干旱区生态环境知识资源中心

Ideally, flood estimation in hydrological studies and applications requires recorded data, but this is missing or insufficient in many cases for a variety of reasons. Approaches to modelling the hydrologic response of ungauged basins may use physical characteristics of watersheds either to directly infer values for the parameters of hydrologic models or to establish regression relationships between watershed structure and model parameters. Most studies, however, are concentrated in humid hydroclimatic zones. In this study, a regional frequency analysis of annual peak maximum series of flood flows from rivers of the arid southern Alborz basins, Iran, was conducted, including identification of suitable regional flood frequency distributions. The peak stream flow for different data periods was estimated based on physiographic as well as climatic parameters. Because of the close relationship between elevation, stream flow regime and hydrological features, the development of a water response model first involved delimiting homogenous hydrological regions by using two-step cluster analysis. Equations were established for each region relating peak flow of different return periods and the basin parameters. The study results also revealed that the regional regression models developed in this study could be applied reasonably well at relatively similar ungauged sites, which is of potential interest in other water-stressed areas of the world, particularly given expected scenarios of population growth, development and climatic change.