A Data-Quality-Aware Machine Learning Framework for Magnitude Homogenization in Indonesia's Earthquake Catalog

Desi  Ramayanti; Bias Yulisa  Geni; Imam  Mulya

doi:10.29207/resti.v10i5.8137

Authors

Desi Ramayanti Universitas Dian Nusantara
Bias Yulisa Geni Universitas Dian Nusantara
Imam Mulya Universitas Dian Nusantara

DOI:

https://doi.org/10.29207/resti.v10i5.8137

Keywords:

instrument-quality metadata, machine learning, magnitude homogenization, selection bias diagnostics, uncertainty calibration

Abstract

Indonesia’s national seismological agency, BMKG, reports earthquake magnitudes under ten distinct, non-interchangeable scale labels in its public catalog (2020–2025, 65,788 events), with no published conversion to a common Mw-equivalent scale. Existing machine learning studies using this catalog have assumed a homogeneous magnitude scale, and instrument-quality metadata in the catalog has never been used as a covariate for magnitude-conversion bias or uncertainty. This study introduces a powerful hybrid magnitude-homogenization framework. For events with a native moment-magnitude anchor (the body-wave family), we perform self-supervised internal calibration and tie it to the global Scordilis relation. For events without such an anchor, the local/regional family, we apply a literature-based conversion using the Egyptian National Seismic Network formula. A quality-aware machine learning model is then trained to estimate homogenized magnitude with calibrated uncertainty, conditioned on instrument-quality features. The validation process uncovers and quantifies a selection bias of approximately 1.8 to 2.1 magnitude units in BMKG’s internal moment-magnitude reporting, reflecting that natively Mw-labeled events are a non-representative, significance-selected subsample rather than ground truth. The quality-aware model reduces point-prediction error (RMSE) by 50 to 70% and achieves better-calibrated 80% prediction intervals than a baseline without quality features. This combination of a practical homogenization solution and an explicit diagnostic framework is intended to be of direct use not only to BMKG and Indonesian seismology, but to any national seismological agency maintaining a similarly heterogeneous magnitude catalog without paired, same-event, multi-scale readings.

Downloads

Download data is not yet available.

References

[1] bmkg, “InaTEWS Earthquake Repository,” bmkg.go.id.

[2] A. Prasetyo, R. F. Aji, and W. S. Wibowo, “Assessing Information Security Awareness Among Indonesian Government Employees: A Case Study of the Meteorology, Climatology, and Geophysics Agency,” Journal of Information Systems Engineering and Business Intelligence, vol. 11, no. 2, 2025, doi: 10.20473/jisebi.11.2.126-142.

[3] Hans Raj Wason, “ A Statistical Analysis of Completeness of Earthquake Data around Dehradun city and its Implications for Seismicity Evaluation,” 15th World Conference on Earthquake Engineering (15WCEE), 2012.

[4] J. Woessner, J. L. Hardebeck, and E. Hauksson, “Theme IV – Understanding Seismicity Catalogs and their Problems What: What is an instrumental seismicity catalog,” Community Online Resource for Statistical Seismicity Analysis, no. September, 2010.

[5] I. Grigoratos, V. Poggi, L. Danciu, and R. Monteiro, “Homogenizing instrumental earthquake catalogs – a case study around the Dead Sea Transform Fault Zone,” Seismica, vol. 2, no. 2, 2023, doi: 10.26443/seismica.v2i2.402.

[6] F. Hisyam, A. Susilo, M. Anshori, and M. F. R. Hasan, “Spatio-Temporal Variation Seismicity Pattern in East Java Between 2002 and 2022 Based on the b-value and Seismic Quiescence z-value,” Trends in Sciences, vol. 21, no. 4, 2024, doi: 10.48048/tis.2024.7608.

[7] N. Mubarak MR, S. Susandri, and A. Zamsuri, “Geodetically-Enhanced Hybrid GRU with Adaptive Dropout and Dynamic L2 Regularization for Earthquake Parameter Prediction in Indonesia,” Jurnal Teknik Informatika (Jutif), vol. 7, no. 2, pp. 1483–1499, Apr. 2026, doi: 10.52436/1.jutif.2026.7.2.5478.

[8] P. Adetokunbo et al., “Statistical and Machine Learning Methods for Magnitude Scale Conversion and Gutenberg-Richter Analysis: Insights from Regional Seismicity of Morrocco,” 2025. doi: 10.2139/ssrn.5139022.

[9] D. Di Giacomo, D. Bindi, S. Parolai, and A. Oth, “Residual analysis of teleseismic P-wave energy magnitude estimates: inter- and intrastation variability,” Geophys. J. Int., vol. 185, no. 3, pp. 1444–1454, Jun. 2011, doi: 10.1111/j.1365-246X.2011.05019.x.

[10] E. Yee and W. Park, “Moment Magnitude Homogenization Relations in the South Korean Region from 1900 to 2020,” Applied Sciences (Switzerland), vol. 12, no. 14, 2022, doi: 10.3390/app12147215.

[11] H. M. Hussein et al., “Integration and magnitude homogenization of the Egyptian earthquake catalogue,” Natural Hazards, vol. 47, no. 3, pp. 525–546, Dec. 2008, doi: 10.1007/s11069-008-9237-3.

[12] S. N. Dybing, W. L. Yeck, H. M. Cole, and D. Melgar, “Rapid Estimation of Single-Station Earthquake Magnitudes with Machine Learning on a Global Scale,” Bulletin of the Seismological Society of America, vol. 114, no. 3, pp. 1523–1538, Jun. 2024, doi: 10.1785/0120230171.

[13] F. Bellalem, A. Talbi, and S. Maouche, “A unified MW parametric earthquake catalog for Algeria and adjacent regions (PECAAR),” Mediterranean Geoscience Reviews, vol. 4, no. 4, pp. 427–443, Dec. 2022, doi: 10.1007/s42990-022-00085-w.

[14] H. M. Hussein et al., “Integration and magnitude homogenization of the Egyptian earthquake catalogue,” Natural Hazards, vol. 47, no. 3, pp. 525–546, Dec. 2008, doi: 10.1007/s11069-008-9237-3.

[15] A. K. Bala and U. Saha, “Exploration of Regression Models for Homogenization of the Earthquake Magnitude Scales to Study Earthquakes in the Andaman-Nicobar Subduction Zone,” Journal of Volcanology and Seismology, vol. 19, no. 5, pp. 490–507, Oct. 2025, doi: 10.1134/S0742046325700228.

[16] A. Costanzo, “A New Catalogue and Insights into the 2022 Adriatic Offshore Seismic Sequence Using a Machine Learning-Based Procedure,” Sensors, vol. 25, no. 1, 2025, doi: 10.3390/s25010082.

[17] V. Anagnostou, E. Papadimitriou, V. Karakostas, and T. Bäck, “Investigating the 2024 Swarm–Like Activity Offshore Kefalonia Island, Aided by Machine Learning Algorithms,” Pure Appl. Geophys., vol. 182, no. 9, 2025, doi: 10.1007/s00024-025-03766-3.

[18] C. Gamboa-Canté et al., “Seismicity in Central America (1520–2020) and Earthquake catalog compilation for seismic hazard assessments,” Bulletin of Earthquake Engineering, vol. 22, no. 15, 2024, doi: 10.1007/s10518-024-02059-9.

[19] A. Boudebouda and A. Athmani, “Expanding moment magnitude dataset for earthquake magnitudes homogenization,” Acta Geophysica, vol. 72, no. 6, 2024, doi: 10.1007/s11600-024-01317-3.

[20] S. Susandri, F. B. Nasution, F. Fajrizal, and Saparudin, “Spatial-Temporal Analysis of Earthquakes in Indonesia with Deep Learning Models: Performance Evaluation of CNN, LSTM, and Hybrid CNN-GRU,” Jurnal RESTI, vol. 9, no. 5, 2025, doi: 10.29207/resti.v9i5.6538.

[21] U. Wijaya, Kusrini, and A. H. Muhammad, “Indonesian Seismic Mitigation using Earthquake Predicted Artificial Intelligence Model,” in ICOIACT 2022 - 5th International Conference on Information and Communications Technology: A New Way to Make AI Useful for Everyone in the New Normal Era, Proceeding, 2022. doi: 10.1109/ICOIACT55506.2022.9972091.

[22] S. S. R. Moustafa, A. M. Faried, and M. H. Yassien, “Characterizing Egyptian National Seismic Network station sites using genetic optimization for microtremor data inversion,” J. Seismol., vol. 28, no. 6, 2024, doi: 10.1007/s10950-024-10245-0.

[23] M. Veisi et al., “Seismic velocity and anisotropy of the uppermost mantle beneath Madagascar from Pn tomography,” Geophys. J. Int., vol. 126, no. 1, 2021.

[24] G. Papacharalampous, H. Tyralis, N. Doulamis, and A. Doulamis, “Uncertainty estimation of machine learning spatial precipitation predictions from satellite data,” Mach. Learn. Sci. Technol., vol. 5, no. 3, p. 035044, Sep. 2024, doi: 10.1088/2632-2153/ad63f3.

[25] A. D. Tsampas, E. M. Scordilis, C. B. Papazachos, and G. F. Karakaisis, “A Homogeneous Earthquake Catalog of Intermediate-Deep Focus Global Seismicity: Completeness and Spatio-Temporal Analysis,” Bulletin of the Geological Society of Greece, vol. 50, no. 3, 2017, doi: 10.12681/bgsg.11833.