A cloud-based mapping approach for assessing spatiotemporal changes in erosion dynamics due to biotic and abiotic disturbances in a Mediterranean Peri-Urban forest

Abstract

During the last decades, the demographic trajectory and the associated urban development have increased the demand for urban and peri-urban green space. Peri-urban forests provide a wide range of goods and services to city dwellers critical to human well-being. However, these ecosystems are particularly vulnerable to biotic and abiotic stresses. This study aimed to assess the effects of forest disturbances on soil erosion dynamics in a peri-urban forest of Northern Greece, between 1995 and 2020. Monitoring of soil erosion dynamic was performed by coupling the Google Earth Engine (GEE) cloud-computing platform and the RUSLE erosion prediction model. The results highlighted the potential of Landsat imagery to efficiently delineate through time forest cover changes due to the influence of biotic and abiotic factors. After a major fire event, average soil erosion showed an increase of 7.7 t/ha/year whilst the emergency hillslope rehabilitation treatments led to a decrease equal to 8.9 t/ha/year. On the contrary, post-fire watershed stabilization measures and selective logging for the removal of the infested individual trees had a negligible effect on soil loss. The temporal changes in soil loss rate are not only justified by forest cover changes but also by the variability in rainfall, which is also considered a dynamic factor of RUSLE model. Monitoring of soil loss and erosion regulation services over peri-urban forests provide essential information for policy-making and management of these valuable natural resources.