Predicting the distribution of small faults in a hydrocarbon reservoir by combining outcrop, seismic and well data

Abstract

Abstract This paper investigates the possibility of identifying small faults in a hydrocarbon reservoir from spatial derivatives of seismic horizons (dip, azimuth, rate of change of dip), combined with well log structural data and analogue models. The analogue models include examples from the literature and a faulted surface constructed from Kilve beach, Bristol Channel, which represent typical surface attributes associated with small faults and artefacts produced by modelling. Well log structural data of good quality were integrated with the seismic data to study the 1D spatial distribution of subseismic faults and their expression on seismic horizons. The results suggest that dip and azimuth modelling can be useful, but the methods need to be combined to achieve reliable interpretations of the individual features. The rate of dip change has less potential to trace individual structures, unless the signal/noise ratio is very high. The ductile strain observed on seismic sections is, to a large degree, assumed to be produced by subseismic faults. The amount of ductile strain may be quantified by calculating the rate of dip change across curved seismic horizons. Exact predictions of fault densities are difficult because of the inaccuracies related to the modelling technique and the variability in fault and fold styles that occurs on a subseismic scale. Detailed dipmeter and core interpretation can be used to calibrate structural dip and spatial frequency of faults.