1-D Occam inversion method for airborne transient electromagnetic data

We analyzed the principle and implementation method of Occam's inversion algorithm and then applied it to estimate airborne time domain transient electromagnetic(ATEM)1-D inversion and imposed a simple search process for regularization factor.The inversion speed can be largely accelerated with improvements of iterative process.Our numerical tests show that the time for 5 iterations to convergence is approximately 20 seconds.In the process of inversion,the initial resistivity datum,representing the initial model,is converted from automatic iterative calculation of achieved induced electromotive force(EMF)and then the depth datum can be obtained from resistivity datum.The tested examples of synthetic models illustrate that the Lagrangian multiplierμ,which minimizes the data misfit,decreases with the increase of iterative numbers and the process is very stable in which generally 5 to 10iterations are required to reach the given tolerance.Our results show that the Occam's inversion can provide the agreement of high-conductivity layer's depth and resistivity value but disagreement of low-conductivity layer's.