Cooperative Inversion Based on Fuzzy c-Means Cluster Analysis – Application to Field Data

Abstract

In many near surface geophysical applications it is now common praxis to collect co-located disparate geophysical data sets. The advantage of such multi-method based exploration strategies lies in its potential to reduce ambiguities and uncertainties in data analysis and interpretation. To reconstruct the physical parameter distributions underlying geophysical data requires in many cases the application of tomographic reconstruction techniques. To improve the reliability of tomographically reconstructed physical parameter models, the information content of all co-located data sets should be considered during the model-generation process. Here, we apply a novel approach based on fuzzy c-means cluster analysis and conventional single-input data set inversion algorithms for the cooperative inversion of crosshole seismic P-wave and S-wave traveltimes. The approach results in a single zoned two-parameter model outlining the major subsurface units and provides additionally P-wave and S-wave velocity fields.