Downloads: 120 | Views: 188
Case Studies | Civil Engineering | India | Volume 4 Issue 5, May 2015
Probability Distribution Analysis of Rainfall Data for Western Maharashtra Region
Swami Shivprasad T. | Anandrao Deshmukh | Ganesh Patil | Sagar Kahar
Abstract: In India rainfall is the major source for the most sought after commodity of water all over the world. India experiences a monsoon season of 4 months making it more precious. Every attempt is therefore made to study and analyse rainfall data in order to understand its distribution, quantity and frequency. The present work is such an attempt where in various probability distributions are be applied to estimate one month and two to seven consecutive cumulative monthly rainfall. As many as 16 different types of continuous probability distributions were tested using goodness of fit tests such as Kolmogorov Smirnov, Anderson Darling and Chi-Square for Western Maharashtra-s districts, India to check which distribution is best. The following analysis of rainfall data is prepared with the help of EASY FIT and MICROSOFT EXCEL software-s. Monthly rainfall data for 35 years (1966-2000) measured by Indian Meteorological Department, Pune was used. It was found that, in Ahmednagar and Solapur district Beta distribution is the best fitted distribution for one month rainfall dataset, Dagum distributions for consecutive cumulative two months rainfall dataset, Dagum and Gen. Extreme distribution for consecutive cumulative three months rainfall dataset, Log Logistic and Gen. Extreme distribution for consecutive cumulative four and five months rainfall dataset, Log Logistic and Nakagami distributions for consecutive cumulative six months rainfall dataset and Log Logistic and Johnson SU distribution for consecutive cumulative seven months rainfall, as per above the best fitted distributions for other districts of western Maharashtra are as follows.
Keywords: Rainfall, Probability distribution, goodness-of-fit tests
Edition: Volume 4 Issue 5, May 2015,
Pages: 1375 - 1379