Seismic Anomaly Analysis Method Based On Outgoing Long-wave Radiation Data

At present,outgoing longwave radiation(OLR),as a type of satellite remote sensing data,has become an important and effective resource in seismic research.However,there are still some problems to be solved:(1)the amount of OLR data is large,and there are few suitable and effective methods for extracting and analyzing sequence features;(2)the lack of verification of a large number of real earthquake cases;(3)lack of consideration of the influence of external factors on experimental results.These problems make it impossible to give full play to the true research value of OLR data.Based on the above problems,we propose three methods of seismic anomaly analysis based on OLR data,extract the hidden feature information in OLR data,and explore the precursory anomaly information and development trends of OLR data.This paper mainly does the following:1.Anomaly analysis method based on random walk algorithm extracts random probability feature sequences of OLR data.Combined with the anomaly calculation method based on martingale theory,it can enlarge the hidden features of OLR data,mine seismic anomaly information in OLR data,and effectively reflect the development of seismic anomalies trend.We use this algorithm to detect abnormal information in the OLR data of two earthquakes in Xinjiang,China.The results show that the proposed random walk method can effectively identify the abnormal change trend of OLR data.Mining the seismic information hidden in OLR data by comparing the maximum CD value of the year when the earthquake occurred with the normal year.We further studied the CD values in the two regions during the 12-year earthquake and the 5 days before and after the earthquake,and found that the abnormal value in the year of the earthquake will be slightly higher than the normal year.We also compared OLR anomalies in Xinjiang and Central China on non-seismic zones from 2006 to 2017.It was found that the anomaly information in Xinjiang was much more pronounced than in Central China.In short,we can conclude that the increase trend of OLR before and after the earthquake is relatively obvious,and it is likely related to seismic activity.2.The quantum walk algorithm has the characteristics of faster convergence.Anomaly analysis method based on quantum walk takes advantage of this.By extracting the quantum probability feature sequence of OLR data,combined with the anomaly calculation method based on martingale theory,the abnormal signal is amplified,and the seismic anomaly information in OLR data is analyzed.We use quantum walk algorithm to analyze OLR data of Wenchuan and Lushan earthquake,and explore the correlation between OLR anomalies and seismic activity.Experiments were added to study earthquakes of magnitude 8.0 and above from 2006 to 2014 at home and abroad.It can be seen from the results that the algorithm can better present the change process of OLR data before and after the earthquake.And the larger the earthquake anomaly,the more obvious.Therefore,the algorithm is suitable for mining anomalies before earthquakes.3.The spatial-temporal analysis method tries to combine the time dimension and the space dimension to extract the OLR feature sequence with spatial-temporal correlation,then calculate the anomaly degree of the study area,and analyze the time and space characteristics in the seismic sequence.This method can effectively reduce the influence of climate and other factors on the experimental results of the study area,and more intuitively reflect the change process and development trend of OLR before and after the earthquake,as well as the temporal and spatial characteristics of the anomalous distribution.We use this method to analyze the Wenchuan earthquake on May 12,2008,the Japan earthquake on March 11,2011 and the Lushan earthquake on April 20,2013.We elaborate the analysis process and experimental results,and explore the hidden information in OLR data.Furthermore,the anomaly occurrence time of 40 earthquakes with magnitude 6.5 and above at home and abroad was further counted,and the temporal and spatial characteristics of the anomaly distribution were analyzed.Some preliminary spatial-temporal distribution laws were found through experiments,which also provided some inspiration for further exploration of the spatial distribution laws of anomalies.