Performance of Machine Learning Techniques for Meteorological Drought Forecasting in the Wadi Mina Basin, Algeria

Water resources, land and soil degradation, desertification, agricultural productivity, food security are all adversely influenced by drought. The prediction of meteorological droughts using the standardized precipitation index (SPI) is crucial for water resource management. modeling results SPI at 3, 6, 9, 12 months based on five types machine learning: support vector (SVM), additive regression, bagging, random subspace, forest. After training, testing, cross-validation folds sub-basin 1, concluded that SVM most effective model predicting different (3, 12). Then, SVM, as best model, was applied 2 timescales it achieved satisfactory outcomes. Its performance validated were achieved. suggested performed better than other models estimating drought sub-basins during testing phase. could be used to predict several timescales, choose remedial measures research basin, assist in management sustainable resources.