Klasterisasi spatio-temporal komparatif untuk deteksi gempa bumi bermagnitudo tinggi: analisis k-means dan DBSCAN terhadap seismisitas zona subduksi

Abstract

Sumatera merupakan salah satu wilayah dengan aktivitas seismik tertinggi di dunia akibat interaksi konvergen Lempeng Indo-Australia dan Lempeng Eurasia serta keberadaan Sesar Besar Sumatera. Pemahaman terhadap pola kegempaan yang bersifat spatio-temporal menjadi penting sebagai dasar mitigasi bencana. Penelitian ini bertujuan mengidentifikasi pola pengelompokan gempa bumi di Sumatera menggunakan dua algoritma eksplorasi data yang berbeda paradigma, yaitu K-Means (berbasis partisi) dan DBSCAN (berbasis kepadatan). Data yang digunakan adalah katalog gempa United States Geological Survey (USGS) periode Januari 2015 hingga Desember 2025 pada rentang 6,5°LU–6,5°LS dan 94°–107°BT. Setelah praproses dan penerapan ambang kelengkapan magnitudo Mc = 4,0, diperoleh 2.834 kejadian. Lima fitur digunakan sebagai masukan clustering, yaitu lintang, bujur, waktu kejadian, kedalaman, dan magnitudo, yang seluruhnya distandardisasi menggunakan z-score. Jumlah klaster optimal K-Means ditentukan melalui Elbow Method dan Davies–Bouldin Index yang keduanya konvergen pada k = 6, sedangkan parameter DBSCAN ditetapkan melalui grafik k-distance dan penelusuran grid dua dimensi sehingga diperoleh ε = 0,7 dan minPts = 20. Hasil penelitian menunjukkan bahwa K-Means membagi kegempaan Sumatera menjadi enam klaster yang secara konsisten memisahkan segmen utara (Aceh–Simeulue) dan segmen selatan (Bengkulu–Lampung), masing-masing terpecah lagi menjadi dua periode waktu (2015–2020 dan 2020–2025), ditambah satu klaster zona Benioff berkedalaman menengah (rerata 173,7 km) dan satu klaster bermagnitudo tinggi (rerata 5,40). DBSCAN menghasilkan tiga klaster dengan 633 titik (22,3%) diklasifikasikan sebagai noise. Temuan paling menonjol adalah bahwa seluruh 27 gempa bermagnitudo M ≥ 6,0 diklasifikasikan sebagai noise oleh DBSCAN, dan magnitudo titik noise secara signifikan lebih besar dibandingkan titik terklaster (uji Mann–Whitney U, p < 0,001). Nilai Silhouette K-Means (0,260) lebih tinggi dibandingkan DBSCAN (0,214), namun DBSCAN unggul pada Davies–Bouldin Index (0,890 berbanding 1,207). Disimpulkan bahwa kedua metode bersifat komplementer: K-Means efektif untuk segmentasi zona seismik, sedangkan DBSCAN efektif sebagai detektor kejadian anomali yang berpotensi merusak.

Keywords
  • Clustering spatio-temporal
  • DBSCAN
  • Gempa bumi
  • K-means
  • Sumatera
Lisensing Information
Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

References
  1. Ansari, A., & Noorzad, A. (2022). Spatio-temporal seismic clustering and declustering of earthquake catalogs using density-based algorithms. Journal of Seismology, 26(4), 785–802. https://doi.org/10.1007/s10950-022-10101-2
  2. Aprianti, A., Jufriansah, A., Donuata, P. B., Khusnani, A., & Ayuba, J. (2023). Comparison of K-Means algorithm and DBSCAN on aftershock activity in the Flores Sea: Seismic activity 2019–2022. Journal of Novel Engineering Science and Technology, 2(3), 77–82. https://doi.org/10.56741/jnest.v2i03.393
  3. Arthur, D., & Vassilvitskii, S. (2007). k-means++: The advantages of careful seeding. Proceedings of the Eighteenth Annual ACM-SIAM Symposium on Discrete Algorithms, 1027–1035.
  4. Bradley, K. E., Feng, L., Hill, E. M., Natawidjaja, D. H., & Sieh, K. (2017). Implications of the 2016 7.8 Mentawai earthquake for intraplate deformation and seismic hazard in Sumatra. Geophysical Research Letters, 44(9), 4102–4109. https://doi.org/10.1002/2017GL072954
  5. Bradley, K., Mallick, R., Andıç, B., & Meltzner, A. J. (2020). The Great Sumatran Fault: A review of its anatomy, kinematics, and earthquake history. Earth-Science Reviews, 211, 103403. https://doi.org/10.1016/j.earscirev.2020.103403
  6. Campello, R. J., Moulavi, D., & Sander, J. (2013). Density-based clustering based on hierarchical density estimates. Advances in Knowledge Discovery and Data Mining (PAKDD 2013), Lecture Notes in Computer Science, 7819, 160–172. https://doi.org/10.1007/978-3-642-37456-2_14
  7. Hayes, G. P., Moore, G. L., Portner, D. E., Hearne, M., Flamme, H., Furtney, M., & Smoczyk, G. M. (2018). Slab2, a comprehensive subduction zone geometry model. Science, 362(6410), 58–61. https://doi.org/10.1126/science.aat4723
  8. Hazarika, D., & Oommen, T. (2021). Spatial and temporal clustering analysis of seismicity in active tectonic zones using machine learning. Computers & Geosciences, 154, 104820. https://doi.org/10.1016/j.cageo.2021.104820
  9. Hubert, L., & Arabie, P. (2001). Assessing cluster validity and partition agreement in spatial-temporal datasets. Journal of Classification, 18(2), 115–130.
  10. Li, C., van de Voorde, T., & Canters, F. (2020). Evaluating K-Means, DBSCAN, and OPTICS for spatial pattern discovery in environmental domain datasets. International Journal of Geographical Information Science, 34(8), 1589–1614. https://doi.org/10.1080/13658816.2020.1726922
  11. McCartney, R. (2009). The tectonic framework of the Sumatran subduction zone. Annual Review of Earth and Planetary Sciences, 37, 345–366. https://doi.org/10.1146/annurev.earth.031208.100212
  12. Mukhopadhyay, S., & Bhowmick, S. (2023). Comparative analysis of density-based and partition-based clustering algorithms on seismic spatial-temporal point processes. Applied Soft Computing, 142, 110345. https://doi.org/10.1016/j.asoc.2023.110345
  13. Natawidjaja, D. H., Sieh, K., Chlieh, M., Galetzka, J., Suwargadi, B. W., Cheng, H., Edwards, R. L., Avouac, J. P., & Ward, S. N. (2006). Source parameters of the great Sumatran megathrust earthquakes of 1797 and 1833 inferred from coral reefs in the Mentawai islands. Journal of Geophysical Research: Solid Earth, 111(B6), B06403. https://doi.org/10.1029/2005JB003825
  14. Pedregosa, F., Varoquaux, G., etc. (2011). Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12, 2825–2830.
  15. Philibosian, B., Sieh, K., Natawidjaja, D. H., Chiang, H. W., Shen, C. C., Suwargadi, B. W., & Daryono, M. R. (2022). Rupture history and seismic potential of the Sunda Megathrust offshore Central Sumatra. Journal of Geophysical Research: Solid Earth, 127(3), e2021JB023412. https://doi.org/10.1029/2021JB023412
  16. Pramono, A., & Wibowo, A. (2021). Spatio-temporal clustering of earthquake activity in Indonesia using ST-DBSCAN. Indonesian Journal of Computing and Cybernetics Systems, 15(2), 145–156.
  17. Putri, M. H., & Sari, D. P. (2025). Earthquake point clustering in Sumatra Island using Spatio-Temporal Density-Based Spatial Clustering Application with Noise (ST-DBSCAN) algorithm. Mathematical Journal of Modelling and Forecasting, 3(1), 8–15. https://doi.org/10.24036/mjmf.v3i1.33
  18. Supendi, P., Nugraha, A. D., Widiyantoro, S., Rawlinson, N., Clarke, D., Purbaspita, J., & Daryono, D. (2023). High-resolution seismic catalog and spatio-temporal seismicity distribution along the Sumatran Fault Zone. Geophysical Journal International, 234(2), 1251–1267. https://doi.org/10.1093/gji/ggad128
  19. Schubert, E., Sander, J., Ester, M., Kriegel, H. P., & Xu, X. (2017). DBSCAN revisited, revisited: Why and how you should (still) use DBSCAN. ACM Transactions on Database Systems (TODS), 42(3), 1–21. https://doi.org/10.1145/3068335
  20. Sieh, K., & Natawidjaja, D. (2000). Neotectonics of the Sumatran fault, Indonesia. Journal of Geophysical Research: Solid Earth, 105(B12), 28295–28326. https://doi.org/10.1029/2000JB900120
  21. Strehl, A., & Ghosh, J. (2002). Cluster ensembles—a knowledge reuse framework for combining multiple partitions. Journal of Machine Learning Research, 3, 583–617.
  22. Triyoso, W., & Suwondo, A. (2022). Probabilistic seismic hazard assessment of Sumatra based on spatial-temporal earthquake distribution and fault slip-rates. Indonesian Journal of Physics, 33(1), 15–28. https://doi.org/10.5614/itb.ijp.2022.33.1.2
  23. United States Geological Survey. (2025). ANSS Comprehensive Earthquake Catalog (ComCat). U.S. Department of the Interior. https://earthquake.usgs.gov/earthquakes
  24. Utami, R. W., Sukamto, S., & Hidayat, T. (2022). Klasterisasi pola kegempaan wilayah Indonesia berbasis Machine Learning menggunakan algoritma K-Means dan Agglomerative Hierarchical Clustering. Jurnal Geofisika Eksplorasi, 8(2), 112–124.
  25. Wiemer, S., & Wyss, M. (2000). Minimum magnitude of completeness in earthquake catalogs: Examples from Alaska, the western United States, and Japan. Bulletin of the Seismological Society of America, 90(4), 859–869. https://doi.org/10.1785/0119990114
  26. Wiemer, S. (2001). A software package to analyze seismicity: ZMAP. Seismological Research Letters, 72(3), 373–382. https://doi.org/10.1785/gssrl.72.3.373
  27. Wijaya, A., & Setiawan, B. (2023). Deteksi anomali seismik dan analisis spasiotemporal gempa bumi di Zona Subduksi Sunda. Jurnal Penginderaan Jauh dan Pengolahan Data Citra Digital, 20(1), 45–58.
  28. Xu, D., & Tian, Y. (2015). A comprehensive survey of clustering algorithms. Annals of Data Science, 2(2), 165–193. https://doi.org/10.1007/s40745-015-0040-1
  29. Yatim, M.,(2024). Analisis Spasial Kegempaan Pulau Sumatra menggunakan visualisasi data geospasial berbasis Python. Jurnal Teknologi Informasi dan Edukasi, 11(2), 88–97.
  30. Zaliapin, I., & Ben-Zion, Y. (2013). Earthquake clusters in southern California I: Identification and stability. Journal of Geophysical Research: Solid Earth, 118(6), 2847–2864. https://doi.org/10.1002/jgrb.50179
  31. Zhang, L., & Ben-Zion, Y. (2023). Automated identification of earthquake clusters and fault structures using spatial-temporal density-based clustering. Journal of Geophysical Research: Solid Earth, 128(7), e2023JB026541. https://doi.org/10.1029/2023JB026541
  32. Zhao, D., & Ohtani, E. (2009). Deep mantle structure and subduction dynamics beneath the Sumatra-Java arc. Tectonophysics, 471(3-4), 302–312.
  33. Zhao, D & Ohtanih, E. (2009) Deep slab subduction and dehydration and their geodynamic consequences: Evidence from seismology and mineral physics. Vol. 16, 401-413. https://doi.org/10.1016/j.gr.2009.01.005

Downloads

PDF

Statistic

Abstract Views
: 34 times
PDF
: 46 times