Spatiotemporal variations analysis of water yield ecosystem service in the Hyrcanian region of northern Iran using the InVEST model

Extended Abstract

Introduction

Ecosystems provide a wide range of benefits known as ecosystem services, which play a vital role in human livelihoods and environmental sustainability. Among these, hydrological services, particularly water yield, are essential in maintaining water security, supporting downstream ecosystems, and regulating hydrological processes. Understanding temporal and spatial variations in water yield is increasingly important in the context of land use change and climate variability. In this regard, modeling tools such as the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) have proven effective in quantifying and mapping ecosystem services, especially water yield, by integrating spatial data on land use, climate, topography, and soil properties. The rationale behind this study stems from the limited research conducted on the long-term and spatial dynamics of water yield ecosystem services in Iran, particularly in ecologically sensitive and hydrologically significant regions such as the Hyrcanian forests. Despite their rich biodiversity and critical ecological functions, these regions have received insufficient attention in ecosystem service assessments using robust modeling frameworks. Therefore, this research aims to evaluate the spatiotemporal variations in water yield services in the Talar Watershed, a representative region within the Hyrcanian forest, over a 25-year period using the InVEST model.



Materials and Methods

The Talar Watershed, located in Mazandaran Province in northern Iran, spans approximately 1,764 km² across the northern slopes of the Alborz Mountains. The elevation in the watershed ranges from 216 to nearly 3,980 meters, contributing to diverse microclimates and land uses. The region has a semi-humid, Mediterranean-like climate influenced by the Caspian Sea, with an average annual precipitation of 547 mm and an average annual evapotranspiration of 446 mm. The InVEST water yield model was employed to estimate annual water production based on biophysical and climatic variables for the years 1989, 2000, and 2014. The input data incuding climatic variables (precipitation, reference evapotranspiration), land use/land cover, digital elevation model (DEM), plant-available water content, root-restricting layer depth, biophysical factors (land cover condition, root depth, and evapotranspiration coefficient), seasonality parameter, watershed boundaries, and water demand for each land use class were prepared for the model in the selected years 1989, 2000, and 2014. In this regards, Annual average precipitation across the study area was estimated based on elevation gradients for the selected years. Reference evapotranspiration was calculated using the modified Hargreaves equation for studied years. Land use/land cover data were derived from Landsat satellite imagery using a supervised classification approach based on the Support Vector Machine (SVM) method in the research years. Plant-available water content was determined using soil texture characteristics and calculated as the volumetric difference between field capacity and permanent wilting point. The depth of the root-restricting layer was assumed equal to soil depth in each land unit, as no significant root-limiting layers were present. Root depth was assigned based on dominant vegetation types in each area. Land cover status was defined as either covered (1) or not covered (0), with all land use types except urban areas classified as having vegetative cover. The evapotranspiration coefficient, used to adjust reference evapotranspiration based on alfalfa as the reference crop in the InVEST model for different land use classes. The seasonality parameter, reflecting the predominantly winter rainfall pattern of the study area's climate, was set to 10.



Results and Discussion

Results indicate that the northern and central parts of the Talār watershed, dominated by dense forests and rangelands, produced the highest water volumes, supported by higher annual precipitation. Findings shows a 23% reduction in water provisioning service from 192.02 to 147.85 million m³ between 1989 and 2014. Validation with hydrometric data indicated a decreasing trend in the ratio of precipitation converted to water supply, likely caused by increased evapotranspiration and land use changes. Among land uses, rangelands produced the highest average annual water, while orchards produced the least. Per hectare, urban areas had the highest water production due to impervious surfaces increasing runoff, and forests had the lowest due to higher infiltration and evapotranspiration. The large extent of rangeland areas and their location in steep, slope regions, leading to reduced infiltration and increased runoff, can be considered the main reasons for the highest water yield observed in this land use type. Overall, protecting natural vegetation, especially in sloped, high-precipitation areas, is vital to maintaining watershed water production. Ecological land use planning is essential for sustainable water and soil resource management in the region.



Conclusions

From a policy and planning perspective, this research underscores the utility of the InVEST model as a decision-support tool for watershed managers and land use planners. The ability to quantify and map water yield variations across time provides valuable insights for identifying priority areas for conservation, designing payment for ecosystem services schemes, and implementing adaptive land management strategies. The model’s outputs can also be integrated into regional climate adaptation frameworks, particularly in semi-humid mountainous regions vulnerable to rainfall variability and water stress. Furthermore, the spatially explicit results facilitate cross-sectoral coordination among forestry, agriculture, and urban planning agencies by identifying synergies and tradeoffs in ecosystem service provision. Finally, this study contributes to the growing body of ecosystem service research in the Middle East and offers a replicable methodology for analyzing other hydrologically sensitive regions under environmental pressure. In conclusion, this research provides a comprehensive and long-term assessment of water yield ecosystem service dynamics in a critical ecological zone of Iran. The combination of empirical data, spatial analysis, and process-based modeling offers a robust foundation for evidence-based decision-making. As land use change and climate variability continue to reshape hydrological processes, integrating ecosystem service assessments into regional planning will be essential for achieving sustainable water resource management and ecological resilience in the Hyrcanian region and beyond.