Scholarlink

<p>With the support of the Earthquake Science Spark Project, the project team has successfully developed an intelligent geomagnetic monitoring terminal suitable for high-density and rapid deployment, in addition to collecting geomagnetic data, this device simultaneously records environmental data such as gravitational acceleration, temperature, and atmospheric pressure, laying the foundation for big data diversity analysis and the construction of datasets correlating observation data features. Currently, the geomagnetic observation demonstration zone in Lincang City, Yunnan Province, has 22 monitoring stations, with an average station spacing of 33 km. Since the demonstration network was established, the project team has analyzed the characteristics of geomagnetic Z-component variations before significant seismic events in the demonstration area using traditional methods such as the loading/unloading response ratio (LURR) and daily variation correlation.Through the aforementioned research, the characteristic patterns of pre-earthquake geomagnetic anomalies have been identified. Leveraging the characteristics of diverse, high-volume time-series data from the demonstration zone’s intelligent geomagnetic monitoring terminals, we introduce a probabilistic time-prediction foundational model to forecast geomagnetic variation trends and provide earthquake probability predictions based on anomaly detection.</p>

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

Research on Magnetic Z-component Anomalies and Probabilistic Time Series Prediction Based on High-Density Geomagnetic Data in Southwest Yunnan

  • 期刊论文

  • 发表日期：

    2025年10月27日

查看全文
  • Lu-Qiang Sun,
  • Tian Song,
  • Chao-Qun Ma,
  • Yan-Li Liu,
  • Ming-Dong Zhang

摘要

With the support of the Earthquake Science Spark Project, the project team has successfully developed an intelligent geomagnetic monitoring terminal suitable for high-density and rapid deployment, in addition to collecting geomagnetic data, this device simultaneously records environmental data such as gravitational acceleration, temperature, and atmospheric pressure, laying the foundation for big data diversity analysis and the construction of datasets correlating observation data features. Currently, the geomagnetic observation demonstration zone in Lincang City, Yunnan Province, has 22 monitoring stations, with an average station spacing of 33 km. Since the demonstration network was established, the project team has analyzed the characteristics of geomagnetic Z-component variations before significant seismic events in the demonstration area using traditional methods such as the loading/unloading response ratio (LURR) and daily variation correlation.Through the aforementioned research, the characteristic patterns of pre-earthquake geomagnetic anomalies have been identified. Leveraging the characteristics of diverse, high-volume time-series data from the demonstration zone’s intelligent geomagnetic monitoring terminals, we introduce a probabilistic time-prediction foundational model to forecast geomagnetic variation trends and provide earthquake probability predictions based on anomaly detection.