Univariate Time Series Analysis of Rainfall in Somali Region, Ethiopia
No Thumbnail Available
Date
2020-06-08
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Addis Ababa University
Abstract
Rainfall is one of the most significant sources of water on earth, supporting the
existence of the majority of living things. It is such a critical variable, particularly
those whose livelihoods depend on rain-fed agriculture. Statistical approaches such as
time series analysis give a way for analyzing and forecasting the patterns of rainfall. In
this study, we have attempted to analyze the pattern of rainfall in the Somali region.
For this purpose, a monthly rainfall data of 26 meteorological stations covering the
region for 34 years of gridded 0.4o *0.4o resolution have been used. A univariate
Box-Jenkins approach was used to build a seasonal ARIMA model to determine the
rainfall pattern of the region. Descriptive results show that, during the years in this
study, the mean annual and monthly rainfall series in the region are 567mm and
34.9mm respectively, with a higher year-to-year variation (CV > 0.30). It also shows
that the region has two rainy periods April-May and August-October, followed by
a prolonged dry season. Using Box-Jenkins approach we have obtained a seasonal
ARIMA (1,0,1)(2,1,2)12 model and then used to establish an 80% and 95% prediction
interval to forecast a 5 year monthly rainfall values. The forecasting results revealed
a very slight decrease of monthly rainfall over the forecast period from January 2017
to December 2021.
Description
Keywords
Univariate, Time Series Analysis, Rainfall, Somali Region, Ethiopia