This winter’s El Niño: why we’re watching California

This is the latest in our series on what a strong El Niño could mean for risk this year. Previous posts looked at why a big El Niño event looked likely in 2026, how seasonal hurricane forecasts can be made useful for risk transfer markets, the hurricane bias El Niño years can bring, and the severe convective storm angle. If you are new to how the El Niño-Southern Oscillation, or ENSO, shifts global risk, those are good places to start. Here, we narrow the focus to flood risk in Southern California.

A forecast pushing into rare territory

The Pacific has now flipped, and it’s shaping up to be one of the strongest El Niño events in history. The latest seasonal forecast from the European Centre for Medium-Range Weather Forecasts (ECMWF) has the tropical Pacific’s sea surface temperature anomaly past 3°C on most ensemble members through winter, peaking around December and holding into early 2027

As we have highlighted before, a strong El Niño redistributes rainfall patterns around the world and gives us clues about where risk may build. It does not let us forecast impacts precisely, but it does change the odds.

What the global pattern says

Some El Niño patterns are more reliable than others. Erik Kolstad, a climate researcher in Bergen, Norway, maintains a set of maps tracking El Niño and La Niña’s global effects on temperature and rainfall. One of those maps shows how three-month average precipitation for December-January-February has historically tracked the Niño 3.4 index over the same period. It is based on decades of reanalysis data, not a forecast, and it tells a familiar story (Figure 2)

Historically, the southern US – including the Gulf Coast and Florida – has tended to be wetter during El Niño winters, as the jet stream shifts south and storm tracks pick up moisture from the Gulf of Mexico. The Maritime Continent – Indonesia, Malaysia and northern Australia – has tended to be drier, as warmer waters in the central and eastern Pacific pull rising air and rainfall eastward, away from the western Pacific.

That is the historical backdrop. More directly for this winter, ECMWF’s seasonal forecast for DJF 2026/27 – based on its current ensemble of model runs rather than past observations – points in the same direction. The southern tier of the US and the Caribbean are flagged with more than a 70% chance of falling in the wettest fifth of the historical record (Figure 3).

Southern California’s signal

Southern California sits partly in the dark red zone in Figure 3, with parts of the coast in the >70% “top 20% wet year” bracket. Daniel Swain covered this recently, and it is worth reading. His point is that the unusual feature is not the wet signal itself, but the confidence ECMWF is showing six months ahead. Seasonal forecasts at this range are usually hedged and noisy. This one is less so. It is also pointing at the season when, according to US National Flood Insurance Program (NFIP) historical records, around 90% of insured flood losses in Southern California have occurred.

The recent history is a useful reminder. In January 2005, Los Angeles County experienced its first large flood since 1938, after a sequence of winter storms pushed the Los Angeles River basin and other Southern California watersheds into flood. Nearly two decades later, in January 2024, San Diego saw record-breaking rainfall and severe flash flooding, with low-lying neighbourhoods such as Southcrest, Shelltown, Mountain View and Encanto among the hardest hit. A further sequence of atmospheric-river storms followed in early February, extending the flood and landslide concern across Southern California. These episodes are not simple ENSO analogues, but they show how quickly a wet winter signal can become a local loss problem when storm tracks, drainage capacity and exposure line up.

What the claims history shows

We then went back to the data: NFIP claims for Southern California, lined up against the Relative Oceanic Niño Index, a measure of ENSO strength (Figure 4). We did not detrend the losses, so if anything, the historical El Niño signal below is understated. Even so, the older loss years still stand out.

Even on that conservative basis, and despite the prevalence of La Niña periods over El Niño during the last decade, the pattern is clear. The winters of 1982/83, 1997/98 and 2015/16 – all strong-to-very-strong El Niño years – sit close to the largest NFIP loss spikes in the fifty-year record.

Conclusion

None of this is certain. El Niño shifts the odds of a wet Los Angeles winter – sometimes sharply, sometimes only slightly. But when three different signals – historical climate data, seasonal forecasts and claims history – all point in the same direction, it is worth paying attention.

California flood risk is worth watching closely this winter. The current signal points to a wet DJF across the southern US, including coastal Southern California. For flood-exposed portfolios in the region, this is a live risk to flag now. It also makes local risk selection more important: when flooding occurs, not every location will be exposed in the same way.

We’ll keep updating our view as the season develops.