Gianfranco Vannucci, Paolo Gasperini, Laura Gulia, Barbara Lolli
{"title":"Earthquakes Parameters from Citizen Testimonies: A Retrospective Analysis of EMSC Database","authors":"Gianfranco Vannucci, Paolo Gasperini, Laura Gulia, Barbara Lolli","doi":"10.1785/0220230245","DOIUrl":null,"url":null,"abstract":"Abstract We aim to compute macroseismic parameters (location and magnitude) using the BOXER code for the first time on the citizen testimonies, that is, individual intensity data points (IDPs) at the global scale collected and made available by the LastQuake system of the European–Mediterranean Seismological Centre (EMSC). IDPs available for different earthquakes are selected to eliminate those that are geographically inconsistent with most data; then they are clustered spatially based on various methods. For each cluster with at least three IDPs, a macroseismic data point (MDP), corresponding to an intensity value assessed for given localities as in classical macroseismic studies, is computed by various central tendency estimators (average, median, and trimmed averages). Finally, macroseismic parameters are obtained by MDP distribution using two location methods of BOXER code. For each earthquake, we used raw and corrected intensities and 132 different combinations of grouping methods, estimators, and BOXER methods. We assigned a ranking to the combinations that best reproduce instrumental parameters and used such a ranking to select preferred combinations for each earthquake. We analyzed retrospectively the reliability of the parameters as a function of time and space. The results are essentially identical using original and corrected intensities and show higher reliability for BOXER’s method 1 than for method 0; they are dependent on the geographical area, and generally improve over time and with the number of IDPs collected. These findings are useful for the future real-time analyses, and for evaluating the location and magnitude of earthquakes whenever a sufficient number of IDPs are available and with a distribution such that MDPs can be derived and the BOXER method applied.","PeriodicalId":21687,"journal":{"name":"Seismological Research Letters","volume":"44 5 1","pages":"0"},"PeriodicalIF":2.6000,"publicationDate":"2023-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Seismological Research Letters","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1785/0220230245","RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"GEOCHEMISTRY & GEOPHYSICS","Score":null,"Total":0}
引用次数: 0
Abstract
Abstract We aim to compute macroseismic parameters (location and magnitude) using the BOXER code for the first time on the citizen testimonies, that is, individual intensity data points (IDPs) at the global scale collected and made available by the LastQuake system of the European–Mediterranean Seismological Centre (EMSC). IDPs available for different earthquakes are selected to eliminate those that are geographically inconsistent with most data; then they are clustered spatially based on various methods. For each cluster with at least three IDPs, a macroseismic data point (MDP), corresponding to an intensity value assessed for given localities as in classical macroseismic studies, is computed by various central tendency estimators (average, median, and trimmed averages). Finally, macroseismic parameters are obtained by MDP distribution using two location methods of BOXER code. For each earthquake, we used raw and corrected intensities and 132 different combinations of grouping methods, estimators, and BOXER methods. We assigned a ranking to the combinations that best reproduce instrumental parameters and used such a ranking to select preferred combinations for each earthquake. We analyzed retrospectively the reliability of the parameters as a function of time and space. The results are essentially identical using original and corrected intensities and show higher reliability for BOXER’s method 1 than for method 0; they are dependent on the geographical area, and generally improve over time and with the number of IDPs collected. These findings are useful for the future real-time analyses, and for evaluating the location and magnitude of earthquakes whenever a sufficient number of IDPs are available and with a distribution such that MDPs can be derived and the BOXER method applied.