Establishment and maintenance of a hydrometric network in any geographical region is required for planning, design and management of water resources projects. Setting up and maintaining a hydrometric network is an evolutionary process, wherein a network is established early in the development of the geographical area; and the network reviewed and upgraded periodically to arrive at the optimum network. This paper presents the methodology for optimization of a hydrometric network using spatial regression in generalized least square framework. The procedures, involving computation of redundancy information using cross-correlation matrix and identification of most influential station using Cook?s D statistics, is applied to the upper Bhima basin upto Ujjani reservoir for illustrative purposes; and results presented. The derived optimum hydrometric network for the basin is evaluated based on WMO guidelines for minimum density of hydrometric network.