Useful indicators and models for assessing erosion control ecosystem service in a semi-arid forest landscape

Forests provide a large array of ecosystem services (ESs) such as wood supply, extreme natural event prevention, and ecotourism opportunities. The quantitative characterization of ESs is a crucial but costly task for environmental managers. The aim of this study was to develop easily applicable models and indicators for assessing erosion control ES in a semi-arid landscape. In order to accomplish this, 107 randomly selected plots were visited for field measurements and topsoil sampling. Several parametric tests were then used to analyze the field data. The findings revealed that (i) normalized difference vegetation index (NDVI), (ii) cover management (C) factor of the Revised Universal Soil Loss Equation (RUSLE), (iii) soil organic matter content, (iv) canopy cover ratio, and (v) land use/land cover (LULC) types could be used as useful performance indicators of erosion control ES. Two regression models were developed based on these indicators and compared to RUSLE results for the study area. Using the first model, we were able to estimate the soil protection performance of different LULC types by NDVI at the pixel level (R²_adj = 0.90, p < 0.05). The second model estimated annual potential soil loss using NDVI and ground slope values (R²_adj = 0.57, p < 0.05). Based on the ES indicators framework, a practical approach was proposed in this study for rapid assessment of the soil erosion problem without running RUSLE. Thus, environmental managers are expected to make well-informed landscape planning decisions and improve their ES provision application capabilities at a reduced cost.