ABSTRACT
The indirect assessment of daily streamflow time series data at some ungauged station requires the selection of a reference stream gauge. Therefore, the accuracy of prediction of daily streamflow at such ungauged stations directly depends upon the degree of correlation of the reference stream gauge with it. Most conventional methods used to determine these ungauged parameters assume the nearest stream gauge station as the reference. However, the nearest catchment may not always be the most correlated one. In the present article, the Map Correlation Method is used to identify the most correlated stream gauge station as the reference of an ungauged station. Another advantage of this method is that it can estimate cross-correlation between the data of a gauged and ungauged site for any location in the study area and need the bare minimum of stream gauges in the network and doesn’t assume the study area is limited by regions of homogeneity or other hydrology boundaries. The efficiency of this method is demonstrated through a case study on the Krishna-Godavari river basin in southern India to predict the catchment parameters for different ungauged sites. Ten years of daily data for 17 stream gauge stations have been considered from both river basins. When this method is applied to the estimation of streamflow in all stream gauges, improved estimation of daily streamflow time series is obtained over the streamflow estimated from the nearest stream gauge approach. It is observed that the suggested techniques could be able to identify the most correlated stream gauge in eight out of the seventeen catchments, and in all other cases, the difference is less.
