Longwave radiation anomalies associated with seismic activity in the northern Tibetan Plateau

Yifei Gong¹Xiao Guo^1*Yuansheng Zhang¹Meijiao Zhong¹Dingnan Guo¹Yuanyan Xu²Qiulin Qin¹Yingwen Fu¹

• ¹Lanzhou Institute of Seismology, Lanzhou, Gansu Province, China
• ²School of Atmospheric Sciences, Nanjing University of Information Science and Technology, Nanjing, China, China

Seismically active northern Tibetan Plateau necessitates the application of remote sensing-driven earthquake precursor detection. We applied wavelet transform and power spectrum analysis to outgoing longwave radiation (OLR) from 37 M  5.0 earthquakes (2009–2024; 89°–105°E, 31°–43°N), establishing a magnitude prediction model. The results show that: (1) Frequency band 6 demonstrates optimal retrospective detection performance for M5.0-7.0 earthquakes, whereas Band 3 captures anomalies associated with M  7.0 events; (2) Anomaly area and maximum ratio quantify energy release in M5.0-6.0 earthquakes, while maximum anomaly ratio estimates magnitudes of M6.0-7.0 events; (3) The Qilian Mountains and Bayan Har Moun-tains exhibit high seismicity due to arid surface conditions, whereas the Qaidam Basin shows elevated OLR power spectrum detection rates attributed to its rigid substrate. Seasonally, summer's warmer temperatures and atmospheric stability result in OLR anomalies accounting for 34.4% of observations, while winter snow cover induces signal attenuation; (4) Normal faults achieve 100% detection via efficient shallow heat transfer from tensile ruptures, contrasting with a 20% missed detection rate in low-dip reverse faults. The proposed split-band multi-parameter remote sensing prediction framework advances medium-to-strong earthquake risk assessment on the Tibetan Plateau, validating thermal infrared remote sensing's unique capability in precursor identification.