无测站流域径流预测区域化方法研究进展

Review of regionalization methods on streamflow prediction in ungauged basins

无测站流域径流预测是地表水文研究的难题与挑战,而区域化方法,即将水文信息从有测站流域移植至无测站或缺测站流域估算径流量的过程,则为该问题提供了有效的解决办法。由于不同流域地文属性及气候特征各不相同,目前并无通用的方法,回归法、空间近似法及物理相似法是径流预测区域化研究中最常用的三种方法。基于此,本文阐明了三种方法的基本理论,综述了其研究进展、适用性及局限性,分析了区域化研究中常用水文模型及流域特征因子遴选的依据,总结了误差检验及不确定性分析的方法,并展望了径流预测区域化研究的未来发展方向。本文可为无测站流域区域化方法的选择提供基本依据与科学参考。

Continuous streamflow prediction in ungauged basins is a difficult issue and challenge in surface hydrology.Regionalization,defined as the process allowing transfer of hydrological information from gauged to ungauged locations,offers the effective solution for this issue.Regionalization methods based on regression,spatial proximity and physical similarity are the most common methods for continuous streamflow prediction.However,there is no universal method so far because the diversity in catchment physiographic attributes and climate variability produces different performances in various basins.This paper presents the basic theory,limitations and advances of the three regionalization methods in last three decades.Based on the reviewed regionalization studies,the most frequently used parameters of hydrological model and the basin attributes are concluded.Moreover,the error indices and uncertainty analysis methods are summed up,as well as the prospects.This paper can offer the scientific reference for the selection of regionalization methods in the ungauged basin.