A New Method to Improve the Accuracy of Depth Prediction for Underlying Strata of Shallow Gas: A Case Study From BZ Oilfield in Bohai Bay Basin
The quality of seismic image of underlying strata is negatively affected by shallow gas. Seismic events below shallow gas are dominated by several special phenomena such as low frequency, drop-down and weak energy. Therefore, it is difficult to predict the depth of target zone accurately. Thus the drilling risk of horizontal wells is greatly raised as well. Currently, the influence of shallow gas on underlying strata is still under qualitative analysis. The purpose of this paper is to quantize the gas influence in order to provide a solution to achieve accurate depth prediction of target zone. In this paper, the influence mechanism of shallow gas on underlying strata is discussed in detail. Based on the discussion we find out main influential factors, including thickness of gas-bearing layer, horizontal scale of gas-bearing layer, buried depth of gas-bearing layer, distance between gas-bearing layer and target layer, receiving cable length and the formation velocity. With certain simplifications, we write a program to calculate the influence range and extent of shallow gas on underlying strata with main influential factors. Although there are some differences between simulation and the real situation, it is available to intuitively present the influence of shallow gas. Furthermore, we build many models based on different gas-bearing formation conditions to achieve pre-stack forward modeling with formation velocities derived from real drilling data in Bohai Bay Basin. Then we complete pre-stack time migration with shot records derived from forward modeling. Finally, we build a series of time correlation templates related to different offsets and depth of underlying strata through event auto-tracking and interpolation. The templates are applied to quantitatively estimate time correlation caused by shallow gas for any point of underlying strata, thus we achieve to improve the accuracy of depth prediction. The application of templates in BZ oilfield shows that we achieve a high-precision depth prediction for underlying reservoir of shallow gas. For instance, the drilled well A shows that the predicted depth error using new method is only 3 meters while conventional predicted error is 10 meters. Therefore, we basically solve the problem of depth prediction and efficiently guide the drilling of development wells in BZ oilfield. We believe that the new method proposed here will be widely used in other oilfields containing shallow gas as well.
AAPG Datapages/Search and Discovery Article #90216 ©2015 AAPG Annual Convention and Exhibition, Denver, CO., May 31 - June 3, 2015