Application Of Multiscale Edge Detection Of Gravity Field In Detecting Buried Faults In Wuhan

The detection of urban active faults is a basic work for earthquake prevention and disaster reduction,it's also an important guarantee for urban geological safety.Wuhan is the only sub-provincial metropolis in six provinces in central China,it's also a core city in the Yangtze River Economic Belt.Wuhan is one of the inland large international cities with the fastest process of urban modernization in China,its permanent resident population has reached 11.212 million by the end of 2019.In the geological evolution,faults are developed owing to multi-stage tectonic movements in Wuhan,it has a geological background for the occurrence of moderately strong earthquakes.In history,Wuhan and its surrounding areas have also experienced many moderately strong earthquakes.With the modernization process of continuous development in Wuhan,the buried deep faults pose an increasingly serious threat to urban security.Therefore,detecting the distribution of buried faults in Wuhan is of positive significance for the overall urban planning and ensuring the city develop in a good state.Based on multiscale edge detection of gravity field,wavelet analysis of gravity anomalies,and power spectrum analysis,this paper studies theoretical models and highprecision gravity data in Wuhan to detect the distribution of main buried faults,the main work and results are as follows.1.Through model tests,the boundary identification received from NTA,Theta Map,ILP and multiscale edge detection of gravity field are compared,it is proved that the multiscale edges of gravity field can extract model boundary more accurately than other methods,and has better anti-noisy performance,it is more suitable for extracting the information of actual fault,at the same time,some characteristics of model boundary's multiscale edge are summarized.2.A 2.5D fault model with multiple strata is developed,based on the previous studies,the wavelet transform's theoretical formula of gravity anomalies and the nonlinear relationship between the model parameters and multiscale edges of the gravity field are derived.Using Quantum Genetic Algorithm,the feasibility of inverse the fault parameters by multiscale edges is verified through model tests.3.The multiscale edges of the gravity field was applied to the actual bouguer gravity anomaly in Wuhan,integrating relevant geological and geophysical data,the distribution and dip of main deep faults are detected.The main deep faults in Wuhan are divided into two groups,NNE and NWW,forming a rhombus network system.The faults inclined NNE include Tuanfeng-Macheng fault,Yueyang-Wuhan fault and Yangtze River fault;The faults inclined NWW include Xiangfan-Guangji fault,Xincheng-Huangpi fault,Tianmenhe fault and Mayangtan fault.4.Based on the wavelet details of bouguer gravity anomaly and the estimation of source depth,the distribution of buried faults at different depths is analyzed.The depths of Xinhuang fault and Xiangguang fault are both between 0 and 30 km.The depths of other main buried faults are between 0 and 30 km.5.Taking the fault depth and the approximate thickness of each layer obtained by wavelet analysis as constraints,combined with the nonlinear relationship between the fault parameters and multiscale edges,the dip angles of main buried faults are calculated through the inversion of the profile gravity anomaly by the second-order wavelet approximation.The dip angles of the Xinhuang fault and Xiangguang fault are both 50°?60°,which is in accordance with the results of previous study.