Soil erosion and sediment sourcing in the Hyrcanian forests, Northern Iran: an integration approach of the G2loss model and sediment fingerprinting technique

This research aimed to study the contribution of main land uses in soil erosion and sediment yield in the Hyrcanian forest watersheds using the integration of the G2loss model and sediment fingerprinting technique. In order to prepare a soil erosion map in the study area, the input factors of the G2loss model were prepared in appropriate spatial and temporal scales using meteorological data, satellite images, application of GIS and RS. In addition, the contribution of sediment sources was performed by geochemical tracers using the FingerPro package in R software in the Talar Watershed as a representative for the study area. The results of the G2loss model showed that the average annual soil erosion in the study area was 13.46 t ha-1, and the highest and lowest monthly soil erosion values of 2.04 and 0.76 t ha-1 correspond to October and June, respectively. The sediment sourcing results showed that Er, Cs, Ti, Ta, Sc, and Ta geochemical tracers in the Talar watershed were selected as the best tracers based on Range tests (RT), Kruskal-Wallis (KW), and Discriminant Function Analysis (DFA). Finally, based on the combination of the G2loss model and sediment sourcing, the three main land uses of rangeland, forest, and agriculture accounted for 73.77, 21.74, and 4.49% of the total annual soil erosion in the study area, respectively. Agricultural lands, followed by forests and rangelands contribute 59.97, 28.15 and 19.84% to annual suspended sediment yield, respectively. The main reasons are that agricultural lands located near the rivers have the highest proportion of silt particles (51.11%), and the lowest value of organic matter (3.88%) compared to the rangeland and forest. In addition, these lands are located in lower elevations with higher rainfall erosivity. These results can be used to complement other more conventional approaches and as an important input variable of multi-criteria planning to propose optimal land use for large areas.