Application Of Several Interface Inversion Methods In Aeromagnetic Data Processing In Northeast China

Inversion of magnetic interface is a very important branch of geophysical exploration,it is very important to look for underground large fault structures,deep resources searching and accurate interpretation for actual measured data.With the rapid development of computer technology,the research of high-precision instruments and the improvement and combination of various exploration methods in recent decades,the accuracy of magnetic data measurement is improving.At the same time,there are more and more inversion methods for magnetic interface-curie surface.Many scholars have derived the inversion algorithm from many aspects and applied it to real data.But all kinds of methods have their own advantages and defects.The purpose of this paper is to make a comparative analysis of three kinds of inversion methods,spectral analysis,Euler algorithm and Parker-Oldenburg inversion method,and summarize their adaptation conditions,so as to provide some references for the selection of interface inversion methods.There are two main steps in the inversion of curie surface: first of all,the deep low-frequency region anomalies caused by the magnetic interface is accurately separated from the total magnetic anomalies measured by the actual measurement;Then,the interface inversion method is applied to the isolated low-frequency field and the depth of the interior is retrieved.In this paper,which is based on the processing of the data of 1:200 000 airborne magnetic data in Northeast China,the fractures in the area are divided and the relative depth of the curie surface is obtained.The specific contents are as follows:1.The basic principles of the multiple field processing methods selected in this paper are summarized-Field separation methods: analytical continuation,the multi-scale wavelet analysis,matched filtering;Interface inversion methods: spectrum analysis,Euler algorithm,the Parker-Oldenburg interface inversion method;The boundary identification methods: tilt derivative method,vertical derivative method,Euler deconvolution.2.The model experiments are carried out for two kinds of field separation methods:the multi-scale wavelet analysis and matched filtering,and the results are compared and analyzed,the paper draws a conclusion that we should choose the more accurate multi-scale wavelet analysis method for the separation of field,but we must pay attention to the selection of order.The logarithmic power spectrum of matched filtering method can be used to extract the average depth information of anomalies.3.The three interface inversion methods are applied to the magnetic interface forward modeling,and the interface depth data are compared and analyzed,and compared with the actual data,the conclusion is as follows: The result of Parker-Oldenburg iterative inversion is almost the same with the depth and contour of the forward model.The error of the spectrum analysis and Euler algorithm is large.4.A variety of field processing methods selected in this paper are applied to the1:20 million aeromagnetic data in Northeast China,and the basement features in the Northeast Region-the most northeastern fracture structure are analyzed.In the study area,11 deep faults were retrieved and compared with previous works,the results were in good agreement.5.The three kinds of interface inversion methods are applied to the data of 1:200 000 airborne magnetic data in Northeast China,and the curie surface is retrieved respectively,and the comparison with the geological data and the previous work results is found that the Parker-Oldenburg iterative inversion method has a better effect,and the depth of the curie surface of the study area is 25km-32 km,which can accurately reflect the undulating state of the curie surface,and the position corresponds well with the location of the deep fault.The spectrum analysis method is based on the isolated column model,and it has better effect on magnetic difference between upper and lower interfaces.Euler algorithm has the advantage of less prior information,but the selection of structural index and the size of window has great influence.