ABSTRACT Intensive physical/chemical changes take place in earthquake preparation zone prior to major earthquake events. These changes result in anomalous variations of atmospheric parameters. In this study, anomalous variations of Surface Latent Heat Flux (SLHF) and Air Temperature (AT) are investigated prior to 17 major earthquake events with magnitude ≥ 5.3 Mw in active tectonic regions of the world for the period from 2004 to 2020 using InterQuartile Range (IQR) and Hierarchal Clustering (HC) statistical techniques. Studied earthquakes are selected from subduction, transform and intraplate locations. The results clearly show anomalous variations in the SLHF and AT atmospheric parameters few days to few weeks prior to the occurrence of the earthquake event. The Honshu earthquake event is captured 54 days prior to the event using SLHF data, whereas Bella Bella earthquake event is observed 39 days prior to the event using AT data. Best anomalies using both SLHF and AT are identified for Pasni and Norcia earthquake events, whereas lowest anomalies using both SLHF and AT are identified for Nsunga and Puerto Quellon events. The anomalous variations of SLHF and AT prior to the earthquake event show a definite precursor pattern. Strong anomalies are observed in subduction/thrust regime whereas, minimum to moderate anomalies are observed along strike slip faults. This study determined the threshold values of 59.52 W/m2 and 2.87 K for SLHF and AT respectively. The HC revealed more than 70% similarity among earthquakes based on depth, magnitude, SLHF and AT. Mechanism of energy exchange between lithosphere and atmosphere can be understood by analyzing significant rise in SLHF and AT. It is essential to monitor these variations continuously to properly identify earthquake related anomalies.

中文翻译：

---

世界活动构造区地震热前兆调查

摘要 在大地震发生之前，地震准备区发生了剧烈的物理/化学变化。这些变化导致大气参数的异常变化。在这项研究中，使用四分位距调查了 2004 年至 2020 年世界活动构造区发生的 17 次 5.3 Mw 级以上大地震事件之前的地表潜热通量 (SLHF) 和气温 (AT) 的异常变化(IQR) 和分层聚类 (HC) 统计技术。研究的地震选自俯冲、转换和板内位置。结果清楚地显示了地震事件发生前几天到几周 SLHF 和 AT 大气参数的异常变化。本州地震事件是在事件发生前 54 天使用 SLHF 数据捕获的，而使用 AT 数据在事件发生前 39 天观察到 Bella Bella 地震事件。使用 SLHF 和 AT 的最佳异常被确定为 Pasni 和 Norcia 地震事件，而使用 SLHF 和 AT 的最低异常被确定为 Nsunga 和 Puerto Quellon 事件。地震事件前SLHF和AT的异常变化显示出明确的前兆模式。在俯冲/逆冲区观察到强烈的异常，而沿着走滑断层观察到最小到中等的异常。本研究确定了 SLHF 和 AT 的阈值分别为 59.52 W/m2 和 2.87 K。根据深度、震级、SLHF 和 AT，HC 显示地震之间的相似性超过 70%。通过分析 SLHF 和 AT 的显着上升，可以理解岩石圈与大气之间的能量交换机制。