REGIONALIZATION OF FLOW DURATION CURVES OF MESO-SCALE CATCHMENTS

Abstract

The demand for water supply is increasing day-by-day due to increasing population. So to meet the needs of water for different purposes like domestic water supply, irrigation, industrial use, hydropower and environmental flow requirements, ideally all the catchments should be gauged. Because all the catchments are not gauged, methods need to be evolved to estimate flows of these ungauged catchments. Apart from the problems of ungauged catchments, sometimes there is a discontinuity in flow measurements of a gauged catchment due to reasons such as washing away of observation sites during floods or discontinuity of flow measurements at the site due to lack of funding etc. Development of flow duration curves (FDCs) at a river site is important for water resources planning and management of any catchment. FDCs of different gauging stations of a catchment will be different depending on the hydrologic characteristics of flow at that site. Apart from the necessity of developing FDCs for ungauged catchments, there is also a necessity to estimate flows at a gauged site, but for the ungauged period. The flow duration curves are changing year to year significantly and it is very important to determine the flow duration curves for the incomplete series, i.e., for the ungauged period. The objective of this study is to develop s procedure for developing FDC for the ungauged period of a gauged site. For this purpose the study is performed for different sub-catchments of Baden-Württemberg and Rhineland-Palatinate in Germany. The rank analysis method is used to develop FDC and this approach is used to develop methodology for the regionalization of FDCs. The study uses specific discharge (cumecs/sq. km) as the variable instead of flow. In this method, FDCs are determined for target years of the target catchments using FDCs of the target years of the reference catchment. The methodology is compared with the linear regression approach to regionalize the FDCs for the study area and it is found that methodology is performing well as compared with the linear regression approach. The linear regression equation is fitted between the specific discharge series of the reference years of both the reference and target catchments to predict the specific discharge series of target years of the target catchment. In Upper Neckar catchment, the linear regression approach is performing well except for low flow conditions. But when the target catchments are in Rhineland-Palatinate, the proposed methodology is performing well in reproducing the observed FDCs in comparison with the linear regression approach. It is inferred from the study that overall the proposed rank analysis approach of developing FDCs is performing better than that of the linear regression approach.