The 1997 El Niño killed 23,000 — we built better defenses, then forgot them

The World Meteorological Organization is warning that El Niño will persist through at least February 2026, disrupting harvests, inflaming drought across vulnerable regions. Destabilizing water supplies in ways that will kill people who did not have to die. The uncomfortable part is that this is not new information arriving. It is old information we have already decided to forget.

In 1997, when El Niño stripped the Pacific of predictability, an estimated 23,000 people died. Floods and droughts and crop failures cascaded through economies that had no early warning, no preparation protocols, no shared understanding of what was coming. The damage toll hit $96 billion. The mortality was not inevitable — it was a failure of infrastructure that, at the time, we did not have. After 1997, we built it.

When the 2015–2016 El Niño arrived with comparable strength, it should have been catastrophic again. Instead, deaths fell dramatically. Larger populations, same phenomenon, vastly better outcomes. That gap exists because networks that did not exist in 1997 were humming by 2016. Early-warning systems. Coordinated response protocols. Real-time data sharing between meteorological agencies. The infrastructure cost money and political will, but it worked with a clarity that almost never happens in climate adaptation.

The question now is brutally simple: Have we maintained it? Kept the funding flowing, the systems updated, the trained staff in place? Or have we done what institutions do when a crisis passes. Let the thing that prevented it slowly erode as budget cycles turn and attention drifts elsewhere? The WMO warning reads as urgent, but urgent about what we should already know how to handle.

We already know how to survive this — we just have to remember that we did. That is not a comforting position to find yourself in: having solved a problem once, then watching yourself slowly forget the solution while the problem returns. It happens in organizations and in lives. The operational discipline that made 2016 survivable was not magic. It was maintenance, attention, the unglamorous work of keeping systems alive. If we have done that work, the coming months will hurt less than they could. If we have not. If we have slipped back into the posture of 1997. Then the warning arriving now is less a forecast than an indictment of what we chose to let lapse.