Dissertation > Astronomy,Earth Sciences > Geophysics > Earth ( rock circles ) physics ( geophysics ) > Seismology > Seismometry

Wavelet-based Methods for Detecting Earthquake Information in Remote Sensing Data

Author XiongPan
Tutor ShenXuHui;BiYaXin
School China Seismological Bureau, Institute of Earthquake Prediction
Course Structural Geology
Keywords Remote seismic Wavelet Method Seismic anomaly OLR DEMETER satellite
CLC P315.6
Type Master's thesis
Year 2009
Downloads 322
Quotes 3
Download Dissertation

Satellite remote sensing technology has been widely used in various fields of earthquake science research, but due to the lack of effective remote seismic data processing technology, resulting in most of the information has not been fully utilized in remote seismic data. Remote seismic data processing technology, the full development and utilization of remote sensing information effectively extract seismic anomaly for earthquake monitoring and forecasting services. The wavelet analysis method is a window size is fixed, but its shape can change the time-frequency localization analysis method. I.e. in the low-frequency portion having a higher frequency resolution and lower time resolution in the high-frequency portion having a high time resolution and a low frequency resolution, and therefore known as a digital microscope. Its wide range of applications, including: signal analysis, image processing, computer classification and identification, fault diagnosis of large machinery and earth science research. Based on the basic principles of the wavelet analysis method, the use of the wavelet maxima graphic method, wavelet coefficients superposition method, wavelet decomposition comprehensive analysis and the Bohe Manchester index wavelet method, analysis and processing of the French DEMETER electromagnetic satellite data longwave radiation data and the U.S. NOAA weather satellites, and combined with the May 12, 2008 Wenchuan 8.0 earthquake and June 3, 2007 Pu'er 6.4 earthquakes and other seismic events, through a series of comparative analysis of experiments to validate these four The validity of the kinds of wavelet method in the earthquake in the remote sensing information extraction. 1, wavelet maxima icon for the NOAA satellite longwave radiation value exception information extraction longwave radiation is radiation absorption of solar radiation, the Earth's surface and turn the energy into the sky to make. Physical constraints of the periodic activities under the Sun-Earth, Earth's outgoing longwave radiation is rendered as a cyclical variation; addition, the activities of the Earth's interior, such as earthquakes, volcanoes and other outward scattering of radiation energy, will be the formation of long-wave radiation, this part of the show aperiodic discontinuous, isolated variation. Based on the physical causes of the longwave radiation, the use of abnormal seismic wavelet maxima icon assay longwave radiation data. The specific research ideas: 1) based on the geological background of the seismic event, the definition of the experimental study area, the epicenter of the center, 300 km range of the experimental study area are divided into nine cells; 2) the use of a one-dimensional continuous wavelet transform and singular property data detection principles calculations of the wavelet maxima longwave radiation in each cell, and draw a visual form of wavelet maxima maps; 3) 9 cell corresponding wavelet maxima distribution FIG spatial sequence stacked together to form a spatially continuous wavelet maxima sequence diagram to identify, by analyzing the distribution form of the the wavelet maximum values ??in the sequence diagram wavelet maxima seismic anomaly in the long-wave radiation; 4) combining four Sichuan The earthquake and the earthquake in Yunnan Pu'er earthquake in the long-wave radiation, through a series of comparative experiments to identify abnormalities. 5) Taking into account the Wenchuan 8.0 earthquake energy larger sphere of influence is relatively large, disturbance area and non-disturbance area for comparative analysis, based on the history of seismic activity and tectonic characteristics, select three experimental zone, are area 1, Zone 2 and Wenchuan earthquake main shock area, the main quake zone covering the area of ??the epicenter of Wenchuan earthquake area near Wenchuan earthquake fault zone, area 2 away from the seismic zone, belonging to non-earthquake zone. The specific location of the three regions: the area of ??the main shock: from 30 ° N to 33 ° N, 103 ° E to 106 ° E, Zone 1: from 28 ° N to 31 ° N, 105 ° E to 108 ° E, about 200 km from the main shock, the Zone 2: from 44 ° N to 47 ° N, 113 ° E to 116 ° E, about 1400 km from the main shock. 6) In order to make a more convincing experimental results to the epicenter of Wenchuan earthquake centered 20 ° × 20 ° defined out experiments in the study area, select a large range of 2000 km from the epicenter Wenchuan earthquake experiments in the study area (21 ° N to 41 ° N, 93 ° E to 113 ° E), the wavelet analysis method for db1, the wavelet order of 16 study of the long-wave radiation data for October 2006 to September 2008, a two-year long-wave radiation mean data. 7) Pu'er earthquake epicenter, 20 ° x 20 ° defined out experiments in the study area, select a large range of 2000 km from the epicenter Wenchuan earthquake experiments in the study area (13 ° N to 33 ° N, 91 ° E to 111 ° E) , adopted wavelet analysis method gaus3, and the wavelet order to 16, the study of the long-wave radiation data for May 2006 to April 2008, two years longwave radiation mean data. The preliminary results of the experiments are as follows: 1) before the Wenchuan earthquake, the earthquake epicenter was 300 km range there are significant long-wave radiation seismic anomaly; Wenchuan earthquake main shock area wavelet maxima the sequence diagram abnormal strength was significantly greater than the two contrast the experimental area, compared to the two experimental area of ??abnormal intensity, area larger than the area 2, area 2 is basically no abnormalities, said the sphere of influence of the Ming Wenchuan earthquake, although large, but with distance weakened and limited. 2) the first eight months of the Wenchuan earthquake (September 2007), the abnormal intensity in the range of 2000 kilometers of the earthquake epicenter large regional decline, abnormal low value continued until January 2008, when the first four months of the Wenchuan earthquake (2008 February), abnormal intensity abrupt and sharp rise, reflecting the the Wenchuan earthquake seismogenic huge energy change; abnormal strength beyond the mean of 141.1% in August 2008, the energy changes that may be caused by a large number of strong aftershocks in Wenchuan earthquake . 3) the Pu'er earthquake before the long-wave radiation values ??three aperiodic, not continuous, abnormal characteristics of isolated changes; Pu'er earthquake longwave radiation data (2006 to 2007 September 3, 2010) push after a year of data (September 3, 2007 to September 3, 2008) compared experimental results show that the annual cycle of the two sets of data wavelet maxima similarity, but Pu'er earthquake year longwave radiation intensity is too large. 4) the first five months of Pu'er earthquake (January 2007), the abnormal intensity in the range of 2000 kilometers of the earthquake epicenter downward trend, abnormal low value, abnormal strength suddenly increased significantly in April 2007, the abnormal strength reaches a maximum, reflect the Pu'er earthquake seismogenic huge energy changes; 5) Overall, the Wenchuan earthquake abnormal strength and areas of influence than Pu'er earthquake, show a positive correlation with the earthquake intensity. 6) The experimental results with previous research conclusions obtained using other methods consistent verify the validity of the the wavelet maxima icon law. Wavelet method in abnormal the DEMETER electromagnetic satellite data information extraction application of the DEMETER electromagnetic satellite data of conventional research methods to determine seismic anomaly by direct visual observation of the two DEMETER satellite data image, this approach is not only likely to cause miscarriage of justice, and not carried out the DEMETER satellite an in-depth analysis of the data, resulting in most of the information in the data have not been fully utilized. Aiming DEMETER satellite data Langmuir probe (ISL) observation of the electron density (Ne), the electron temperature (Te), Ion Density (Ni), plasma analyzer (IAP) observation of ion temperature (Ti) and other 4 kinds of parameters, by DEMETER satellite two preliminary interpretation of the image data, filter out this study, four earthquake cases 10 satellite orbit superposition method using wavelet coefficients, wavelet decomposition comprehensive analysis and small the Poch Manchester index etc. the three wavelet method to extract the exception information in the track, summarize abnormal features. The research ideas: The wavelet coefficients superposition method is calculated according to the number of one-dimensional continuous wavelet transform wavelet order wavelet coefficients superimposed wavelet coefficients of each order to highlight the exception of the original data; the wavelet decomposition comprehensive analysis of the raw data signal decomposed into high-frequency and low-frequency part of the identification of the relevant parts of each band seismic events seismic anomaly; small the Bohe Manchester index using a one-dimensional complex Morlet wavelet transform and linear regression mathematical methods to calculate the small Poch Manchester index extract hidden in seismic anomaly in the raw data. Preliminary understanding and conclusions: 1) after processing by the wavelet coefficients of the superposition method, the raw data of the Wenchuan earthquake before the DEMETER the Electromagnetic satellite half track 20595-1 4 parameters, within the range of 2000 km of the epicenter can be found obvious exception; 2) Figure Pu'er earthquake before 5 the DEMETER electromagnetic satellite orbit data, after more than three wavelet method after processing can extract the obvious exception information after half track 15440-1 after the wavelet decomposition comprehensive analysis of the results of each spectrum segments in abnormal before the earthquake, the 4 parameters small Bohe Manchester index after the results of figure also indicates the the Pu'er earthquake before the abnormal.

Related Dissertations
More Dissertations