A "Godzilla-strength" El Niño is forming — and the alarm system that would warn the food world is being switched off

Food & Farming 4 min 80 sources

Forecasters say there is a 63% chance this El Niño reaches "very strong," among the worst since 1950. The last one this size, in 1877, helped kill tens of millions. The difference now is that we can see it coming — if the sensors stay on.

Key takeaways

A strong El Niño is forming, with a 63% chance of becoming "very strong" — the kind of event that can fail several of the world's breadbaskets at once.
The difference between today and the catastrophic 1877 famine is advance warning from ocean sensors — and the US nearly dismantled a key part of that network this season.
Closer to home, farm costs are heading for record highs and the White House is asking for another $11 billion in aid, even as the US corn and wheat crops look healthy for now.

The biggest food story this week wasn’t a price or a harvest. It was a forecast. The world’s ocean-watchers say a strong El Niño is forming in the Pacific, and the tools that turn that warning into action are under threat.

The forecast, and why a warm ocean moves your grocery bill

This month the United States ocean agency confirmed an El Niño has formed in the tropical Pacific [44]. El Niño is a periodic warming of the surface of the eastern Pacific Ocean. That warm water shifts where rain falls across the whole planet — pulling it away from some farming regions and dumping too much on others. Forecasters give it a 63% chance of reaching the “very strong” threshold, which would place it among the strongest events since records began in 1950 [44].

That matters at the table because a few regions grow most of the world’s staple food. When El Niño dries out southern Asia or floods parts of the Americas at the wrong moment, harvests fail in several breadbaskets at once — and there is no spare crop sitting somewhere to fill the gap. United Nations experts this week warned of a “compound shock effect”: the El Niño landing on top of the economic fallout from the Middle East conflict, which together could put millions of tonnes of the world’s food at risk [73].

South-east Asia is the place watching most closely. The region grows close to a tenth — about 9% — of the world’s farm exports, and a “Godzilla strength” El Niño would hit its rice and other crops directly [73].

What history says happens when no one sees it coming

The last El Niño this size, in 1877, is a warning written in lives. As much of the world dried out, harvests collapsed in India, China, parts of Africa, and Brazil. The drought, made worse by the colonial trade policies of the time, became the “Great Famine,” which killed an estimated 30 to 60 million people — about 3% of everyone alive [44].

The thing that separates us from the people of 1877 is not luck. It is data. Modern ocean monitoring gives weeks to months of advance warning — enough time for governments to release grain reserves, for traders to reprice calmly instead of in a panic, for aid to move before the hunger, not after it [44].

The part that should worry an eater

Here is the turn in the story. While this El Niño builds, the United States has been trying to pull apart one of its best ocean-watching networks. The National Science Foundation began “descoping” — bureaucratic language for dismantling — the Ocean Observatories Initiative, a system of more than 900 sensors built over a decade at a cost of about $386 million [44].

Those particular sensors don’t spot El Niño directly; they measure how much heat the deep ocean is soaking up, which feeds the wider forecasts. Independent researchers estimate that switching off United States ocean observations would raise the error in annual ocean-heating estimates by 163% — blurring exactly the early warnings that help a country prepare [44]. In 2025, weather and climate disasters cost the United States $115 billion [44].

After a rare unanimous Senate vote, the agency paused the teardown and began redeploying sensors it had already pulled [44]. But a paused dismantling is not the same as an uninterrupted record. Data already has gaps. The alarm was nearly unplugged the same season the fire started building.

Meanwhile, the cost of growing food keeps climbing

On the ground, the squeeze is already here. The White House this week asked Congress for more than $11 billion in extra aid for farmers struggling with high fuel and fertiliser costs, on top of $12 billion already paid out this year [36]. The USDA expects farm production costs — what it takes to plant and harvest a crop — to hit record highs in 2027 [8].

The United States crop itself looks healthy for now. As of mid-June, 68% of the corn crop was rated good or excellent, and winter wheat harvest was running 16 percentage points ahead of its five-year pace [4][9]. But “for now” is doing a lot of work. Nebraska still had 44% of the state in extreme or exceptional drought even after a weekend of rain and crop-stripping hail [10]. A healthy crop in June is not a harvest in the bin.

A quieter story worth ending on

Off the west coast of Scotland, a patch of seabed that was torn up by an illegal trawler in 2019 is starting to come back [38]. Divers using an underwater drone have found sea cucumbers, cat sharks, crabs, and returning seaweed inside a protected zone where dredging is now kept out [38].

It is slow — experts say full recovery may take a decade [38]. But it is a reminder of the same principle running under the El Niño story: the systems that feed us can repair and protect themselves, but only if we keep the protections switched on. A reef recovers when you stop dragging gear across it. A famine is foreseen when you keep the sensors in the water.

02 · Lesson · why it matters

The disaster you can see coming is a different disaster

The same drought killed tens of millions in 1877 and far fewer today — not because the world got kinder, but because we learned to see it arriving in time to move.

Two famines, one ocean

In 1877 the Pacific Ocean warmed, the rains moved, and harvests failed across India, China, Africa, and Brazil at the same time. Tens of millions of people died. They never knew it was coming. The ocean was sending the same signal it sends today, but no one could read it.

This year the same ocean is warming again, possibly to the same strength. We know. We have known for weeks. That single difference — knowing — is the most valuable thing the food system owns, and it is invisible. It doesn’t grow in a field or sit in a warehouse. It is a number arriving early enough to act on.

What “early” is actually worth

A harvest failure with no warning and a harvest failure with three months’ warning are not the same event. They produce the same empty fields. They produce wildly different outcomes for people.

With warning, a government opens its grain reserves before prices spike. Traders reprice slowly instead of in a stampede. Aid moves toward the hunger before it arrives, not after the photographs. The crop still fails — but the famine doesn’t have to follow, because the time between the signal and the shortage is when every useful decision gets made.

Take that time away and you don’t just lose information. You lose every choice the information would have unlocked. The failure becomes a surprise, and surprises are what kill people.

The signal travels further than you think

It is tempting to file this under “their problem” — a drought in southern Asia, a flooded field in Brazil, somewhere far away. But the ocean doesn’t respect that map.

A warm patch of Pacific water sets the rain that waters the rice that a few countries grow most of. When those harvests stumble, there is no spare crop anywhere to cover the gap, so the price of that grain rises on every continent at once. The number on the shelf where you shop is partly set by water you will never see, watched by sensors you will never visit. The Nebraska farmer praying for rain, the Vietnamese rice grower, and the person reading a receipt at a till are all standing inside the same weather system, pretending they are in separate rooms.

The warning is not a courtesy extended to distant strangers. It is the thing that keeps your own shelf stocked and your own bill steady. You are downstream of the sensors too.

The alarm is a choice, not a fact of nature

Here is the part that is easy to miss because it looks like plain bureaucracy. The network of ocean sensors that makes the warning possible was built by people, paid for by people, and can be switched off by people. This season, one of the best of those networks was nearly dismantled — not by a storm, but by a budget decision, before a rare unanimous vote paused it.

We tend to treat early warning as if it were weather itself: just there, like the tide. It isn’t. It is a structure someone chose to build and keeps choosing to maintain. The cost of building it shows up on a ledger every year. The cost of not having it shows up only once, all at once, as a famine no one saw coming — and by then the choice has already been made.

Calling that out is not picking a villain. A sensor network is genuinely expensive, and the people deciding its fate are weighing real budgets against a risk that hasn’t happened yet. That is exactly what makes it dangerous: the savings are visible now and the cost is invisible until it isn’t.

What seeing the whole leaves you holding

The honest thing to sit with is how thin the line is. Between us and 1877 stands a string of buoys in the water and the agreement to keep funding them. Not destiny, not progress as a law of nature — a maintenance decision, renewed or not.

And even with every sensor running, no one is watching the whole ocean. The forecaster sees the Pacific but not the Nebraska hailstorm. The farmer feels the drought but not the fertiliser ship stuck in a distant strait. The trader reads the price but not the field. Each seat sees a real piece and mistakes it for the whole picture. The warning works only when those partial views are stitched together — which is itself a fragile, human, fundable thing.

We are not above this system, watching it from a safe height. We are inside it, eating from it, depending on an alarm we mostly forget is switched on. The most useful thing knowing can teach us is how much of our safety we never had to think about — and how easily a thing that quiet gets turned off.

03 · Lab · your turn

The Warning Window

Rehearse how the same harvest failure becomes a manageable shortage or a famine, depending only on how early the warning arrives.

04 · Hope · carry this

The people of 1877 could only watch the rains fail; we built buoys that let us see the same disaster coming in time to act, and a torn-up Scottish seabed is quietly growing back where we simply stopped the harm. The systems that feed us can heal and forewarn — the work is just deciding to keep them switched on.