Analyzing correlations between GNSS retrieved precipitable water vapor and land surface temperature after earthquakes occurrence

• Published in: The Science of the total environment 2023-05, Vol.872, p.162225-162225, Article 162225
• Main Authors: Guo, Ao, Xu, Yan, Jiang, Nan, Wu, Yuhao, Gao, Zhaorui, Li, Song, Xu, Tianhe, Bastos, Luísa
• Format: Article
• Language: English
• Subjects: Coupling; Earthquake; GNSS; Precipitable water vapor; Thermal anomaly
• ISSN: 0048-9697; 1879-1026
• DOI: 10.1016/j.scitotenv.2023.162225

Earthquake is a common and destructive natural disaster. The enormous amount of energy released from seismic events can result in anomalous land surface temperature (LST) and catalyze the accumulation of water vapor in the atmosphere. The majority of previous works are not consensual concerning precipitable water vapor (PWV) and LST after the earthquake. Here, we utilized multi-source data to analyze the changes of PWV and LST anomaly after three Ms 4.0–5.3 crustal earthquakes at low depth (8–9 km) that occurred in Qinghai-Tibet Plateau. Firstly, PWV retrieval using Global Navigation Satellite System (GNSS) technology is performed, showing that its root mean square error (RMSE) is less than 1.8 mm against radiosonde (RS) data or European Centre for Medium-Range Weather Forecasts (ECMWF) Reanalysis 5 (ERA5) PWV data. The PWV change derived from the nearest GNSS stations around the hypocenter during the earthquakes shows anomalies, and the results reveal that PWV anomalies occurred after the earthquakes, mainly obeying a trend of increasing first and then decreasing. In addition, LST increases three days before PWV peak with a thermal anomaly of 12 °C higher than that of previous days. Robust Satellite Technique (RST) algorithm and ALICE index on Moderate Resolution Imaging Spectroradiometer (MODIS) LST products are introduced to analyze the correlation between the abnormality of LST and PWV. Based on ten-year background field data (2012–2021), the results show that LST during the earthquake has more thermal anomaly occurrences than in previous years. The more severe the LST thermal anomaly is, the higher the probability of a PWV peak occurring. [Display omitted] •The RMSE of GNSS-PWV against RS-PWV or ERA5-PWV were less than 1.8 mm during three earthquakes.•GNSS-PWV increased and subsequently decreased over the epicenter after the major events.•Thermal anomalies of land surface temperature were detected in two temporal scales after the earthquakes.