Browse:
/
/
NEW JOURNAL PUBLICATION: Backbone ground motion model through simulated records and XGBoost machine learning algorithm: An application for the Azores plateau (Portugal):

NEW JOURNAL PUBLICATION: Backbone ground motion model through simulated records and XGBoost machine learning algorithm: An application for the Azores plateau (Portugal):

In alignment with the objectives articulated in Work Package 1 (WP1) of the S4H project, the collaborative efforts of the S4H team, in partnership with Amirhossein Mohammadi from the University of Minho and Dr Alexandra Carvalho from LNEC, have resulted in the recent publication of a paper in the journal of Earthquake Engineering and Structural Dynamics. This publication intricately delves into the team’s investigation into the efficacy of integrating stochastic ground motion simulations, uncertainty propagation and advanced machine learning (ML) techniques, specifically Extreme Gradient Boosting (XGBoost). 

The Azores Islands undergo seismic activity due to the tectonic structure of the area. Since the 15th century, they have experienced periodic impacts from approximately 33 moderate to strong earthquakes, with the most recent occurrence recorded in 1998 (Magnitude = 6.2). However, the absence of comprehensive instrumental seismic data has led to a lack of a consistent database of real records in the region. 

The primary objective of this research is to forecast the full-time series of ground motions for scenario earthquake events, specifically concentrating on the Azores region in Portugal. Through the integration of these simulations, incorporating modelling uncertainty, and proposing a cohesive simulated ground motion dataset, the study puts forward an innovative ground motion model utilising the XGBoost algorithm. This model is designed to predict the intensity levels of ground motion for potential events in the Azores Plateau.

Additionally, the researchers have taken a step further by developing an online platform that enhances accessibility to the models for interested users. The source codes for these models are readily available at [https://github.com/amirxdbx/GMM_Azores]. For more comprehensive information, please visit [https://amirxdbx-gmm-azores-deploy-36glao.streamlit.app/]. The published version of the paper can also be accessed through the following link [https://stand4heritage.org/publications/backbone-ground-motion-model-through-simulated-records-and-xgboost-machine-learning-algorithm-an-application-for-the-azores-plateau-portugal/].

Share this post