A Hybrid Machine Learning Model for Rainfall Prediction in Tropical Regions
DOI:
https://doi.org/10.64479/iarci.v2i1.75Keywords:
Rainfall prediction, hybrid machine learning, LSTM, gradient boosting, tropical climate, meteorological forecastingAbstract
Rainfall prediction in tropical regions remains a significant challenge due to high atmospheric instability, convective precipitation patterns, and limited historical data quality. This study develops and validates a hybrid machine learning model that combines Long Short Term Memory networks with Gradient Boosting Decision Trees for daily rainfall prediction across three tropical locations in Indonesia. The research utilized ten years of meteorological data from 2014 to 2023, including temperature, humidity, wind speed, cloud cover, and sea surface temperature anomalies. The proposed hybrid architecture processes temporal dependencies through LSTM layers while capturing nonlinear feature interactions through gradient boosting mechanisms. Experimental results demonstrate that the hybrid model achieves a root mean square error of 2.34 mm and an accuracy of 89.7 percent for binary rainfall events, outperforming standalone LSTM, Random Forest, and Support Vector Regression models by margins of 12 to 18 percent. The novelty lies in the adaptive weighting mechanism that dynamically adjusts contributions from both components based on seasonal patterns. Key contributions include a robust preprocessing framework for tropical climate data, a computationally efficient hybrid architecture suitable for operational forecasting, and empirical evidence regarding feature importance variations across wet and dry seasons. The findings provide practical implications for agricultural planning, disaster preparedness, and water resource management in tropical nations.
Downloads
References
[1] P. Settu And M. Ramaiah, “A Data Driven Comparison Of Hybrid Machine Learning Techniques For Soil Moisture Modeling Using Remote Sensing Imagery.” 2025. Doi: 10.1038/S41598-025-27225-0.
[2] R. Svs And R. Mukkamala, “A Cnn Bidirectional Lstm Framework For Predicting Monsoon Rainfall In India.” 2023. Doi: 10.31577/Ahs-2023-0024.02.0024.
[3] J. Jumadi Et Al., “Ai-Driven Ensemble Learning For Spatio-Temporal Rainfall Prediction In The Bengawan Solo River Watershed, Indonesia.” 2025. Doi: 10.3390/Su17209281.
[4] J. Jumadi Et Al., “Performance Of Machine Learning And Deep Learning Models For Predicting Rainfall In A Large Watershed: Case Study Of Bengawan Solo River Basin, Indonesia.” 2026. Doi: 10.21163/Gt_2026.212.05.
[5] N. Mahavik And Et Al., “Optimizing Rainfall Prediction In Central Thailand With Weather Radar And Machine Learning During The Monsoon,” Environ. Earth Sci., Vol. 84, P. 137, Feb. 2025.
[6] N. Salaeh And Et Al., “Comparative Analysis And Enhancing Rainfall Prediction Models For Monthly Rainfall Prediction In The Eastern Thailand,” Results Eng., Vol. 14, P. 103094, Jun. 2025.
[7] A.-C. Akazan, V. R. Mbingui, G. L. R. N’guessan, And I. Karambal, “Localized Weather Prediction Using Kolmogorov-Arnold Network-Based Models And Deep Rnns,” Arxiv, May 2025.
[8] J. Seidu, “Gated Residual Learning Framework For Daily Rainfall Prediction In Tropical Monsoon Regions: A Case Study Of Coastal Ghana,” Front. Clim., Vol. 8, 2026.
[9] S. Shetty, Dhanyashree, D. Dharmendra, S. Bankapur, And P. P. B. M, “Hydrostack: A Hybrid Meta-Ensemble Machine Learning Framework For Accurate Annual Rainfall Prediction,” In 2024 8th International Conference On I-Smac (Iot In Social, Mobile, Analytics And Cloud) (I-Smac), Oct. 2024, Pp. 1926–1934. Doi: 10.1109/I-Smac61858.2024.10714739.
[10] Salehi, Sohrab ; Akbar Salehi Neyshabouri, Seyed Ali, “A Hybrid Deep Learning Framework For Improving Short-Term Precipitation Forecasts,” Environ. Model. Softw., Vol. 182, 2025.
[11] C. A. M. Vallejo And R. L. Manzione, “Multidimensional Forecasting Of Precipitation And Potential Evapotranspiration In The Paranapanema River Basin Using Neural Network Time Series.” 2024. Doi: 10.1016/J.Jsames.2024.104961.
[12] Y. J. Praveen, D. P. Isravel, J. P. M. Dhas, And G. Deepthi, “Machine Learning Based Hybrid System For Rainfall Prediction And Crop Recommendation.” 2024. Doi: 10.1109/Amathe61652.2024.10582105.
[13] P. Laon Et Al., “Hybrid Unsupervised–Supervised Learning Framework For Rainfall Prediction Using Satellite Signal Strength Attenuation.” 2026. Doi: 10.3390/S26020648.
[14] S. Nippani, V. K. Sharma, A. Yadav, Sitender, And B. K. Bhadra, “High Rainfall Event Identification Using Remote Sensing Satellite Data Integrated With A Hybrid Deep Learning Framework.” 2024. Doi: 10.1007/S43538-024-00286-X.
[15] A. Saleh, M. L. Tan, Z. Yaseen, And F. Zhang, “Integrated Machine Learning Models For Enhancing Tropical Rainfall Prediction Using Nasa Power Meteorological Data.” 2024. Doi: 10.2166/Wcc.2024.719.
[16] O. Wardhani, R. M. A. Akbar, And Y. A. Natawijaya, “A Review On Trends And Effectiveness Of Rainfall Prediction Models For Smart Irrigation: Toward Future Development,” Intellect Indones. J. Learn. Technol. Innov., Vol. 4, No. 1, Pp. 30–41, Jun. 2025.
[17] R. Aprianto, A. Putra, And H. Nufus, “Predictive Modelling Of Rainfall Using Deep Learning Based On Satellite Data For Flood And Drought Disaster Mitigation In Sumbawa Regency,” Jtam J. Teori Dan Apl. Mat., Vol. 9, No. 4, Oct. 2025.
[18] D. D.Ganesh, M. M.D.Niveedha, M. M.Deepa, M. G.Harika, M. G. R. Reddy, And D. M. S. Kumar, “Application Of Hybrid Machine Learning Approaches For Effective Rainfall Prediction.” 2025. Doi: 10.1109/Icdsaai65575.2025.11011750.
[19] T. Trinh And H. Tran, “Coupling Regional Climate Model And Machine Learning To Model High-Resolution Precipitation.” 2021. [Online]. Available: Https://Www.Semanticscholar.Org/Paper/5f5be71ab534c3aed0afe2bc89ed54b3872bad57
[20] J. W. Lockwood, T. Loridan, N. Lin, M. Oppenheimer, And N. Hannah, “A Machine Learning Approach To Model Over Ocean Tropical Cyclone Precipitation.” 2023. Doi: 10.1175/Jhm-D-23-0065.1.
[21] S. T. Da Silva Et Al., “Rainfall Forecast In Brazil Using Machine Learning.” 2025. Doi: 10.1063/5.0259222.
[22] C. Funk, P. Peterson, M. Landsfeld, D. Pedreros, J. Verdin, And J. Michaelsen, “The Climate Hazards Infrared Precipitation With Stations—A New Environmental Record For Monitoring Extremes,” Sci. Data, Vol. 2, P. 150066, Dec. 2015.
[23] M. Abadi And Et Al., “Tensorflow: A System For Large-Scale Machine Learning,” In Proc. Osdi, Savannah, Ga, Usa, Nov. 2016, Pp. 265–283.
[24] J. Jumadi And Et Al., “Ai-Driven Ensemble Learning For Spatio-Temporal Rainfall Prediction In The Bengawan Solo River Watershed, Indonesia,” Sustainability, Vol. 17, No. 20, P. 9281, Oct. 2025.
[25] T. Chen And C. Guestrin, “Xgboost: A Scalable Tree Boosting System,” In Proc. Kdd, San Francisco, Ca, Usa, Aug. 2016, Pp. 785–794.
[26] S. Bai, J. Z. Kolter, And V. Koltun, “An Empirical Evaluation Of Generic Convolutional And Recurrent Networks For Sequence Modeling,” Arxiv, Mar. 2018.
[27] W. C. Low, Y. H. Lee, A. N. Biswas, And W. T. Yeo, “Rainfall Prediction Using Deep Learning Method For Tropical Region.” 2024. Doi: 10.1109/Igarss53475.2024.10641877.
[28] N. Madala, S. D. S. P. D. Tallapudi, V. S. Malladi, D. M. Muthinti, S. Tokala, And M. K. Enduri, “Improving Weather Forecast Accuracy Using Hybrid Machine Learning Algorithms.” 2024. Doi: 10.1109/Cicn63059.2024.10847449.
[29] S. Shetty, Dhanyashree, D. Dharmendra, S. S. Bankapur, And P. P. B. M, “Hydrostack: A Hybrid Meta-Ensemble Machine Learning Framework For Accurate Annual Rainfall Prediction.” 2024. Doi: 10.1109/I-Smac61858.2024.10714739.
[30] W. C. Low, Y. H. Lee, A. N. Biswas, And W. T. Yeo, “Rainfall Prediction Using Deep Learning Method For Tropical Region.” 2024. Doi: 10.1109/Igarss53475.2024.10641877.
[31] J. Wang, R. K. W. Wong, M. Jun, C. Schumacher, And R. Saravanan, “Statistical And Machine Learning Methods Applied To The Prediction Of Tropical Rainfall.” 2020. Doi: 10.1002/Essoar.10504402.1.
[32] S. Kumar, S. Hupesh, N. Ketan, I. Kurnool, And Of Ece, “Rainfall Prediction Using Hybrid Cnn–Lstm Model With Seasonal Trend Decomposition.” 2026. Doi: 10.1109/Aiei69164.2026.11496929.
[33] M. M. Rahman Et Al., “Hybrid Deep Learning Framework For Rainfall Prediction: Integrating Wavelet-Arima, Ceemdanlstm, And Cnn-Bilstm For Enhanced Climate Variability Analysis.” 2025. Doi: 10.1109/Ic2nc67409.2025.11376471.
[34] R. Satya, R. Singh, B. V. V. S. R. K. K. Pavan, S. S. Madhav, M. S. Sarvagna, And Y. Rao, “Spatio-Temporal Predictions Of Rainfall Patterns Using Integrated Geospatial Data.” 2026. Doi: 10.1109/Iatmsi68868.2026.11465199.
