Extended Abstract
Introduction:Soil loss and sediment yield are among the main challenges in watershed management, especially in mountainous areas, which, in addition to reducing soil fertility and ecosystem functioning, also threaten water quality and the sustainability of hydrological services. In this context, watershed hydrological services, such as sediment retention, are of high importance because they control soil erosion and reduce downstream sediment transport, thereby improving water quality. The topographic, climatic, and human conditions of mountainous regions in northern Iran, especially the Hyrcanian forests, make these areas particularly susceptible to soil degradation, increasing the need for monitoring ecosystem services and analyzing the spatiotemporal dynamics of sediment retention to assess changes in ecosystem functioning and the effectiveness of management measures. In this regard, the InVEST model, with a simple structure, limited input requirements, and high interpretability, is an effective tool for analyzing these services at the watershed scale. This model, using sediment delivery ratio equations and spatial data, allows examination of the spatiotemporal changes in soil erosion, sediment yield, and sediment retention. Given the need for comprehensive studies in Iran, especially in the Hyrcanian forests, the present study aims to investigate the spatiotemporal changes of soil loss, sediment retention, and sediment yield over a 25-year period in the Talar mountainous watershed. Additionally, sensitivity analysis of model input factors was conducted to identify the most influential variables affecting sediment delivery ratio results.
Materials and methods: The study area is the Talar watershed (Mazandaran Province), spanning over 1,700 km² on the northern slopes of the Alborz mountain range, with an elevation range from 216 to 3,983 meters. The SDR model from the InVEST toolkit was applied to estimate annual soil loss, sediment retention, and sediment yield. Average soil loss was calculated using the InVEST soil erosion equation based on key factors including rainfall erosivity, soil erodibility, vegetation cover, land management, and topography across three study periods. Additional parameters such as the sediment connectivity index, K factor, and maximum sediment delivery ratio were defined based on regional characteristics and expert knowledge. A sensitivity analysis of the model input factors was also conducted to identify the most influential variables affecting the sediment delivery ratio outcomes.
Results and Discussion: According to the results, over the 25-year study period, soil loss decreased from 896,535 to 758,981 ton, sediment retention from 1.297 to 0.757 million  ton, and sediment yield from 113,643 to 57,426  ton. The main cause of this reduction was a 28.8 % decrease in rainfall erosivity during the study period. The analysis showed that pasture land had the highest sediment retention capacity (415,978  ton), whereas orchards had the lowest (3,676 t). Meanwhile, the highest sediment yield occurred in rainfed agriculture and pasture, while forests and orchards had the lowest amounts. Sensitivity analysis indicated that rainfall erosivity, with a relative sensitivity coefficient of 0.99, was the most important factor affecting soil loss, sediment retention, and sediment yield, explaining more than 47 % of model output variations. The K parameter also influenced sediment retention and yield, though it had no direct effect on soil loss. Overall, the InVEST model is more sensitive to natural parameters such as rainfall than to management variables, highlighting the necessity of integrating climatic data with human interventions in management.
Conclusion: The findings indicate that the InVEST model, with high spatiotemporal analysis capability, is a suitable tool for assessing watershed hydrological services, especially in mountainous regions. The observed reduction in soil loss, sediment retention, and sediment yield over the study period was mainly due to decreased rainfall erosivity and climatic changes. Dense vegetation cover, such as forests and pastures, played a key role in sediment retention and soil erosion reduction, while human land uses such as agriculture contributed more to sediment yield. Sensitivity analysis emphasized the importance of natural factors, particularly rainfall, in watershed management. The findings can be applied to identify critical areas for conservation, prioritize sensitive land uses, design monitoring systems, and develop future management scenarios. It is recommended that future studies utilize long-term data, complementary models, and scenario-based analyses under climate change to enhance the accuracy of assessments and the applicability of results in ecosystem management decisions.