Analysis and Simulation of Improved Seismic Data using Adaptive Processing by National Data Centre Iraq

Abstract

Abstract All States Parties have convenient access to all International Monitoring System (IMS) data, International Data Center (IDC) products, and all applications and scientific studies programs used in the IDC of the Comprehensive Nuclear-Test-Ban Treaty Organization (CTBTO). Integrating machine learning with seismic exploration is crucial for obtaining accurate and essential information about subsurface formations' stratigraphic structure, lithology, and porosity. However, there still needs to be a clear understanding of which algorithm produces the most accurate earthquake detection. Consequently, this study aims to perform a comparative analysis of the effectiveness of LSTM, CNN, MLP, and SVM algorithms in earthquake detection. This study has used various earthquake datasets from Arabian Sea earthquake seismic event on 26 October 2022 at 23:00:07 which was detected by two IMS monitoring technologies and non-IMS which is implemented in this work is the inclusion of analysis data using HA1 IMS (hydroacoustic station) with event location integration with seismic data for stations near that seismic event. As well as seismic event which was also studied and evaluated in Turkey, detected by IMS and non-IMS stations in IRIS on 23 November 2022 at 01:08:15 and analyzed the performance of each algorithm on these datasets by applying numerous performance metrics related to accuracy, precision, recall, F1-score, and others. The output performance results have demonstrated that the CNN network outperforms all other algorithms. In conclusion, this study provides a comprehensive evaluation of the existing literature on digital signal processing techniques employed in the analysis and detection of seismic waves.