Author: Eshev, Sobir Samatovich; Egamov, Mirshohid Xolmurodovich

Annotation: This scientific article presents an assessment of the extreme water discharges of the Kashkadarya River using statistical distribution models based on probability theory. The study involved a comparative analysis of Gumbel, Log-normal, Pearson III, and Normal distribution models using the Kolmogorov–Smirnov test and root mean square error (RMSE) as evaluation criteria. The results indicate that the Gumbel distribution showed the lowest RMSE (5.8 m³/s) and the highest p-value (0.45), making it the most statistically appropriate model for extreme event estimation. Specifically, the 100-year return period maximum discharge was estimated at 200 ± 10 m³/s. This statistical approach can serve as an effective predictive tool for assessing flood risks, defining safety parameters, and designing water infrastructure on a scientific basis.

Keywords: extreme water discharge, Gumbel distribution, probability theory, statistical modeling, Kashkadarya River, flood risk.

Pages in journal: 341 - 348