Daily fire weather projections for mid- and late-century climate in Canada and the United States

Abstract The Canadian Fire Weather Index System (CFWIS) is widely used to characterize fire weather conditions and support wildfire risk assessments in real time as well as in historical trend analysis. To extend its application to future climate scenarios, we generated a daily CFWIS dataset for the 2050 s (2041–2070) and 2080 s (2071–2100) across Canada and the United States at 0.25° spatial resolution. The dataset is based on four CMIP6 global climate models (CanESM5, EC-Earth3, GFDL-ESM4, and UKESM1) and multiple Shared Socioeconomic Pathways (SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5). Daily meteorological inputs were generated by applying monthly CMIP6 climate change factors to ERA5 reanalysis data (1991–2020). This approach preserves the daily variability and temporal sequencing of the ERA5 baseline while incorporating projected changes in mean climate conditions. Compared with raw daily GCM data, ERA5-derived inputs more closely reproduce observed meteorological distributions and produce substantially more realistic CFWIS component distributions. The dataset provides internally consistent daily fire-weather sequences suitable for comparative and scenario-based analyses across Canada and the United States.