Evaluation and Selection of Rain Gauge Network using Entropy

Establishment and maintenance of a rain gauge network (RGN) in any geographical region is required for planning, design and management of water resources projects so as to reduce chances of project failure and to minimize the economic risk. Setting up and maintaining a RGN is an evolutionary process, wherein a network is established early in the development of the geographical area; and the network is reviewed and upgraded periodically to arrive at the optimum network. The study shows the computation of transinformation index from marginal and conditional entropy values adopting normal distribution, which is used for evaluation of RGN with 25 rain gauge stations to derive the optimum network. Kolmogorov–Smirnov test is used for checking the adequacy of fitting of normal distribution to the recorded rainfall data. The paper presents that the derived network consisting of nineteen rain gauge stations with network density of 774 km2 per gauging station is considered as the optimum RGN for upper Bhima basin, which satisfied the WMO recommended value of 600–900 km2 per station.