GIbran, Maulana Teuku Raffly (2026) Analisis Klastering Otomatis Gempa Vulkanik di Gunung Ibu Berdasarkan Korelasi Silang Menggunakan REDPy: Studi Kasus Periode 31 Oktober-29 November 2024. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5001221037_Undergratuate Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5001221037_Undergratuate Thesis.pdf - Accepted Version Restricted to Repository staff only Download (16MB) | Request a copy |
Abstract
Gunung Ibu adalah gunung api stratovolkano aktif di bagian barat laut Pulau Halmahera yang men catat lebih dari 2.000 kejadian letusan pada tahun 2024 dan menjadikannya gunung api paling aktif di Indonesia pada tahun tersebut. Penelitian seismologi kuantitatif terhadap Gunung Ibu masih terbatas dan belum mencakup pemanfaatan metode klastering otomatis berbasis korelasi silang bentuk gelombang. Penelitian ini menerapkan perangkat lunak Repeating Earthquake Detector in Python (REDPy) untuk mendeteksi dan mengklasterkan gempa berulang dari rekaman seismik kontinu dua stasiun PVMBG (GIN komponen EHZ dan IBKC komponen HHZ) pada periode 31 Oktober hingga 29 November 2024. Optimasi parameter dilakukan melalui pengujian sepuluh konfigurasi yang memvariasikan bandpass fil ter, algoritma STA/LTA, ambang korelasi silang, dan jumlah stasiun. Evaluasi kuantitatif dilakukan terhadap katalog manual yang berisi 1.593 event pada subset 4 hingga 10 November 2024 menggunakan metrik precision, recall, F1-Score, dan weighted score gabungan. Konfigurasi optimal (ibufinal) meng gunakan bandpass 1 sampai 15 Hz, algoritma classic STA/LTA, dua stasiun, dan ambang korelasi silang 0,70. Konfigurasi ini memperoleh F1-Score 57,0%, recall 89,6%, dan precision 41,8% yang diinter pretasi sebagai batas bawah karena keterbatasan kelengkapan ground truth. Konfigurasi optimal meng hasilkan 3.312 event yang terklasterkan ke dalam 152 famili dan terklasifikasi ke tujuh tipe gempa standar PVMBG dengan dominasi Letusan 79,9%, Hembusan 9,7%, Vulkanik Dangkal 5,9%, Low-Frequency 2,4%, Vulkanik Dalam 0,9%, Tektonik Lokal 0,7%, dan Tektonik Jauh 0,4%. Distribusi tersebut mencer minkan karakter sistem vulkanik open-vent dengan kubah lava dasit yang aktif. Hubungan temporal antar tipe gempa terlihat dari gempa Vulkanik Dalam yang berpuncak pada 5 November dan mendahului pun cak kejadian gempa Letusan pada 7 November 2024. Hal ini mengindikasikan bahwa kejadian letusan di Gunung Ibu tidak hanya berasal dari sumber dangkal, namun juga masih dikontrol oleh suplai magma di kedalaman.
================================================================================================================================================
Mount Ibu is an active stratovolcano located on the northwestern part of Halmahera Island that recorded more than 2,000 eruption events in 2024, making it the most active volcano in Indonesia that year. Quantitative seismological studies of Mount Ibu remain limited and have not yet utilized automated clustering methods based on waveform cross-correlation. This study applied the Repeating Earthquake Detector in Python (REDPy) software to detect and cluster repeating earthquakes from continuous seis mic recordings of two PVMBG stations (GIN with EHZ component and IBKC with HHZ component) during the period of 31 October to 29 November 2024. Parameter optimization was performed by test ing ten configurations that varied the bandpass filter, STA/LTA algorithm, cross-correlation threshold, and number of stations. Quantitative evaluation was conducted against a manual catalog containing 1,593 events for the subset of 4 to 10 November 2024 using precision, recall, F1-Score, and a com bined weighted score. The optimal configuration (ibufinal) employed a bandpass of 1 to 15 Hz, classic STA/LTA algorithm, two stations, and a cross-correlation threshold of 0.70. This configuration achieved an F1-Score of 57.0%, recall of 89.6%, and precision of 41.8% which is interpreted as a lower bound due to ground truth completeness limitations. The optimal configuration produced 3,312 events clustered into 152 families and classified into seven standard PVMBG earthquake types, dominated by Eruption events at 79.9%, followed by Emission 9.7%, Shallow Volcanic 5.9%, Low-Frequency 2.4%, Deep Vol canic 0.9%, Local Tectonic 0.7%, and Distant Tectonic 0.4%. This distribution reflects the character of an open-vent volcanic system with an actively growing dacitic lava dome. The temporal relationship between earthquake types is shown by the peak of Deep Volcanic events on 5 November preceding the peak of Eruption events on 7 November 2024. This indicates that eruption events at Mount Ibu do not originate solely from shallow sources but are also controlled by magma supply from depth.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Frequency Index, Gempa Berulang, Gunung Ibu, Klastering, Klasifikasi Gempa Vulkanik, REDPy Clustering, Frequency Index, Mount Ibu, REDPy, Repeating Earthquake, Volcanic Earthquake Classification |
| Subjects: | Q Science > QC Physics Q Science > QE Geology Q Science > QE Geology > QE538.8 Earthquakes. Seismology |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Physics > 45201-(S1) Undergraduate Thesis |
| Depositing User: | Maulana Teuku Raffly Gibran |
| Date Deposited: | 01 Sep 2026 03:28 |
| Last Modified: | 01 Sep 2026 03:28 |
| URI: | http://repository.its.ac.id/id/eprint/141819 |
Actions (login required)
 |
View Item |