Probability Analysis of Rainfall of Bankura for Design of Soil and Water Conservation Structures

Abstract

Daily rainfall data of 21 years (from 1983 to 2003) for Bankura district of West Bengal were analyzed to ascertain their fit to several probability distributions. The series generated by using Weibull's plotting position formula was treated as the Observed series. The expected values were estimated by Gumbel, Log Pearson Type -III and Log Normal probability distribution functions. The goodness of fit was determined by X2 test. Linear probability model was also developed using the parameter Frequency Factor for the best distribution. The analysis indicated that Log Pearson Type -III gave the closest fit to the observed data and it is recommended to forecast the annual maximum I-day rainfall for the design return period. The theoretical annual maximum one day rainfall values using Log Pearson Type -III for 5, 10, 15,25 and 50 years return period were 116, 155, 165, 185 and 207 mm, respectively. These expected I-day maximum rainfall values for the desired return period could be used for hydrological design of different soil and water conservation structures in watershed development projects being implemented under different programme in Bankura district. In this paper, the usefulness of predicted annual maximum I-day rainfall for the desired return period has been verified in the design of a masonry check dam in Kashibedia micro- watershed, which is located at Bankura I block of Bankura district.