Good well-to-seismic ties are key to proper seismic reservoir rock property characterisation. Acoustic log measurements are significantly affected by unstable hole conditions that can impact on the usual porosity computation from the sonic and density logs within these bad hole intervals. It is commonly assumed that sonic logs are not significantly impacted by these bad hole conditions.

In this study, sonic, density and caliper logs were each taken from one field located in the eastern and western Niger delta. The acoustic logs were recreated by a Neural Network technique in intervals where the hole indicates extensive wash out. Cross-plots of the acoustic logs against calipers were made for both the measured and recreated logs. Well-to-seismic ties were also completed using both the measured and the recreated logs.

In both fields, a dependence of sonic velocity on hole size was observed for the measured logs. Across washout regions, an improved well to seismic tie was obtained from the sonic and density logs when corrected for the effect of borehole washout. Cross-plots of p-impedance versus density using uncorrected logs showed no significant differentiation between sands and shales. However, sands and shales showed good separation in the reconstructed logs, enabling more accurate lithologic calls. It is concluded that hole washout affects sonic and density log readings more substantially than is commonly recognized, and that the quality of the sonic log has a significant impact on the quality of subsequent seismic to well ties and inversion results used in reservoir characterisation.

You can access this article if you purchase or spend a download.