Spatiotemporal assessment of wildfire impact and burn severity in southern California using multi-sensor remote sensing and Google Earth Engine (GEE)

Wildfires are increasingly intense and frequent in California, fueled by anthropogenic climate change and extreme weather conditions. This study presents an integrated geospatial workflow using multi-source Earth observation data and cloud-based analysis to evaluate wildfire dynamics across Southern California during the December 2024–January 2025 outbreak. Using Google Earth Engine (GEE), we synthesized MODIS, Landsat-9, ERA5, and GPM-IMERG datasets to assess vegetation loss, relevant climatic factors, and fire severity. Pre- and post-fire NDVI and NBR indices revealed significant vegetation loss and fire-induced stress, while dNBR mapping classified burn severity using USGS thresholds. Seasonal anomalies in temperature, precipitation, and soil moisture indicated concurrent drought and heat stress across fire-prone zones. Hourly ERA5 temperature and wind speed time series revealed meteorological conditions conducive to fire ignition and spread. These results underscore the utility of multi-sensor remote sensing for characterizing wildfire risk through spatiotemporal analysis of ecological and atmospheric variables.