Assessing the Effects of Climate Change and Land-Use Changes on Extreme Discharge in the Western Watershed of Lake Urmia, Iran

Abstract

This study investigates the impacts of climate change and land-use changes on peak discharge and runoff behavior in the western watersheds of Lake Urmia, Iran. Employing machine learning algorithms (e.g., SVM), stochastic models (e.g., CA-MARKOV), ERA5 reanalysis climate data, and the large-scale hydrological VIC model, we assessed these effects across multiple sub-basins. Our analysis revealed that a 2°C rise in average minimum winter temperature over the past 50 years has reduced frost days by nearly 80 days, diminished mountain snow volume, and shifted precipitation from snow to a mix of snow, rain, and snowmelt. Consequently, peak discharge timing has advanced from May to March and April, amplifying flood intensity and frequency, with some sub-basins showing up to 30% higher peak flows. Furthermore, land-use change assessments indicated that expanding orchards and irrigated and rain-fed agricultural lands will significantly elevate future peak discharge, with differences exceeding 10 m³/s in sub-basins like Nazlochay and Barandozchay, driven by reduced infiltration and increased runoff rates. Modeling with these tools confirmed that climate and land-use changes synergistically alter flood dynamics, a pattern consistent with regional studies. These findings underscore the urgent need to integrate these factors into flood management strategies for this flood-prone region, offering a robust framework for sub-basin-scale hydrological planning.