Statistical analysis on variability of Annual Daily Maximum Rainfall of South Australia

Abstract

Rainfall variability assessment is crucial for effective planning of water resource management in areas with diverse climatic conditions like Australia. The objective of this study is to examine rainfall variability patterns in South Australia (SA) by analyzing Annual Daily Maximum Rainfall (ADMR) and Annual Rainfall (AR) data. Four weather stations were selected: two from the eastern and two from the southern part of SA to consider spatial variation in rainfall distribution. 30 years of rainfall data were acquired from the website of the Australian Government, Bureau of Meteorology. Statistical Package for the Social Sciences (SPSS) software was used to perform Mann-Whitney and Kruskal-Wallis tests because the data were not normal. These tests help to find the spatial variation of rainfall. Kruskal-Wallis tests is suitable for three or more comparison groups whereas Mann-Whitney is suitable for two groups. Besides, the correlation and regression analysis were conducted to establish the relationship between ADMR and AR. It was found that there are similar distributions on ADMR, and variation on AR data of the study area. The annual planning for water resources management may be different in these areas. ADMR can be used to estimate the AR or vice versa as there is a positive and moderate correlation (AR=266.236 + 2.716ADMR). The statistical relationship established between daily maximum and total yearly rainfall can be utilized for better planning of water resource management in SA, which considers regional variation.