There are thermal infrared anomalies(TIA)before earthquake, and TIA has become one of the important parameters for assessing regional earthquake risk. However, not all of the surface infrared anomalies are related to tectonic activities or earthquakes. How to eliminate the influence of non-structural factors and extract the weak signals from strong disturbances is the key and difficult point for tectonic activities studies based on the thermal infrared remote sensing techniques. Land surface temperature(LST)background field is the basis for thermal infrared anomalies extraction. However, the established background fields in previous researches cannot eliminate the influence of climate changes, so the accuracy of thermal anomaly extraction is limited. Now an improved method is proposed. Combined with the periodic character of LST time series, harmonic analysis is lead into the process of LST background field establishment. Specifically, the yearly trend of LST is fitted based on Fourier Approximation method. As a new background field, the yearly trend is dynamic, includes the local and the yearly information. Then, based on the rule of "kσ", the earthquake anomalies, calculated by RST with the yearly trend of LST, can be extracted. At last, the effectiveness of the algorithm can be tested by the quantitative analysis of anomalies with anomaly area statistics, anomaly intensity statistics and distance index statistics. The Wenchuan earthquake was discussed again based on the proposed algorithm with MODIS land temperature products in 2008. The results show that, there were obvious pre-earthquake thermal anomalies along the Longmen Mountains faults with a longer time; but there were no anomalies when the earthquake happened; and the post-earthquake thermal anomalies occurred with much smaller amplitudes and scopes. Compared with the results derived from the traditional RST which is based on the spatial average of LST values, the TIA extracted by the new RST, which is based on the yearly trend of LST, is more fit with the active faults, and the process of the anomalies occurring and removing can be described in more detail. Therefore, as the background field to extract earthquake anomalies, the yearly trend of LST is more reliable.

Earthquake prediction is one of the key areas of earthquake research. Thermal infrared abnormity, which is the abnormally increased land surface temperature, is universal before earthquake and has complex nonlinear relation with the three elements of earthquake. Combining the advantage of neural network, this paper provides a method for earthquake prediction by taking thermal anomaly as information source and constructing a neural network to carry out the test. Based on the MODIS data which has synthesis of eight days with 1km resolution, taking a 10°×10° rectangle, whose center is the epicenter, as research area, and a two-month time before earthquake as the time range, we used RST algorithm to extract thermal anomaly information before earthquake. Considering the time-space relationship between thermal anomaly information and the fault zone, thinking carefully about the information of the neural network input neurons, we constructed BP neural network and used 100 earthquake cases with magnitudes larger than 5, and 70 random samples without earthquake in the research region for training and simulating. According to the statistical analysis, the prediction accuracy is 80%, missing prediction rate is 20%, and false prediction rate is 13.3%. Prediction accuracy of magnitude with error within magnitudes of 2 is 69%, prediction accuracy of earthquake origin time with error within 30 days is 87.5%, and prediction accuracy of epicenter location with error within 3°is 81.2%. The result shows that BP neural network-based earthquake prediction is feasible by using thermal infrared abnormal precursor extracted by RST method. However, in this experiment, the determination of the start time of thermal abnormity, the origin point and range of research area are based on the known epicenter location and time. In fact, the result depicts a non-linear relationship between earthquake and thermal abnormity, and the accuracy of prediction reflects the correlation degree. Therefore, the prediction accuracy of future earthquake may be not as large as our result. For future earthquake prediction, accurate selection of research area and neuron number of hidden layer in neural network has great influence on prediction result.