The Role of Postfire Rainfall and Drought Timing in Vegetation Recovery and Soil Erosion Risk

In wildfire-affected landscapes, climate variability and fire burn severity jointly determine how long the soil surface remains exposed and how quickly vegetation recovers. In practice, annual basis soil loss estimates often overlook the seasonal dynamics between exposure and stabilization, which are strongly driven by rainfall and drought timing. This study examines how rainfall and drought impact vegetation recovery and the risk of soil erosion in the Upper Cache Creek watershed, following the 2018 Ranch Fire in the southwestern United States. We monitored biome-specific vegetation cover dynamics and postfire recovery using seasonal time series of the Normalized Difference Vegetation Index (NDVI) derived from Landsat-8 imagery spanning 2016 to 2024. We mapped fire burn severity using the differenced Normalized Burn Ratio (dNBR) and quantified drought stress from the Gridded Surface Meteorological (GRIDMET) dataset using the Standardized Precipitation–Evapotranspiration Index. Rainfall erosivity density was reconstructed by integrating long-term mean annual precipitation from the Parameter-elevation Regressions on Independent Slopes Model (PRISM) with a high-temporal-resolution station-based erosivity dataset, thereby providing an updated rainfall erosivity factor for the study period. We further derived high-resolution soil erodibility and topographic parameters from the Soil Survey Geographic database and the Shuttle Radar Topography Mission digital elevation model, respectively. Geospatial data processing and model parameterization were conducted using Google Earth Engine and Quantum Geographic Information System. Results show that forest and shrubland exhibited an exponential recovery pattern in moderate-to-high-burn-severity areas, with a half-recovery period of about 2 to 3 years. By the sixth year following the fire, the two vegetation types had significantly rebounded, reaching 69% recovery in forests and 76% in shrublands compared to prefire conditions. Grasslands responded erratically, marked by rapid greening during the first postfire wet season and declines in subsequent drought years. In line with these vegetation trends, RUSLE estimates indicate that the largest erosion pulse occurred in the first postfire year, when high rainfall erosivity coincided with widespread soil exposure.  As a result, mean annual soil loss rates in fire-affected areas were up to fourfold relative to the prefire values. With rainfall erosivity closer to baseline conditions in 2024, erosion remained double due to intervening drought years, which suppressed early recovery gains by up to 40%. RUSLE soil loss estimates were validated with observed sediment yield at the watershed’s outlet and showed strong agreement across most postfire years. The observed sediment yield in 2017 remained notably higher relative to the estimated value. This anomaly was likely influenced by seasonal sediment-flushing operations, given the presence of two large upstream reservoirs. The results show that the interaction between rainfall and drought events governs postfire recovery and erosion, highlighting the importance of accounting for their timing, especially in annual assessments of postfire erosion.

Keywords: Wildfire, Soil Erosion, Vegetation Recovery, RUSLE, Rainfall Erosivity