Origins of Thermal Anomalies Prior to Earthquakes: Lithosphere, Atmosphere, or Ionosphere?

Thermal infrared (TIR) anomalies preceding earthquakes have been widely studied as potential precursors, yet their detection remains challenged by atmospheric noise and methodological limitations. This study advances the analysis of thermal anomalies by integrating a three-dimensional (3D) “heating core” filter with National Centers for Environmental Prediction (NCEP) multilayer temperature data across 26 isobaric levels (1000-10 hPa). We systematically assess the spatiotemporal relationship between 3D thermal anomalies and earthquakes (ML≥3.0,4.0,5.0) in mainland China (2008-2019). Results reveal distinct vertical stratification: thermal anomalies in the troposphere ( ≥ 200 hPa) exhibit a pyramidal structure originating near the epicenter, correlating strongly with seismic events, while anomalies in the mesosphere (≤ 50 hPa) suggest a secondary source potentially linked to lithosphere-atmosphere-ionosphere coupling via VLF-induced heating. The 3D approach significantly outperforms 2D methods, reducing non-seismic noise and achieving optimal parameterization (NL= 4,μ = 0.8–1.0 ) for magnitude-dependent anomaly penetration. These findings highlight the critical role of vertical profiling in enhancing precursor reliability and provide statistical evidence for multi-layer thermal dynamics preceding earthquakes.