Prediksi Data Curah Hujan Dengan Menggunakan Statistika Non Parametrik

Abstract

Rainfall data analysis is the first stage of a water resource planning. One of  rainfall data analysis method is using rain frequency analysis. In this research, rainfall frequency analysis is used to prediction the probability of occurrence from hydrological event. The maximum monthly rainfall frequency distribution is affects to rainfall during high repeat periods. Rainfall is the amount of water that falls on a flat surface during certain repetitive periods. Secondary data is got from Temindung Station of  Samarinda City on 2007 to 2016. The type of distribution are used Normal, Gumbel, Log Pearson Type III, and Log Normal. Compatibility test of Non Parametric Statistics using Chi Square method. The results showed if the estimated rainfall at the highest repeating period of 2, 5, 10, 25, 50, and 100 years is Log Normal distribution. The distribution that requirement of qualify criteria is Log Normal and Gumbel distribution. The distribution that fit from Chi Square test is Gumbel distribution is 3,5177 and Log Normal distribution is 6,8945. From Kolmogorov Smirnov test the value of  Gumbel distribution is 0, and Log Normal distribution is 0,0805. Rainfall patterns for Normal distribution, Gumbel distribution, Pearson Log distribution Type III and Log Normal distribution are horizontal patterns.