Masilin Gudoshava

East African March–May rainfall (MAM) remains difficult to predict despite its importance for agriculture, water resources, and disaster preparedness. This study identifies pre-season physical drivers of MAM rainfall and tests their value for probabilistic seasonal prediction. Predictor basins were derived from December and January sea surface temperature (SST), 2 m air temperature (T2), and sea-level pressure (SLP) anomalies relative to 1991–2020, using correlations with the leading mode of East African MAM rainfall and subsequent SHAP-based feature selection. The selected basin-derived indices were applied in Random Forest (RF) and Extreme Gradient Boosting (XGB) models. The dominant predictors appear to be the southern Indian Ocean T2 tendency, Australian and Eurasian T2 gradients, South Pacific and Antarctic T2 signals, Atlantic Niño tendency, and the Euro–African SLP gradient. T2-related predictors dominate both the January and December initialisations, showing that near-surface thermal gradients provide useful information in addition to SST memory. Walker-circulation diagnostics show that these drivers influence rainfall through pressure-gradient changes, tropical overturning, and upper-level wave-train development. For January initialisation, RF and XGB achieve spatially averaged Brier Skill Scores of 0.48 and 0.41, respectively, while the corresponding Area Under the Receiver Operating Characteristic Curve values amounting to 0.72 and 0.65. These results demonstrate that physically constrained machine learning provides promising probabilistic skill for East African MAM rainfall prediction.

East African March–May rainfall (MAM) remains difficult to predict despite its importance for agriculture, water resources, and disaster preparedness. This study identifies pre-season physical drivers of MAM rainfall and tests their value for probabilistic seasonal prediction. Predictor basins were derived from December and January sea surface temperature (SST), 2 m air temperature (T2), and sea-level pressure (SLP) anomalies relative to 1991–2020, using correlations with the leading mode of East African MAM rainfall and subsequent SHAP-based feature selection. The selected basin-derived indices were applied in Random Forest (RF) and Extreme Gradient Boosting (XGB) models. The dominant predictors appear to be the southern Indian Ocean T2 tendency, Australian and Eurasian T2 gradients, South Pacific and Antarctic T2 signals, Atlantic Niño tendency, and the Euro–African SLP gradient. T2-related predictors dominate both the January and December initialisations, showing that near-surface thermal gradients provide useful information in addition to SST memory. Walker-circulation diagnostics show that these drivers influence rainfall through pressure-gradient changes, tropical overturning, and upper-level wave-train development. For January initialisation, RF and XGB achieve spatially averaged Brier Skill Scores of 0.48 and 0.41, respectively, while the corresponding Area Under the Receiver Operating Characteristic Curve values amounting to 0.72 and 0.65. These results demonstrate that physically constrained machine learning provides promising probabilistic skill for East African MAM rainfall prediction.

East African March–May rainfall (MAM) remains difficult to predict despite its importance for agriculture, water resources, and disaster preparedness. This study identifies pre-season physical drivers of MAM rainfall and tests their value for probabilistic seasonal prediction. Predictor basins were derived from December and January sea surface temperature (SST), 2 m air temperature (T2), and sea-level pressure (SLP) anomalies relative to 1991–2020, using correlations with the leading mode of East African MAM rainfall and subsequent SHAP-based feature selection. The selected basin-derived indices were applied in Random Forest (RF) and Extreme Gradient Boosting (XGB) models. The dominant predictors appear to be the southern Indian Ocean T2 tendency, Australian and Eurasian T2 gradients, South Pacific and Antarctic T2 signals, Atlantic Niño tendency, and the Euro–African SLP gradient. T2-related predictors dominate both the January and December initialisations, showing that near-surface thermal gradients provide useful information in addition to SST memory. Walker-circulation diagnostics show that these drivers influence rainfall through pressure-gradient changes, tropical overturning, and upper-level wave-train development. For January initialisation, RF and XGB achieve spatially averaged Brier Skill Scores of 0.48 and 0.41, respectively, while the corresponding Area Under the Receiver Operating Characteristic Curve values amounting to 0.72 and 0.65. These results demonstrate that physically constrained machine learning provides promising probabilistic skill for East African MAM rainfall prediction.

East African March–May rainfall (MAM) remains difficult to predict despite its importance for agriculture, water resources, and disaster preparedness. This study identifies pre-season physical drivers of MAM rainfall and tests their value for probabilistic seasonal prediction. Predictor basins were derived from December and January sea surface temperature (SST), 2 m air temperature (T2), and sea-level pressure (SLP) anomalies relative to 1991–2020, using correlations with the leading mode of East African MAM rainfall and subsequent SHAP-based feature selection. The selected basin-derived indices were applied in Random Forest (RF) and Extreme Gradient Boosting (XGB) models. The dominant predictors appear to be the southern Indian Ocean T2 tendency, Australian and Eurasian T2 gradients, South Pacific and Antarctic T2 signals, Atlantic Niño tendency, and the Euro–African SLP gradient. T2-related predictors dominate both the January and December initialisations, showing that near-surface thermal gradients provide useful information in addition to SST memory. Walker-circulation diagnostics show that these drivers influence rainfall through pressure-gradient changes, tropical overturning, and upper-level wave-train development. For January initialisation, RF and XGB achieve spatially averaged Brier Skill Scores of 0.48 and 0.41, respectively, while the corresponding Area Under the Receiver Operating Characteristic Curve values amounting to 0.72 and 0.65. These results demonstrate that physically constrained machine learning provides promising probabilistic skill for East African MAM rainfall prediction.

East African March–May rainfall (MAM) remains difficult to predict despite its importance for agriculture, water resources, and disaster preparedness. This study identifies pre-season physical drivers of MAM rainfall and tests their value for probabilistic seasonal prediction. Predictor basins were derived from December and January sea surface temperature (SST), 2 m air temperature (T2), and sea-level pressure (SLP) anomalies relative to 1991–2020, using correlations with the leading mode of East African MAM rainfall and subsequent SHAP-based feature selection. The selected basin-derived indices were applied in Random Forest (RF) and Extreme Gradient Boosting (XGB) models. The dominant predictors appear to be the southern Indian Ocean T2 tendency, Australian and Eurasian T2 gradients, South Pacific and Antarctic T2 signals, Atlantic Niño tendency, and the Euro–African SLP gradient. T2-related predictors dominate both the January and December initialisations, showing that near-surface thermal gradients provide useful information in addition to SST memory. Walker-circulation diagnostics show that these drivers influence rainfall through pressure-gradient changes, tropical overturning, and upper-level wave-train development. For January initialisation, RF and XGB achieve spatially averaged Brier Skill Scores of 0.48 and 0.41, respectively, while the corresponding Area Under the Receiver Operating Characteristic Curve values amounting to 0.72 and 0.65. These results demonstrate that physically constrained machine learning provides promising probabilistic skill for East African MAM rainfall prediction.

Accurate and reliable rainfall data is crucial for climate-sensitive applications in Kenya. This study evaluates the performance of nine satellite rainfall estimate (SRE) datasets. The study aims to identify the most suitable dataset for various applications in the country. The datasets include Africa Rainfall Climatology (ARC2), NOAA’s Rainfall Estimation Version 2 (RFEv2), Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS), Integrated Multi-satellitE Retrievals for GPM (IMERG-Early), Climate Prediction Center Morphing Technique (CMORPH), Tropical Applications of Meteorology using SATellite and ground-based observations (TAMSAT), Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks (PERSIANN-CDR), Climate Hazards Infrared Precipitation (CHIRP), and Multi-Source Weighted-Ensemble Precipitation (MSWEP). The datasets are assessed based on their spatial and temporal accuracy against ground-based observations in Kenya. Daily rainfall station data is obtained from the Kenya Meteorological Department, which maintains an updated archive of quality-controlled observed rainfall datasets across the country. Metrics are evaluated based on how well satellite estimates capture ground-based observed rainfall. The metrics used include the correlation coefficient (r), root mean square error, mean error, mean absolute error, bias, probability of detection (POD), false alarm ratio (FAR), and the Heidke skill score (HSS). Comparing the different products, RFEv2, CMORPH and ARC2 provided the best overall compromise across the metrics utilized in the study, while CHIRP did not perform very well, highlighting the importance of merging with ground stations.

East African March–May rainfall (MAM) remains difficult to predict despite its importance for agriculture, water resources, and disaster preparedness. This study identifies pre-season physical drivers of MAM rainfall and tests their value for probabilistic seasonal prediction. Predictor basins were derived from December and January sea surface temperature (SST), 2 m air temperature (T2), and sea-level pressure (SLP) anomalies relative to 1991–2020, using correlations with the leading mode of East African MAM rainfall and subsequent SHAP-based feature selection. The selected basin-derived indices were applied in Random Forest (RF) and Extreme Gradient Boosting (XGB) models. The dominant predictors appear to be the southern Indian Ocean T2 tendency, Australian and Eurasian T2 gradients, South Pacific and Antarctic T2 signals, Atlantic Niño tendency, and the Euro–African SLP gradient. T2-related predictors dominate both the January and December initialisations, showing that near-surface thermal gradients provide useful information in addition to SST memory. Walker-circulation diagnostics show that these drivers influence rainfall through pressure-gradient changes, tropical overturning, and upper-level wave-train development. For January initialisation, RF and XGB achieve spatially averaged Brier Skill Scores of 0.48 and 0.41, respectively, while the corresponding Area Under the Receiver Operating Characteristic Curve values amounting to 0.72 and 0.65. These results demonstrate that physically constrained machine learning provides promising probabilistic skill for East African MAM rainfall prediction.

East African March–May rainfall (MAM) remains difficult to predict despite its importance for agriculture, water resources, and disaster preparedness. This study identifies pre-season physical drivers of MAM rainfall and tests their value for probabilistic seasonal prediction. Predictor basins were derived from December and January sea surface temperature (SST), 2 m air temperature (T2), and sea-level pressure (SLP) anomalies relative to 1991–2020, using correlations with the leading mode of East African MAM rainfall and subsequent SHAP-based feature selection. The selected basin-derived indices were applied in Random Forest (RF) and Extreme Gradient Boosting (XGB) models. The dominant predictors appear to be the southern Indian Ocean T2 tendency, Australian and Eurasian T2 gradients, South Pacific and Antarctic T2 signals, Atlantic Niño tendency, and the Euro–African SLP gradient. T2-related predictors dominate both the January and December initialisations, showing that near-surface thermal gradients provide useful information in addition to SST memory. Walker-circulation diagnostics show that these drivers influence rainfall through pressure-gradient changes, tropical overturning, and upper-level wave-train development. For January initialisation, RF and XGB achieve spatially averaged Brier Skill Scores of 0.48 and 0.41, respectively, while the corresponding Area Under the Receiver Operating Characteristic Curve values amounting to 0.72 and 0.65. These results demonstrate that physically constrained machine learning provides promising probabilistic skill for East African MAM rainfall prediction.

Accurate and reliable rainfall data is crucial for climate sensitive applications in Kenya. This study evaluates the performance of nine satellite rainfall estimates (SREs) datasets. The study aims to identify the most suitable dataset for various applications in the country. The datasets include Africa Rainfall Climatology (ARC2), NOAA’s Rainfall Estimation Version 2 (RFEv2), Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS), Integrated Multi-satellitE Retrievals for GPM (IMERG), Climate Prediction Center Morphing Technique (CMORPH), Tropical Applications of Meteorology using SATellite and ground-based observations (TAMSAT), Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks (PERSIANN), Climate Hazards Infrared Precipitation (CHIRP), and Multi-Source Weighted-Ensemble Precipitation (MSWEP). The datasets are assessed based on their spatial and temporal accuracy against ground-based observations in Kenya. The daily rainfall station data is obtained from the Kenya Meteorological Department which maintains an updated archive of quality-controlled observed rainfall datasets across the country. Metrics are evaluated on how well satellite estimates capture ground-based observed rainfall. The metrics used include correlation coefficient (r), root mean square error, mean error, mean absolute error, bias, probability of detection (POD), false alarm ratio (FAR), and Heidke Skill Score (HSS). Results indicate that the RFEv2 is the best-performing satellite rainfall estimate dataset with r=0.58, HSS=0.64 and FAR=0.26. Therefore, it can be applied especially in cases of scarce data. In terms of performance, RFEv2 is followed closely by CMORPH with r=0.63, HSS=0.64 and FAR=0.32.