Re-parameterization of the Cole-Cole Model for Improved Spectral Inversion of Induced Polarization Data
Gianluca Fiandaca, Line Meldgaard Madsen, Pradip Kumar Maurya · Proceedings · 2017
Summary The induced polarization phenomenon, both in time-domain (TD) and frequency-domain (FD), is often parameterized using the empirical Cole-Cole model. We here suggest three re-parametrizations of the Cole-Cole model, namely the maximum phase angle (MPA) model, the maximum imaginary conductivity (MIC) model and the minimum imaginary resistivity (MIR) model. We test these new parameterizations against the classic Cole-Cole model using a Markov Chain Monte Carlo (MCMC) inversion algorithm. The MCMC method allows us to study the probability distributions of each model parameter and gives us direct information on the parameter uncertainty. We show that models that are poorly resolved from inversion with the Cole-Cole model, e.g. due to low C values or strongly correlated parameters, can be resolved well with the new parameterizations and that the parameterizations work equally well for TDIP and FDIP data.