Forecasting Seasonal Drought Using Spatio-SPI and Machine Learning Algorithm: The Case of Borana Plateau of Southern Oromia, Ethiopia
Received Date: Mar 07, 2023 / Published Date: Sep 20, 2023
Abstract
Drought is a natural phenomenon that occurs in all parts of the world. Hence, drought monitoring and forecasting have been fundamental issue for developing and implementing a proactive drought mitigation plan. In the process of monitoring and forecasting drought occurrences; the major decisive factors are drought identification/quantification and selection of appropriate forecasting models. This study model seasonal drought forecasting using the Spatio-standardized Precipitation Index (SPI) data in the Borana plateau of Southern Oromia region of Ethiopia with the techniques of machine learning algorithm. Adjusted native resolution based historical rainfall data of 1981 to 2021 of the study area were used from NASA power project climate data repository website in the January 2022. Quantifications of SPI seasonal drought were done using SPEI package with in the RStudio software. The nonlinear outoregressive neural neuron network (NAR network) based Levenberg-Marquardt Back Propagation algorithm (LMBP) was used to model spatio-SPI seasonal drought forecasting of some sites in the study area using MATLAB software. The findings of this study showed SPI 3 months and SPI 6 months ANN based seasonal drought prediction model performance evaluation value of MSE ranged between 0.0022 and 5.5752 which were in the excellent acceptable range of validation. SPI 3 months and SPI 6 months model performance evaluation value of correlation coefficient (R) of all the study sites were above 0.9034 which was also in the excellent range of validation. The study results revealed that ANN modeling could works effectively for forecasting seasonal drought/SPI 3 months and SPI 6 months/ ahead of two months and five months lead times, respectively, in all the districts in the study area. This study identified that, actual/observed and ANN based Ganna and Hagayya seasons SPI 3 months and SPI 6-months prediction value of 1981–2021 discovered that Borana’s zone rainfall seasons on which communities rely for their entire life supporting systems were/ are drought prone seasons.
Keywords: SPI seasonal; RStudio; ANN; Matlab; Drought forecast
Citation: Dinsa AB, Wakjira FS, Demmiese ET, Negash TT (2023) Forecasting Seasonal Drought Using Spatio-SPI and Machine Learning Algorithm: The Case of Borana Plateau of Southern Oromia, Ethiopia. J Earth Sci Clim Change 14: 722.
Copyright: © 2023 Dinsa AB, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution and reproduction in any medium, provided the original author and source are credited.
Share This Article
Recommended Journals
Open Access Journals
Article Usage
- Total views: 672
- [From(publication date): 0-2023 - Nov 23, 2024]
- Breakdown by view type
- HTML page views: 600
- PDF downloads: 72