Modeling Flood-Prone Areas Using Statistical and GIS Approaches: Insights from the Yertis River Basin, Kazakhstan

Floods remain one of the most frequent and destructive natural phenomena, the scale and consequences of which are exacerbated by climate variability and anthropogenic pressure on river systems. The Yertis water basin (Kazakhstan) is an area with high exposure to flood risks, where the dense concentration of settlements and infrastructure is within floodplain areas. This study applies an integrated approach based on the integration of statistical methods for hydrological analysis and GIS-based spatial modeling to assess and delimit potential flood zones. Long-term series of maximum water levels and discharges from hydrological stations for the period 1974–2025 were analyzed using probability distribution functions, including the log-normal, Pearson Type III, and Gumbel distributions. The optimal distribution model for each station was selected based on the Kolmogorov–Smirnov goodness-of-fit test and the Akaike information criterion. Exceedance-probability curves for extreme hydrological events were constructed for 0.1%, 1%, and 10% probabilities. Spatial flood modeling was performed in the ArcGIS 10.8 environment using a hydrologically corrected digital elevation model and interpolated flood levels. The resulting flood zone maps allow for the identification of the highest-risk areas and serve as a tool for scientifically based planning of emergency prevention measures and floodplain area management. The study contributes to the methodological development of probabilistic floodplain mapping through the integration of statistical frequency analysis and GIS technologies and demonstrates the applicability of this approach for flood hazard assessment in large transboundary river systems under conditions of climatic and hydrological variability.