Full text loading...
-
Improved Visualization of Hydrocarbon Migration Pathways to Detect Shallow Gas - Dutch North Sea Case Study
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, 77th EAGE Conference and Exhibition 2015, Jun 2015, Volume 2015, p.1 - 5
Abstract
Vertical hydrocarbon migration is recognized in normally processed seismic data as vertically aligned zones of chaotic low amplitude seismic response. Because of their diffuse character, these features are often difficult to map in three dimensions. Thus, a method has been developed to detect them using a supervised neural network. The result is a “chimney” probability volume. There a two main difficulties using these volumes. First, shallow velocity anomalies in the shallow section may cause seismic artifacts that are difficult to distinguish from true HC migration. Second, chimneys have a very different seismic character, depending on the rock properties of the stratigraphic interval. To address these issues new workflows have been developed in both the neural network training and validation process. We show a case study from the Dutch North Sea A15 Block, where improved imaging of chimneys shows hydrocarbons originating from the Carboniferous are charging shallow Miocene gas sands. This petroleum system has important implications for deep exploration in the area.