Past seismic events worldwide demonstrated thatdamage and death toll depend on both the strong ground motion (i.e., source effects) and the local site effects. The variability of earthquake ground motion distribution is caused bythe local stratigraphic and/or topographic setting and buriedmorphologies (e.g., irregular sub-interface between soft andstiff soils) that can give rise to amplification and resonanceswith respect to the ground motion expected at the referencesite. Therefore, local site conditions can affect an area withdamage related to the full collapse or loss in functionality offacilities, roads, pipelines, and other lifelines. To this concern, the near-real-time prediction of ground motion variation over large areas is a crucial issue to support the rescue and operational interventions. A machine learning approach was adopted to produce ground motion predictionmaps considering both stratigraphic and morphological conditions. A set of about 16 000 accelerometric data points andabout 46 000 geological and geophysical data points wasretrieved from Italian and European databases. The intensity measures of interest were estimated based on nine input proxies. The adopted machine learning regression model(i.e., Gaussian process regression) allows for improving boththe precision and the accuracy in the estimation of the intensity measures with respect to the available near-real-timeprediction methods (i.e., ground motion prediction equationand ShakeMaps). In addition, maps with a 50 m × 50 m resolution were generated, providing a ground motion variabilityin agreement with the results of advanced numerical simulations based on detailed subsoil models.
Ground motion prediction maps using seismic-microzonation data and machine learning
Federico Mori;Amerigo Mendicelli;Gaetano Falcone;Gianluca Acunzo;Rose Line Spacagna;Massimiliano Moscatelli
2022
Abstract
Past seismic events worldwide demonstrated thatdamage and death toll depend on both the strong ground motion (i.e., source effects) and the local site effects. The variability of earthquake ground motion distribution is caused bythe local stratigraphic and/or topographic setting and buriedmorphologies (e.g., irregular sub-interface between soft andstiff soils) that can give rise to amplification and resonanceswith respect to the ground motion expected at the referencesite. Therefore, local site conditions can affect an area withdamage related to the full collapse or loss in functionality offacilities, roads, pipelines, and other lifelines. To this concern, the near-real-time prediction of ground motion variation over large areas is a crucial issue to support the rescue and operational interventions. A machine learning approach was adopted to produce ground motion predictionmaps considering both stratigraphic and morphological conditions. A set of about 16 000 accelerometric data points andabout 46 000 geological and geophysical data points wasretrieved from Italian and European databases. The intensity measures of interest were estimated based on nine input proxies. The adopted machine learning regression model(i.e., Gaussian process regression) allows for improving boththe precision and the accuracy in the estimation of the intensity measures with respect to the available near-real-timeprediction methods (i.e., ground motion prediction equationand ShakeMaps). In addition, maps with a 50 m × 50 m resolution were generated, providing a ground motion variabilityin agreement with the results of advanced numerical simulations based on detailed subsoil models.File | Dimensione | Formato | |
---|---|---|---|
prod_489883-doc_204045.pdf
accesso aperto
Descrizione: Ground motion prediction maps using seismic-microzonation data and machine learning
Tipologia:
Versione Editoriale (PDF)
Licenza:
Creative commons
Dimensione
28.14 MB
Formato
Adobe PDF
|
28.14 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.