Abstract
Populations living in arid and semi-arid regions characterized by highly variable precipitation regimes are frequently exposed to both droughts and flood events, which directly affect water resource availability. Surface runoff harvesting represents a long-established water management strategy used under water-scarcity conditions to meet increasing water demand, mitigate climate change impacts, and reduce land degradation and desertification processes. This study integrates remote sensing data and geographic information systems within an analytical hierarchy process framework to develop a cost-effective and reliable approach for identifying suitable locations for farm reservoirs in the steppe zone of Northern Kazakhstan. The analysis is based on six key thematic spatial layers, including hydrogeology, terrain slope, drainage density, land use and land cover, soil characteristics, and snow water equivalent. These factors play a critical role in determining water availability, runoff generation, infiltration processes, and the accumulation of snowmelt and floodwaters. The results indicate that the most suitable sites are typically associated with moderate drainage density, gentle slopes, high snow water equivalent values, and areas dominated by flood-prone vegetation. Furthermore, harvesting snowmelt runoff from approximately 30 % of the study area could potentially enable deficit irrigation across nearly one-quarter of the agricultural land within the region. This approach may contribute to reducing flood risk, stabilizing farm income during dry years, and supporting the cultivation of higher-value crops. Overall, the findings highlight the strong potential of farm reservoirs for snowmelt water storage as an effective adaptation measure to mitigate agricultural drought and seasonal spring flooding.