Processing gravity gradients to detect kimberlite pipes

Abstract

A modelling and pattern recognition-based approach is applied to processing airborne gravity gradient data for kimberlite exploration. The carrot-like bodies with low density crater facies that typify kimberlite pipes are particularly amenable to this treatment. Results for small and medium-sized pipes buried deeply beneath nominal geologic clutter are promising. Details regarding various error rates provide valuable input to exploration programs and the framework can include any data type. A three-class problem is formulated to address the case of false alarms. A first example is worked for low density shallow depressions that closely mimic gravity images of pipes, providing insight to what is needed from a survey fidelity standpoint to effectively mitigate false alarms.