Spatial estimation of soil erosion and sediment yield using the RUSLE and SEDD models in Chehelchay watershed, Golestan Province

Extended Abstract
Introduction
Soil erosion is one of the most serious and pervasive environmental challenges worldwide, directly and significantly threatening the stability of natural ecosystems, the quality of drinking water and agricultural resources, and long-term food security. In many mountainous regions of Iran, the combination of climatic conditions, steep slopes, geological instability, and increasing anthropogenic pressures has exacerbated this process. The Chehelchay watershed, located in Golestan Province, is a prime example of a watershed susceptible to erosion. This basin, covering an area of 25683 hectares and receiving an average annual rainfall of 766 mm, has witnessed extensive land-use changes, with a significant portion of its natural forest cover being converted into agricultural lands. Despite numerous studies conducted in this region and similar areas in the past, limited efforts have been made to integrate empirical models (such as RUSLE) with process-based or distributed models (such as SEDD) for the simultaneous estimation of the spatial pattern of soil erosion and sediment transfer, using reliable spatial data. Therefore, this study aims (i) to accurately quantify the spatial distribution of soil erosion and (ii) to estimate sediment transfer rates, and to identify priority and critical erosion areas within the Chehelchay watershed, by integrating the Revised Universal Soil Loss Equation (RUSLE) model and the Sediment Delivery Distributed (SEDD) model. Few studies have simultaneously mapped soil erosion and sediment delivery using integrated empirical and distributed models in data‑limited mountainous watersheds. This integrated approach, which simultaneously accounts for processes of surface sediment generation and its transfer along the river network, provides a robust and comprehensive framework for supporting conservation and management planning in data-limited watersheds.
Materials and Methods
To estimate surface erosion and sediment yield in the Chehelchay watershed, the RUSLE and the SEDD model were employed. For estimating surface erosion, the RUSLE model was utilized, incorporating five factors: Rainfall erosivity (R), Soil erodibility (K), Slope-Length (LS), Land cover and management (C), and Support practices (P). Spatial input maps were generated for each RUSLE factor in ArcGIS. The R-factor map was extracted using long-term daily precipitation data and the modified Fournier index. Physicochemical properties of the soil formed the basis for preparing the K-factor map. The LS factor map was calculated by processing the Digital Elevation Model (DEM) in a GIS environment. The interpretation of Landsat 8 satellite imagery and vegetation cover maps enabled the preparation of the C-factor map. The P-factor was determined based on land use type and farming practices. After preparing the RUSLE parameter layers, the annual water erosion map was generated by multiplying these layers in ArcGIS software. Subsequently, the SEDD model was used to quantitatively estimate sediment yield and its transfer ratio to the watershed outlet. This model, by spatially calculating the Sediment Delivery Ratio (SDR), establishes a link between the erosion calculated by RUSLE and the sediment data measured at the Lezoreh hydrometric station. Suspended sediment data from the Lezoreh station were analyzed using the sediment rating curve method and an intermediate data approach. The accuracy of the curve calibration was confirmed with a coefficient of determination R² = 0.79. Finally, the sediment yield, SDR, and erosion maps derived from the models were integrated and spatially analyzed to identify priority areas with severe erosion and high sediment yield potential.
Results and Discussion
The RUSLE-based assessment indicated that the average annual soil erosion rate in the Chehelchay watershed is approximately 4 t ha⁻¹ yr⁻¹. However, its spatial distribution exhibited considerable heterogeneity and was found to be significantly controlled by topographic factors, particularly the LS factor (slope and slope length), and land use type. The highest erosion rates were observed in agricultural fields situated on steep slopes; this pattern is primarily attributed to minimal soil conservation measures and the direct exposure of soil to intense rainfall events. In contrast, forested areas displayed strong protective effects, substantially reducing surface erosion due to dense vegetation cover and soil structural stability. These findings underscore the critical importance of preserving and expanding forest cover in mountainous regions. Based on data collected from the Lezoreh sediment monitoring station, the average annual sediment yield was estimated at 1.78 t ha⁻¹ yr⁻¹. The overall watershed SDR was calculated to be approximately 44%. This relatively high SDR indicates the presence of efficient sediment transfer pathways, largely facilitated by steep slopes and discontinuous vegetative cover across the watershed. Integrated mapping of erosion potential and sediment delivery ratios revealed that areas with the highest erosion potential and sediment generation are primarily concentrated in the middle and upper sections of the watershed, where slope gradients exceed 30% and agricultural expansion has replaced natural forest cover. The spatial correspondence between high erosion potential and elevated SDR highlights the crucial importance of simultaneously considering both sediment generation (erosion) and sediment transport (SDR) processes when designing management interventions.
Conclusion
Ultimately, the outcomes of this research can serve as a practical guide for developing and implementing sustainable water and soil resource management plans in the Chehelchay watershed and in other regions with comparable ecological and data conditions. The overall study results emphasize the pivotal role of topographic factors and land use; specifically, terrain slope, land use type, and vegetation condition were identified as the most determinant factors of erosion intensity. Notably, the negative consequences of land use changes, particularly the degradation of forest cover and its conversion to agricultural or other uses on steep slopes, lead to a significant increase in soil erosion and, consequently, a higher volume of sediment transported downstream. This study affirms the critical importance of principled land use planning, taking into full account the inherent topographical constraints of mountainous regions. The effective implementation of conservation measures, including the preservation and restoration of natural vegetation and the application of sustainable management practices in agricultural lands, is deemed essential for controlling and reducing sediment transport in priority areas. Despite the satisfactory accuracy of the employed models, limitations such as sensitivity to the quality and precision of input data and the lack of comprehensive field data for complete validation still persist. These limitations create valuable opportunities for future research to significantly improve prediction accuracy by integrating extensive field sampling, high-resolution soil data, and the application of more advanced process-based modeling approaches.