Scholarlink

Total electron content (TEC) data from GPS are used as a precursor of an impending earthquake. The extraction of earthquake-induced signatures based on day-to-day variations in TEC peak and profile shape is challenging since this parameter needs elaborate processing because it involves filtration of data with respect to disturbed day variations, contribution from multipath effects and also normal day-to-day fluctuations even during quiet days. In this research, an attempt is made by utilizing matrix profile based machine learning technique for detection of anomaly, if any, in TEC time series data before an impending earthquake. A few case studies using this algorithm are presented in this paper. For analysis, earthquake events of equatorial anomaly region are considered. The results demonstrate the effectiveness of the proposed approach in detecting earthquake precursors. The model exhibit high accuracy, sensitivity, and specificity in identifying seismic-related anomalies in the ionospheric data.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Machine Learning-Based Detection of Earthquake Precursors in Ionospheric Time Series Data at Anomaly Crest Station

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Santanu Kalita,
  • Mridupaban Bora,
  • Bornali Chetia,
  • Hirak Jyoti Goswami,
  • Naba Jyoti Sarmah

摘要

Total electron content (TEC) data from GPS are used as a precursor of an impending earthquake. The extraction of earthquake-induced signatures based on day-to-day variations in TEC peak and profile shape is challenging since this parameter needs elaborate processing because it involves filtration of data with respect to disturbed day variations, contribution from multipath effects and also normal day-to-day fluctuations even during quiet days. In this research, an attempt is made by utilizing matrix profile based machine learning technique for detection of anomaly, if any, in TEC time series data before an impending earthquake. A few case studies using this algorithm are presented in this paper. For analysis, earthquake events of equatorial anomaly region are considered. The results demonstrate the effectiveness of the proposed approach in detecting earthquake precursors. The model exhibit high accuracy, sensitivity, and specificity in identifying seismic-related anomalies in the ionospheric data.