Comparative analysis of the effects of climate change and land use on runoff and its prediction in a mountainous watershed in Northwestern Iran

Sustainable water resource management in semi-arid mountainous watersheds is increasingly critical due to the combined impacts of climate change and land use dynamics on hydrological processes. The Zolachai watershed, a vital sub-basin of the Lake Urmia watershed in northwestern Iran, spans 2,258 km² and serves as a primary water source for agricultural, urban, and ecological needs. Rapid population growth and unsustainable land use practices, particularly agricultural expansion on steep slopes, have intensified environmental pressures, resulting in forest degradation, soil erosion, reduced agricultural productivity, and compromised drinking water quality due to elevated surface runoff. This study conducts a comparative analysis of the effects of climate change and land use changes on runoff generation and water retention in the Zolachai watershed, employing the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) model to provide actionable insights for sustainable watershed management.



The study integrated multiple datasets to assess hydrological processes, including climate projections, land use data, soil hydrological properties, and runoff curve numbers (CN). Climate data were sourced from the ACCESS-CM2 model under the Coupled Model Intercomparison Project Phase 6 (CMIP6), utilizing two Shared Socioeconomic Pathways (SSP) scenarios: SSP2-4.5 (moderate emissions) and SSP5-8.5 (high emissions). Precipitation data from 1966 to 2021, collected from rain gauge stations within the watershed and the Climate4impact database, were used to project precipitation patterns for 2023 and 2030. The SSP5-8.5 scenario, selected to estimate maximum runoff potential, projected an increase in heavy precipitation days by 2030, heightening flood risks, particularly in areas with low water retention capacity. This scenario underscores the potential for intensified rainfall events, which could exacerbate runoff, soil erosion, and sedimentation in downstream water bodies, threatening water security.



Land use and land cover (LULC) data were derived from Sentinel-2 satellite imagery for 2016, 2020, and 2023, with a 10-meter spatial resolution. Images captured during the late June growing season, ensuring minimal cloud cover, were processed using object-based classification and the Support Vector Machine (SVM) algorithm in eCognition Developer 10.3 software. Seven land use categories were identified: water bodies, bare soil (areas with less than 5% vegetation cover), irrigated agriculture and orchards, rainfed agriculture, salt flats around Lake Urmia, rangelands, and residential areas. The classification for 2023 was validated against a predicted map, achieving high accuracy indices (Kno: 0.87, Klocation: 0.93, Klocationstrata: 0.93, Kstandard: 0.84). Land use change predictions for 2030 were generated using the Markov and CA-Markov models, with a Kappa coefficient of 0.91, indicating robust predictive accuracy. Results revealed a significant expansion of irrigated agriculture and orchards (from 221.11 km² in 2016 to 528.18 km² in 2030) and residential areas (from 9.08 km² to 24.39 km²), alongside declines in rangelands (from 857.95 km² to 724.52 km²), bare soil, and water bodies. These shifts reflect increased water consumption in upstream agricultural areas and reduced surface flows downstream, exacerbating water scarcity.



Soil hydrological groups (A, B, C) were determined using field sampling data from the Soil and Water Research Institute for plain areas, supplemented by SoilGrids.org data for mountainous regions. Clay, silt, and sand percentages were integrated using SAGA GIS and ArcGIS software, based on the USDA soil texture triangle. The central watershed was dominated by groups A and B, while group C prevailed in the northwestern, southern, and western parts. Curve Number (CN) values, ranging from 36 to 93, were calculated using the USDA Soil Conservation Service method, incorporating land use and soil hydrological data. Residential areas, characterized by impervious surfaces and CN values of 70–90, exhibited the highest runoff potential (359.4–647 mm), while irrigated agriculture and orchards, with CN values of 36–80, showed the lowest runoff (137.2–333.8 mm) due to high soil permeability and vegetation cover. These CN estimates were validated against studies like Haghdadi et al. (2018) and Birhanu et al. (2019), confirming their reliability.



The InVEST Water Yield model integrated these datasets to produce spatially explicit runoff and water retention maps for 2016, 2023, and 2030. Widely recognized for its simplicity and efficiency, the model has been applied in studies such as Bai et al. (2013) and Reheman et al. (2023). In the Zolachai watershed, runoff volumes ranged from 137.2–359.4 mm in 2016, 179.5–418.8 mm in 2023, and 333.8–647 mm in 2030. Residential areas and hydrological soil group C exhibited the highest runoff potential, while irrigated agriculture and orchards demonstrated superior water retention due to high infiltration rates. The central and southern watershed areas showed moderate to high runoff potential, while the northeastern parts had lower potential. Projections for 2030 indicate a decline in runoff volume (from 61.91 million m³ in 2023 to 53.59 million m³), driven by increased water retention in expanding agricultural areas.



The expansion of irrigated agriculture and orchards, driven by economic incentives and modern irrigation techniques, increases water consumption, reducing surface flows and exacerbating downstream water scarcity. Climate change, under SSP5-8.5, intensifies rainfall, amplifying runoff and flood risks in residential and rangeland areas. These findings align with global studies, such as Reheman et al. (2023) in the Tian Shan Mountains, and regional research by Emlaei et al. (2021) and Azizi et al. (2022). The InVEST model’s ability to evaluate management scenarios supports prioritizing conservation in high-runoff areas and promoting sustainable practices like afforestation and drip irrigation. This study provides a robust framework for sustainable management of semi-arid mountainous watersheds, contributing to water security, improved livelihoods, and biodiversity preservation. Future research should focus on comprehensive ecosystem service assessments to refine runoff regulation strategies and address the complex interplay of climatic and anthropogenic factors.