Ensemble and Hybrid Machine Learning Models for Seasonal Water Consumption Forecasting Under Climate Variability
Downloads
The objective of this paper is to improve the forecasting of monthly water consumption under climate variability by combining ensemble and hybrid modelling with a season-aware design. Monthly consumption and meteorological data from 2003 to 2024 were utilized in this study. Four models were evaluated: (i) a stacking ensemble with STL-trend plus residual learning; (ii) a hybrid machine-learning–physics model with differentially-evolved weights; and (iii–iv) season-specific stacked models for wet and dry periods. Robustness was assessed with time-aware validation and residual diagnostics (Shapiro–Wilk, Breusch–Pagan, Durbin–Watson, Ljung–Box). The findings indicate that across models, ensembles captured nonlinear climate–demand variations while maintaining linear structure. The ensemble and hybrid model achieved strong accuracy with low errors while the season-specific models attained high fit (wet R²≈0.998; dry R²≈0.991) with stable residual behavior. Sensitivity to temperature and humidity aligns with expected physical behavior. Precipitation shows a diminishing-returns effect on water use, where moderate rainfall leads to higher consumption, while heavy rainfall tends to reduce demand. The framework innovatively combines decomposition-assisted stacking, physics-informed hybridization, and seasonal ensemble modelling. Overall, the approach provides highly accurate, interpretable, and climate-aware water demand forecasts for tropical regions, offering a practical basis for utility-scale implementation.
Downloads
[1] Beharry, S. L., & Clarke, R. M. (2023). Estimations of future reservoir volumes under different climate scenarios for a tropical reservoir in a small Caribbean Island, Trinidad. Environmental Monitoring and Assessment, 195(5), 590. doi:10.1007/s10661-023-11207-8.
[2] Hemati, A., Rippy, M. A., Grant, S. B., Davis, K., & Feldman, D. (2016). Deconstructing demand: The anthropogenic and climatic drivers of urban water consumption. Environmental Science & Technology, 50(23), 12557–12566. doi:10.1021/acs.est.6b02938.
[3] Mycoo, M. A., & Roopnarine, R. R. (2024). Water resource sustainability: Challenges, opportunities and research gaps in the English-speaking Caribbean Small Island Developing States. PLOS Water, 3(1), e0000222. doi:10.1371/journal.pwat.0000222.
[4] Saeed, F. H., Al-Khafaji, M. S., Al-Faraj, F. A. M., & Uzomah, V. (2024). Sustainable Adaptation Plan in Response to Climate Change and Population Growth in the Iraqi Part of Tigris River Basin. Sustainability (Switzerland) , 16(7), 2676. doi:10.3390/su16072676.
[5] Alikhani, M. R., & Moeini, R. (2025). Predicting the urban water demand by equipping intelligent-based methods with discrete wavelet transform function. Applied Water Science, 15(2), 38. doi:10.1007/s13201-025-02368-7.
[6] Hao, W., Cominola, A., & Castelletti, A. (2024). Combining wavelet-enhanced feature selection and deep learning techniques for multi-step forecasting of urban water demand. Environmental Research: Infrastructure and Sustainability, 4(3), 35005. doi:10.1088/2634-4505/ad5e1d.
[7] Skanupong, N., Xu, Y., Yu, L., Wan, Z., & Wang, S. (2024). The convolutional neural network for Pacific decadal oscillation forecast. Environmental Research Letters, 19(12), 124022. doi:10.1088/1748-9326/ad8be2.
[8] Shu, J., Xia, X., Han, S., He, Z., Pan, K., & Liu, B. (2024). Long-term water demand forecasting using artificial intelligence models in the Tuojiang River basin, China. PLOS ONE, 19(5), e0302558. doi:10.1371/journal.pone.0302558.
[9] Kulaczkowski, A., & Lee, J. (2024). Harnessing the Power of Random Forest for Precise Short-Term Water Demand Forecasting in Italian Water Districts †. Engineering Proceedings, 69(1), 81. doi:10.3390/engproc2024069081.
[10] Li, X., Wu, X., Sun, M., Yang, S., & Song, W. (2022). A Novel Intelligent Leakage Monitoring-Warning System for Sustainable Rural Drinking Water Supply. Sustainability (Switzerland), 14(10), 6079. doi:10.3390/su14106079.
[11] Liu, J., Zhou, X. L., Zhang, L. Q., & Xu, Y. P. (2023). Forecasting Short-term Water Demands with an Ensemble Deep Learning Model for a Water Supply System. Water Resources Management, 37(8), 2991–3012. doi:10.1007/s11269-023-03471-7.
[12] Ji, F. (2024). Time Series Prediction Method for Meteorological Data Based on the ARIMA-LSTM Model. Academic Journal of Science and Technology, 13(1), 193–196. doi:10.54097/rx3d5d70.
[13] Rajballie, A., Tripathi, V., & Chinchamee, A. (2022). Water consumption forecasting models - a case study in Trinidad (Trinidad and Tobago). Water Supply, 22(5), 5434–5447. doi:10.2166/ws.2022.147.
[14] Bakhshipour, A. E., Namdari, H., Koochali, A., Dittmer, U., & Haghighi, A. (2024). An Ensemble Data-Driven Approach for Enhanced Short-Term Water Demand Forecasting in Urban Areas. Engineering Proceedings, 69(1), 69. doi:10.3390/engproc2024069069.
[15] Herrera, M., Torgo, L., Izquierdo, J., & Pérez-García, R. (2010). Predictive models for forecasting hourly urban water demand. Journal of Hydrology, 387(1–2), 141–150. doi:10.1016/j.jhydrol.2010.04.005.
[16] Johnson, R. C., Burian, S. J., Halgren, J., Irons, T., Baur, E., Aziz, D., Hassan, D., Li, J., Kirkham, T., Stewart, J., & Briefer, L. (2024). Preparing municipal water system planning for a changing climate: Integrating climate-sensitive demand estimates. Journal of the American Water Resources Association, 60(1), 211–225. doi:10.1111/1752-1688.13165.
[17] Wu, W., & Kang, Y. (2024). Ensemble Empirical Mode Decomposition Granger Causality Test Dynamic Graph Attention Transformer Network: Integrating Transformer and Graph Neural Network Models for Multi-Sensor Cross-Temporal Granularity Water Demand Forecasting. Applied Sciences (Switzerland), 14(8), 3428. doi:10.3390/app14083428.
[18] Banda, P., Bhuiyan, M., Zhang, K., & Song, A. (2021). Multivariate monthly water demand prediction using ensemble and gradient boosting machine learning techniques. Proceedings of the International Conference on Evolving Cities, 14. doi:10.55066/proc-icec.2021.14.
[19] Zubaidi, S. L., Al-Bdairi, N. S. S., Ortega-Martorell, S., Ridha, H. M., Al-Ansari, N., Al-Bugharbee, H., Hashim, K., & Gharghan, S. K. (2023). Assessing the Benefits of Nature-Inspired Algorithms for the Parameterization of ANN in the Prediction of Water Demand. Journal of Water Resources Planning and Management, 149(1), 04022075. doi:10.1061/(asce)wr.1943-5452.0001602.
[20] Hekimoğlu, M., Çetin, A. İ., & Kaya, B. E. (2023). Evaluation of Various Machine Learning Methods to Predict Istanbul’s Freshwater Consumption. International Journal of Environment and Geoinformatics, 10(2), 1–11. doi:10.30897/ijegeo.1270228.
[21] Tiwari, M. K., & Adamowski, J. F. (2015). Medium-Term Urban Water Demand Forecasting with Limited Data Using an Ensemble Wavelet–Bootstrap Machine-Learning Approach. Journal of Water Resources Planning and Management, 141(2), 4014053. doi:10.1061/(asce)wr.1943-5452.0000454.
[22] Bachour, R., Maslova, I., Ticlavilca, A. M., Walker, W. R., & McKee, M. (2016). Wavelet-multivariate relevance vector machine hybrid model for forecasting daily evapotranspiration. Stochastic Environmental Research and Risk Assessment, 30(1), 103–117. doi:10.1007/s00477-015-1039-z.
[23] Xu, Y., Zhang, J., Long, Z., & Chen, Y. (2018). A novel dual-scale deep belief network method for daily urban water demand forecasting. Energies, 11(5), 1068. doi:10.3390/en11051068.
[24] GOBTT. (2021). National Water Resources Management Strategy. Ministry of Public Utilities, Government of the Republic of Trinidad and Tobago, Port of Spain, Trinidad & Tobago.
[25] WRA. (2020). Water Resource Assessment for Trinidad and Tobago. Water Resources Agency, Water and Sewerage Authority (WASA), Port of Spain, Trinidad & Tobago.
[26] McSweeney, C., New, M. & G. Lizcano, G. (2010). UNDP Climate Change Country Profiles: Trinidad and Tobago. United Nations Development Programme, New York, United States.
[27] GCCA+. (2019). Trinidad and Tobago National Adaptation Strategy: Climate Resilience in the Water Sector. Global Climate Change Alliance Plus, United Nations Development Programme, New York, United States.
[28] SIDS. (2017). Impacts of Climate Change on Settlements and Infrastructure in the Coastal and Marine Environments of Caribbean Small Island Developing States (SIDS). Science Review 2017, Caribbean Marine Climate Change Report Card. Available online: https://www.gov.uk/government/publications/commonwealth-marine-economies-cme-programme-caribbean-marine-climate-change-report-card-scientific-reviews (accessed on January 2026).
[29] Hyndman, R. J., & Athanasopoulos, G. (2018). Forecasting: principles and practice. OTexts, Melbourne, Australia.
[30] Box, G. E. P., & Cox, D. R. (1964). An Analysis of Transformations. Journal of the Royal Statistical Society Series B: Statistical Methodology, 26(2), 211–243. doi:10.1111/j.2517-6161.1964.tb00553.x.
[31] Gong, Q., Li, Q., Gavrielides, M. A., & Petrick, N. (2020). Data transformations for statistical assessment of quantitative imaging biomarkers: Application to lung nodule volumetry. Statistical Methods in Medical Research, 29(9), 2749–2763. doi:10.1177/0962280220908619.
[32] Venables, W. N., & Ripley, B. D. (2002). Modern Applied Statistics with S. Statistics and Computing. Springer, New York, United States. doi:10.1007/978-0-387-21706-2.
[33] Yao, D., Zhang, B., Li, X., Zhan, X., Zhan, X., & Zhang, B. (2023). Applying negative sample denoising and multi-view feature for lncRNA-disease association prediction. Frontiers in Genetics, 14. doi:10.3389/fgene.2023.1332273.
[34] Donoho, D. L., & Johnstone, I. M. (1995). Adapting to unknown smoothness via wavelet shrinkage. Journal of the American Statistical Association, 90(432), 1200–1224. doi:10.1080/01621459.1995.10476626.
[35] Percival, D. B., & Walden, A. T. (2000). Wavelet Methods for Time Series Analysis. Cambridge University Press, Cambridge, United Kingdom. doi:10.1017/cbo9780511841040.
[36] Cleveland, R. B., Cleveland, W. S., McRae, J. E., & Terpenning, I. (1990). STL: A seasonal-trend decomposition. Journal of Official Statistics, 6(1), 3-73.
[37] Breiman, L. (2001). Random Forests. Machine Learning, 45(1), 5–32. doi:10.1023/a:1010933404324.
[38] MacKay, D. J. C. (1992). Bayesian Interpolation. Neural Computation, 4(3), 415–447. doi:10.1162/neco.1992.4.3.415.
[39] Christopher, M. B. (2006). Pattern recognition and machine learning. Springer, New York, United States.
[40] Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 13-17-August-2016, 785–794. doi:10.1145/2939672.2939785.
[41] Wolpert, D. H. (1992). Stacked generalization. Neural Networks, 5(2), 241–259. doi:10.1016/S0893-6080(05)80023-1.
[42] Storn, R., & Price, K. (1997). Differential Evolution - A Simple and Efficient Heuristic for Global Optimization over Continuous Spaces. Journal of Global Optimization, 11(4), 341–359. doi:10.1023/A:1008202821328.
[43] Daw, A., Karpatne, A., Watkins, W. D., Read, J. S., & Kumar, V. (2022). Physics-guided neural networks (pgnn): An application in lake temperature modeling. Knowledge guided machine learning, Chapman and Hall/CRC, London, United Kingdom. doi:10.1137/1.9781611974973.33.
[44] Moriasi, D. N., Arnold, J. G., Van Liew, M. W., Bingner, R. L., Harmel, R. D., & Veith, T. L. (2007). Model Evaluation Guidelines for Systematic Quantification of Accuracy in Watershed Simulations. Transactions of the ASABE, 50(3), 885–900. doi:10.13031/2013.23153.
[45] Durbin, J., & Watson, G. S. (1950). Testing for serial correlation in least squares regression. Biometrika, 37(3–4), 409–428. doi:10.1093/biomet/37.3-4.409.
[46] Box, G. E. P., Jenkins, G. M., & Reinsel, G. C. (2008). Time Series Analysis. Wiley Series in Probability and Statistics, John Wiley & Sons, Hoboken, United States. doi:10.1002/9781118619193.
[47] James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An Introduction to Statistical Learning. In Springer Texts in Statistics. Springer, New York, United States. doi:10.1007/978-1-4614-7138-7.
[48] Alsulaili, A., Alkandari, M., & Buqammaz, A. (2022). Assessing the impacts of meteorological factors on freshwater consumption in arid regions and forecasting the freshwater demand. Environmental Technology & Innovation, 25, 102099. doi:10.1016/j.eti.2021.102099.
[49] Dimkić, D. (2020). Temperature Impact on Drinking Water Consumption. Environmental Sciences Proceedings, 2(1), 31. doi:10.3390/environsciproc2020002031.
[50] Breusch, T. S., & Pagan, A. R. (1979). A Simple Test for Heteroscedasticity and Random Coefficient Variation. Econometrica, 47(5), 1287. doi:10.2307/1911963.
[51] Koenker, R. (1981). A note on studentizing a test for heteroscedasticity. Journal of Econometrics, 17(1), 107–112. doi:10.1016/0304-4076(81)90062-2.
[52] SHAPIRO, S. S., & WILK, M. B. (1965). An analysis of variance test for normality (complete samples). Biometrika, 52(3–4), 591–611. doi:10.1093/biomet/52.3-4.591.
[53] Khashei, M., & Bijari, M. (2011). A novel hybridization of artificial neural networks and ARIMA models for time series forecasting. Applied Soft Computing, 11(2), 2664–2675. doi:10.1016/j.asoc.2010.10.015.
[54] Moosavi, S. F., Salehnia, N., Seifi, A., AsgharpourMasouleh, A., & Salehnia, N. (2023). Designing and calibrating an agent-based platform to evaluate the effect of climate variables on residential water demand. Water and Environment Journal, 37(3), 604–615. doi:10.1111/wej.12864.
[55] Makpiboon, C., Pornprommin, A., & Lipiwattanakarn, S. (2020). Impacts of weather variables on urban water demand at multiple temporal scales. International Journal of GEOMATE, 18(67), 71–77. doi:10.21660/2020.67.5758.
[56] Sadok, W., Lopez, J. R., & Smith, K. P. (2021). Transpiration increases under high-temperature stress: Potential mechanisms, trade-offs and prospects for crop resilience in a warming world. Plant Cell and Environment, 44(7), 2102–2116. doi:10.1111/pce.13970.
[57] Farah, E., & Shahrour, I. (2025). Forecasting Urban Water Demand Using Multi-Scale Artificial Neural Networks with Temporal Lag Optimization. Water (Switzerland), 17(19), 2886. doi:10.3390/w17192886.
[58] Timotewos, M. T., Barjenbruch, M., & Behailu, B. M. (2022). The Assessment of Climate Variables and Geographical Distribution on Residential Drinking Water Demand in Ethiopia. Water (Switzerland), 14(11), 1722. doi:10.3390/w14111722.
[59] Willard, J., Jia, X., Xu, S., Steinbach, M., & Kumar, V. (2020). Integrating physics-based modeling with machine learning: A survey. arXiv preprint, arXiv:2003.04919, 1-34. doi:10.1145/1122445.1122456.
[60] Reichstein, M., Camps-Valls, G., Stevens, B., Jung, M., Denzler, J., Carvalhais, N., & Prabhat. (2019). Deep learning and process understanding for data-driven Earth system science. Nature, 566(7743), 195–204. doi:10.1038/s41586-019-0912-1.
[61] DeGuzman, K., Knappenberger, T., Brantley, E., & Olshansky, Y. (2023). Estimating runoff probability from precipitation data: A binomial regression analysis. Hydrological Processes, 37(11), e15029. doi:10.1002/hyp.15029.
[62] Donkor, E. A., Mazzuchi, T. A., Soyer, R., & Alan Roberson, J. (2014). Urban Water Demand Forecasting: Review of Methods and Models. Journal of Water Resources Planning and Management, 140(2), 146–159. doi:10.1061/(asce)wr.1943-5452.0000314.
[63] Tucker, J., MacDonald, A., Coulter, L., & Calow, R. C. (2014). Household water use, poverty and seasonality: Wealth effects, labour constraints, and minimal consumption in Ethiopia. Water Resources and Rural Development, 3, 27-47. doi:10.1016/j.wrr.2014.04.001.
[64] San Diego County Water Authority. (2021). Water Demand Forecasting Model Update 2020. County Water Authority, San Diego, United States.
[65] Tipping, M. E. (2000). Sparse Bayesian Learning and the Relevance Vector Machine. Journal of Machine Learning Research, 1, 211–244. doi:10.1162/15324430152748236.
[66] Zhu, X., Miao, P., Qin, J., Li, W., Wang, L., Chen, Z., & Zhou, J. (2023). Spatio-temporal variations of nitrate pollution of groundwater in the intensive agricultural region: Hotspots and driving forces. Journal of Hydrology, 623, 129864. doi:10.1016/j.jhydrol.2023.129864.
[67] Sollmann, R. (2024). Estimating the temporal scale of lagged responses in species abundance and occurrence. Ecosphere, 15(1), e4704. doi:10.1002/ecs2.4704.
[68] Sazib, N., Bolten, J., & Mladenova, I. (2020). Exploring spatiotemporal relations between soil moisture, precipitation, and streamflow for a large set of watersheds using google earth engine. Water (Switzerland), 12(5), 1371. doi:10.3390/w12051371.
[69] Msigwa, A., Chawanda, C. J., Komakech, H. C., Nkwasa, A., & Van Griensven, A. (2022). Representation of seasonal land use dynamics in SWAT+ for improved assessment of blue and green water consumption. Hydrology and Earth System Sciences, 26(16), 4447–4468. doi:10.5194/hess-26-4447-2022.
[70] Xenochristou, M., & Kapelan, Z. (2020). An ensemble stacked model with bias correction for improved water demand forecasting. Urban Water Journal, 17(3), 212–223. doi:10.1080/1573062X.2020.1758164.
[71] Lovaglio, P. G. (2025). Cross-Learning With Panel Data Modeling for Stacking and Forecast Time Series Employment in Europe. Journal of Forecasting, 44(2), 753–780. doi:10.1002/for.3224.
- Authors retain all copyrights. It is noticeable that authors will not be forced to sign any copyright transfer agreements.
- This work (including HTML and PDF Files) is licensed under a Creative Commons Attribution 4.0 International License.
