In 1877, North Americans experienced an unusually mild winter – it was known as the “year without a winter”. It coincided with one of the strongest El Niño events ever recorded. Scientists suspect the same El Niño was a major factor in one of the worst environmental disasters in history. As much of the world was enveloped in drought, harvests collapsed in India, China, parts of Africa, and Brazil. The drought, compounded by colonial and other socioeconomic policies, led to the “Great Famine”, which killed between 30 and 60 million people, about 3% of the world’s population at the time.

What distinguishes us from the victims of 1877 is not luck but data. When I served as deputy administrator of the National Oceanic and Atmospheric Administration, I saw modern ocean monitoring and forecasting provide the advance warnings the Victorians lacked. This lead time saves thousands of lives and billions of dollars each year. Today, we can anticipate climate shocks before they arrive.

This month, NOAA confirmed the formation of El Niño in the tropical Pacific and issued an official advisory. Forecasters expect it to strengthen through the winter of 2026–27, with a 63% chance it will reach the “very strong” threshold, placing it among the strongest events in the modern record dating back to 1950. In a world already experiencing record heat, such an event could bring more dangerous extremes: drought, wildfires, flooding, and in the Pacific, a more active hurricane season. And, as is always the case, these events disproportionately affect the most vulnerable.

In the face of this evolving threat, the Trump administration has sought to cripple our forecasting capabilities. This spring, the National Science Foundation (NSF) began “descoping” the Ocean Observatories Initiative, a network that delivers real-time ocean data from more than 900 sensors. “Descoping” is bureaucratic language for dismantling the program. The agency announced plans to pull all sensors, buoys and other equipment from four of the program’s five sites. These arrays span from the Gulf of Alaska to the Irminger Sea between Greenland and Iceland, and down to the waters off North Carolina. Built over a decade at a cost of approximately $386m, the system is among the most advanced ocean-observing networks in the world.

Make no mistake. Pulling these arrays was not a budgetary exercise. Rather, NSF’s actions are more properly viewed as an extension of the Trump administration’s broader assault on federal climate science. The objective is apparently to weaken the programs that measure climate change and then claim the problem is “uncertain”. But turning off the alarm does not put out the fire.

The scientific community and members of Congress reacted to NSF’s actions with fierce objections. In a rare display of bipartisan unity, the Senate unanimously passed a bill introduced by senators Lisa Murkowski and Jeff Merkley to prohibit the use of federal funds to dismantle the network until a thorough review is conducted. Last week, the NSF announced it would stop the removal, keep the system running, and redeploy the sensors that had been pulled out of the water.

But this is merely a temporary reprieve, and the system remains at risk. The NSF has paused its actions and deferred the network’s future to a yet-to-be-convened panel. Sensors have already been removed, and data streams have been interrupted. Their redeployment after removal is not equivalent to uninterrupted operation. We should not have to rely on last-minute interventions to preserve systems critical to protecting lives and property.

While these sensors don’t detect El Niño formation, they measure deep-ocean temperature, the best gauge of how much excess heat the planet is absorbing. Independent researchers warn that removing US observations like these would increase the error in annual ocean-heating estimates by 163%, degrading forecasts and early-warning systems that help the country prepare for disasters. In 2025 alone, those disasters cost the US $115bn. The same data informs the management of fisheries that support 2.1m American jobs and $319bn in annual sales, as of 2023. The Trump administration was prepared to put all of that at risk to dismantle a system that costs just $56m a year to run. That is the scale of its recklessness.

If we allow these systems to remain vulnerable to political whims, an extreme event will eventually catch us unawares. An unexpected flood, a failed harvest or a catastrophic storm will be dismissed as an unavoidable act of nature. The famines of 1877 were also attributed to fate, but they largely resulted from failures to anticipate and respond.

Last week, the system was saved. For now. The panel NSF plans to convene should recommend permanent protection, and Congress should write that protection into law, so the instruments we rely on to understand the ocean are not at the mercy of an election outcome. The ocean stores most of the excess heat that shapes storms, marine heatwaves and climate shocks such as El Niño events. We now have the ability to measure it, issue forecasts based on what it tells us, and brace for what may be coming. We came far too close to throwing it away.

• Terry Garcia is a former deputy administrator of the National Oceanic and Atmospheric Administration