Post-fire Vegetation and Hydrologic Recovery in a Mediterranean Climate

Abstract:

Accurate field data is required to predict elevated runoff and sediment transport, which can aid short- and long-term planning after wildfire to reduce risk downstream. Sources of elevated erosion include soil hydrophobicity, in-channel sediment loading through dry ravel, extensive rill networks, and exposed bare soil. Recovery of these processes are related to vegetation conditions before, during, and after wildfire. This can be quantified using satellite-based vegetation indices to understand the recovery of the burned area and hydrologic response. The main objectives of this work are to evaluate regrowth patterns with respect to 1) soil burn severity patterns for the entire burned area and 2) streamflow generation using a paired catchment analysis after the 2018 Holy Fire for WY 2014-2020 (Water Year Oct 1, 2013 to Sep 30, 2020). The Landsat 8 Collection 1 Tier 1 8-day, 30-meter enhanced vegetation index (EVI) composite product was analyzed pre- (WY 2014-2018) and post-fire (WY 2019-2020) for the entire burned area (94 km²). ECOSTRESS products (70-meter), Evapotranspiration (ET) and Evaporative Stress Index (ESI) - both generated from the Priestley-Taylor Jet Propulsion Laboratory algorithm, were available after July 2018 and were also used to assess catchment recovery. On average for the entire burned area, high soil burn severity had the greatest pre-fire EVI compared to less severe burn categories, indicating the influence of fuel load on the spatial heterogeneity of soil burn severity. Coldwater Creek (burned) and Santiago Creek (unburned), two geographically and geologically similar catchments, were used for the paired catchment analysis where annual cumulative EVI (∑EVI) pre-fire averages were similar (0.23% difference). The burned catchment had a 45% decrease in ∑EVI during the first post-fire year compared to pre-fire average, while ∑EVI for the unburned catchment recorded an 8% increase during the same period. Satellite-based vegetation indices were compared to hydrologic signatures (time to peak, slope of rising limb, annual flow volume, and rainfall-runoff ratio) to understand recovery with respect to time. Results demonstrate a suite of satellite-based vegetation indices (EVI, ET, ESI) that can be used by planners to assess hydrologic recovery following wildfire.