Development of an Automatic Process Chain for Estimating Post-fire Soil Erosion Susceptibility

Abstract

Soil erosion is a global problem because it has negative influences on both society and the environment. Amongst other factors, fire increases the susceptibility of soils to erosion due to the removal of vegetative cover and changes in soil properties. In order to take preventive measures against soil erosion, areas of increased post-fire soil erosion susceptibility need to be determined first. Therefore, this study aims to develop an automatic process chain for estimating post-fire soil erosion susceptibility. The approach is based on the USLE model family, and applied qualitatively. The resulting Soil Erosion Susceptibility Index (SESI) provides classes of soil erosion susceptibility instead of absolute estimates. Datasets and equations are chosen to make the approach applicable to study areas worldwide and to be simple and fast. Suggestions for suitable calculation variants are given to run the approach in a default mode. To assess changes in soil erosion susceptibility, a bi-temporal difference grid between pre- and post-fire soil erosion susceptibility is calculated. The automatic process chain is tested for three exemplary study areas in California (USA 2018), Maule (Chile 2017), and La Palma (Canary Islands 2016). In contrast to Maule study area, California study area is not only characterized by increases but also by decreases in post-fire soil erosion susceptibility due to quick vegetation recovery. La Palma study area has already high pre-fire soil erosion susceptibility, but the post-fire soil erosion susceptibility decreases to nearly pre-fire values only one year after the fire. The results of this study show that the automatic process chain is well suited for giving a qualitative overview of the extent and changes in post-fire soil erosion susceptibility. However, validation of the results remains problematic, and the application reveals several limitations and points of improvement.