Forecasting Of Climate Variables In Mbeya Region

ABSTRACT

Tanzania, like other countries in the world, is affected by climate change. There is enough evidence to substantiate this that comes from the increase of temperatures across the country over the past thirty years. Despite the mitigations approved to attack the problem of climate change in the country, forecasting of climate variables is a prerequisite. Studies about climate change that have been conducted in Tanzania have mainly focused on assessing the weakness and adaption to the effects of climate change in Tanzania while using either temperature or rainfall variables. Further, the studies have depended on the information at the national level rather than the regional level. To fill this gap, this study forecast the climate variables in Mbeya Region. Specifically, the study analyzed the pattern of climate in Mbeya region with the help of the fit climate model using time series techniques and the fitted model. The climate variables included in the study are temperature, rainfall, evaporation, humidity, and wind speed. The study used secondary data from 1981-2019 collected from Tanzania Meteorological Authority (TMA), Mbeya center. The analysis of the data revealed that the patterns of climate variables are seasonal for monthly data while observing the yearly data trend. Either, there was no data to show irregularities and cyclic components. Basing on the theory of Box and Jenkins, SARIMA were fitted and used to forecast rainfall and temperature respectively. The study recommends Box and Jenkins to be used to get the ARIMA model, either seasonally or non-seasonally.