Abstract: The transient electromagnetic response sometimes becomes negative at late times, which may be caused by the induced polarization(IP) effect. Because of the fact that there is no consideration of the IP effect in the traditional real resistivity inversion, the interpretation of the negative part of the observed data has always been a problem which affects the accuracy and reliability of the inversion. In this paper, we first introduced the Cole-Cole complex resistivity model into the forward modeling code and tested its accuracy. Then, one-dimensional TEM Occam inversion algorithm was implemented based on the damped least-squares method. We constructed the inversion equations involving Occam smoothness constraint matrix and inversed simultaneously the four complex resistivity parameters, i.e. DC resistivity, chargeability, time constant and frequency dependence. In order to test the inversion algorithm, we established two different typical theoretical polarized layered models and analyzed their inversion effects. The results show that the complex resistivity parameters can be better recovered in the layer with low resistivity and high chargeability. Comparing with the traditional real resistivity inversion, the complex resistivity inversion can not only perform like the real resistivity inversion, but also solve the fitting problem in the case of the negative response.

Key words: transient electromagnetic (TEM), IP effect, Cole-Cole model, Occam inversion