Understanding Drivers of Monthly Rainfall in West Africa Using SHAP and PDP Analyses
- Authors
-
-
R. A. Akum
West African Science Service Centre on Climate Change and Adapted Land Use (WASCAL), Federal University of Technology, Akure, Nigeria. Department of Geography, SDD University of Business and Integrated Development Studies, Wa, Ghana.
Author
-
P. G. Oguntunde
West African Science Service Centre on Climate Change and Adapted Land Use (WASCAL), Federal University of Technology, Akure, Nigeria.
Author
-
R. A. Balogun
West African Science Service Centre on Climate Change and Adapted Land Use (WASCAL), Federal University of Technology, Akure, Nigeria.
Author
-
- Keywords:
- Array, Array, Array, Array, Array, Array
- Abstract
-
Rainfall in West Africa exhibits strong spatial and temporal variability, with significant implications for agriculture, water resources, and socio-economic activities. Improving our understanding of its drivers is critical for enhancing predictive capabilities. This study applies SHapley Additive exPlanation (SHAP) and partial dependence plot (PDP) techniques to quantify and interpret the drivers of monthly rainfall across West Africa using LightGBM models. Utilising monthly ERA5 predictors of 23 years spanning between 2000 and 2022 and GPM-IMERG rainfall data for the same period, the model captures 71% of the monthly rainfall variance. The dominant predictors across the region include convective available potential energy (CAPE), convective inhibition (CIN), vertical integral of eastward cloud liquid water flux (P88.162), mean sea-level pressure (MSL), evaporation (E), and geographic coordinates. In the Guinea Coast, rainfall is most strongly modulated by P88.162, MSL, CIN, and E. In the Savannah and Sahel, CIN and CAPE are the primary drivers. CAPE consistently shows a nonlinear positive association with rainfall, with saturation thresholds near 1000 J/kg across zones. CIN acts as a strong negative constraint, particularly in the drier northern zones. MSL, although positively associated with rainfall on monthly scales, reflects climatological rather than synoptic dynamics, with threshold behaviour near 1012 hPa. Evaporation is inversely related to rainfall, especially in the Guinea Coast, reflecting seasonal shifts in cloud cover and surface radiation. The integration of SHAP and PDP methods provides complementary insights into the magnitude, direction, and nonlinearity of these relationships. Identifying these thresholds and spatial variations enhances understanding of rainfall processes and offers potential improvements for sub-seasonal to seasonal forecasting in West Africa.
- Author Biographies
- References
-
Danso, D. K., Anquetin, S., Diedhiou, A., Kouadio, K., and Kobea, A. T. (2020). Daytime low-level clouds in West Africa–occurrence, associated drivers and shortwave radiation attenuation. Earth System Dynamics Discussions, 2020, p.1-32.
Dogbey, F., Asilevi, P. J., Dzrobi, J. F., Koffi, H. A., and Klutse, N. A. B. (2022). Modelling cloud cover climatology over tropical climates in Ghana. Atmosphere, vol. 13(8), p.1265.
Fischer, E. M., and Knutti, R. (2015). Anthropogenic contribution to global occurrence of heavy-precipitation and high-temperature extremes. Nature climate change, vol. 5(6), p.560-564.
Gao, T., Wang, H. J., and Zhou, T. (2017). Changes of extreme precipitation and nonlinear influence of climate variables over monsoon region in China. Atmospheric Research, vol. 197, p.379-389.
Giorgi, F., Coppola, E., and Raffaele, F. (2018). Threatening levels of cumulative stress due to hydroclimatic extremes in the 21st century. Npj Climate and Atmospheric Science, vol. 1(1), p.18.
Grandey, B. S., Gururaj, A., Stier, P., and Wagner, T. M. (2014). Rainfall‐aerosol relationships explained by wet scavenging and humidity. Geophysical Research Letters, vol. 41(15), p.5678-5684.
Hersbach, H., Bell, B., Berrisford, P., Hirahara, S., Horányi, A., Muñoz‐Sabater, J., ... and Thépaut, J. N. (2020). The ERA5 global reanalysis. Quarterly journal of the royal meteorological society, vol. 146(730), p.1999-2049.
Janicot, S., Mounier, F., Hall, N., Leroux, S., and Sultan, B. (2011). Intraseasonal variability of the West African monsoon. Atmospheric Science Letters, vol. 12(1), p.58–66. https://doi.org/10.1002/asl.280.
Krishnaswamy, J., Vaidyanathan, S., Rajagopalan, B., Bonell, M., Sankaran, M., Bhalla, R. S., and Badiger, S. (2015). Non-stationary and non-linear influence of ENSO and Indian Ocean Dipole on the variability of Indian monsoon rainfall and extreme rain events. Climate Dynamics, vol. 45, p.175-184.
Lavaysse, C., Flamant, C., Janicot, S., Parker, D. J., Lafore, J. P., Sultan, B., and Pelon, J. (2009). Seasonal evolution of the West African heat low: a climatological perspective. Climate Dynamics, vol. 33, p.313-330.
Lin, X., Fan, J., Hou, Z. J., and Wang, J. (2023). Machine learning of key variables impacting extreme precipitation in various regions of the contiguous United States. Journal of Advances in Modeling Earth Systems, vol. 15(3), p.e2022MS003334.
Lundberg, S. M., and Lee, S. I. (2017). A unified approach to interpreting model predictions. Advances in neural information processing systems, p. 30.
Nicholson, S. E. (2013). The West African Sahel: A review of recent studies on the rainfall regime and its interannual variability. International Scholarly Research Notices, vol. 2013(1), p.453521.
Nwachukwu, P. N., Satge, F., Yacoubi, S. E., Pinel, S., and Bonnet, M. P. (2020). From TRMM to GPM: how reliable are satellite-based precipitation data across Nigeria?. Remote Sensing, vol. 12(23), p.3964.
Reichstein, M., Camps-Valls, G., Stevens, B., Jung, M., Denzler, J., Carvalhais, N., and Prabhat, F. (2019). Deep learning and process understanding for data-driven Earth system science. Nature, vol. 566(7743), p.195-204.
Roscher, R., Bohn, B., Duarte, M. F., and Garcke, J. (2020). Explainable machine learning for scientific insights and discoveries. Ieee Access, vol. 8, p.42200-42216.
Saeed, F., Bethke, I., Fischer, E., Legutke, S., Shiogama, H., Stone, D. A., and Schleussner, C. F. (2018). Robust changes in tropical rainy season length at 1.5 C and 2 C. Environmental Research Letters, vol. 13(6), p.064024.
Sylla, M. B., Nikiema, P. M., Gibba, P., Kebe, I., and Klutse, N. A. B. (2016). Climate change over West Africa: Recent trends and future projections. Adaptation to climate change and variability in rural West Africa, p.25-40.
Tuckman, P., Agard, V., and Emanuel, K. (2023). Evolution of convective energy and inhibition before instances of large CAPE. Monthly Weather Review, vol. 151(1), p.321-338.
Zhang, Lily N., Daniel J. Short Gianotti, and Dara Entekhabi. (2023). Land surface influence on convective available potential energy (CAPE) change during interstorms. Journal of Hydrometeorology vol. 24.8 (2023): p.1365-1376.
- Downloads
- Published
- 2025-06-30
- Section
- Articles