Segmentation characteristics of deep, low-frequency tremors in Shikoku, Japan using machine learning approaches
Abstract Shikoku island, southwestern Japan lies in the western Nankai Trough and showcases along-strike segmentation of slow earthquake behavior. Whether the spatial variation of tremor behavior reflects the regional differences in structure/source properties and how much such differences can be re...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2023-03-01
|
| Series: | Earth, Planets and Space |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s40623-023-01776-w |
| _version_ | 1797865343358599168 |
|---|---|
| author | Kate Huihsuan Chen Hao-Yu Chiu Kazushige Obara Yi-Hung Liu |
| author_facet | Kate Huihsuan Chen Hao-Yu Chiu Kazushige Obara Yi-Hung Liu |
| author_sort | Kate Huihsuan Chen |
| collection | DOAJ |
| description | Abstract Shikoku island, southwestern Japan lies in the western Nankai Trough and showcases along-strike segmentation of slow earthquake behavior. Whether the spatial variation of tremor behavior reflects the regional differences in structure/source properties and how much such differences can be recognized by the seismic signals themselves are two questions addressed in this paper. Taking advantage of advanced methods in recognizing and classifying signals using machine learning approaches, we attempt to answer them by conducting signal classification experiments in Shikoku. Based on the tremor catalog from 1 June 2014 to 31 March 2015, the tremors recorded in four different areas were treated as different classes and segmented into 60-s-long signals. The number of tremors in four different areas (A to D, from west to east) reached 15,000, 31,000, 10,000, and 16,000, respectively. To efficiently distinguish between tremors from different areas, we applied a k-nearest neighbor (k-NN) classifier with Fisher’s class separability criteria to select the optimal feature subset. The resulting classification performance reached more than 90% at all 12 stations. We further designed a triangle test to select the features that can better represent the differences in source properties between areas. We found that the most efficient features were associated with (1) the number of peaks in the temporal evolution of discrete Fourier transforms and (2) the energy distribution in the autocorrelation function (ACF). To match the difference in behavior revealed by the ACF, the size of the tremor zone, which mainly controls how long the seismic energy lasts in a tremor episode, was determined to be largest in Area B and smallest in Area C. The heterogeneity of the asperities in a tremor zone, which may control how spiky the tremor signals developed over time, was determined to be strong in Areas B and C. Together with previously documented variations in slow earthquake behavior in the same area, we finally propose a conceptual model that provides a better understanding of the regional differences in the tremor source properties in Shikoku, Japan. Graphical Abstract |
| first_indexed | 2024-04-09T23:06:31Z |
| format | Article |
| id | doaj.art-9fc5f5761ccb4ad792917c6e624fa903 |
| institution | Directory Open Access Journal |
| issn | 1880-5981 |
| language | English |
| last_indexed | 2024-04-09T23:06:31Z |
| publishDate | 2023-03-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | Earth, Planets and Space |
| spelling | doaj.art-9fc5f5761ccb4ad792917c6e624fa9032023-03-22T10:39:53ZengSpringerOpenEarth, Planets and Space1880-59812023-03-0175111910.1186/s40623-023-01776-wSegmentation characteristics of deep, low-frequency tremors in Shikoku, Japan using machine learning approachesKate Huihsuan Chen0Hao-Yu Chiu1Kazushige Obara2Yi-Hung Liu3Department of Earth Sciences, National Taiwan Normal UniversityDepartment of Geosciences, National Taiwan UniversityEarthquake Research Institute, the University of TokyoDepartment of Mechanical Engineering, National Taiwan University of Science and TechnologyAbstract Shikoku island, southwestern Japan lies in the western Nankai Trough and showcases along-strike segmentation of slow earthquake behavior. Whether the spatial variation of tremor behavior reflects the regional differences in structure/source properties and how much such differences can be recognized by the seismic signals themselves are two questions addressed in this paper. Taking advantage of advanced methods in recognizing and classifying signals using machine learning approaches, we attempt to answer them by conducting signal classification experiments in Shikoku. Based on the tremor catalog from 1 June 2014 to 31 March 2015, the tremors recorded in four different areas were treated as different classes and segmented into 60-s-long signals. The number of tremors in four different areas (A to D, from west to east) reached 15,000, 31,000, 10,000, and 16,000, respectively. To efficiently distinguish between tremors from different areas, we applied a k-nearest neighbor (k-NN) classifier with Fisher’s class separability criteria to select the optimal feature subset. The resulting classification performance reached more than 90% at all 12 stations. We further designed a triangle test to select the features that can better represent the differences in source properties between areas. We found that the most efficient features were associated with (1) the number of peaks in the temporal evolution of discrete Fourier transforms and (2) the energy distribution in the autocorrelation function (ACF). To match the difference in behavior revealed by the ACF, the size of the tremor zone, which mainly controls how long the seismic energy lasts in a tremor episode, was determined to be largest in Area B and smallest in Area C. The heterogeneity of the asperities in a tremor zone, which may control how spiky the tremor signals developed over time, was determined to be strong in Areas B and C. Together with previously documented variations in slow earthquake behavior in the same area, we finally propose a conceptual model that provides a better understanding of the regional differences in the tremor source properties in Shikoku, Japan. Graphical Abstracthttps://doi.org/10.1186/s40623-023-01776-wMachine learningSignal classificationDeep low-frequency tremork-nearest neighborHi-netShikoku |
| spellingShingle | Kate Huihsuan Chen Hao-Yu Chiu Kazushige Obara Yi-Hung Liu Segmentation characteristics of deep, low-frequency tremors in Shikoku, Japan using machine learning approaches Earth, Planets and Space Machine learning Signal classification Deep low-frequency tremor k-nearest neighbor Hi-net Shikoku |
| title | Segmentation characteristics of deep, low-frequency tremors in Shikoku, Japan using machine learning approaches |
| title_full | Segmentation characteristics of deep, low-frequency tremors in Shikoku, Japan using machine learning approaches |
| title_fullStr | Segmentation characteristics of deep, low-frequency tremors in Shikoku, Japan using machine learning approaches |
| title_full_unstemmed | Segmentation characteristics of deep, low-frequency tremors in Shikoku, Japan using machine learning approaches |
| title_short | Segmentation characteristics of deep, low-frequency tremors in Shikoku, Japan using machine learning approaches |
| title_sort | segmentation characteristics of deep low frequency tremors in shikoku japan using machine learning approaches |
| topic | Machine learning Signal classification Deep low-frequency tremor k-nearest neighbor Hi-net Shikoku |
| url | https://doi.org/10.1186/s40623-023-01776-w |
| work_keys_str_mv | AT katehuihsuanchen segmentationcharacteristicsofdeeplowfrequencytremorsinshikokujapanusingmachinelearningapproaches AT haoyuchiu segmentationcharacteristicsofdeeplowfrequencytremorsinshikokujapanusingmachinelearningapproaches AT kazushigeobara segmentationcharacteristicsofdeeplowfrequencytremorsinshikokujapanusingmachinelearningapproaches AT yihungliu segmentationcharacteristicsofdeeplowfrequencytremorsinshikokujapanusingmachinelearningapproaches |