Identification And Comparison Of Thermal Anomalies Of Two Ms7.3 Earthquakes In Yutian Based On MODIS Data

Earthquake is the most serious and destructive sudden solid geographical disaster to human beings.Earthquake activity monitoring and early warning has become one of the most important contents to maintain national public security.A large number of observation facts and studies show that seismic activity is accompanied by thermal anomalies.The thermal anomalies detected by satellite thermal infrared technology provide an important scientific basis for identifying earthquake precursors.However,not all of the thermal anomaly information is related to earthquake or tectonic movement.How to eliminate the influence of the information that has nothing to do with tectonic movement or earthquake on the surface thermal infrared and extract the weak anomaly information from the complex and comprehensive background has always been one of the important steps of using thermal infrared remote sensing technology to study earthquake.The construction of LST background field is the foundation of extracting seismic thermal anomaly.Therefore,in this paper,the thermal anomaly is extracted from the 8-day synthetic LST data of MODIS MOD11A2 product from 2000 to 2016,and the background field is improved in the extraction process.Then,the two Ms7.3 earthquakes occurred in Hotan Yutian County on March 21,2008 and February 12,2014 are taken as the main research objects,combined with the periodic characteristics of the surface temperature time series,Then,the Mean,Range,Slope,Stdv and MK mutation test are used to analyze the temporal and spatial distribution characteristics and evolution process of LST anomalies before and after earthquakes,and evaluate the adaptability of different seismic thermal anomaly information and extraction methods.The results show that:1)According to the analysis and comparison of the characteristics of land surface temperature(LST)variation with time of two Ms7.3 earthquakes in Yutian County,the four indexes(Mean,Range,Slope and Stdv)in pixel by pixel statistical method show different degrees of thermal anomaly information before and after the earthquake,The Mean index is more obvious than other indexes(Range,Slope,Stdv)before and after the earthquake,which is suitable for extracting seismic thermal anomaly information.2)Among the four indexes(Mean,Range,Slope and Stdv)in the pixel by pixel statistical method used to extract the thermal anomaly information,the three indexes of Mean(Mean)are: the difference between daytime surface temperature and daytime surface temperature mean(Td-Tdm),daily range of surface temperature(Td-Tn),daily range of surface temperature(Td-Tn)In the difference between the nighttime surface temperature and the nighttime surface temperature average(Tn-Tnm),the difference between the daytime surface temperature and the daytime surface temperature average(Td-Tdm)is the most suitable for extracting the anomaly information of the two 7.3ms earthquakes in Yutian county;3)MK mutation point test is used to analyze the surface temperature data of the Ms7.3 earthquake occurred on February 12,2014.The average index(Mean)of the pixel by pixel statistical method is the difference between daytime surface temperature and daytime surface temperature average(Td-Tdm).According to the time of abnormal mutation point before the earthquake,it is more suitable to predict the approximate occurrence time of the earthquake.4)Based on the comparative analysis of the time series and spatial distribution characteristics of the abnormal information of the two MS7.3 earthquakes in Yutian area,the same point is obtained: in the lst time series,the abnormal information of the surface temperature will appear around the earthquake source one month before the earthquake,and the amplitude of the abnormal information reaches the peak 16-20 days before the earthquake,Within 24-25 days after the earthquake,the warming amplitude gradually decreases;In the spatial distribution of LST,the distribution of orange red spots,which represent the information of thermal anomaly,began to appear one month before the earthquake,and gradually widened with time,until 16-20 days before the earthquake,and showed a decreasing trend after the earthquake.The Mean,Range,Slope and Stdv of the daytime / nighttime LST in the seismogenic and non seismogenic periods were calculated pixel by pixel,and the MK mutation test was used,The three results obtained from the analysis of the temporal and spatial distribution characteristics and evolution process of surface temperature anomalies before and after earthquakes may have corresponding reference value for the research of extracting anomaly information before and after earthquakes.