Stefania Gentili, Giuseppe Davide Chiappetta, Giuseppe Petrillo, Piero Brondi, Jiancang Zhuang
{"title":"使用改进版 NESTORE 机器学习算法预报日本强震后续地震","authors":"Stefania Gentili, Giuseppe Davide Chiappetta, Giuseppe Petrillo, Piero Brondi, Jiancang Zhuang","doi":"arxiv-2408.12956","DOIUrl":null,"url":null,"abstract":"The advanced machine learning algorithm NESTORE (Next STrOng Related\nEarthquake) was developed to forecast strong aftershocks in earthquake\nsequences and has been successfully tested in Italy, western Slovenia, Greece,\nand California. NESTORE calculates the probability of aftershocks reaching or\nexceeding the magnitude of the main earthquake minus one and classifies\nclusters as type A or B based on a 0.5 probability threshold. In this study,\nNESTORE was applied to Japan using data from the Japan Meteorological Agency\ncatalog (1973-2024). Due to Japan's high seismic activity and class imbalance,\nnew algorithms were developed to complement NESTORE. The first is a hybrid\ncluster identification method using ETAS-based stochastic declustering and\ndeterministic graph-based selection. The second, REPENESE (RElevant features,\nclass imbalance PErcentage, NEighbour detection, SElection), is optimized for\ndetecting outliers in skewed class distributions. A new seismicity feature was\nproposed, showing good results in forecasting cluster classes in Japan. Trained\nwith data from 1973 to 2004 and tested from 2005 to 2023, the method correctly\nforecasted 75% of A clusters and 96% of B clusters, achieving a precision of\n0.75 and an accuracy of 0.94 six hours after the mainshock. It accurately\nclassified the 2011 T\\=ohoku event cluster. Near-real-time forecasting was\napplied to the sequence after the April 17, 2024 M6.6 earthquake in Shikoku,\nclassifying it as a \"Type B cluster,\" with validation expected on October 31,\n2024.","PeriodicalId":501270,"journal":{"name":"arXiv - PHYS - Geophysics","volume":"21 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Forecasting Strong Subsequent Earthquakes in Japan using an improved version of NESTORE Machine Learning Algorithm\",\"authors\":\"Stefania Gentili, Giuseppe Davide Chiappetta, Giuseppe Petrillo, Piero Brondi, Jiancang Zhuang\",\"doi\":\"arxiv-2408.12956\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The advanced machine learning algorithm NESTORE (Next STrOng Related\\nEarthquake) was developed to forecast strong aftershocks in earthquake\\nsequences and has been successfully tested in Italy, western Slovenia, Greece,\\nand California. NESTORE calculates the probability of aftershocks reaching or\\nexceeding the magnitude of the main earthquake minus one and classifies\\nclusters as type A or B based on a 0.5 probability threshold. In this study,\\nNESTORE was applied to Japan using data from the Japan Meteorological Agency\\ncatalog (1973-2024). Due to Japan's high seismic activity and class imbalance,\\nnew algorithms were developed to complement NESTORE. The first is a hybrid\\ncluster identification method using ETAS-based stochastic declustering and\\ndeterministic graph-based selection. The second, REPENESE (RElevant features,\\nclass imbalance PErcentage, NEighbour detection, SElection), is optimized for\\ndetecting outliers in skewed class distributions. A new seismicity feature was\\nproposed, showing good results in forecasting cluster classes in Japan. Trained\\nwith data from 1973 to 2004 and tested from 2005 to 2023, the method correctly\\nforecasted 75% of A clusters and 96% of B clusters, achieving a precision of\\n0.75 and an accuracy of 0.94 six hours after the mainshock. It accurately\\nclassified the 2011 T\\\\=ohoku event cluster. Near-real-time forecasting was\\napplied to the sequence after the April 17, 2024 M6.6 earthquake in Shikoku,\\nclassifying it as a \\\"Type B cluster,\\\" with validation expected on October 31,\\n2024.\",\"PeriodicalId\":501270,\"journal\":{\"name\":\"arXiv - PHYS - Geophysics\",\"volume\":\"21 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-08-23\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - PHYS - Geophysics\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2408.12956\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - PHYS - Geophysics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2408.12956","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Forecasting Strong Subsequent Earthquakes in Japan using an improved version of NESTORE Machine Learning Algorithm
The advanced machine learning algorithm NESTORE (Next STrOng Related
Earthquake) was developed to forecast strong aftershocks in earthquake
sequences and has been successfully tested in Italy, western Slovenia, Greece,
and California. NESTORE calculates the probability of aftershocks reaching or
exceeding the magnitude of the main earthquake minus one and classifies
clusters as type A or B based on a 0.5 probability threshold. In this study,
NESTORE was applied to Japan using data from the Japan Meteorological Agency
catalog (1973-2024). Due to Japan's high seismic activity and class imbalance,
new algorithms were developed to complement NESTORE. The first is a hybrid
cluster identification method using ETAS-based stochastic declustering and
deterministic graph-based selection. The second, REPENESE (RElevant features,
class imbalance PErcentage, NEighbour detection, SElection), is optimized for
detecting outliers in skewed class distributions. A new seismicity feature was
proposed, showing good results in forecasting cluster classes in Japan. Trained
with data from 1973 to 2004 and tested from 2005 to 2023, the method correctly
forecasted 75% of A clusters and 96% of B clusters, achieving a precision of
0.75 and an accuracy of 0.94 six hours after the mainshock. It accurately
classified the 2011 T\=ohoku event cluster. Near-real-time forecasting was
applied to the sequence after the April 17, 2024 M6.6 earthquake in Shikoku,
classifying it as a "Type B cluster," with validation expected on October 31,
2024.