Seasonal Outlooks
One to six months ahead from the world's seasonal forecast systems via Copernicus (C3S): ECMWF, the UK Met Office, Météo-France, DWD, CMCC, NCEP, JMA and Australia's BOM, plus their multi-model mean; NOAA's North American Multi-Model Ensemble (NMME); and Canada's CanSIPS straight from Environment Canada, which runs 12 months ahead. Maps show each model's ensemble-mean anomaly against its own past forecasts. Seasonal forecasts shift the odds rather than predict specific storms, and their skill in the West leans heavily on ENSO; see the Pattern Heatmaps for how that plays out in the Sierra. For the next two weeks, use Model Forecasts.
Loading seasonal maps…
How to read seasonal outlooks
Each map is a model's ensemble-mean anomaly for the chosen month or 3-month season: how much warmer or colder, wetter or drier, or how much higher or lower the 500 mb heights it forecasts compared with its own past forecasts for the same time of year. Measuring against the model's own hindcasts removes its built-in biases, so models can be compared side by side. The multi-model mean blends the Copernicus systems and is usually more reliable than any single model.
Seasonal forecasts shift the odds; they do not predict individual storms. Their skill over California and the Sierra Nevada comes largely from ENSO and fades in neutral years and at longer leads. A wetter-than-normal signal means a wet season is more likely than usual, not that every month will be wet. Outlooks update each month as new runs arrive.
For how past ENSO states played out in the Sierra, see the Composite Maps and Snowfall Explorer. For the next two weeks, use the model forecasts.