PROJECTIONS
What models agree on, what they don't, and the one place where the disagreement is the finding.
A different kind of page
READ THIS FIRSTEvery other page here is computed from a data table in this repository, and the build fails if the prose and the table disagree. This page is not. It carries no downscaled projections, no scenario ensembles and no numbers of its own — it is a summary of published work, with links, and it is kept in a separate room from the measured record for that reason. Nothing on this page is plotted on the same axis as an observation, and the shield above says zero computed numbers because that is the honest count.
The reason for the separation is that the two kinds of claim fail differently. An observation can be wrong because an instrument moved. A projection can be wrong because the world took a different emissions path, because the model's cloud physics is imperfect, or because the thing being asked about is genuinely not predictable. Putting them on one chart makes the second look like the first.
What the models agree on
HIGH CONFIDENCEContinued warming
Every model, under every scenario, warms California further. The disagreement is about magnitude and timing, not sign — and the spread across scenarios is wider than the spread across models, which is to say the dominant uncertainty is what people do rather than what the physics does. The observed record is already running at more than +5°F per century since 1970.
Less snow, for the same storms
The most confident regional result, and it barely depends on precipitation at all. A warmer atmosphere raises the elevation at which snow falls rather than rain, so a fixed amount of winter precipitation yields less snowpack and melts it earlier. California stores roughly a third of its water supply in that snowpack, which makes this a water-management result rather than a scenic one.
Drier soils at the same rainfall
The mechanism already visible on the drought page: warmer air pulls more moisture out of soil and vegetation, so a given dry year does more damage than the same dry year did a century ago. This is why drought severity can worsen without any change in precipitation, and why the flat rainfall trend is not the reassurance it appears to be.
Sharper swings between wet and dry
Models generally project an intensification of California's already-extreme year-to-year variability — wetter wet years from more intense atmospheric rivers, drier dry years, and faster transitions between them. The observed record shows a clustering of such reversals in recent decades that is consistent with this, though fourteen events is far too few to call it confirmation.
The disagreement that is the finding
TOTAL PRECIPITATIONModels do not agree on whether California gets more or less total precipitation this century. The state sits on the boundary between a subtropical dry zone that is expected to expand poleward and a mid-latitude storm track that is expected to intensify, and where exactly that boundary lands is close to the hardest thing to get right in a global model.
Because it is a genuine result and not a gap waiting to be filled. It means a plan that only works if California gets wetter, and a plan that only works if it gets drier, are both bets on something the science does not currently support either way. The robust results above — hotter, less snow, drier soils, sharper swings — hold under both, which makes them the ones worth planning against. A page that resolved this uncertainty to make a cleaner story would be more satisfying and less true.
Where to get the actual projections
SOURCESAll of them are free, and all of them carry the scenario and ensemble detail this page deliberately does not reproduce.
- Cal-Adapt — California's own downscaled projection portal, built for the state's Climate Change Assessments. The right first stop for a question about a specific county or watershed.
- California's Climate Change Assessment — the state's periodic synthesis, including its regional reports.
- IPCC AR6 Working Group I — the physical science basis, with the Atlas chapter and its regional fact sheets.
- Fifth National Climate Assessment — the U.S. assessment, with a Southwest chapter covering California.
- NIDIS / Drought.gov — current conditions and outlooks, which is a different question again.