Climate Science and Policy

In the last decade, research on atmospheric rivers (ARs) has surged due to their potential to cause extreme flooding and alleviate drought stress in the Western United States. Concurrently, wildfires in Southern California have become increasingly costly and destructive. While the timing of the first fall AR is known to influence wildfire risk in Southern California, the impact of ARs on spring vegetation growth and subsequent fall wildfire risk remains poorly understood, especially at the fine scale of Southern California ecosystems.

To investigate this link, we analyzed forty years of AR-associated Integrated Vapor Transport (IVT) and winter precipitation data across ten Level IV Ecoregions in Southern California. We examined the correlations between these metrics and the Enhanced Vegetation Index (EVI) and the Evaporative Demand Drought Index (EDDI). Our analysis revealed that productivity during the growing season is linked to the previous winter's total AR-associated IVT, and even more strongly to winter precipitation totals. We found positive correlations between AR-associated IVT and precipitation with EVI in early peak fire season (defined as August through October), particularly in chaparral-dominated areas, while conifer forests and grassland ecosystems displayed weaker correlations. 

We also found that the 90-day EDDI in early peak fire season was influenced by winter precipitation in some ecoregions, and that EDDI also had significant correlations with early peak fire season EVI. 

Our findings suggest that wet winters may increase wildfire risk in fuel-limited ecosystems like grasslands due to higher biomass accumulation, whereas in chaparral-dominated areas, increased moisture may reduce wildfire risk via maintaining higher fuel moistures in the fire season. These results underscore the need for further research to clarify the importance of fuel abundance versus fuel dryness, especially in conifer forest and sage scrub ecoregions for accurate Southern California wildfire risk assessment.