with no substitute. A farmer who cannot afford fertilizer grows a smaller crop. A farmer who cannot afford diesel harvests late. A farmer with no water grows nothing, and no amount of capital, technology or policy will change that arithmetic.

For most of the last seventy years the world has hidden that fact from itself by borrowing. It pumped aquifers faster than rain could refill them, built cities and export industries on the difference, and recorded the resulting harvests as productivity rather than as debt. Seventy percent of the world’s major aquifers are now in decline. The lake behind Hoover Dam is one-quarter full. The wheat belt of the southern Plains is farming on fifty feet of saturated sand. Punjab is pumping half again more groundwater each year than its aquifers can sustainably yield.

The people who will pay are not the people who spent it. The pastoralist in Somalia whose Deyr harvest came in 83 percent short did not drain an aquifer. The Guatemalan smallholder whose growing season failed did not build a diversion canal. The mother in Phoenix looking at the price of ground beef, and the farmworker in the Imperial Valley whose employer has just been paid to leave a field dry, did not sign the 1922 Compact. But all of them are standing downstream of the same ledger, and it is coming due in the currency it always comes due in: skipped meals, sold livestock, mounting debt, a decision made at a kitchen table about whether to stay.

What makes this crisis different from a drought or a strait or an El Niño is that it does not surprise anyone. Aquifer decline is measured in monitoring wells, published in federal reports, and projected decades in advance. There will be no moment at which the world can say it did not know. There is only the question of whether the people who hold the water rights, write the farm bills, fund the humanitarian budgets and set the pumping limits will act while acting is still cheap.

Call your representatives, in your statehouse and in Washington. Ask them three questions: what is being done to measure the aquifer your food comes from, what is being done to reduce the draw on it before the wells fail, and what happens to the families who cannot afford to leave when they do. Then ask them again in October, when the SNAP cost-shift takes effect, and again in January, when the Colorado River framework begins. The water is being spent now. So is the time.

Endnotes

1. Lake Mead fell to 1,040.46 feet between midnight and 1 a.m. on August 7, 2026. A Reclamation spokesperson gave the previous historic low as 1,040.50 feet, recorded July 28, 2022; UPI and Western Water report the prior record as 1,040.58 feet on the same date. This publication has not been able to reconcile the two, and has described the margin as inches rather than choosing between them. Reported: Las Vegas Review-Journal, “Lake Mead sets a new record low,” August 7, 2026, https://www.reviewjournal.com/news/environment/lake-mead-sets-a-new-record-low-its-only-the-beginning-3860660/. The projection of successive monthly record lows through 2028 is from the same report.

2. Bureau of Reclamation, Lower Colorado River Operations, daily reservoir levels, August 8, 2026: Lake Mead 1,040.34 ft / 7,001,000 AF / 26.8% of capacity; Lake Powell 3,521.09 ft / 5,323,000 AF / 22.8%. https://www.usbr.gov/lc/region/g4000/hourly/levels.html. Combined Mead–Powell storage is at its lowest since 1957. Reclamation’s July 15, 2026 24-Month Study most-probable projection puts Lake Powell at 3,507.68 ft at the end of December 2026 and 3,501.25 ft at the end of January 2027, within eighteen feet, then eleven feet, of the 3,490 ft minimum power pool at Glen Canyon Dam, below which the dam ceases to generate. https://www.usbr.gov/lc/region/g4000/24mo.pdf

3. U.S. Department of the Interior, “Reclamation Publishes Final Environmental Impact Statement for the Future of the Colorado River,” July 31, 2026, https://www.doi.gov/pressreleases/reclamation-publishes-final-environmental-impact-statement-future-colorado-river. The framework provides for Lower Basin reductions of up to 3.0 million acre-feet through 2036; the seven basin states missed the February 14, 2026 deadline to agree on a consensus alternative, which transferred the decision to the Secretary of the Interior. The allocation among the three Lower Basin states, the Secretary’s legal authority to impose it, the separate annual shortage-tier mechanism already in force, and the objections now being raised are set out in full in “The Water in the American Basket” below, at notes 32 through 35.

4. Madani, K. et al. (2026). Global Water Bankruptcy: Living Beyond Our Hydrological Means in the Post-Crisis Era. United Nations University Institute for Water, Environment and Health, January 20, 2026. DOI 10.53328/INR26KAM001. https://unu.edu/inweh/collection/global-water-bankruptcy

5. The cattle and wheat figures in this paragraph are USDA NASS, Cattle, January 30, 2026 (all cattle and calves 86.2 million head as of January 1, 2026) and USDA World Agricultural Supply and Demand Estimates, WASDE-673, July 10, 2026 (2026/27 all-wheat production of 1,536 million bushels against 1,985 million bushels in 2025/26, a decline of 22.6 percent). Both are treated in full, with their sourcing caveats, in “The Water in the American Basket” below. On the fertilizer shock, see Fertilizer Famine (this publication, 2026).

6. Jasechko, S., Seybold, H., Perrone, D., Fan, Y., Shamsudduha, M., Taylor, R.G., Fallatah, O., & Kirchner, J.W. (2024). “Rapid groundwater decline and some cases of recovery in aquifers globally.” Nature 625, 715–721. DOI 10.1038/s41586-023-06879-8. Note the paper’s own counter-finding: 16 percent of aquifers reversed prior declines following policy intervention, managed recharge, or surface-water substitution.

7. Steward, D.R. et al. (2013). “Tapping unsustainable groundwater stores for agricultural production in the High Plains Aquifer of Kansas, projections to 2110.” PNAS 110(37): E3477–86. DOI 10.1073/pnas.1220351110. See also USDA Climate Hubs, Impacts to the Ogallala Aquifer (Yamazaki & Suhr Pierce, December 2024): the region accounted for at least 9 percent of U.S. crop sales in 2019; recharge supplies 15 percent of current pumping; refilling a depleted aquifer would take 500–1,300 years. https://www.climatehubs.usda.gov/sites/default/files/Ogallala%20Overview%20241202.pdf

8. McGuire, V.L. & Strauch, K.R. (2024). Water-Level and Recoverable Water in Storage Changes, High Plains Aquifer, Predevelopment to 2019 and 2017 to 2019. USGS Scientific Investigations Report 2023-5143, published February 13, 2024. DOI 10.3133/sir20235143. State figures are from Table 2 of the source PDF; note that the HTML rendering of this report at pubs.usgs.gov returns different and incorrect state-level values, an error that has propagated into secondary sources. https://pubs.usgs.gov/sir/2023/5143/sir20235143.pdf

9. High Plains Underground Water Conservation District No. 1, 2026 water-level measurements: average annual change −0.57 feet; district average saturated thickness 51 feet. https://www.hpwd.org/2026-water-level-measurements-are-complete

10. Deines, J.M., Schipanski, M.E., Golden, B., Zipper, S.C., et al. (2020). “Transitions from irrigated to dryland agriculture in the Ogallala Aquifer: Land use suitability and regional economic impacts.” Agricultural Water Management 233, 106061. Study area 450,660 km² across eight states.

11. Central Ground Water Board / Ministry of Jal Shakti, Dynamic Ground Water Resources Assessment 2025; national figures via PIB release, December 15, 2025, https://www.pib.gov.in/PressReleasePage.aspx?PRID=2204095. Punjab state figures as tabled in the Rajya Sabha, July 2026, and reported by The Tribune, https://www.tribuneindia.com/news/punjab/punjab-extracting-56-more-groundwater-than-limit-centre-tells-rajya-sabha/

12. The mechanism: aquifer sands are separated by fine-grained clay and silt layers, or aquitards, whose pore water helps carry the weight of everything above. Withdraw the water faster than it is replaced and that support is lost; the fine layers compact, the ground surface drops, and where compaction is inelastic the aquifer’s storage capacity is destroyed rather than merely emptied. The U.S. Geological Survey estimates that more than 80 percent of documented land subsidence in the United States is a consequence of groundwater use, affecting more than 17,000 square miles across 45 states. USGS, “Land Subsidence,” https://www.usgs.gov/mission-areas/water-resources/science/land-subsidence; USGS Fact Sheet FS-103-03, https://pubs.usgs.gov/fs/fs-103-03/

13. Payne, J. et al. (2025). “Widespread Extent of Irrecoverable Aquifer Depletion Revealed by Country-Wide Analysis of Land Surface Subsidence Hazard in Iran.” Journal of Geophysical Research: Solid Earth, August 27, 2025. DOI 10.1029/2024JB030367.

14. Mercy Corps (2025). Kabul’s Water Crisis: An Inflection Point for Action, April 2025. The report’s projection that Kabul’s aquifers could run dry by 2030 is a UNICEF modeling estimate, not an observation. https://www.mercycorps.org/sites/default/files/2025-05/crisis-analysis-team-kabul-water-april-2025.pdf

15. Long, D., Xu, Y., Cui, Y., Scanlon, B.R., et al. (2025). “Unprecedented large-scale aquifer recovery through human intervention.” Nature Communications 16, 7296, published August 7, 2025. DOI 10.1038/s41467-025-62719-5.

16. Segall, P. (1989), “Earthquakes triggered by fluid extraction,” Geology 17(10): 942–946, on why withdrawal is locally stabilising. Amos, C.B. et al. (2014), “Uplift and seismicity driven by groundwater depletion in central California,” Nature 509: 483–486, DOI 10.1038/nature13275, models San Andreas unclamping at 0.07–0.18 kPa per year; co-author Roland Bürgmann: “The stress is very small, much less than you need to build up stress on a fault toward an earthquake.” The most-cited contrary case, González, J.J. et al. (2012), Nature Geoscience 5: 821–825, DOI 10.1038/ngeo1610, was contested on publication, and its lead author accepted that the Lorca earthquake would probably have occurred regardless.

17. Sirorattanakul, K. & Avouac, J-P. (2026), Science Advances, March 25, 2026, DOI 10.1126/sciadv.adz5711: seismicity in northern California increases by up to roughly 10 percent in association with groundwater changes, peaking about half a month after the peak change. The comparison with tidal stresses of similar amplitude, which produce no comparable effect, is the authors’ own. Figures here are drawn from the California Institute of Technology’s summary of the paper, https://www.caltech.edu/about/news/seismic-activity-in-california-varies-with-the-seasons; this publication has not independently accessed the paywalled article.

18. HiQuake, the Human-Induced Earthquake Database, Durham University (1,377 cases as of December 2025): hydraulic fracturing 31 percent, mining 25 percent, reservoir impoundment 17 percent, conventional oil and gas 11 percent, groundwater extraction 0.9 percent. https://inducedearthquakes.org/. The database’s compilers note that it includes proposed cases “without regard to the strength of the case made,” so the groundwater share is if anything generous. On the Groningen bound: Boitz, N. et al. (2024), Nature Communications 15: 329, DOI 10.1038/s41467-023-44485-4, roughly 28 MPa of pore-pressure depletion produced a maximum event of ML 3.6, and the authors put the maximum possible induced magnitude at around Mw 4.

19. Oklahoma magnitude-3-and-above events rose from a background of one to two a year before 2009 to 888 in 2015; the largest was the Mw 5.8 Pawnee earthquake of September 3, 2016. The cause is wastewater disposal rather than hydraulic fracturing: USGS states that “only 1-2% of the earthquakes can be linked to hydraulic fracturing operations” and that over 90 percent of injected wastewater is a byproduct of oil extraction. Rates have since fallen close to historical levels following mandated plug-backs and volume limits. USGS, “Myths and Misconceptions about Induced Earthquakes,” https://www.usgs.gov/programs/earthquake-hazards/myths-and-misconceptions-about-induced-earthquakes. Subsidence and earth-fissure figures in the following paragraph: USGS as at the note above, and Arizona Geological Survey, Earth Fissures and Subsidence, https://azgs.arizona.edu/geohazards/arizona-geohazards-resources/earth-fissures-subsidence

20. Ntawuguranayo, S. et al. (2024), “Stem traits promote wheat climate-resilience,” Frontiers in Plant Science 15: 1388881, DOI 10.3389/fpls.2024.1388881: stem reserves of water-soluble carbohydrate contribute roughly 10 to 40 percent of final grain weight in favourable conditions, rising to 40 to 70 percent under stress, with remobilisation elevated under terminal drought, the plant drawing down savings because current photosynthesis has failed. The elevator discount for shrivelled grain and low test weight is a matter of grain-grading practice rather than of published research; test weight is a grade-determining factor in the U.S. wheat and corn standards administered by USDA’s Federal Grain Inspection Service.

21. Dong, B., Zheng, X., Liu, H., Able, J.A., Yang, H., Zhao, H., Zhang, M., Qiao, Y., Wang, Y., & Liu, M. (2017). “Effects of Drought Stress on Pollen Sterility, Grain Yield, Abscisic Acid and Protective Enzymes in Two Winter Wheat Cultivars.” Frontiers in Plant Science 8: 1008. DOI 10.3389/fpls.2017.01008. Figures cited are for severe stress at the young microspore stage in cultivars Jinmai47 (tolerant) and Shiluan02-1 (sensitive).

22. Christian, J.I., Martin, E.R., Basara, J.B., Furtado, J.C., Otkin, J.A., Lowman, L.E.L., Hunt, E.D., Mishra, V., & Xiao, X. (2023). “Global projections of flash drought show increased risk in a warming climate.” Communications Earth & Environment 4, 165. DOI 10.1038/s43247-023-00826-1. Cropland figures are for SSP5-8.5, 2015 to 2100.

23. Kraklow, V.A., Paff, K., Comeau, D., Solander, K., Pitts, T.R., Price, S.F., Xu, C., et al. (2026). “Impact of drought on global food security by 2050.” Nature Communications 17, 1099. DOI 10.1038/s41467-025-67862-7. Modeling uses a process-based crop model within an Earth system model under SSP2-4.5 across four ESMs. The frequently quoted “3.6 percent” figure is the maximum for soybean alone, not a combined-staple global loss.

24. The mechanism is osmotic before it is toxic: dissolved salts lower the water potential of the soil solution, so a root must generate more suction to draw water than it can produce, and the plant experiences physiological drought in visibly wet soil. Sodium and chloride ion toxicity and induced nutrient deficiency follow. FAO and the Intergovernmental Technical Panel on Soils, Soil Salinization as a Global Major Challenge (Soil Letter No. 3, 2021), https://www.fao.org/global-soil-partnership/resources/highlights/detail/en/c/1412475/

25. Reclaiming a salinised field requires applying water beyond the crop’s needs so that the excess percolates through the root zone carrying dissolved salts with it, the leaching requirement, together with functioning subsurface drainage to remove that water; without drainage, leaching raises the water table and returns the salt by capillary rise. Both are standard irrigation-management practice and both are water-intensive. FAO Irrigation and Drainage Paper 29, Water Quality for Agriculture (Ayers & Westcot), and FAO, Global Status of Salt-Affected Soils (2024).

26. Food and Agriculture Organization of the United Nations (2024). Global Status of Salt-Affected Soils. Global Soil Partnership / International Network of Salt-Affected Soils, launched December 11, 2024, the first global assessment in fifty years. https://www.fao.org/newsroom/detail/fao-launches-first-major-global-assessment-of-salt-affected-soils-in-50-years/en. Note: this assessment supersedes the widely circulated figure of 1.1 billion hectares.

27. Qadir, M. et al. (2014). “Economics of salt-induced land degradation and restoration.” Natural Resources Forum 38(4): 282–295. DOI 10.1111/1477-8947.12054. The figure is in 2013 dollars and covers lost crop production only; infrastructure damage, health effects, land abandonment and lost farm employment have never been priced globally. FAO’s Global Status of Salt-Affected Soils (2024) contains no economic figure, and FAO’s Global Soil Partnership still quotes the 2014 estimate. The calorie figure is Russ, J. et al. (2020), Salt of the Earth, World Bank Policy Research Working Paper 9144, a working paper, not peer-reviewed. Note that the widely quoted claim that 2,000 hectares are lost to salt daily is not independent corroboration: it is the press release for the same 2014 study, the same arithmetic of $441 per hectare across 62 million hectares, describing an average over 1990–2013 rather than a current rate.

28. Drechsel, P., Bartram, J., Qadir, M. & Medlicott, K. (2024), npj Clean Water 7: 67, restating Thebo, A.L. et al. (2017), Environmental Research Letters 12(7): 074008, DOI 10.1088/1748-9326/aa75d1. An important qualification: these are modelled proximity-and-dilution estimates that include indirect use, a farmer pumping from a river carrying a city’s discharge, not a census of deliberate sewage irrigation. No global burden-of-disease estimate for wastewater irrigation exists, and the widely repeated figure of “200 million farmers” has no traceable primary source; neither is used here.

29. Qadir, M., Drechsel, P. & Jones, E.R. (2025), Discover Water 5: 74, reporting 2023 data: 267.5 billion cubic metres of municipal wastewater generated globally, 54.7 percent treated and 45.3 percent discharged untreated. By income group the treated shares are 91.5 percent (high income), 48.0 percent (upper-middle), 29.4 percent (lower-middle) and 5.8 percent (low income); sub-Saharan Africa stands at 17.2 percent. Only 37 of 231 countries report agricultural reuse at all.

30. Amoah, I.D. et al. (2016), PLOS Neglected Tropical Diseases 10(12): e0005161. Wet-season Ascaris prevalence among wastewater-irrigating farmers in Kumasi was 15.77 percent against 6.00 percent in matched controls (odds ratio 3.99, 95% CI 1.15–13.86). The corresponding hookworm comparison (12.73 versus 2.00 percent) did not reach significance, and dry-season hookworm showed no excess; this publication cites the Ascaris result only. A frequently quoted pooled odds ratio of 5.49 from a later review conflicts with the primary data, including its own co-author’s, and is not used.

31. Shuval, H. (1993), Water Science and Technology 27(3–4): 167. Some 13,500 hectares of salad crops around Santiago were irrigated with raw sewage; typhoid incidence peaked at 210 cases per 100,000 in 1977 and again in 1982, and Salmonella typhi and Vibrio cholerae were both isolated from irrigation canals. In the April 1991 cholera outbreak, 68 percent of the first 41 cases had eaten raw wastewater-irrigated salad vegetables. Chile prohibited the practice and incidence fell.

32. Soil metal enrichment after roughly 90 to 100 years of untreated wastewater irrigation in the Mezquital Valley: Ziegler Rivera, F.R.A. et al. (2023), reporting chromium 7.7 to 8.6 times and cadmium 3.0 to 4.7 times background. On antimicrobial resistance and pharmaceutical residues in the same soils: Heyde, B.J. et al. (2025), npj Clean Water 8: 6, reporting sul1 resistance-gene abundance roughly two orders of magnitude higher in wastewater-irrigated soil, together with triclosan at 5.7–16.7 mg/kg and carbamazepine at 3.5–7.0 mg/kg.

33. On Israel: World Bank (2017, using 2015 data), 87 percent of treated effluent reused, roughly half of national irrigation water, priced at $0.22–0.34 per cubic metre against $0.66 for fresh water. Higher figures circulate, EPA says “nearly 90 percent,” Qadir et al. (2025) say 99, and this publication uses the World Bank number with its year attached. On the nutrient value and the Mezquital farmers’ objection to the Atotonilco treatment plant: Heyde et al. (2025), as above, valuing the nitrogen at US$179–451 per hectare per year, with nitrogen use efficiency of about 65 percent against under 50 for synthetic fertiliser, and soil carbon roughly 1.5 times higher after forty years.

34. Henderson, J., Gastelle, J., & Caffarelli, P. (2024). Transportation of U.S. Grains: A Modal Share Analysis, 1984–2022 Update. USDA Agricultural Marketing Service, September 2024. 2022 export modal shares: all grains 44% barge, 45% rail, 11% truck. https://www.ams.usda.gov/sites/default/files/media/TransportationofUSGrainsModalShare1984_2022.pdf On the figure this corrects: the claim that barges carry 92 percent of U.S. agricultural exports originates on the National Park Service’s “Mississippi River Facts” page, which states that the industry developed in the river basin, a 31-state drainage: produces 92 percent of the nation’s agricultural exports. It is a statement about farmland, not barges, and the page carries no citation for it. https://www.nps.gov/miss/riverfacts.htm

35. American Farm Bureau Federation Market Intel, “Low Mississippi River Levels Drive Up Grain Transportation Costs,” September 28, 2023. https://www.fb.org/market-intel/low-mississippi-river-levels-drive-up-grain-transportation-costs. Memphis gauge records via NOAA/AP.

36. The Ohio River flow contribution is from NOAA/NIDIS, “Drought and Water Update: Mississippi River Basin,” September 18, 2025, https://www.drought.gov/drought-status-updates/drought-and-water-update-mississippi-river-basin-2025-09-18. The shipment figures are from American Farm Bureau Federation Market Intel, “Low Mississippi River Levels Again Jeopardize Farm Income,” September 24, 2025, citing USDA AMS: total southbound Mississippi grain shipments fell from 2.4 million short tons to 502,000 in little more than a month, a 79 percent drop, within which corn movements fell 72 percent. https://www.fb.org/market-intel/low-mississippi-river-levels-again-jeopardize-farm-income

37. Mitchell, J. & Biram, H. (2025). Analysis of 2022 low-water impacts on Arkansas and Mississippi soybean basis, Applied Economic Perspectives and Policy, June 16, 2025. Summary: University of Arkansas Division of Agriculture, https://www.uaex.uada.edu/media-resources/news/2025/june/06-16-2025-ark-low-mississippi-soybeans.aspx

38. USDA Agricultural Marketing Service, Grain Transportation Report, issue dated August 6, 2026 (week ending August 1, 2026): St. Louis barge rate $25.46/ton, 638% of tariff, up 39% year over year; Cairo–Memphis $18.56/ton, 591% of tariff, up 47%. https://www.ams.usda.gov/sites/default/files/media/GTR08062026.pdf

39. U.S. Drought Monitor (NDMC/USDA/NOAA), Percent Area of Contiguous U.S., valid August 4, 2026: D1--D4 48.54%, versus 29.85% on August 5, 2025. Note the honest counterpoint: D1+ coverage stood at 60.92% three months earlier, on May 5, 2026. https://droughtmonitor.unl.edu/

40. USDA, “Agriculture in Drought,” August 4, 2026. https://www.usda.gov/sites/default/files/documents/AgInDrought.pdf

41. USDA World Agricultural Supply and Demand Estimates, WASDE-673, July 10, 2026: 2026/27 all-wheat production 1,536 million bushels on 32.1 million harvested acres, against 1,985 million bushels on 37.2 million acres in 2025/26; ending stocks 722 million bushels. https://esmis.nal.usda.gov/publication/world-agricultural-supply-and-demand-estimates/2026-07-10

42. USDA NASS Crop Progress, week ending August 2, 2026: corn 61% good/excellent versus 73% a year earlier; soybeans 63% versus 69%. https://www.nass.usda.gov/Publications/National_Crop_Progress/

43. USDA NASS, Cattle, January 30, 2026: all cattle and calves 86.2 million head as of January 1, 2026; 2025 calf crop 32.9 million. https://www.nass.usda.gov/Newsroom/2026/01-30-2026.php. The characterization of this as the smallest inventory since 1951 is a reading of the NASS historical series by trade analysts; NASS’s own release does not use that framing. The July 1, 2026 report put all cattle and calves at 94.2 million, with beef cows down 1 percent year over year and replacement heifers up 1 percent: early, fragile signs of herd rebuilding that further tighten near-term beef supply.

44. U.S. Department of the Interior, “Reclamation Publishes Final Environmental Impact Statement for the Future of the Colorado River,” July 31, 2026. https://www.doi.gov/pressreleases/reclamation-publishes-final-environmental-impact-statement-future-colorado-river

45. Richter, B.D. et al. (2024). “New water accounting reveals why the Colorado River no longer reaches the sea.” Communications Earth & Environment 5, 134. DOI 10.1038/s43247-024-01291-0. The 52 percent figure uses a denominator including reservoir evaporation and riparian evapotranspiration; the 74 percent figure counts direct human consumption only. Both describe agriculture; they are not interchangeable.

46. Bureau of Reclamation, Post-2026 Colorado River Operations, Final Environmental Impact Statement, July 31, 2026. The seven basin states missed the February 14, 2026 deadline for a consensus agreement, transferring the decision to the Interior Secretary. https://www.usbr.gov/ColoradoRiverBasin/post2026/index.html

47. DLA Piper, “New Federal Framework Charts the Colorado River’s Next Decade,” August 2026, analyzing the Final EIS. https://www.dlapiper.com/en-us/insights/publications/2026/08/new-federal-framework-charts-the-colorado-rivers-next-decade. For 2027–2028 the framework assumes a 1.5 maf Lower Basin shortage, with cuts of 760,000 AF (Arizona), 440,000 AF (California) and 50,000 AF (Nevada).

48. Allocation of Lower Basin reductions (approximately 60 percent Arizona, 35 percent California, 4 percent Nevada), the October 1, 2026 effective date, and state reactions: Deseret News, “Winners and losers of the Interior Department’s Colorado River water proposal,” August 5, 2026, https://www.deseret.com/politics/2026/08/05/winners-and-losers-of-interior-departments-colorado-river-water-proposal-federal-cuts-to-arizona-and-lower-basin-water-use/. Note an unresolved discrepancy this publication has not been able to settle: reporting on the Final EIS gives the framework an October 1, 2026 effective date, coinciding with the expiry of the current guidelines, while legal analyses of the same document describe implementation beginning January 1, 2027. Readers should treat the precise commencement date as unsettled pending the Record of Decision.

49. The Secretary of the Interior’s role as water master of the Lower Colorado River derives from the Boulder Canyon Project Act of 1928 and the Supreme Court’s decree in Arizona v. California, 376 U.S. 340 (1964), under which the Secretary contracts for and delivers all mainstream water below Lee Ferry. The annual shortage-tier mechanism operates under the 2007 Interim Guidelines and the 2019 Drought Contingency Plan: Reclamation’s August 24-Month Study projects Lake Mead’s January 1 elevation, which determines the shortage tier and the resulting delivery reductions for the following calendar year. As of this writing Reclamation had published the Final EIS but had not signed a Record of Decision.

50. Yuma figures: Arizona Farm Bureau, “Yuma Agriculture: A Powerhouse of Economic Impact and Efficiency,” April 15, 2025, citing Dr. George Frisvold, University of Arizona. Imperial Valley production, employment and unemployment figures: Capital & Main, “A Drying Colorado River Threatens Imperial Valley’s Future,” April 2, 2026. https://capitalandmain.com/a-drying-colorado-river-threatens-imperial-valleys-future

51. Yuma figures: Arizona Farm Bureau, “Yuma Agriculture: A Powerhouse of Economic Impact and Efficiency,” April 15, 2025, citing Dr. George Frisvold, University of Arizona. Imperial Valley production, employment and unemployment figures: Capital & Main, as at note 50., the Imperial Irrigation District holds the single largest entitlement on the river with roughly 97 percent going to agriculture; agriculture is one job in six in Imperial County; county unemployment is about 19 percent.

52. Public Policy Institute of California, “Policy Brief: The Future of Agriculture in the San Joaquin Valley,” February 2023. The 900,000-acre figure is an explicitly worst-case, no-adaptation scenario; roughly 500,000 acres is the projection under active management. https://www.ppic.org/publication/policy-brief-the-future-of-agriculture-in-the-san-joaquin-valley/. Probationary subbasin status and fee schedules from the California State Water Resources Control Board, https://www.waterboards.ca.gov/sgma/

53. Lees, M. & Knight, R. (2024). Communications Earth & Environment, November 19, 2024. DOI 10.1038/s43247-024-01778-w. Friant-Kern Canal figures: SJV Water, “Eight years, ten miles and $325 million later, first phase of Friant-Kern Canal fix celebrated,” https://sjvwater.org/eight-years-ten-miles-and-325-million-later-first-phase-of-friant-kern-canal-fix-celebrated/

54. USDA Economic Research Service, Food Price Outlook, updated July 24, 2026: all food +3.1%; beef and veal +10.7% (prediction interval 7.2–14.6%); fresh vegetables +6.8%; eggs −30.7%. https://www.ers.usda.gov/data-products/food-price-outlook/summary-findings

55. Rabbitt, M.P., Reed-Jones, M., Hales, L.J., Suttles, S., & Burke, M.P. (2025). Household Food Security in the United States in 2024. USDA Economic Research Service, Economic Research Report No. 358, December 2025. 13.7 percent of households (18.3 million) food insecure; 5.4 percent with very low food security. Population figures via FRAC’s reading of the same report. https://ers.usda.gov/sites/default/files/_laserfiche/publications/113623/ERR-358.pdf

56. USDA announced discontinuation of the Household Food Security report series in September 2025. Alliance to End Hunger, “USDA’s Final Household Food Security Report,” January 12, 2026. https://alliancetoendhunger.org/usdas-final-household-food-security-report-a-current-snapshot-and-a-loss-for-the-future/

57. Urban Institute, Food Insecurity Remained High in 2025, As Safety Net Cuts Loom (March 2026), reporting its Well-Being and Basic Needs Survey fielded December 2025. https://www.urban.org/research/publication/food-insecurity-remained-high-2025-safety-net-cuts-loom. The divergence from USDA is not a denominator effect; USDA’s own adult-level rate is roughly 13 percent, close to its household rate. It is a survey-mode effect: the Urban survey is self-administered on a probability-based panel (Ipsos KnowledgePanel, address-recruited, including households without internet), lacks the Current Population Survey’s income screener, and respondents report food hardship more readily without an interviewer present. Neither Feeding America nor food-bank surveys support a national one-in-four adult figure: Map the Meal Gap 2025 reports 14.3 percent of individuals and is a downscaling model of the federal data, not an independent survey; the Greater Boston Food Bank’s April 2026 report finds 40 percent of Massachusetts households food insecure, a state figure, and a household one.

58. SNAP provisions of H.R. 1 (2025), per Feeding America Action FAQ, updated June 24, 2026: approximately $200 billion in reductions over ten years; state administrative cost share rising from 50 to 75 percent effective October 1, 2026; benefit cost-sharing of up to 15 percent from October 1, 2027; ABAWD age ceiling raised from 54 to 64. https://feedingamericaaction.org/wp-content/uploads/FAQ_HR1.pdf

59. This is a claim about the record, and it is documented. The Internal Displacement Monitoring Centre began recording drought as a displacement trigger only in 2017, and attributes recent year-on-year declines in recorded drought displacement partly to reduced reporting rather than reduced displacement. Sudden-onset disasters generate evacuation orders and camp registrations; drought generates a household that sells its assets and leaves, recorded, if at all, as an economic migrant. See “What This Publication Cannot Tell You” below.

60. Hoffmann, R., Abel, G., Malpede, M., Muttarak, R., & Percoco, M. (2024). “Drought and aridity influence internal migration worldwide.” Nature Climate Change. DOI 10.1038/s41558-024-02165-1.

61. Linke, A., Leutert, S., Busby, J., Duque, M., Shawcroft, M., & Brewer, S. (2023). “Dry growing seasons predicted Central American migration to the US from 2012 to 2018.” Scientific Reports. DOI 10.1038/s41598-023-43668-9. The headline effect is 70.7 percent; the range across model specifications is 8.9 to 257.0 percent. The authors note that apprehension data excludes single adults and successful crossings, and cannot confirm that drought-affected households are the same individuals arriving at the border.

62. Dupre, S.I., Harvey, C.A., & Holland, M.B. (2022). “The impact of coffee leaf rust on migration by smallholder coffee farmers in Guatemala.” World Development 156, 105918. DOI 10.1016/j.worlddev.2022.105918.

63. International Organization for Migration, via UN News, April 2026: nearly 62,000 drought displacements in Somalia since the start of 2026; three of every four new displacements in Somalia drought-related; a 22 percent year-over-year increase, with a further 125,000 projected by June 2026 if conditions persisted. https://news.un.org/en/story/2026/04/1167379

64. World Food Programme and the Federal Government of Somalia: 6.5 million people projected in IPC Phase 3 or above by March 2026, including 2 million in Phase 4; 1.84 million children acutely malnourished; southern Somalia’s Deyr cereal harvest 83 percent below the 1995–2025 average. https://www.wfp.org/news/somalias-humanitarian-crisis-worsening-65-million-people-facing-high-levels-hunger-federal

65. IOM Displacement Tracking Matrix, Climate Emergency Tracking, Iraq: 28,379 families (170,274 individuals) displaced by drought 2016–2024 across 513 locations in twelve central and southern governorates. https://iraq.iom.int/stories/navigating-rivers-and-new-cities-migration-and-water-scarcity

66. Abdolkarim Hosseinzadeh, Iran’s Vice President for Rural Development and Disadvantaged Areas, statement of January 6, 2026: of Iran’s 69,000 villages, 38,000 remain inhabited and 31,000 are deserted. The figure is cumulative over roughly five decades and is not attributable to water scarcity alone. https://www.kurdistan24.net/en/story/885570/

67. Michel, D., Todman, W., & Jun, J. (2025). Satellite Imagery Shows Tehran’s Accelerating Water Crisis. Center for Strategic and International Studies, November 25, 2025: reservoir declines exceeding 70 percent at Lar and Latyan between June and November 2025; Tehran Province losing 101 million cubic meters of groundwater annually; roughly one-third of supply lost to leaks and theft. President Masoud Pezeshkian has publicly warned that evacuation of parts of Tehran may become necessary. https://www.csis.org/analysis/satellite-imagery-shows-tehrans-accelerating-water-crisis

68. Clement, V., Rigaud, K.K., de Sherbinin, A., Jones, B., Adamo, S., Schewe, J., Sadiq, N., & Shabahat, E. (2021). Groundswell Part 2: Acting on Internal Climate Migration. World Bank, September 13, 2021. The 216 million figure is the pessimistic reference scenario (RCP 8.5 + SSP4); the full modeled range is 44 to 216 million. The Bank estimates concerted action could reduce the scale by as much as 80 percent.

69. European Commission Joint Research Centre (2025). Forecasting Climate Migration; How Much Do We Really Know? JRC140538. https://knowledge4policy.ec.europa.eu/sites/default/files/JRC140538_Climate_modeling_factsheet_final_final.pdf

70. Internal Displacement Monitoring Centre, Global Report on Internal Displacement 2026, May 2026. IDMC began recording drought as a displacement trigger only in 2017. Total 2025 internal displacements: 62.2 million, of which 29.9 million were disaster-related; 82.2 million people living in internal displacement at year-end. https://www.internal-displacement.org/global-report/grid2026/

71. Tárraga, J.M. et al. (2024). “Causal discovery reveals complex patterns of drought-induced displacement.” iScience. DOI 10.1016/j.isci.2024.110628. Analysis covers 74 Somali districts, weekly, 2016–2023; 74 percent of surveyed households at arrival sites named food as their primary need.

72. Food Security Information Network / Global Network Against Food Crises, Global Report on Food Crises 2026, April 24, 2026: 266 million people in 47 countries at IPC/CH Phase 3 or above; conflict the primary driver for 147.4 million in 19 countries; extreme weather the primary driver for 87.5 million in 16 countries, down from 96.6 million in 2024. https://www.wfp.org/news/acute-food-insecurity-and-malnutrition-remain-alarmingly-high-crises-deepen-un-eu-and-partners

73. Kelley, C.P., Mohtadi, S., Cane, M.A., Seager, R., & Kushnir, Y. (2015). “Climate change in the Fertile Crescent and implications of the recent Syrian drought.” PNAS 112(11): 3241–3246. Critique: Selby, J., Dahi, O.S., Fröhlich, C., & Hulme, M. (2017). “Climate change and the Syrian civil war revisited.” Political Geography 60: 232–244. DOI 10.1016/j.polgeo.2017.05.007.

74. Eklund, L., Theisen, O.M., Baumann, M., Tollefsen, A.F., Kuemmerle, T., & Nielsen, J.Ø. (2022). “Societal drought vulnerability and the Syrian climate-conflict nexus are better explained by agriculture than meteorology.” Communications Earth & Environment. DOI 10.1038/s43247-022-00405-w.

75. Kelley, C. et al. (2017). “Commentary on the Syria case: Climate as a contributing factor.” Political Geography. DOI 10.1016/j.polgeo.2017.06.013. The authors now describe 1.3 to 3 million people as affected by the drought.

76. Cattaneo, C. & Peri, G. (2016). “The migration response to increasing temperatures.” Journal of Development Economics 122: 127–146. DOI 10.1016/j.jdeveco.2016.05.004. 115 countries, 1960–2000. See also Foresight (2011), Migration and Global Environmental Change, UK Government Office for Science: “Environmental change is equally likely to prevent migration as it is to cause migration.”

77. Link, A-C., Oakes, R., Durand-Delacre, D., Thalheimer-Prezyna, L., & van der Geest, K. (2025). “To what extent do climatic stressors drive human mobility in the world’s drylands? A systematic review of empirical evidence.” Population and Environment 47(1), art. 16. DOI 10.1007/s11111-025-00486-7. Of the 183 studies reviewed, 137 examined climatic effects on mobility; of those, 15 percent found climatic stressors decreased mobility. Only ten of the 183 addressed involuntary immobility at all, and only one made it the primary focus.

78. Imperial Irrigation District has conserved more than 9 million acre-feet since 2003 and expanded its 2026 conservation program to 800,000 acre-feet, compensating farmers through its Deficit Irrigation Program. Federal Lower Basin System Conservation payments are widely reported at $400 per acre-foot, with IID’s December 2023 agreement reported at a substantially higher rate; this publication has not independently verified that figure against a Reclamation contract document. https://www.western-water.com/2026/05/16/imperial-irrigation-district-expands-colorado-river-conservation/

79. USDA announced in September 2025 that it would discontinue Household Food Security in the United States after the 2024 report, ending an annual series running roughly thirty years; the final edition appeared in December 2025. Alliance to End Hunger, January 12, 2026, https://alliancetoendhunger.org/usdas-final-household-food-security-report-a-current-snapshot-and-a-loss-for-the-future/. On the currency of the aquifer data, see note 8: the most recent aquifer-wide USGS assessment reports water levels as of 2019.

80. Steward, D.R. et al. (2013), as at note 5: “Water use reductions of 20% today would cut agricultural production to the levels of 15–20 y ago, the time of peak agricultural production would extend to the 2070s, and production beyond 2070 would significantly exceed that projected without reduced pumping.”

81. Drysdale, K. & Hendricks, N. (2016). Sheridan County 6 Local Enhanced Management Area, ACCC Fact Sheet No. 7, Kansas State University Department of Agricultural Economics. Farmers in the SD-6 LEMA reduced applied inches per acre by 34 percent, primarily by reducing water intensity rather than idling acres. https://accc.k-state.edu/research/factsheets/FS7_LEMA.pdf. On the current renewal: Kansas Department of Agriculture, Division of Water Resources, Greater GMD4 LEMA, the Chief Engineer issued an Initial Public Hearing Order on August 5, 2026 for the 2028–2032 term. https://www.agriculture.ks.gov/divisions-programs/division-of-water-resources/managing-kansas-water-resources/local-enhanced-management-areas/gmd4-lema

82. Refugees International (2026), A Generational Collapse: Tracking the Toll of Humanitarian Aid Cuts: U.S. humanitarian support fell from approximately $14 billion to $3.7 billion between 2024 and 2025. https://www.refugeesinternational.org/reports-briefs/a-generational-collapse-tracking-the-toll-of-trumps-humanitarian-aid-cuts/

83. Price forecasts are USDA Economic Research Service, Food Price Outlook, updated July 24, 2026: beef and veal +10.7 percent, fresh vegetables +6.8 percent, all food +3.1 percent. The SNAP provisions are those of H.R. 1 (2025): the state share of administrative costs rises from 50 to 75 percent on October 1, 2026, and states begin contributing up to 15 percent of benefit costs on October 1, 2027. Both are sourced in full earlier in this article.

A flash drought in 2012 dried out soil, harming crops in Kansas and several other states. John Moore/Getty ImagesThe Crop Hazard That Skips the Forecast

The last time this publication described a climate driver reshaping the world’s food supply, the story was measured in months. In Scorched Harvest, we wrote about an El Niño building in the tropical Pacific on what we called a twelve-month fuse: a slow, hemisphere-spanning reorganization of where rain falls, its worst harvest failures a year out. Extreme heat is the opposite kind of threat. It gives no fuse, no warning at all.

2024 was the hottest year in the instrumental record, the first full calendar year more than 1.5°C above the pre-industrial baseline, and the capstone of the hottest ten-year stretch ever measured in the history of humankind.¹ 2025 came in second or third (depending on how temperatures around the world are measured), cooler than 2024 only in the sense that one burning furnace runs slightly cooler than another. The rolling three-year average has now crossed 1.5°C for the first time, and the heat waves that swept Europe, Asia, and North America were, by the reckoning of the scientists who study them, the deadliest weather events of the year and, in several cases, roughly ten times more likely than they would have been only a decade ago.² A single European summer put land temperatures past 46°C on the Iberian Peninsula and is estimated to have killed on the order of sixteen thousand people.³

The temptation will be to file this under weather. File it instead under food, and under the speed at which that whole crops can now be lost.

Drought creeps. Floods arrive and recede. An El Niño telegraphs its intentions half a year in advance. Heat is the driver that collapses the warning window to nothing, and just as importantly, it does its damage not at one stage of food production but at every stage at once: in germination, in flowering, in the field as it grows, in the barn, in the soil, in the river, and, on a lag of a single growing season, in prices of the food on your grocer’s shelf.

“A drought you can see coming for months. A heat wave you can see coming only for days. By the time you can feel it, the pollen is already dead, the milk has already stopped, and the forecast of food production for the season has already dropped precipitously.”

The Season That Fails Before It Begins

Every impact to grains that follows, begins with a seed. Heat can reach the crop before the crop even exists. A seed is a living thing running on a slow clock, and the warmth of the new season winds that clock forward toward its birth and growth. The relationship is regular enough that seed scientists have reduced it to a “rule of thumb” nearly a lifetime ago: Within ordinary limits, a seed’s viable life is cut roughly in half for every one per cent rise in its moisture content and then halved for every ten degrees Fahrenheit of storage heat above normal.⁴ The seed held over for next season’s planting does not wait patiently in a warm storeroom. When exposed to high heat before it is planted, the seed ages in fast-forward, its reserves burning down through accelerated respiration and the oxidation of its own oils, until, when it finally goes into the ground, there may be too little of it left for it to wake. It is the same ten-degree arithmetic that will later govern the grain in the storehouse (Figure 5): heat halves the usable life of the seed at the start of the season and then dictates how long grains can be stored at the end of the harvest.

“A seed kept too warm is a seed already spending itself and reducing its viability. By planting time, if too much of it has already been used up a seed may not be able to bring forth a live plant.”

If seeds have been exposed to too much moisture or too high a temperature, the loss shows in the field as thin growth and patchy gaps that a farmer must either replant, at cost and delay, or simply accept as a smaller crop. This loss occurs due to excessive heat even before the first leaf has unrolled. The seed’s ordeal is not confined to the storage bin. Too much heat during the weeks that a seed ripens in the Fall on the mother plant, can leave it weak and short-lived even before it is ever gathered as seed. Drying the seeds after they are harvested at too high a temperature does further, invisible harm. The next risk is that if there is too much heat in the Spring at the time of germination, germination itself can falter in soil that is too hot or too dry, precisely the ground conditions that a heat wave creates.⁵

For a commercial grower, heat impact to seed production through germination is merely a cost and an aggravation, absorbed in part by paying a higher price for a certified seed bought fresh each year. For the many millions of smallholders who save their own seeds and keep them in whatever warmth the season provides, with no cold room and no dehumidifier, the seeds are at risk. The hotter the year, the less of their saved seeds survive to plant the next season. This means that a punishing season leads to a smaller harvest and also reduces the viability of the very seed needed to be used to begin the recovery the next year. Heat, in other words, does not wait for a crop to grow before it begins subtracting from it. It is already at work in the seed bag, and the season can be diminished before the seeds are ever sown.

Figure 1. Warm storage runs the seed’s clock down. Germination remaining after prolonged storage falls sharply as storage temperature rises, so seed held over in a hot storage container loses much of its viability before it is ever planted. The same ~10°F rule that governs the grain in the bin (Figure 5) applies to how many seeds will be viable. The curve is illustrative, anchored to reported rice-storage trials.

Every Agricultural Crop Has a Fertilization Number at Risk in a Heat Dome

Start with the grains, because the grains are the least forgiving part of our food system. Every staple cereal has a temperature above which its reproduction simply fails, and those numbers are lower, and closer to being exceeded, than most people who eat the grain in a bowl of cereal, in bread, in pasta, in a rice bowl or on a cob of corn will ever guess. We truly were in a “Goldilocks” climate before we started burning fossil fuels and emitting CO₂ into the atmosphere.

For the three crops that feed most of humanity, agronomists have measured the critical thresholds for seed set at flowering with unsettling precision: roughly 37.2°C for rice, 37.9°C for maize and just 27.3°C for wheat.⁶ When agronomists talk about “seed set at flowering,” they are specifically looking at the critical pollination window when the crop blooms. During this short phase, if temperature thresholds are exceeded, the extreme heat damages the flower’s delicate reproductive organs rendering pollen sterile or preventing fertilization.

Read the wheat number in that figure again. Wheat, the crop the temperate world treats as its dependable staple, begins to lose fertility at a temperature that a spring afternoon now routinely surpasses in Kansas, Punjab, or the Australian wheat belt. Rice yields begin sliding once the season sits above about 33°C.⁶

What happens to crop production above those set temperatures is not a gentle taper. Flowering is the crop’s single most heat-sensitive hours or days, and heat sabotages fertilization mechanically: it dries and closes the flower, it kills or sterilizes the pollen, it arrests the pollen tube before it can reach the ovule. In maize, the most heat-vulnerable of the three, only a few hot days landing precisely when pollen is going down the silk to the kernels can stall pollen-tube growth and abort kernels wholesale, so that an ear which looked healthy in July fills only halfway by September.⁷ Almost everyone has peeled the husks off an ear of corn only to find many kernels have not filled out. The cruelty of it is the timing: a crop can be perfectly watered, perfectly fertilized, and still be rendered barren by a two-day heat spike that happens to coincide with the wrong week of the life of the wheat, rice or corn plant. The grain is not lost to scarcity of anything. It is solely lost to a thermometer reading above its critical threshold.

Figure 2. Every crop has a number. The air temperature at which seed set fails during flowering (27.3°C for wheat, 37.2°C for rice, 37.9°C for maize) set against the 40°C afternoons a modern heat wave now clears routinely. Solid bars mark viable temperature; hatched zones, sterility. Thresholds from the reproductive-stage heat-tolerance literature.

Interfering with reproduction is not the only way heat causes dramatic crop loss. The heat wave does not need to come during the fertilization period to steal the vitality of the grains. Heat also cuts into crop yield in three slower ways that are just as devastating and just as predictable.

1) The first is haste. Grain fills the corn, rice and wheat kernels during a fixed window between flowering and maturity, and a plant that is exposed to high temperatures that speeds up its growth processes is a plant that matures too fast: high temperatures accelerate development, shorten that grain-fill window, and rush the crop to a premature, lightweight maturity. Exposed to excess heat, the plant checks every growth hallmark on the calendar but still delivers less, because it spent fewer days pouring starch into each kernel. A field that looks like it ripened early did not ripen. It surrendered.

2) The second is thirst, not the crops being thirsty, but the atmosphere and soil’s thirst. Every degree of additional heat widens the vapor-pressure deficit, the gap between how much moisture the air can hold and how much it actually does hold, and that gap is the engine that pulls water out of soil and the plants’ leaves. This is why a heat wave can hollow out a harvest even where the rain gauge reads normal. The rain fell; but the heat simply took back the rain before the roots could use it, evaporating soil moisture and forcing the plant to close its pores and stop growing to keep from desiccating. Rainfall near average is no defense against an atmosphere that is poised to evaporate soil moisture faster than the rain can fall.

3) The third is the water demand that no one can create out of thin air. Extreme heat drives irrigation demand to its annual peak at the exact moment water is already being drawn down from the reservoirs, aquifers, rivers, and snowpack that irrigation depends on. Demand spikes hit as supply falls, and the two curves cross at the worst possible time which causes the farmer with a well or a pump to discover that due to the excess heat, the season now asks for more water than the snowfall, landscape and aquifer have to give.

The preview to the damage extreme heat can cause is already on the record. Europe’s farm sector has been losing an estimated €28 billion a year to extreme weather, only a fraction of it insured, and the summer of 2025 alone extreme heat reportedly damaged a large share of France’s corn crop, took roughly half of some regions’ carrots, and killed poultry by the flock.^{8 9}

The record has a human face. In Punjab, Gurjinder Singh watched the March heat of 2022 rush his wheat to a premature, hollow ripeness; his land, which normally gives 23 to 25 quintals an acre, yielded 14 to 15, what he called “a substantial 40 percent loss.”¹⁰

The Flash Droughts That Destroy a Crop in a Fortnight

There is now a name for the fast version of what the last section described, and it belongs almost entirely to the vocabulary of this century. Scientists call it a “flash drought.” This is something different from the slow, seasonal deficit that creeps across a landscape over years. A flash drought is a drought that ignites and matures in the space of a few weeks, when a run of extreme heat, wind, and cloudless sun drives the atmosphere’s thirst so high that it pulls moisture from the soil faster than normal rain can replace it. The term only was coined in 2002, and for a decade it sat as a specialist’s curiosity; monitoring it now means watching not the rain gauge but the rate at which the air is drying, a jump of half again in evaporative demand over two weeks is enough to declare one.¹¹

The fast droughts appear to be winning and being the prominent kind of drought. The first global study to settle the question, published in Science in 2023, found that over the past sixty-four years droughts have begun to intensify measurably faster across nearly three-quarters of the world’s land regions, with a clear statistical shift from the slow-onset kind of drought toward the flash kind, a transition the authors trace directly to human-caused warming, through the very amplification of evaporation and rainfall deficits that a hotter atmosphere produces. The slowest of the old climate disasters is learning to sprint. The acceleration is sharpest over Europe, northern and eastern Asia, the Sahel, and the western coast of South America, and it is projected to spread across most of the planet’s land if warming continues.¹² The old kind of drought is still happening as well.

“A slow drought is a thing you watch approach for a season. A flash drought is a thing you discover has already happened in a matter of weeks.”

The bill for a drought that moves this fast is steep, precisely because speed removes the time in which anyone might have adapted. The 2012 event that first seized the country’s attention carved more than thirty billion dollars from the American economy in a single summer; a 2017 flash drought across the Dakotas and Montana took an estimated $2.6 billion from agriculture alone, and the frequency of flash droughts is set to climb. On the high-emissions path, the annual risk of flash drought on North American cropland is projected to rise from roughly a third today to about half by the end of the century, and the increases are projected to be higher still in Europe.¹³

These are not projections waiting on the future; the events are already arriving in series. In June 2023 a flash drought spread across the U.S. Midwest until it covered close to two-thirds of the nation’s corn, catching the crop just as it entered pollination; by that October the Mississippi River had fallen to record-low stages, halting barge traffic and drawing saltwater upstream into the drinking water of southern Louisiana. That same autumn a second flash drought seized the Southeast, wilting newly planted wheat, rye, and hay and forcing so many ranchers onto bought feed that the USDA paid its largest livestock-forage disaster relief on record, concentrated in Mississippi, Tennessee, and Alabama. The following summer heat settled over the mid-Atlantic, the rain stopped, and corn from the Carolinas to Pennsylvania shriveled in a matter of weeks.¹⁴

Each of these droughts matured too quickly for the farmers to do anything to prepare, which is the whole of the danger. The speed and regional impacts of these droughts are the reason flash droughts belong at the center of this account rather than its margins. It is heat’s signature written into the water cycle itself: the same argument as every section before it, arriving from a different direction. The hazard that once granted a season’s warning now grants only a fortnight to prepare.

Behind the numbers are the people who watched it happen. Greg Lee, a North Dakota farmer who had worked the same ground for forty years, lived through that 2017 Northern Plains flash drought; asked what it did to his fields, he put it in one flat sentence: “The crops just didn’t come out of the ground.”¹⁵

Figure 3. The drought that moves at heat-speed. A conventional drought lowers soil moisture to drought levels over a season or more; a flash drought, driven by extreme heat and evaporative demand, can do it in two to four weeks, collapsing the window in which a farmer might respond. The curves are a schematic of the two onset regimes; one gives no chance of survival.

Animals Cannot Sweat Off Extreme Heat When It Comes

Then there is what happens in the barn, where heat is a matter of animals in distress.

A dairy cow is a walking furnace: a high-producing animal generates enormous metabolic heat, and unlike a human, a cow sweats at only about a tenth the rate, far too little to shed the heat it makes. Cows begin to suffer heat stress at temperatures that feel merely warm to the people milking them. The cows that produce the greatest amount of milk can begin to experience reduced milk production if temperatures in the barn reach even the mid-60s Fahrenheit (17-19 degrees Celsius). As the combined heat-and-humidity load climbs, cows will eat less, stand more, and give less milk. Mild heat stress alone can cut milk by a couple of pounds per cow per day; a hard summer with weeks of high heat can pull daily yields down by double digits.^{16 17} If the heat-humidity index in the barn is raised past the high 80s Fahrenheit for even several hours, with no cool night for the cows to recover in, vulnerable milking cows do not merely produce less, they die, in the barn, in the feedlot and in the pasture, too.¹⁸

The losses are not hypothetical and not small. Heat stress already costs U.S. livestock producers on the order of $2.9 billion a year in today’s dollars, the majority of the losses borne by the dairy industry.¹⁷ A heat wave, in other words, does not wait for the harvest to translate itself into less food, it takes the milk that afternoon and it takes the beef of the cow later that week if the high temperatures hold. The heat wave that enveloped northern India in 2022 reportedly cut milk output by as much as 15 percent at the same time it reduced wheat yields by anywhere from a tenth to a third across several Indian states. A single excess heat event depletes the food system at both ends at once.⁸The same arithmetic runs in American barns. In Fremont, Wisconsin, Jacob Hoewisch, who milks a fifth-generation herd, loses three to ten pounds of milk from each cow on the hottest days, and watches the heat reach past the milk to the animal itself: a cow whose “body overheated for too long of a period”, he says, can lose the calf she is carrying.¹⁹

Figure 4. The animals cannot sweat it off. Milk yield falls as the temperature - humidity index climbs, well before the barn feels dangerous to the people in it, and above roughly THI 84, with no cool night to recover in, vulnerable cattle begin to die. The curve is illustrative of the measured response; anchor values are drawn from dairy-science field studies.

The barn is not only a dairy barn. Birds have no sweat glands at all; they shed heat only by panting, and a fast-growing broiler runs so hot inside that a single bad afternoon can kill it. Chickens, turkeys, and ducks are all exposed, and turkeys, the largest of the three, are among the most heat-fragile birds on any farm; ducks cope only where they can reach the water, they use to cool themselves. When a heat wave swept France in the summer of 2026, poultry sheds turned into ovens and birds died by the million, with turkeys taking the heaviest toll. It was not the first time. A comparable heat wave in 2003 killed an estimated four to five million French chickens and turkeys, about two percent of the national flock.²⁰ Sheep and goats aren’t much better off, because they cool by panting instead of sweating, and heat and drought wipe out both their grazing pastures and the lamb and kid crop that herding families in the Global South live on. Because chicken is now the most-eaten meat on Earth, a bird engineered to fatten fast and left without any way to sweat sits close to the center of the food supply, not at its edge.

The Harvest Lost After the Harvest

Every section so far has followed heat into the field and the barn and then stopped at the moment of harvest, as though a crop once cut was a crop secured. It is not. Heat keeps working on food after it leaves the ground, and the losses it inflicts in the bin, the barn, and the cold chain are large, accelerating, and almost entirely absent from the public conversation about climate and hunger.

Begin in the grain bin, where a harvest is supposed to be safe. Stored grain is not inert; it breathes, and warm, damp grain breathes hard, generating its own heat and moisture until pockets of it form “hot spots” that bloom with mold, draw insects, and cake into spoiled, unsellable crust. The governing rule is unforgiving: for every roughly ten degrees Fahrenheit that grain temperature rises, the safe storage time is cut about in half.²¹ It is the very rule that governs the viability of the seed at the other end of the season, too (Figure 1). Insects that lie dormant in cool grain breed freely between 70 and 90°F, precisely the range a warming climate now delivers for much of the year. The mold that follows does not merely reduce weight; it strips nutritional value and, worse, can leave behind mycotoxins, and mycotoxins, once formed, are heat-stable, surviving the milling and cooking that kill the fungus that made them.²¹

The most dangerous of those toxins is now migrating with the heat. Aflatoxin, the most acutely toxic and most tightly regulated of the mycotoxins, and a known liver carcinogen, is made by a fungus that flourishes on heat- and drought-stressed crops in the field and on warm, humid grain in storage. For most of history it was a problem of the tropics and the American South. That is changing: a landmark 2016 study projected that just two degrees of warming, a level now expected between roughly 2030 and 2050 rather than in some distant future, would make aflatoxin a genuine food-safety problem in European maize, driving its contamination index up more than ninety per cent and pushing its range northward into countries that had never had to reckon with it.²² The projection is now being met by observation. Aflatoxin-producing fungi, once rare in temperate Europe, are established across the south and center of the European continent; and a 2026 analysis of more than seventeen thousand samples in China found contamination measurably higher than a decade earlier, with temperature alone explaining nearly half of the increase.²²

The barn tells the same story in a different crop. Hay baled even slightly too wet when baled or stored is a constant hazard when heat and humidity compress the drying window, the hay begins within days to mold and ferment, its sugars and proteins are consumed by microbes and locked, through the browning of the Maillard reaction, into forms livestock cannot digest. A bale of hay can shed much of its feed value while still smelling sweet. Push the internal temperature past about 170°F, as dense wet bales can, and the same microbial heat that spoils the forage causes spontaneous combustion and sets the barn on fire.²³ Rice, the staple of half the planet, most of it warehoused in the hottest, most humid corners of the world, deteriorates faster the warmer it is held, turning rancid, yellowing, shedding milling yield and cooking quality, and, above about sixteen per cent moisture, growing the very molds that make aflatoxin.²⁴

Then there is everything that must stay cold. Roughly a third of all the food the world produces is already lost or wasted, and about a seventh of it has gone bad before it ever reaches a retailer and much of that for want of refrigeration that a hotter climate makes necessary.²⁵ Meat, dairy, fish, and fresh produce spoil in hours, not weeks, when a heat wave overwhelms a cold chain, or when the refrigeration or the power that runs it fails for any reason. Without refrigeration, warmth speeds the bacteria that turn spoiled food dangerous and is a driver behind hundreds of thousands of deaths from foodborne illness each year.

The newest evidence closes the loop, identifying the rest of the crop losses caused by climate-driven heat. A 2026 study found that heat stress raises the share of the crop lost at the moment of harvest itself, a direct climate channel into food loss that had simply gone unmeasured.²⁵ It was the first study to show, with data, that high temperatures sharply increase food loss during harvest, finding that heat-induced losses are three to four times greater in manual than in mechanical harvesting. This means smallholder farms across the Global South are at greatest risk of crop spoilage during a hot harvest.

None of this shows up in a yield statistic. A crop can be grown, survive the flowering heat and the flash drought and the exhausted river, be cut and counted as a good harvest and still be halved while it is being transported to market and sold to people who need food, by the same warmth that menaced the crop in the field. It is the last stage heat attacks, and the most invisible, and it arrives at the same counter, in the same currency, as all the rest of the attacks heat mounts on food production around the world.

Figure 5. Every ten degrees halves the storage time. Grain in storage is not inert; it self-heats, and the time it can be held safely falls by roughly half for each ~10°F rise in grain temperature, so a hot summer bin turns a season’s reserve toward spoilage far faster than a cool one; this is the same clock that governs seed viability at planting (Figure 1). The curve is a schematic of that rule of thumb.

Why Heat is Different Than Other Climate Shocks

Put the pieces together and the argument writes itself. Heat is not one hazard on a list of climate hazards. It is the master variable, the climate driver that attacks the field, the soil, the water, and the animals over wide areas simultaneously, and does it faster than any warning system was built to detect and sound the alarm.

Consider what separates heat from every other climate shock this 4Hunger.org tracks. The fertilizer crisis we documented in Fertilizer Famine moves on a one-season lag. The American megadrought in Parched Ground took many years to form. The El Niño in Scorched Harvest announces itself half a year ahead with the worst impacts predicted a year in advance. Every one of those is a slow emergency you can, in principle, prepare for.

Heat is the fastest climate impact. The pollen in wheat, corn and rice can die within hours if heat hits during the days when the flower is open. The production of milk drops in cows the same week the barn overheats. The soil gives up its moisture across a single scorching fortnight. There is no season of grace between the forecast of heat domes and the loss, because for a crop at flowering, or at other stages of growth, the forecast of extreme heat can become an unstoppable loss.

That speed is precisely why heat has become the biggest climate driver of agricultural loss right now. In a food system already stripped of its buffers, with reserves drawn down, fertilizer in short supply, and with the American food production shock absorber running on empty, the hazard that can erase a harvest in an afternoon is the one the world is least equipped to catch. Heat is the climate impact that is arriving most often, on a planet that has now made 40°C afternoons a routine feature of the growing season on four continents.

“Every other threat to the harvest gives you time to act. Heat only gives you a chance if you have already acted. That is the whole of the difference, and it is the difference between a hungry year prevented and a hungry year explained only afterward.”

Figure 6. One shock, every stage at once. Unlike a drought or an El Niño, extreme heat strikes crops, soil, water, and livestock simultaneously and within days and then travels downstream to the harvest that season, to prices in the months after, and, on a lag of roughly a year, to hunger.

What It Means at the Checkout Line

No American eater should mistake a cool morning for immunity from the consequences of a heat wave. The price of food in an American grocery store is set in a global market, and when wheat bakes in Australia, when heat sterilizes rice across Asia, when Europe’s corn and dairy come in short, U.S. commodity prices rise regardless of the weather in Kansas. This is not because American grain is scarce, but because American grain sells into world prices. The damage from a heat dome does not travel at the speed of a headline; it moves at the speed of a growing season, and it moves, reliably and predictably, to the consumer’s receipt everywhere in the world.

The people that the price increases from heat damage moves toward, are the ones least able to pay. Roughly one in four American adults now lives in a food-insecure household, with the lowest-income fifth of households spending a third of their income on food.²⁶ Congress has made cuts to the very safety net that stands between price volatility and hunger for tens of millions of Americans. Every percentage point of heat-driven food inflation that lands in 2027 will land on families who did little to warm the planet and can least afford the rising prices.

Gurjinder Singh, the Punjab wheat farmer, watching his crop shrivel in a March that arrived with temperatures not normally seen until May; Jacob Hoewisch, the Wisconsin dairyman, running fans on electricity he cannot afford over cows that have all but stopped giving milk; the shopper in Ohio staring at the price of ground beef at $6.82 a pound, none of them chose this. But all of them are downstream of the same warming air that is traveling around the world. All of them will pay for the excess heat in the currency the poor always pay in: skipped meals, skipped medicine, sold assets, mounting debt; the quiet emergency that hunger always causes.

This is not a forecast. As of the summer of this writing, excess heat has already carved into the 2026 harvest across the Northern Hemisphere, Western and Central Europe’s maize, the United States’ spring wheat and barley, China’s corn and rice, while a developing El Niño pushes heat and drought into the crops of South and Southeast Asia and Australia (Figure 7).²⁷ The losses are still regional rather than systemic, cushioned by a record 2025 harvest and full reserves; but they are exactly the kind that travel, one growing season later, to the prices on the supermarket shelf.

Figure 7. Where excess heat is hitting crops in 2026. Countries shaded where heat-driven crop stress was reported through early August: confirmed losses (red) across Western and Central Europe, the United States, and China; emerging El Niño heat-and-dry stress (orange) across South and Southeast Asia and Australia. Country-level and indicative, not sub-national.

How Can the Impacts of Extreme Heat Be Mediated?

Heat’s speed is its cruelest trait, but it is also, perversely, its weakness: because heat maps onto the calendar of the crops it threatens, we know in advance which weeks of which growing seasons carry the most risk. What is required is not incremental adjustment. It is deliberate action on four fronts, taken before the season that fails, not after. This means taking action to protect the poor and working classes around the world at a time when all of the major food donors are cutting back both internationally and at home. This will make hundreds of millions of people suffer if it is not reversed.

First, fund heat adaptation and anticipatory relief before the harvest, not after it. Cash, drought- and heat-tolerant seed, and pre-positioned food released on forecast triggers rather than on famine declarations cost a fraction of late-stage emergency response and save lives that late response cannot. Those trigger systems already exist. However, they are being defunded at exactly the moment the heat is intensifying, and that reversal is a choice this budget cycle can still unmake.

Second, protect and expand SNAP and other domestic nutrition programs. The food-price wave from a hot growing season reaches American shelves the following year, stacked on inflation, missing fertilizer and impact of tariffs already in the pipeline. Cutting the primary safety net standing between people dealing with price volatility and hunger, in the same eighteen months that volatility is set to spike, is not fiscal discipline; it is a decision to hand the bill for a warming atmosphere to the households least able to pay it. It is a war on the poor and working classes.

Third, pass a 2026 Farm Bill built for heat volatility. American farmers will plant the crop the world will need facing record input costs and a climate that now delivers a season’s worth of damage in a single week. Crop insurance designed for gradual bad luck, and disaster programs that pay two years late, are not up to responding to a hazard that strikes in an afternoon. Build the farmers safety net for extreme heat explicitly, and build it to reward the diversified, heat-resilient farms most able to survive it - not only the monocrops which are most vulnerable to heat’s disasters.

Fourth, protect the people and the water that grow the food. The same heat that sterilizes the pollen makes the field lethal to the workers standing in it. Enforceable farmworker heat standards are food-security policy, not merely labor policy. The water that irrigation leans on hardest in a heat wave is the water heat is draining fastest, which makes serious investment in water storage, efficiency, and shared drought planning a frontline defense of America’s breadbasket and our harvests, not an afterthought. Rivers, lakes, dams and aquifers that are being used for agriculture have to be protected and their use sustainably divided so that they do not go dry and create humanitarian disasters.

The Emergency That Arrives in an Afternoon

The great climate disasters of the past announced themselves slowly enough that the world could at least pretend to be surprised. Heat has taken away that excuse. We know the thresholds now, crop by crop and degree by degree; we know which weeks of the calendar carry the greatest danger; we can watch the temperature climb toward the number that ends a season in real time, in open data, free to anyone who cares to look.

That knowledge is either a warning or an indictment, and which one it becomes will be decided by whether anyone acts on it. The heating air has delivered its message with a precision no disaster in history has matched. Whether the next hot summer becomes a harvest quietly protected or hunger and famines loudly regretted; belongs, as it always does, to those in power and to the pressure the rest of us are willing to put on them before the thermometer rises again, not after.

So put pressure on your elected officials now, while the window is still open.

Call your elected officials, state and federal, today, tomorrow and the days after that; and demand that the food safety net be restored, the Farm Bill be passed with measures to protect against the impacts of climate change for farmers and for the poor and working classes in America, and the domestic and international food aid budgets be built for the fastest-moving threat that the food system faces before the next heat wave writes another season’s hunger in a single afternoon, in your state, in this country, and across the world.

Endnotes

1. World Meteorological Organization / Copernicus Climate Change Service, confirmed via the Center for Climate and Energy Solutions (2026): 2024 was the hottest year in the instrumental record, with global average temperatures more than 1.5°C above the 1850–1900 baseline, and 2015–2024 the hottest decade on record.

2. World Meteorological Organization, State of the Global Climate update (Nov. 2025): 2025 set to rank as the second- or third-warmest year on record, with ocean heat content exceeding record 2024 values. World Weather Attribution / reporting in CBS News and PBS NewsHour (Dec. 2025): the 2023–2025 three-year mean crossed 1.5°C for the first time; heat waves were assessed as 2025’s deadliest extreme-weather category, with several studied events roughly ten times more likely than a decade earlier owing to human-caused warming.

3. 2025 European heatwaves, summary record (peak 46.6°C at Mora, Portugal, June 29, 2025); attribution analyses cited therein estimate on the order of 16,500 heat-related deaths across the summer linked to climate change; Spain recorded its hottest summer on record.

4. Seed longevity follows “Harrington’s Rule” (J. F. Harrington, 1963/1972): within roughly 0–50°C and about 5–14% seed moisture, a seed’s storage life is approximately halved for every 1% increase in moisture content or every ~10°F (5.6°C) increase in storage temperature. The related “James’ Rule” holds that the storage temperature in °F plus the ambient relative humidity in per cent should remain below 100. Heat and moisture shorten viability by accelerating respiration, lipid oxidation, and membrane and DNA damage, and by enabling storage fungi. FAO seed-storage guidance and university seed-conditioning references.

5. High soil temperature and moisture stress at sowing reduce germination and seedling establishment; high temperatures during seed maturation on the plant, and excessive-heat mechanical drying after harvest, further reduce viability and vigor. Controlled-storage studies show rice germination falling from roughly 88–95% toward zero over several months at high storage temperature, versus little change under cool storage. Many of the world’s smallholder farmers rely on farmer-saved seed held without temperature or humidity control, heightening their exposure. Seed-physiology and storage-temperature literature; FAO/CGIAR seed-systems sources.

6. Critical high-temperature thresholds for seed set during flowering (rice 37.2°C ±0.2, wheat 27.3°C ±0.5, maize 37.9°C ±0.4), from the review From the floret to the canopy: High temperature tolerance during flowering (Plant Communications / ScienceDirect, 2023), which synthesizes the reproductive-stage heat sensitivity of the three leading cereals. Rice production declines have been documented above roughly 33°C.

7. On maize as the most heat-sensitive of the three cereals, and on heat-induced spikelet/silk closure, pollen-tube arrest, and kernel abortion during silking and early grain fill: Heat-stress-induced ROS in maize silks cause late pollen tube growth arrest and sterility (iScience, 2024); Male Sterility in Maize after Transient Heat Stress during the Tetrad Stage (Plant Physiology, 2019); Heat Stress After Pollination Reduces Kernel Number in Maize by Insufficient Assimilates (2021).

8. Regional impacts of extreme heat on agriculture: northern India’s 2022 heat wave was associated with milk-output declines of up to about 15% and state-level wheat-yield losses of roughly 9–34%, after which India restricted wheat exports in May 2022; estimates vary by state.

9. European Commission and insurance-sector estimates (2024–2025): extreme weather costs EU agriculture roughly €28 billion a year, about 6% of crop and livestock output, with only an estimated 20–30% of losses insured. Early tallies from the summer 2025 European heat pointed to substantial damage to French corn, to roughly half of some regions’ carrot production, and to mass poultry mortality.

10. Gurjinder Singh, a wheat farmer in Punjab, India, quoted in FairPlanet, “Saving India’s Wheat Yield from Rising Heatwaves” (August 2022): Singh reported his 2022 wheat output falling from about 23–25 quintals per acre to 14–15, roughly a 40% loss, after the heat wave that began in March. India’s March–April 2022 heat, its hottest March in 122 years, is estimated to have cut wheat yields 10–35% across the northern wheat bowl of Punjab, Haryana, and Uttar Pradesh.

11. The term flash drought was introduced by Svoboda and colleagues at the U.S. National Drought Mitigation Center in 2002 and later formalized by the American Meteorological Society as an unusually rapid-onset drought marked by a multi-week period of accelerated intensification culminating in sectoral impacts. Operational monitoring relies on evaporative-demand indices; one common threshold is a 50% or greater increase in atmospheric evaporative demand over two weeks, sustained thereafter. NOAA / National Integrated Drought Information System (drought.gov).

12. Xing Yuan et al., “A global transition to flash droughts under climate change,” Science 380 (2023): drought intensification has accelerated over subseasonal timescales, with a transition toward more flash droughts across roughly 74% of the IPCC SREX global regions over the past 64 years, attributed to anthropogenic climate change via amplified evapotranspiration and precipitation-deficit anomalies; the shift is fastest over Europe, northern and eastern Asia, the Sahel, and western South America, and is projected to expand across most land areas under higher-emissions scenarios.

13. Jordan I. Christian et al., “Global projections of flash drought show increased risk in a warming climate,” Communications Earth & Environment 4 (2023): annual flash-drought risk over cropland is projected to rise from about 32% (2015) toward 49% in North America and 53% in Europe by 2100 under the highest-emissions scenario. Loss figures: the 2012 U.S. flash drought exceeded $30 billion in economic losses (per reporting on Yuan et al. 2023); the 2017 northern Plains (Dakotas–Montana) flash drought caused an estimated $2.6 billion in U.S. agricultural losses (Dilling et al. 2017; Neelam et al., Geophysical Research Letters, 2024).

14. Event accounts from NOAA / NIDIS drought status updates and retrospectives (drought.gov): the June 2023 Midwest flash drought (dryness covering roughly 64% of U.S. corn production at the onset of pollination; record-low Mississippi River stages on 13–18 October 2023, with barge-load restrictions and saltwater intrusion in southern Louisiana); the 2023 Southeast fall flash drought (record Livestock Forage Disaster Program payouts, concentrated in Mississippi, Tennessee, and Alabama); and the June–July 2024 mid-Atlantic flash drought affecting corn from the Carolinas to Pennsylvania.

15. Greg Lee, a dryland farmer and rancher in Divide County, North Dakota, quoted in Meera Subramanian, “The Flash Drought Brought Misery, but Did It Change Minds on Climate Change?” Inside Climate News, July 17, 2018. The 2017 flash drought across the U.S. Northern Plains caused an estimated $2.6 billion in agricultural losses.

16. University of Minnesota Extension, Heat stress in dairy cattle: mild heat stress reduces milk output by roughly 2.5 lb per head per day, with larger losses as stress intensifies; high-producing cows can begin to experience heat stress at air temperatures as low as the mid-60s °F. Australian grazing-dairy study (Univ. of Melbourne robotic dairy, 2020) recorded a ~14% drop in daily milk production from low to high temperature-humidity index.

17. Washington State University Veterinary Medicine Extension (2025) and Frontiers in Veterinary Science (2023): U.S. livestock heat-stress losses estimated at up to ~$2.3 billion/year (2003), equivalent to roughly $2.9 billion in 2024 dollars, with dairy cattle accounting for roughly 63% of U.S. livestock heat-stress losses.

18. Journal of Dairy Science (2003) and heat-stress mortality literature: several hours of temperature-humidity index above ~84 with little or no nighttime recovery (THI falling below ~74) can cause death in vulnerable feedlot and dairy cattle; on-farm mortality rises measurably during multi-day heat waves (≥3 days at or above ~32°C).

19. Jacob Hoewisch, owner and operations manager of the fifth-generation Hoewisch Dairy Farm in Fremont, Wisconsin, in reporting by WLUK-Fox 11 (July 2026). Above roughly 70°F, cows can give 3–10 pounds less milk per day, a 5–15% drop, which for a 240-cow farm can mean losses of up to about $13,000 in a month; prolonged overheating can also cause a cow to lose a pregnancy.

20. Poultry and small ruminants: birds have no sweat glands and cool only by panting, which leaves them acutely vulnerable to heat (University of Minnesota and University of Tennessee Extension); turkeys are among the most heat-sensitive poultry because of their size. Reporting on the June and July 2026 European heat wave described French poultry deaths in the millions, with turkeys hardest hit, and France’s 2003 heat wave killed an estimated 4 to 5 million chickens and turkeys, about 2% of the national flock. St-Pierre et al. (2003) put U.S. poultry heat-stress losses at roughly $128 to $165 million a year. Chicken is now the world’s most-consumed meat by volume. On small ruminants, heat and drought reduce pasture and cut lambing and kidding in the pastoral systems on which many of the world’s poorest households depend.

21. On stored-grain deterioration: warm, moist grain respires and self-heats, forming hot spots that promote mold growth, insect infestation, dry-matter loss, reduced nutritional value, and loss of seed germination; as a rule of thumb, allowable storage time falls by roughly half for each ~10°F (5.5°C) rise in grain temperature, and storage insects are most active between about 70 and 90°F. Mycotoxins produced by storage fungi are chemically stable and are not reliably removed by milling or cooking. North Dakota State University Extension; Bayer Crop Science; reviews of stored-grain fungi and mycotoxins.

22. Aflatoxin (produced by Aspergillus flavus) is the most acutely toxic and most tightly regulated mycotoxin and a recognized liver carcinogen, favored by heat- and drought-stressed crops in the field and by warm, humid storage. Paola Battilani et al., “Aflatoxin B1 contamination in maize in Europe increases due to climate change,” Scientific Reports 6:24328 (2016): under a +2°C scenario (the most probable near-term case), the mean aflatoxin index for European maize rose about 92% and the high-risk zone shifted northward. Van der Fels-Klerx et al. (2016) and subsequent work document A. flavus becoming established in southern and central Europe; Yu et al. (2022) describe a comparable northward shift in the United States. A 2026 analysis of 17,263 peanut samples in China (2009–2022) found aflatoxin B1 contamination significantly higher than a decade earlier, with temperature explaining roughly 49% of the increase (npj Science of Food).

23. Hay baled above roughly 20% moisture supports rapid microbial growth that consumes sugars and proteins, degrades feed value (partly through the Maillard/browning reaction, which binds protein into indigestible forms), and generates internal heat; above about 170°F spontaneous combustion becomes a risk, with most fires occurring within roughly six weeks of baling. Mississippi State, South Dakota State, and Missouri Extension services.

24. High-temperature storage accelerates rice deterioration: rising fatty-acid value (rancidity) and lipid peroxidation, loss of palatability and head-rice yield, discoloration, and loss of germination and nutritional quality, with damage increasing at higher storage temperature and moisture content; above ~16% moisture, warm storage promotes Aspergillus and Fusarium growth and aflatoxin accumulation. Storage-temperature studies in the Journal of Stored Products Research, Journal of Food Processing and Preservation, and Cereal Chemistry.

25. Post-harvest and cold-chain losses: the FAO estimates that roughly one-third of food produced is lost or wasted, with about 14% lost between harvest and retail and on the order of 12% (≈526 million tonnes) attributable to insufficient refrigeration; inadequate cold chains contribute to foodborne illness, which the World Health Organization links to roughly 420,000–600,000 deaths a year. Rising temperatures and extreme heat increase cold-chain energy demand and failure risk and accelerate spoilage. Yin et al., “Heat stress increases food loss during harvest: empirical evidence from China,” Journal of Stored Products Research (2026), identifies heat strain at harvest as a direct, previously overlooked climate channel into food loss; FAO Food Loss and Waste data indicate an average cereal loss of about 5.2% at harvest.

26. Estimates of U.S. food insecurity differ substantially by survey instrument, and the divergence is itself part of the story. The Urban Institute’s Well-Being and Basic Needs Survey, a nationally representative internet panel, found 24.2% of adults 18 and older reporting household food insecurity in the previous twelve months as of December 2025, and 27.7% of adults 18 to 64 (Karpman, Waxman, Gupta, Gonzalez and Kennedy, “Food Insecurity Remained High in 2025, As Safety Net Cuts Loom,” Urban Institute, March 2026). The USDA’s Current Population Survey Food Security Supplement, the long-standing federal measure, put household food insecurity at 13.7% in 2024, some 47.9 million people (Rabbitt et al., ERR-358, December 2025); Feeding America’s Map the Meal Gap 2026, which models local rates from that same federal data, reports 14.4% of individuals. The Urban Institute attributes its higher figures to the absence of the CPS income screener, a six-item rather than eighteen-item scale, and survey mode, since respondents report food hardship more readily in self-administered internet surveys than to an interviewer. State-level survey work runs higher still: the Greater Boston Food Bank and Mass General Brigham found 40% of Massachusetts households food-insecure in 2025, up from 19% in 2019. USDA discontinued the federal series in September 2025, making 2024 the last official national baseline. Food spending share: USDA Economic Research Service, Food Prices and Spending, drawing on Bureau of Labor Statistics Consumer Expenditure Survey data. In 2024, households in the lowest income quintile spent an average of $5,498 on food, 33.0% of before-tax income, against 12.2% for the middle quintile and roughly 8% for the highest. The share is volatile for poorer households, ranging between 28.8% and 42.6% over the past two and a half decades, because their incomes are less stable and grocery prices move.

27. Figure 7 and 2026 conditions. Western and Central Europe: a record late-June heat dome and repeated heatwaves cut the 2026 harvest, with COCERAL’s out-of-cycle forecast lowering EU-and-UK grain by 23.4 million tonnes and French maize near 9.4 million tonnes (its lowest in decades), and EU JRC MARS bulletin reductions for grain maize and sunflower. United States: USDA Agricultural Weather Highlights on record Northern Plains heat (about 105°F at Bismarck, North Dakota) and Plains drought (June–July 2026). China: reporting on the early-August 2026 heat wave across northern and eastern farming regions (Bloomberg, citing Vaisala). South and Southeast Asia and Australia: Reuters and ABARES on El Niño-driven heat and below-normal rainfall affecting India, Thailand, Vietnam, Indonesia, Malaysia, the Philippines, and Australia. Country shading marks where heat-driven crop stress was reported, not sub-national extent; figures are as reported through early August 2026.

Gaza’s Children Can’t Afford to Wait: All children of the Gaza Strip under 5, more than 320,000 children, are at risk of becoming acutely malnourished. When a child becomes acutely malnourished, the body enters a cascade of survival responses that, if not interrupted, progressively shut down its systems leading to death. It’s worth understanding because it explains both why acute malnutrition kills so fast and why it’s so treatable when food aid and medical treatment arrive in time. Photo credit: World Food Programme

Comforting Global Hunger Number Arrived from the U.N. at the Worst Possible Moment

On July 21, five UN agencies released the 2026 edition of The State of Food Security and Nutrition in the World, and the sentence that traveled among the major media was a soothing one: global hunger had declined for a third straight year, to about 645 million undernourished people in 2025, some 14 million fewer than the year before.¹ The head of the FAO gave the wires their quote: “We can bend the hunger curve” and within hours the headline was set, in near-identical stories, across dozens of outlets: global hunger is easing.²

Here is what the good news headline sat next to, in the same month, from the same agencies:

1) two famines were confirmed in a single year for the first time in modern times, in Gaza and Sudan;

2) four countries are at risk of famine right now: South Sudan, Mali, Haiti, Yemen;

3) A record 266 million people are in acute, life-threatening hunger;

4) And the sharpest humanitarian funding collapse in a generation helped create this new acute hunger and push global hunger numbers to their records.³

The comforting number did not just miss that suffering; it is being used to deny it.

“An accurate number, stripped of its date and its caveats, is not information. It is politically useful cover for the people and governments defending recent aid reductions.”

This is not a complaint about a statistic. The statistic is sound. It is a warning about what is being done with it because a headline that says hunger is falling is the most useful thing in the world to anyone who wants the aid cuts made when USAID was dismantled and other European donor countries joined in cutting foreign aid to look painless. These reductions in foreign aid are not painless. This piece is about the people the number leaves out, and about why leaving them out is a choice with consequences. Hunger is a policy choice, and donor governments have enacted policies that leave tens of millions with less humanitarian assistance and extreme hunger and other humanitarian disasters.

The People the Headline Cannot See

The falling number is a measure of chronic hunger, that is the slow, structural kind of hunger, estimated from food-supply balance sheets and surveys that arrive a year or more late. It is the wrong instrument for the emergency now underway. That does not make the measure wrong—it makes it unsuitable as a description of today’s emergency; it is pointed at the wrong year, 2025. The emergency is acute hunger, the kind that kills within a season, and it is not falling - it is at a record high and climbing.³

Behind that record are people with names and coordinates. In El Fasher and Kadugli, the two Sudanese towns where the Famine Review Committee confirmed famine, people starved under siege while the world’s attention followed this cheerier statistic and ignored the hunger crisis.⁴ Across Sudan, nearly 4.2 million cases of acute malnutrition are expected in 2026, including more than 800,000 severe cases of acute malnutrition among children where it is the deadliest.⁵ In South Sudan, one of four countries at risk of famine, aid is being rationed so tightly that agencies can only reach those already in catastrophe, leaving millions in crisis to slide toward famine untouched.⁶ None of these people are in the 645-million headline, because the instrument that counts them is a different one, and the press quoted the headline that looks away from today’s global hunger crisis.

The Cuts Are the Cause, and We Can Prove It

The good-news framing treats hunger as a problem quietly solving itself. The truth is that the fastest-moving force in global hunger right now is a policy decision: the abrupt dismantling of humanitarian and food aid and its fingerprints are on the record in a way that is hard to ignore or wave away.

Start with the funding. After the January 2025 executive order suspending U.S. foreign aid and the dismantling of USAID that followed, the World Food Programme expects to run on roughly 40 percent less funding or about $6.4 billion, down from $10 billion in 2024 to deal with increasing humanitarian crises.⁷ The June 2026 hunger warning came against a 59 percent drop in food-aid funding.⁸ That is not a rounding error. It represents the removal of one of the principal safeguards standing between for tens of millions of people and severe hunger and famine.

Now the effect. In Nigeria, funding shortfalls forced WFP to scale down so far that it could reach only 72,000 people in February 2026, down from 1.3 million during the 2025 lean season, while malnutrition in several northern states deteriorated from “serious” to “critical.”⁹ Across Uganda, Kenya, and Bangladesh, rations were halved or suspended for millions of refugees which occurred alongside documented deaths in Kenyan refugee camps, according to Refugees International.¹⁰

Perhaps the strongest field evidence comes from Mali. In Mali, WFP ran what amounts to a natural experiment: where in some parts of the country families kept receiving full rations, (the number of calories and nutrients recommended by public health research) acute hunger fell 34 percent. In other parts of the country where not all of the aid was received and as a result, rations were cut, acute hunger surged 64 percent. While not a randomized trial, the comparison offers unusually strong real-world evidence under humanitarian conditions. These rates of hunger were in same country, same season, opposite outcomes, split primarily by whether adequate levels of survival rations of food arrived.¹¹ This is as close to a causal relationship available of causation as field data ever gets, and it points in one direction: the foreign aid cuts made by the United States and European countries have substantially increased hunger in many affected regions around the world.

“Same country. Same season. Where the food arrived, hunger fell by a third. Where it was cut, hunger surged by two-thirds. That is not a correlation. That is evidence strongly consistent with causation.”

The human toll of the cuts has been estimated through modeling, and the numbers are staggering even at the low end. The Center for Global Development puts humanitarian and food-aid cuts at the potential for roughly 490,000 lives lost per year, part of a wider toll it places near 1.6 million annually; a Lancet study projects more than 14 million additional deaths by 2030, 4.5 million of them children, if the U.S. and European aid defunding is not reversed.¹²These are models, with real uncertainty. But they are consistent with the same reality the Mali experiment measures directly, and they do not describe a world where hunger is quietly easing.

Figure 1. The two hunger lines the coverage collapsed into one. In blue, the headline: chronic undernourishment, easing to about 645 million in 2025 (left axis) a lagging average. In red, the climate-driven slice of acute hunger, up nearly fivefold since 2020 to 87.5 million (right axis) the current emergency. Much of the media coverage emphasized the blue line while giving comparatively less attention to the red one. Sources: SOFI 2026 (FAO/IFAD/UNICEF/WFP/WHO); GRFC 2020–2026 (GNAFC). Chronic values for 2022-24 approximate.

The Data Is Vanishing Along with the Aid

There is a second cruelty folded into the first. The same budget cuts that are deepening hunger are also defunding the systems that count how many people are hungry, which means the reassuring number is not only late, but also built on the thinnest evidence in a decade according to the 2026 Global Report on Food Crises and will miss more hungry people as time passes.

The Global Report on Food Crises said so plainly: funding cuts and blocked access produced the lowest availability of food-security data in ten years.¹³ In 2025, 18 countries and territories lacked comparable data, including Burkina Faso, the Republic of Congo, and Ethiopia, which together had more than 27 million acutely food-insecure people in 2024 who are simply missing from the totals.¹⁴ A number can fall because fewer people are hungry; it can also fall because fewer people are counted. When the counting is being dismantled in the same act that the aid to feed hungry people is being dismantled a headline that reported the decline has mistaken a blind spot for good news may be cited by the people and governments defending aid reductions as evidence that current policies have not significantly worsened hunger and potentially to keep cutting aid even as tens of millions of people are in desperate hunger.

How an Accurate Number Tells a Lie

It is worth naming exactly how the distortion works, because it will happen again with next year’s report unless it is called out now. Nobody fabricated anything. The issue is not fabrication. It is framing. The lie is assembled from true sentences, in four moves.

● The date is dropped. “Hunger fell in 2025” becomes “hunger is falling,” and a year-old average is passed off as a live trend.

● The instruments are swapped. The chronic-hunger number stands in for the acute-crisis number, so the falling line covers the rising one where millions are in the worst stages of hunger.

● The caveats are cut. The report’s own warnings including the data gaps, the regional reversals, data limitations, the lack of fertilizer, the excluded El Niño and other qualifying context rarely survive the trip to the headline.¹⁵

● A hope is printed as a fact. “We can bend the curve” is a plea for action; it gets reported as a description of progress.

The result is a story true in every particular and false in its total effect. This is the hardest kind of misinformation to correct, because every sentence passes a fact-check while the whole misinforms. In this year, the false impression is doing political work: it is being used by the people and countries defending recent aid reductions to suggest the humanitarian consequences have been limited, at the exact moment the famines say otherwise. By definition, famine represents the most extreme stage of food insecurity, a famine is “an extreme and widespread shortage of food that causes severe malnutrition, starvation, and death.” [1, 2]

What Is Actually True

None of this means burying the SOFI number. It means telling the whole truth which it is only a part of all of which is available from the very same reports:

Chronic hunger, on a lagging estimate, edged down for a third year, a real, slow, statistic carried by a few countries with revised data. Although the simultaneous decline in comparable data warrants caution when interpreting long-term trends, at the same time, acute hunger is at a record: 266 million people, two confirmed famines, four more countries at risk of famine. The improving line od chronic hunger describes the past; the worsening one describes the present acute hunger and with El Niño coming the near future. Both rest on the sparsest data in a decade. One major factor driving the worsening acute crisis is a choice Western Donor Countries made to cut the aid, which the Mali evidence suggests can be substantially mitigated where adequate food assistance is restored tomorrow, because where the food still arrives, hunger still falls.

That is not despair. It is the report, read to the end, with the people left out the frame. Much of the major media coverage emphasized the opening finding while giving comparatively little attention to the report’s discussion of acute hunger, data gaps, and humanitarian funding. They ran the just ran the first sentence and stopped.

What You Can Do About It

This is the rare humanitarian catastrophe that is being measured in advance, in public, in detail. Many of the deaths that may follow are foreseeable and mostly preventable through different policy choices. The window to change the decision is open now, and it does not require heroics. It requires refusing the comfortable version of the story and acting on the true one where aid cuts are making people hungry all over the world. The reduction in crops from the fertilizer that never arrives is being modeled and mapped in advance. The OECD–FAO Outlook projects reduced fertilizer use cutting cereal harvests in low-income countries, and satellite systems like the FAO’s GIEWS and the EU’s ASAP track where those losses are concentrated. The hunger this will cause is being forecast even now; the only thing missing is the will to act on the forecast. An El Niño is forecast six to twelve months ahead, the single most predictable pattern in the climate system, tracked instrumentally for about 150 years. And where its droughts will harden into hunger is charted in advance, region by region, by the early-warning systems built for exactly that purpose. Famine is on the calendar before it happens. Now it is a question of whether the world humanitarian organizations will have the funding to prepare for what we know is coming.

Restore the funding, this budget cycle. The single most direct lever is restoring the humanitarian and food-aid that collapsed in 2025. The Mali comparison illustrates the central argument: where the rations arrive, hunger falls. Tell your representatives, both state and federal, that reversing the cuts is not charity, it could substantially influence whether famine conditions worsen or are contained, and whether anticipatory funding is in place for southern Africa’s planting season and the four famine-risk countries which cannot wait for a distant or even the next fiscal year.

Challenge misleading headlines. When you see a major media outlet report that “global hunger is falling,” challenge them which hunger and which year. Share the acute hunger numbers alongside the chronic ones. Tell people that aid reductions are worsening humanitarian food crises and increasing famine risk in multiple countries meaning people will starve to death from lack of food. A public that knows the difference between a lagging average and a live famine is much easier to encourage into action.

Fund the counting, not just the feeding. The vanishing data is its own emergency. The monitoring systems called the IPC, FEWS NET, the crisis reports are what make famine foreseeable and therefore preventable. Modern famine early-warning systems expanded after the devastating famines of the 1980s, reflecting a growing recognition that timely data saves lives. Defending them is cheap and critical for understanding what is happening with global hunger.

“The famines were forecast. The cuts were chosen. The headline was optional. Every one of those is still ours to change but only if we stop mistaking a year-old average for the end of the emergency that has just started and is only going to get worse as more aid is withheld.”

Check It Yourself

Every figure here is public and free, released by the same agencies whose headline this piece disputes. Hold both the report and its coverage to account:

● The chronic number, in full: the SOFI 2026 report and the agencies’ launch release. Please read past the headline into the caveats.

● The acute crisis: the Global Report on Food Crises 2026 for the 266-million figure, the climate driver, and the data-gap warning.

● The famines, live: the IPC country analyses for the formal classifications and the FAO–WFP Hunger Hotspotsoutlook.

● The cuts and their toll: Refugees International’s tracking of the aid-cut impact and the Lancet defunding study.

● Hunger in near-real time: WFP’s HungerMap Live and FEWS NET, this system could correct any number a year out of date if they were operational.

The Sentence That Should Have Traveled

There was a true headline available, and it fit: “Chronic hunger inched down on a year-old estimate, even as acute hunger hits a record high, two famines were confirmed, and the aid cut driven humanitarian crises are still deepening.” It is less comforting than “hunger is falling,” but it has the advantage of being true and not leaving out the people dying of starvation or the tens of millions of people who have been made hungry due to the aid cuts.

The child in El Fasher, the family in northern Nigeria whose ration was cut from a program that once reached 1.3 million and now reaches 72,000, the refugee in Kenya who died after the food basket was halved, neither of them are in the 645-million headline, and all of them are in the world it claims to describe. They are not a rounding error in an improving trend. They are the trend; in the places where the trend is measured honestly and the humanitarian alarm bells are ringing. A press that reports the average and skips the emergency has not lied. It has done something quieter and, this year, much more deadly: it has handed the people and countries who cut the aid a headline that says nobody got hurt and let the rest of us believe it.

Donate to international humanitarian aid agencies now!

Where to Give: Top-Rated Humanitarian Organizations

Every group below is active in the hunger and famine responses described above and earns top marks from independent watchdogs (Charity Navigator, Candid/GuideStar, or CharityWatch). Ratings and details were current as of July 2026; confirm at each organization’s site before giving.

World Food Program USA — 4-star Charity Navigator; Platinum Transparency (Candid)
Web: wfpusa.org | Donate: wfpusa.org/give
Phone: (202) 627-3939
1750 H Street NW, Suite 500, Washington, DC 20006
Mail donations: P.O. Box 96316, Washington, DC 20090-6316

Oxfam America — Long-standing famine-relief organization (name originates in the 1942 Oxford Committee for Famine Relief)
Web: oxfamamerica.org | Donate: oxfamamerica.org/donate
Phone: 1-800-77-OXFAM (1-800-776-9326)
26 Causeway Street, 5th Floor, Boston, MA 02114

International Rescue Committee (IRC) — 4-star Charity Navigator (96%); Platinum Seal (Candid)
Web: rescue.org | Donate: rescue.org/donate
Donor services: 1-855-9RESCUE (1-855-973-7283) | DonorServices@Rescue.org
122 East 42nd Street, New York, NY 10168

Save the Children — Top-rated; emergency nutrition and child-hunger programs
Web: savethechildren.org | Donate: support.savethechildren.org
Phone: 1-800-728-3843
501 Kings Highway East, Suite 400, Fairfield, CT 06825

CARE — Top-rated for humanitarian and hunger relief
Web: care.org | Donate: my.care.org
Phone: 1-800-521-CARE (1-800-521-2273)
151 Ellis Street NE, Atlanta, GA 30303

Doctors Without Borders / Médecins Sans Frontières (MSF-USA) — 4-star Charity Navigator; Gold Seal (Candid)
Web: doctorswithoutborders.org | Donate: give.doctorswithoutborders.org
Donor services: (888) 392-0392 | donations@newyork.msf.org
40 Rector Street, 16th Floor, New York, NY 10006
Mail donations: P.O. Box 5030, Hagerstown, MD 21741-5030

Action Against Hunger USA — 4-star Charity Navigator; Platinum Seal (Candid); $0.90 of each dollar to programs
Web: actionagainsthunger.org | Donate: actionagainsthunger.org/take-action/donate
Phone: (212) 967-7800
One Whitehall Street, 2nd Floor, New York, NY 10004

Mercy Corps — Highest watchdog ratings (Charity Navigator, Candid, CharityWatch)
Web: mercycorps.org | Donate: mercycorps.org/donate
Phone: (888) 842-0842
45 SW Ankeny Street, Portland, OR 97204

Before giving, you can check any charity’s current rating at charitynavigator.org or charitywatch.org.

Endnotes

1. FAO, IFAD, UNICEF, WFP and WHO. 2026. The State of Food Security and Nutrition in the World 2026 (SOFI). Released 21 July 2026. An estimated 645 million people (7.8 percent of the global population) faced hunger in 2025, down from 8.1 percent in 2024 and 8.6 percent in 2022 — about 14 million fewer than 2024 and 43 million fewer than the 2022 peak. The metric (prevalence of undernourishment) is a modeled estimate built from food-supply, population, and household-survey inputs that arrive with substantial lag.

2. FAO Director-General QU Dongyu, remarks at the SOFI 2026 launch, 21 July 2026: “We can bend the hunger curve.” The launch release ran, in near-identical wording, across FAO, WFP, WHO, UNICEF and downstream outlets as “Global hunger levels ease for third consecutive year.”

3. FSIN and Global Network Against Food Crises. 2026. Global Report on Food Crises 2026 (GRFC): 266 million people (22.9 percent of the analysed population) faced high levels of acute food insecurity (IPC/CH Phase 3 or above) across 47 countries in 2025; acute food insecurity has nearly doubled over the past decade; and two famines (parts of the Gaza Strip and the Sudan) were confirmed in the same year for the first time in the report’s history.

4. IPC Famine Review Committee (2025): Famine (IPC Phase 5) confirmed in El Fasher (North Darfur) and Kadugli (South Kordofan), Sudan, as of September 2025, with a risk of Famine in some 20 additional areas.

5. IPC / UN OCHA (2026): nearly 4.2 million cases of acute malnutrition projected in Sudan in 2026, including more than 800,000 severe cases; SMART surveys found GAM levels exceeding the Famine threshold in parts of North Darfur in December 2025.

6. IPC South Sudan Projection Update, April–July 2026: 7.8 million people (55 percent of the population) in IPC Phase 3 or above, with a risk of Famine in Luakpiny/Nasir and Ulang (Upper Nile) and Nyirol and Akobo (Jonglei); limited resources have forced programming to concentrate on populations already in Catastrophe.

7. WFP funding: the agency projected receiving roughly 40 percent less funding in 2025, about $6.4 billion against $10 billion in 2024. WFP, “Africa: Humanitarian Funding Cuts Pushing Millions Into Hunger” (October 2025).

8. FAO–WFP Hunger Hotspots (June 2026) and associated reporting cited a 59 percent drop in food-aid funding against the June–November 2026 outlook.

9. UN News / WFP (January 2026): in Nigeria, WFP could reach only 72,000 people in February, down from 1.3 million assisted during the 2025 lean season, as malnutrition in several northern states deteriorated from “serious” to “critical.”

10. Refugees International, A Generational Collapse: Tracking the Toll of the Humanitarian Aid Cuts (February 2026): ration suspensions and cuts across Uganda, Kenya, and Bangladesh; in Kenya, the minimum food basket was cut ~40 percent for nearly 800,000 refugees, with documented deaths.

11. WFP, “Humanitarian aid cuts push millions deeper into hunger … West and Central Africa” (January 2026): in Mali, areas where families received reduced rations saw a ~64 percent surge in acute hunger (IPC 3+) since 2023, while communities that received full rations saw a 34 percent decrease.

12. Center for Global Development, “Update on Lives Lost from USAID Cuts” (December 2025): humanitarian and food-aid cuts alone associated with the potential for ~490,000 lives lost per year, within a wider total near 1.6 million per year. Cavalcanti, Rasella et al., “Evaluating the impact of two decades of USAID interventions and projecting the effects of defunding on mortality up to 2030,” The Lancet 406:283–294 (2025): more than 14 million additional deaths projected by 2030 (95% UI 8.5–19.7 million), including ~4.5 million children under five, under a sustained-defunding scenario. These are modeled projections with substantial uncertainty.

13. GRFC 2026: funding cuts and limited access to conflict areas produced the lowest availability of food-security data in a decade.

14. GRFC 2026: in 2025, 18 countries and territories lacked comparable data, including Burkina Faso, the Republic of Congo, and Ethiopia, together accounting for more than 27 million acutely food-insecure people not captured in comparable totals.

15. SOFI 2026: the report’s five-year projections do not account for the effects of weather extremes such as the 2026–27 El Niño, which it notes could further compound the numbers; on current trends roughly 510–512 million people are projected to remain chronically undernourished by 2030, most in Africa, missing Sustainable Development Goal 2. Hunger also continued to rise across most of Africa and western Asia even as the global average fell.

A note on sourcing. Every figure above is drawn from the 2026 SOFI report, the 2026 Global Report on Food Crises, the IPC country analyses, the June 2026 Hunger Hotspots outlook, WFP operational reporting, Refugees International, or the peer-reviewed mortality literature — all public. The mortality figures are modeled projections and are labeled as such. The complaint is not with the agencies’ data. It is with which sentence of it the coverage chose to run.

The Year the Averages Lied

Halfway through 2026, the food system tells two stories, and only one is reassuring. The reassuring one is in the aggregates: the FAO Food Price Index sat at 130.3 in June, barely above a year earlier and nearly a fifth below its 2022 peak, with the world’s cereal harvest tracking toward its second-largest ever on full reserves.¹ The other is written crop by crop: vegetable oils prices were 23 percent higher than a year ago and meat set an all-time record; a flood erased a fifth of a Bangladeshi rice season; drought sat over half the world’s cocoa belt; Spain gathered its smallest orange crop in sixteen years; two thirds of U.S. winter wheat was not healthy enough to harvest.² The averages held. The specifics did not.

Figure 1. The calm on the surface. The aggregate readings behind the reassuring story of mid-2026 — a food-price index below its 2022 peak, a near-record cereal harvest, rebuilt stocks. All axes begin at zero, so the bars sit almost level; the damage is not visible at this altitude. Source: FAO, 2026.

An annual average is a place to hide. It is the crop, the region, and the week that tell you where the climate actually landed and in 2026 it landed unevenly, and hard.

What follows takes stock of what the first six months have already done to real crops in real fields, and a glance at the landscape on which the coming El Niño will land.

The Crops That Failed First

The clearest losses fell on the specialty and tree crops that smallholders grow and no national average can rescue. A pre-monsoon flood took the Bangladeshi boro rice at harvest peak; drought hit the West African cocoa belt during pod formation even as the price fell on a demand glut, the falling price is not the result of a healthy crop; and the vegetable-oil index, showing prices up 23 percent to its highest since 2022, is where the year’s bad weather actually shows up on the shelf due to limited supply.³

The People at the Growing End

Every figure has a household behind it, and for specialty crops it is usually a smallholder with no cushion. In Bangladesh the April floods hit more than 236,000 farming families and destroyed some 214,000 tonnes of rice with no chance to replant; in West Africa, when futures collapsed, Ivory Coast cut its farmgate price by more than half, to about $2.13 a kilogram, so growers caught the downside of a boom they never shared; and Vietnam’s coffee farmers will face a confirmed El Niño before they have recovered from the last drought.⁴

A staple crop failing is a statistic in a report. A specialty crop failing is a family with one income, one harvest, and no cushion.

Meat: A Record and Its Mirror

One line in the world ledger never stopped climbing: the FAO Meat Price Index reached a record 131.0 in June, about a third above its 2014–2016 base. But an index is not a butcher’s ticket, and June’s record was written mostly in poultry — higher Brazilian export prices and firm lamb quotations — while global beef and pork prices actually eased.

The American cattle story is the mirror, and it runs the other way. The U.S. herd has shrunk to 86.2 million head, the smallest since 1951, after eight years of contraction and drought-driven liquidation that no single season can reverse. It shows up not in the global index but on the domestic shelf, where retail ground beef has climbed past $6.80 a pound; and because rebuilding a herd takes three to four years, relief is unlikely before 2028.⁵

The heat reached every animal. Sheep and goats lose lambing on burnt pasture, even the drought-proof camels of the Horn of Africa have begun dying, and a late-June European heatwave killed an estimated 2.5 to 3 million French broiler chickens.⁶

Grain: The Averages Lied Again

If one crop proves the thesis, it is European maize. A third heatwave since late May burned France’s crop to about 9.4 million tonnes, a third below last year and its smallest in over three decades, and COCERAL cut EU-and-UK grain by 23.4 million tonnes.⁸ A fourth heatwave is hitting France right now. Kenya’s maize fell on a June dry spell and Brazil–Thailand sugar is seen down a tenth. Gathered in a table the failures are dramatic; gathered in the world total they nearly vanish. That gap is the whole argument of the impacts of climate change on hunger.⁹

The Southern Hemisphere’s Dry Turn

The El Niño’s dry phase has reached the south before the harvest even began. Australia’s dry autumn cut canola area by nearly a third and trimmed the wheat crop toward 26.7 million tonnes for 2026–27, 25% below last year, with ranchers selling cattle in record numbers; across the Pacific, Argentina’s early-2026 drought scare eased and Brazil banked a record soybean crop even as its safrinha corn went into the dry season on thin soil moisture.¹⁰

What the Ledger Says, and What Comes Next

Add the six months together: the shocks were real but regional, sharp but not systemic, concentrated in the crops least able to average out and cushioned by a record 2025 harvest and full reserves.¹¹

So far, 2026 has bruised the food system without breaking it. The open question is what happens when the next blow as a super El Niño lands before the bruises heal.

Every shock so far landed in the first half of a year whose defining event does its worst work in the second. On June 11 the U.S. Climate Prediction Center confirmed an El Niño observed, not forecast, with a better-than-96-percent chance of persisting and a peak near +3.6°C that would be the strongest El Niño ever to impact the world since records have been kept. It will be landing on the growing seasons of the Southern Hemisphere and the tropics. ¹² The sharpest forward risks include Philippine rice, a quarter to a third at risk; a weak Indian monsoon and a held sugar-export ban; and palm oil, where Indonesia and Malaysia hold 88 percent of supply on thin stocks. The pivotal case is southern Africa, but two La Niña years left near-record stocks, so even a real drought should bite more softly than 2016.¹³

The distribution has shifted toward the harder outcomes. The crops at risk are already named, and the reserves are being drawn down before the peak arrives. The forecast is conditional. The window to act on it is not.¹⁴

The Humanitarian Map of the Second Half

The crop ledger reads one commodity at a time; the humanitarian sector reads one region at a time, and the two converge.UMCOR’s June outlook resolves the confirmed El Niño into a redistribution of drought, flood, and heat across ten regions, with a window of greatest concern from July 2026 through March 2027, exactly the stretch the crop calendar leaves exposed, and a three-phase timeline that is paid out in crop failure, displacement, and disease by early 2027. ¹⁵

The El Nino is coming as aid agencies struggle to provide food for the tens of millions of people in need in Iran, Suda, Gaza, Haiti, Myanmar, Syria and elsewhere.

Figure 3. The same second half, drawn as a humanitarian map. UMCOR’s June 2026 outlook resolves the El Niño into a redistribution of drought (reds), flood (blues), and heat across ten regions, with a concern window of Q3 2026 – Q1 2027. Source: UMCOR, drawing on WMO, NOAA CPC, FAO GIEWS, and WFP.

The call for early action arrives in the worst funding environment in a generation. The OECD reported 2025 aid down 23 percent, humanitarian aid down more than a third, and core UN funding down 27 percent after an 87-percent U.S. cut, with the 2026 appeal now reaching barely a third of those in need. The warning is not the problem; money has always moved after the photographs arrive, and the institutions built to change that are the ones being defunded.¹⁶

The scale of that retreat is now measurable, and it is still deepening. The OECD’s preliminary 2025 figures show official development aid falling for a third consecutive year, to its lowest level since 2014, with a further decline of roughly 7 percent projected for 2026 that has barely begun to land. The contraction is strikingly concentrated with the five largest donors accounting for 96 percent of the reduction in aid, led by a 57 percent cut from the United States. The cuts in aid fall hardest where the cushion is thinnest: bilateral aid to the least-developed countries and to sub-Saharan Africa each dropped more than a quarter in 2025 and is forecast to fall by double digits again this year, while global health financing is on course to run as much as 63 percent below its 2022 peak. What remains is being redrawn by geopolitics rather than need, with more aid flowing to Ukraine in 2025 than to all the least-developed countries combined. The humanitarian system downstream is, in its coordinator’s words, “overstretched, underfunded and under attack.” Roughly a fifth (only 20%) of its appeals have been funded this year, and more than 320 aid workers have been killed in a single year. ¹⁷

The humanitarian map of the coming year, shows a redistribution of drought, flood, and heat across five continents and a calendar that says the reckoning arrives after the existing food reserves have already been spent.

How to Read the Ledger Yourself

Every figure here comes from open data updated through the year:Every figure here comes from open data updated through the year: USDA’s Crop Explorer and GEOGLAM for crop condition and the WASDE for official world numbers; the FAO Food Price Index for the split beneath the headline; the ICCO, ICO, and IOCfor the tree and tropical crops; NOAA’s ENSO page and Australia’s Bureau of Meteorology for the El Niño driver; and, at home, the U.S. Drought Monitor and USDA’s weekly Crop Progress.

Endnotes

1. FAO, June 2026: Food Price Index 130.3 (~18.7% below the 2022 peak); 2026 world cereals ~2,983 MMT, the second-largest ever, on comfortable stocks. Source: FAO.

2. FAO sub-indices, June 2026: vegetable oils +~23% y/y (highest since 2022); meat a record 131.0. Source: FAO.

3. Crops: an April flood submerged 46,000+ ha of ripe boro in the Bangladeshi haor (loss ~20% haor / ~10% national); West African cocoa drought during cherelle formation (Ivory Coast next crop seen down to −17.5%) even as futures fell >30% on a glut; FAO oil index highest since 2022; Spain’s smallest orange crop in 16 years (−11.6%); record Brazilian arabica vs drought-hit Vietnamese robusta. Sources: Bangladesh DAE; Fitch; ICCO; FAO; Conab; USDA.

4. Growers: in Bangladesh, 236,000+ families affected and ~214,000 t of rice lost with no replant; Ivory Coast cut its 2026 cocoa farmgate price >50% to ~$2.13/kg; Vietnam’s coffee sector had not recovered from 2023–24 before a confirmed El Niño. Sources: Bangladesh Min. of Agriculture; Reuters.

5. Meat prices and the U.S. herd: the FAO Meat Price Index averaged a record 131.0 in June 2026 (base 2014–2016 = 100), up 4.0% year on year — an index of international export quotations, led in June by poultry (stronger Brazilian prices) and lamb, with global beef and pork easing. USDA-NASS's January 2026 Cattle report put all U.S. cattle and calves at 86.2 million head, the smallest herd since 1951, and beef cows at 27.6 million, the lowest since 1961, in the eighth year of contraction. U.S. retail 100% ground beef averaged about $6.82/lb in June 2026 (BLS); herd rebuilding takes roughly 3–4 years, so tight domestic supply is expected to persist into 2027–2028. Sources: FAO; USDA-NASS (Cattle, 30 Jan. 2026); U.S. BLS.

6. Livestock: camels saw 2026 mortality even in the Horn of Africa; a late-June European heatwave (all but impossible without climate change) killed ~2.5–3M French broilers and cut western French milk ~15–20%; U.S. April lamb ~$262/cwt (+~53% y/y). Sources: BBC; FEWS NET; Le Monde/ANVOL; WWA; MSU.

7. China pork: Q1 2026 output +4.2% to 16.7 MMT; hog price ~11.3 yuan/kg (−27% y/y), an eight-year low, on oversupply not weather; a March directive cut the sow herd (39.6→36.5M) and the state bought reserves. Sources: China NBS; AHDB; SCMP.

8. France/EU maize: France ~9.4 MMT (COCERAL), ~a third below 2025 and according to USDA WASDE, the lowest in 30+ years; COCERAL cut EU+UK grain by 23.4 MMT; EU maize 52.7 MMT, lowest since 2007. Sources: COCERAL; Expana; USDA WASDE.

9. Kenya/sugar/China grain: USDA cut Kenya’s 2026 maize on a June dry spell (some ~a third below average); BofA sees Brazil–Thailand sugar −10% in 2026/27; China took a record 2025 harvest (~715 MMT) into 2026 and cut corn imports ~90%. Sources: USDA; Bank of America; SunSirs; FAO.

10. Southern Hemisphere: Australia’s dry autumn cut canola area ~32% and NSW wheat ~13%; national wheat ~26.7 MMT for 2026–27 (vs ~36), with record cattle sell-offs; Argentina soy held ~48 MMT after a scare; Brazil banked a record ~180 MMT soy crop with safrinha corn at risk. Sources: ABARES; DTN; Kpler; USDA FAS.

11. Midyear balance: shocks concentrated in crops least able to average out with flood-prone rice, tree crops, tropical specialties being hit the hardest while grain aggregates held on a record 2025 harvest and full reserves. Source: FAO.

12. El Niño: on 11 June 2026 the U.S. CPC issued an El Niño Advisory (observed; >96% chance of persisting), with a Nov 2026–Jan 2027 peak, ~63% odds of a “super” event and a pooled peak near +3.6°C — the strongest on record if realized. Sources: NOAA CPC; WMO.

13. Forward risks and southern Africa: a severe event could cut Philippine rice 25–35%; India’s monsoon ~92% of average with its sugar ban expected to hold; Indonesia+Malaysia hold ~88% of palm oil on thin stocks. Southern Africa’s maize peaks with the El Niño, but two La Niña years left near-record stocks (food inflation ~3.2% vs 10.8% in 2016), so a drought should bite more softly. Sources: PAGASA; IMD; Marex/Helios; S&P Global.

14. Scenarios: from a mild close (regional specialty shocks over intact grains) to a central strong-to-very-strong case (reduced Australian wheat, cushioned southern African maize, tighter palm/sugar, a second leg up in oils into 2027) to a severe case in which shocks outrun reserves, pressuring the ~333M already in acute food insecurity. Sources: NOAA CPC; USDA; WFP.

15. UMCOR: “El Niño 2026–2027: Global Humanitarian Risk Outlook,” 22 June 2026, on WMO/NOAA/FAO/WFP guidance; concern window Q3 2026–Q1 2027 across ten regions, in three phases ending in peak impact (Jan–Mar 2027: crop failure, displacement, disease). Source: UMCOR.

16. Funding: OECD reported 2025 aid −23.1% (steepest ever), humanitarian aid down >a third, core UN funding −27% (an 87% U.S. cut after USAID’s dismantling); the 2026 Iran war diverted money; the UN’s $33bn 2026 appeal reaches barely a third of those in need. Sources: OECD; UN OCHA; CERF.

17. The current state of international aid: the OECD’s preliminary 2025 data (April 2026) show DAC official development assistance fell 23.1% to $174.3 billion , the steepest drop on record and a third consecutive annual decline, its lowest level since 2014, with a further ~6.9% projected for 2026; the five largest donors account for 96% of the fall, led by the United States (−57%). Bilateral aid to least-developed countries and sub-Saharan Africa fell 25.8% and 26.3% in 2025 and is projected to drop a further ~10.9% and ~11.6% in 2026, while health financing is projected as much as 63% below its 2022 peak. DAC members’ 2025 aid to Ukraine exceeded their combined aid to all least-developed countries. The UN’s 2026 Global Humanitarian Overview seeks $33 billion for 135 million people after 2025 funding hit a decade low (~$12–15 billion) and reached 25 million fewer people; UN relief chief Tom Fletcher called the system “overstretched, underfunded and under attack,” with only about 20% of appeals funded and more than 320 aid workers killed in 2025. Sources: OECD (Preliminary 2025 ODA Data; ODA Projections for 2026, April 2026); UN OCHA (Global Humanitarian Overview 2026); The New Humanitarian.

The July forecast runs are in, and they describe something no working forecaster has ever seen. According to 667 ensemble members from 14 seasonal prediction systems, the multi-model median now puts this winter’s El Niño at a peak of 3.6°C in the Niño 3.4 region of the tropical Pacific. This is roughly 0.8°C beyond the strongest El Niño event in a century and a half of observation. ¹ Ninety-one percent of the ensemble members predicted that this El Niño will exceed the all-time record. The ocean itself is not waiting for the models: daily sea surface temperatures in the region are already running around 2°C above their era-adjusted average in mid-July, warmer at this date than in 1997 or 2015, the two most explosive El Niño onsets ever measured, and by a margin that is not close. ²

The onset race: daily sea surface temperature anomalies in the Niño 3.4 region for 2026, against 1997 and 2015, the two most explosive El Niño onsets previously measured. The dashed line is the multi-model forecast median; the shading spans the middle 80 percent of the 667-member ensemble.

The temptation will be to file this under meteorology. File it instead under food, and under hunger.

An El Niño is not a storm. It does not arrive with wind and fury and then recede. It is a reorganization of where rain falls on Earth, a slow, hemisphere-spanning transfer of moisture away from the croplands of Southeast Asia, southern Africa, Central America, and northeastern Brazil, and toward places that cannot use the rain to produce food such as the open Pacific, the deserts of coastal Peru, the flood plains of the Horn of Africa. The warm water is only the mechanism; the failed harvests are what people will actually live through. Since the global food system runs on a handful of staple crops grown in a handful of climate-exposed regions, a sufficiently large El Niño is not a weather anomaly at all. It is going to be a global food shock on a twelve-month fuse. ³

“An El Niño does not destroy one harvest in one place. It reorganizes the rain over half the planet, on a schedule the world’s croplands were never meant to adapt to.”

4Hunger.org has spent the past several months documenting the two crises already squeezing the food system: the fertilizer supply shock that followed the closure of the Strait of Hormuz in February (see Fertilizer Famine), and the two-year drought that has driven the American cattle herd to its smallest size since 1951 and the U.S. wheat crop to its smallest since 1972 (see Parched Ground). The forecast now on the table is the third shock, and it is aimed squarely at the harvests of the next eighteen months all around the world.

A Map of What Comes Next

Strong El Niños do not damage the world’s agriculture randomly. They damage it in a pattern that has been documented across 150 years of El Niño events, and the pattern is brutally consistent. ⁴ To understand the geography of what is now forecast, start with where previous events did their damage, because they do it in the same places again and again.

Indonesia, the Philippines, and mainland Southeast Asia sit first in the line of fire, and the impacts are already arriving there. Strong El Niños suppress the rains across the Maritime Continent. Rice planting is delayed, reservoir levels fall, and the drained peatlands of Sumatra and Kalimantan become tinder: the 1997–98 El Niño event burned millions of hectares of Indonesian forest and peat and forced the largest rice imports in the country’s history. ⁵ Indonesia and the Philippines are, respectively, the world’s first- and second-largest rice importers in a bad year. Their bad year has already started.

India’s southwest monsoon which is the single most important weather system for human food security on Earth, watering the kharif rice, pulse, and oilseed crops on which 1.4 billion people depend is statistically reduced in strong El Niño years. The relationship is probabilistic, not mechanical; monsoons have defied El Niños before. But the monsoon now underway is running against the strongest Pacific heat forcing ever forecast, and the last time an El Niño of merely ordinary strength developed, in 2023, India’s response to a modest monsoon shortfall was to ban the export of non-basmati rice within weeks which sent shockwaves through the global rice supply. ⁶ India alone accounts for roughly 40 percent of the world’s traded supply.

Australia’s wheat belt faces its canonical El Niño outcome: a hot, dry austral spring landing precisely on the grain-fill and harvest window of the 2026–27 crop, in a year when the world has no American surplus to fall back on.⁷

Southern Africa is where the arithmetic turns cruelest. The region’s single maize season grows from December through March, the exact months of the forecast peak of the El Niño event. In 2015–16, a far weaker event than the one now projected, the drought that followed was southern Africa’s worst in 35 years: 40 million people became food-insecure, national disaster declarations were issued across the region, and maize harvests in Zimbabwe and Malawi fell by nearly half.⁸ The 2023–24 event repeated the pattern and again Zambia, Zimbabwe, and Malawi all declared national drought disasters.⁹ The season that begins this December will be planted by farmers who, thanks to the Hormuz fertilizer shock, are applying less nitrogen than at any time in a decade, into soil that the models say the rain will not reach.

Central America’s Dry Corridor including Guatemala, Honduras, El Salvador, Nicaragua loses its primera and postrera maize and bean seasons to El Niño droughts with grim regularity; the 2015 event left 3.5 million people there needing food assistance and helped drive a wave of northward migration.¹⁰ Northeast Brazil’s smallholder belt faces the same equation on a different calendar.

In contrast, El Niños cause other places to drown rather than to burn. The Horn of Africa’s October–December short rains typically come in torrents in El Niño years that cause flooding that destroys standing crops, drowns livestock, displaces communities, spreads cholera, and creates the wet breeding grounds from which desert locust outbreaks have historically erupted. Coastal Peru and Ecuador face the floods that gave El Niño its name, it also causes the collapse of the Peruvian anchoveta fishery, the largest single-species fishery on Earth and the anchor of the global fishmeal supply that feeds farmed fish, poultry, and hogs. When the anchoveta disappear, as they did in 1997–98, feed costs rise everywhere.¹

¹Figure 1. The harvest calendar in the crosshairs: canonical strong-El Niño impact windows for the world’s most exposed food-producing regions, July 2026 through December 2027, against the forecast November–January peak. Teleconnections are probabilistic, not certainties — but this is the pattern every major event since 1877 has broadly followed.

The Rice That Won’t Be Planted

Every staple has its own exposure to the effects of El Niño, but rice is where El Niño’s geography and humanity’s geography overlap most completely. Rice is the primary calorie source for more than 3.5 billion people. It is grown, overwhelmingly, in precisely the monsoon Asia that El Niño dries out. Unlike other crops only about a tenth of production is traded across borders as most rice is consumed is consumed in the country where it is grown. This means small production shortfalls produce large price increases, and large increases in price can produce panic. ¹²

The panic has a playbook, and the world has already experience rice market panic twice this century. In 2007–08, export restrictions cascading from one rice producer to the next tripled world prices in six months without any global production shortfall at all. In 2023–24, a moderate El Niño was enough to trigger India’s export ban and push rice prices to fifteen-year highs. The event now forecast is not moderate. If the El Niño causes the monsoon to be less than average or fail, and Southeast Asia’s rice paddies get late or insufficient monsoons, the question is how fast and what happens to Indonesia, the Philippines, Bangladesh, and the import-dependent nations of West Africa when they do.

In a global food system where a handful of export prices set the reference for everyone, having countries issue export bans will impact prices all over the world.

Three Shocks, One Table

What makes 2026 different from 1997 or 2015 is not only the size of the forecast. It is what the forecast is landing on.

The first shock came through a 21-mile strait. The closure of Hormuz in February severed a fifth of the world’s traded nitrogen fertilizer from its markets, sent urea prices up more than 60 percent, and guaranteed that farmers from Iowa to Uttar Pradesh would apply less nitrogen this year which will cause yield losses that arrive, on roughly a one-season lag, exactly as El Niño’s droughts do.¹³ A smallholder who cannot afford fertilizer loses 40 to 60 percent of her yield in a normal rainfall year.¹⁴ The models say this will not be a normal rainfall year.

The second shock is already in the ground. Sixty-one percent of the contiguous United States is in drought. The U.S. winter wheat harvest is only 1.561 billion bushels, the smallest since 1972. The American cattle herd is the smallest it has been since 1951 mean that the country that has served as the world’s agricultural shock absorber for eight decades is entering this event with no slack of its own.¹⁵ World grain reserves are the buffer between a bad harvest and a hungry year, and they are being drawn down even before the El Niño harvest failures begin.

The third shock is the ocean. Each of these crises would be manageable alone. The fertilizer shortage could be absorbed by a good growing year. The American drought could be offset by strong harvests abroad. An El Niño could be cushioned by full granaries and cheap inputs. Instead, each has arrived to find the others already there.

“A food system already starved of fertilizer and parched of rain is about to be struck by the strongest ever El Niño forecast. Any one of these is a crisis. All three at once is something the modern food system has never had to survive.”

Figure 2. The transmission chain from sea surface temperature to food insecurity. Global temperature lags the Pacific by three to five months; harvests lag by a season; prices lag harvests; hunger lags prices. The worst of what is now forecast lands on the world’s tables in 2027 and the lean seasons of early 2028.

What It Means in America

Honesty requires a complication here, and this publication has never shied from one: for parts of the drought-stricken American Southwest and Southern Plains, El Niño is not the villain of this story. A strong event typically shifts the winter storm track south, delivering above-normal cool-season precipitation across the southern tier from California to Texas. After six years of drought stress in the Southern Plains, a wet winter would be an unambiguous mercy that will recharge soil moisture ahead of the 2027 winter wheat crop and offering rangeland its first real drink since 2024. Credit where credit is due: if the pattern holds, some American farmers will owe this El Niño a debt.¹⁶

But no American eater should mistake regional rainfall for national immunity, for three reasons.

First, the price of food in an American grocery store is set in a global market. When Australian wheat fails and India restricts rice and southern Africa’s maize collapses, U.S. commodity prices rise regardless of the weather in Kansas, not because American grain is scarce, but because American grain traders sell into world prices. That transmission chain does not move at the speed of a headline. It moves at the speed of a growing season. But it moves, reliably and historically, to the consumer’s receipt.¹⁷

Second, the inputs. The anchoveta collapse and a global feed-grain squeeze land on a livestock sector already at a 75-year inventory low, with ground beef already migrating from staple to luxury. Beef prices were projected to rise 6.3 percent in 2026 before this forecast existed. And El Niño winters, for all their southern rain, raise flood risk across California’s Central Valley, the source of a third of America’s vegetables, where the line between drought relief and levee failure has always been thin.

Third, and most important: the Americans with the least room to absorb any of this. The United States entered 2026 with 47.4 million people, that is one in eight Americans already food-insecure, with the lowest-income fifth of households already spending nearly 30 percent of their budgets on food, and with Congress still weighing more SNAP cuts that would strip benefits from up to 8 million people.¹⁸ Every percentage point of El Niño-driven food inflation that arrives in 2027 will land on families who did not cause a war in the Gulf, did not cause a megadrought, and did not warm the Pacific.

What the Last Great El Niños Did

If the forecast holds, 2026–27 will join a short and terrible list, and it is worth remembering what the previous entries on that list actually did.

The 1877-78 El Niño was until this year, arguably the strongest ever observed caused the collapse of the monsoons of India and the rains of northern China and northeastern Brazil simultaneously. The famines that followed, amplified catastrophically by the colonial and market policies of the era, killed tens of millions of people. Historians call them the Late Victorian Holocausts. ¹⁹ The 1982-83 El Niño brought Australia’s worst drought of the century and Peru’s worst floods. The 1997-98 El Niño killed an estimated 23,000 people, caused some $40 billion in losses, set Indonesia’s forests alight, and broke the anchoveta fishery. ²⁰ The 2015-16 El Niño left 60 million people in need of humanitarian assistance across four continents which was the largest El Niño response operation ever mounted, and one the UN itself judged to have started too late.²¹

Figure 3. Peak intensity of the great El Niño events since 1877 (era-adjusted Niño 3.4 anomaly) and their human consequences, against the 2026–27 multi-model forecast median of 3.6°C. The whisker spans the middle 80 percent of the 667-member forecast ensemble — the bottom of which grazes the all-time record.