A RECALL-ORIENTED STACKING ENSEMBLE FOR EXTREME RAINFALL EARLY WARNING IN TERNATE

Achmad Fuad; Muhammad Sabri Ahmad; Muhammad Ridha Albaar; Yasir Muin

doi:10.33387/jiko.v9i2.11857

Authors

Achmad Fuad Teknik informatika universitas khairun
Muhammad Sabri Ahmad Teknik informatika universitas khairun
Muhammad Ridha Albaar Teknik informatika universitas khairun
Yasir Muin Teknik informatika universitas khairun

DOI:

https://doi.org/10.33387/jiko.v9i2.11857

Abstract

Extreme-rainfall detection in small tropical islands is difficult because extreme events are rare and false negatives have substantial disaster-mitigation consequences. This study aims to (1) compare imbalance-aware machine-learning and deep temporal models, (2) determine whether an out-of-fold stacking ensemble can maximize sensitivity, and (3) explain model behavior using SHAP. ERA5 hourly data for a representative Ternate grid point were collected from 1 January 2005 to 6 February 2026 and aggregated into 7,707 daily records; only 12 days met the extreme threshold of 75 mm/day. A 30-day sequence was used to predict the next-day class, and three expanding temporal test folds were evaluated. Weighted loss, focal loss, balanced tree learning, validation-based threshold selection, and leakage-controlled stacking were applied. Across 2,229 pooled test days containing four extreme events, the stacking model detected all events (recall 1.000) but produced low precision (0.0023), F2-score 0.0115, specificity 0.2256, and 1,723 false positives. Balanced Random Forest provided the best overall trade-off, with recall 0.750, F2-score 0.0316, specificity 0.7951, and MCC 0.0570. Therefore, stacking is useful as a high-sensitivity screening layer, but it is not yet suitable as a standalone operational warning model. Local observations and additional event samples are required for calibration.

Downloads

Download data is not yet available.

Author Biography

Muhammad Sabri Ahmad, Teknik informatika universitas khairun

Dosen pns di universitas khairun ternate, dengan jabatan asisten ahli

References

B. Dong et al., “City-scale high-resolution flood nowcasting based on high-performance hydrodynamic modelling,” Int. J. Disaster Risk Reduct., vol. 125, p. 105584, 2025, doi: 10.1016/j.ijdrr.2025.105584.

H. Hersbach et al., “The ERA5 global reanalysis,” Q. J. R. Meteorol. Soc., vol. 146, no. 730, pp. 1999–2049, 2020, doi: 10.1002/qj.3803.

H. He and E. A. Garcia, “Learning from Imbalanced Data,” IEEE Trans. Knowl. Data Eng., vol. 21, no. 9, pp. 1263–1284, 2009, doi: 10.1109/TKDE.2008.239.

S. Hochreiter and J. Schmidhuber, “Long Short-Term Memory,” Neural Comput., vol. 9, no. 8, pp. 1735–1780, 1997, doi: 10.1162/neco.1997.9.8.1735.

M. S. Ahmad, H. Hadiyanto, and R. Sanjaya, “Mathematical Modelling of Engineering Problems An Explainable Hybrid Long Short-Term Memory and Extreme Gradient Boosting Framework for Daily Stock Price Forecasting with Contextual Latent Features,” vol. 13, no. 3, pp. 513–520, 2026.

A. Vaswani et al., “Attention Is All You Need,” in Advances in Neural Information Processing Systems, 2017. doi: 10.5555/3295222.3295349.

S. M. Lundberg and S. I. Lee, “A Unified Approach to Interpreting Model Predictions,” in Advances in Neural Information Processing Systems, 2017, pp. 4765–4774.

M. Oprea, “A general framework and guidelines for benchmarking computational intelligence algorithms applied to forecasting problems derived from an application domain-oriented survey,” Appl. Soft Comput., vol. 89, p. 106103, 2020, doi: 10.1016/j.asoc.2020.106103.

S. An, T.-J. Oh, E. Sohn, and D. Kim, “Deep learning for precipitation nowcasting: A survey from the perspective of time series forecasting,” Expert Syst. Appl., vol. 268, p. 126301, 2025, doi: 10.1016/j.eswa.2024.126301.

L. Mdegela, E. Municio, Y. De Bock, E. Luhanga, J. Leo, and E. Mannens, “Extreme Rainfall Event Classification Using Machine Learning for Kikuletwa River Floods,” Water, vol. 15, no. 6, p. 1021, 2023, doi: 10.3390/w15061021.

S. Chkeir, A. Anesiadou, A. Mascitelli, and R. Biondi, “Nowcasting extreme rain and extreme wind speed with machine learning techniques applied to different input datasets,” Atmos. Res., vol. 282, p. 106548, 2023, doi: 10.1016/j.atmosres.2022.106548.

F. J. Hop, R. Linneman, B. Schnitzler, A. Bomers, and M. J. Booij, “Real time probabilistic inundation forecasts using a LSTM neural network,” J. Hydrol., vol. 635, p. 131082, 2024, doi: 10.1016/j.jhydrol.2024.131082.

Y. Wu et al., “Effective LSTMs with seasonal-trend decomposition and adaptive learning and niching-based backtracking search algorithm for time series forecasting,” Expert Syst. Appl., vol. 236, p. 121202, 2024, doi: 10.1016/j.eswa.2023.121202.

C. C. C. Service, “ERA5 hourly data on single levels from 1940 to present,” 2023. doi: 10.24381/cds.adbb2d47.

S. Bao, C. Yu, and Z. Wan, “Spatial Mismatch and Attribution Analysis of Flood Risk and Resilience in Megacities: Insights from a ‘Risk-Resilience-Effect’ Framework and an Interpretable XGBoost-SHAP Model,” Sustain. Cities Soc., vol. 131, no. August, p. 106758, 2025, doi: 10.1016/j.scs.2025.106758.

S. He, Z. Li, and X. Liu, “An improved GEV boosting method for imbalanced data classification with application to short-term rainfall prediction,” J. Hydrol., vol. 617, p. 128882, 2023, doi: 10.1016/j.jhydrol.2022.128882.

T. Chen and C. Guestrin, “XGBoost: A Scalable Tree Boosting System,” in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016, pp. 785–794. doi: 10.1145/2939672.2939785.

Y. Xie, W. Sun, M. Ren, S. Chen, Z. Huang, and X. Pan, “Stacking ensemble learning models for daily runoff prediction using 1D and 2D CNNs,” Expert Syst. Appl., vol. 217, p. 119469, 2023, doi: 10.1016/j.eswa.2022.119469.

F. Zhu et al., “A hybrid process-data driven framework for real-time hydrological forecasting with interpretable deep learning,” J. Hydrol., vol. 662, p. 134082, 2025, doi: 10.1016/j.jhydrol.2025.134082.

A. B. Upadhyay, S. R. Shah, and R. A. Thakker, “Advanced rainfall nowcasting using 3D convolutional LSTM networks on satellite data,” J. Comput. Math. Data Sci., vol. 16, p. 100125, 2025, doi: 10.1016/j.jcmds.2025.100125.

F. Piadeh, V. Bakhtiari, and F. Piadeh, “Automated novel real-time framework for rainfall data imputation in flood early warning systems,” Eng. Appl. Artif. Intell., vol. 164, p. 113348, 2026, doi: 10.1016/j.engappai.2025.113348.