Assessment of streamflow in ungauged basin by using physical similarity approach

In the context of hydrological phenomena interpretation and water resource management, estimating runoff in ungauged basins presents a significant challenge. This study implements the physical similarity method, a recognized regionalization approach, supplemented by climatic attribute consideration—an integration uncommon in existing literature—for enhanced catchment classification and donor catchment determination. Using the K-mean clustering algorithm, ten catchments were categorized into three distinct groups. Each group’s performance was evaluated based on the GR4J hydrological model’s effectiveness. Selection of the donor catchment and assessment of model performance hinged on the statistical measures of KGE, R, and Pbias. Parameters derived from the donor catchment were subsequently transferred to the target catchment to enable streamflow assessment and performance evaluation. Results illustrate that catchments with lower mean elevation and larger areas yield more favourable outcomes with the GR4J model. Among the three groups, the PA group, which integrates both physical and climatic attributes, outperformed the others. These findings underscore the importance of comprehensive attribute inclusion for effective catchment classification and accurate runoff estimation in ungauged basins.