Frequency Analysis of Rainfall Data of Dharamshala Region

The present study evaluates the rainfall magnitude for different return periods and also to ascertain the type of probability distribution that best fits the rainfall data of Dharamshala (H. P. ), India. The study uses the 20 years of annual rainfall data that are useful for the prediction of annual one day maximum rainfall and two to seven days consecutive days maximum rainfall corresponding to return period varying from 2 to 20 years are to be used for the economic planning, by design engineers and by hydrologists, design of small and medium hydrologic structures and determine the drainage coefficient for agricultural fields. Most probability distributions (viz: Normal, Log Normal and Gamma distribution) and transformations are applied to estimate one day and two to seven consecutive days annual maximum rainfall of various return periods in the Dharamshala (H. P. ) region. The mean value of one-day annual maximum rainfall at Dharamshala is found to be 142.9 mm with standard deviation and coefficient of variation of 54.8 and 51.34 respectively. The coefficient of skewness is 1.1. For 2 to 7 days consecutive annual maximum rainfall range values for mean, standard deviation, coefficient of variation and coefficient of skewness are 201 393.4 mm, 70.17 146.5, 41.65 30.47 and 0.726 1.593. It is observed that all distribution fitted function significantly. A maximum of 191.1 mm in 1 day, 262.7 mm in 2 days, 314.27 mm in 3 days, 370.24 mm in 4 days, 419.7 mm in 5 days, 477.4 mm in 6 days and 522.32 mm in 7 days is expected to occur at Dharamshala, Himachal Pradesh every 2 years. For a recurrence interval of 20 years, the maximum rainfall expected in 1 day, 2, 3, 4, 5, 6 and 7 days is 277.7 mm, 373.6 mm, 445.11 mm, 518.62 mm, 589 mm, 680.3 mm and 753.79 mm respectively. Various probability distributions and transformations are applied to estimate one day and two to seven consecutive days annual maximum rainfall of various return periods. Three commonly used probability distributions (viz: Normal, Log Normal and Gamma distribution) are tested by comparing the Chi-square, Kolmogorov-Smirnov and Anderson Darling values.