Understanding the relative contribution of climate and human factors to wildfires is critical for managing risk across California’s diverse ecosystems, in the United States (US).

We propose a model that distinguishes between proximate and ultimate drivers of fire regimes and apply it to a century of fire and climate data to assess regional variation in causal mechanisms.

We analyzed fire statistics (1910–2021) alongside climate and weather data, stratifying the state by 10 ecoregions.

Northern forests had the strongest correlation with the proximate factor fuel aridity, ultimately due to climate. Fire rotation intervals exceeded 100 years, implicating woody fuel accumulation as an additional factor. Lightning ignitions occurred in decadal bursts, with dense strike events potentially overwhelming fire-fighting resources. Lower elevation/latitude foothill ecoregions experienced highest fire activity following wet winters and springs, implicating control by herbaceous fuel loads and a negative effect of global warming on future fires. Human ignitions dominate in these ecoregions, and population growth contributes to expansion of powerlines, a major ignition source.

While climate change may increase fire activity in forested ecoregions, its role is less pronounced in non-forested ecoregions, where human ignition sources are the dominant factor.

Different areas within ecoregions may require different management actions that reflect the specific proximate and ultimate factors at play.