SOTEM数据一维OCCAM反演及其应用于三维模型的效果

摘要

We propose an OCCAM scheme for inversion of SOTEM data based on the vertical magnetic field component,and applied it to synthetic 3D data and field measured data.This inversion scheme works well for the most 1D models and the inversion results are independent of offsets.The noise has a great influence on the OCCAM inversion,but the results still have good accuracy as long as the noise level is less than 5％.However,if there is a thick conductive layer in the shallow subsurface,the resolution of the OOCAM inversion will be reduced and only the average resistivity of lower strata can be obtained.The 1D inversion scheme applied to 3D data will produce a large error,especially in the edges of the abnormal bodies.The degree of such influence is related to the resistivity contrast between the anomaly bodies and the background.But for the most real earth models whose resistivity approximately changes continuously,the 1D OCCAM inversion still works well.At last,a case study in Shaanxi province is presented to verify the research results of this work.%本文基于垂直磁场分量研究了SOTEM数据的一维OCCAM反演方法,并将其应用于理论三维数据及野外实测数据的反演.对于大部分一维模型,OCCAM反演可取得较好的反演效果,且反演结果不依赖于偏移距;噪声对SOTEM数据的OCCAM反演具有较大影响,但当信号含噪水平不超过5％时,反演结果仍具有较好的准确性;若浅层存在较厚的低阻层,OCCAM反演结果对下部地层的分辨能力下降,仅能获得具有平均效应的电阻率.将一维算法应用于SOTEM三维数据的反演,会产生较大的误差,尤其是在异常体边缘地带影响最为严重.该影响程度与异常体和背景电阻率之间的差异有关,对于大多数电性近似呈连续变化的真实大地而言,一维OCCAM反演算法仍可获得较好的效果.最后通过陕西某煤田深部富水性调查的实测SOTEM数据反演验证了本文的研究成果.