Shallow Gas In The Oseberg, Brage And Troll Fields North Sea, 60°30' N

Abstract An integrated approach using geological, seismic, geotechnical and well log data have been used to investigate the presence of shallow gas in the Oseberg, Brage and Troll fields. This study have been motivated by the low rate of success in the prediction of shallow gas from the interpretation of seismic reflection anomalies alone. The main conclusion is that the integrated approach offers a broader data base and a more rational approach to evaluate the shallow gas within the frame of a geological model rather than from scattered "reflection anomalies" in seismic profiles. Introduction Shallow gas is a term that has been used rather loosely to indicate potentially hazardous gas accumulations that might be encountered in the initial drilling phase of a well (i.e. before setting a blow-out preventer). In the North Sea, gas accumulations shallower than 1000 m, but more typically 200-600 m, are refered as shallow gas. Shallow gas has been reported in 150 of 558 wildcat and appraisal wells drilled in the Norwegian Continental shelf. Seven blow-outs and a number of smaller kicks caused by shallow gas have been recorded (Aamodt, NPD, 1987). Currently, shallow gas prediction at a drilling site is interpreted from "reflection anomalies" in multichannel high resolution seismic profiles. This approach alone is not sufficient to predict the existence of shallow gas in the complex stratigraphy of the Cenozoic layers in the northern North Sea. This paper discuss the presence of shallow gas in the Cenozoic layers of the Oseberg, Brage and Troll fields (see locations in figure 1). We propose an integrated approach to predict shallow gas using geological, seismic, geotechnical and well log data. Integrated approach for shallow gas prediction In general the main criteria used to predict the presence of shallow gas is to pick "amplitude anomalies" from an inspection of the seismic profile along with other features such as "polarity reversal", "diffraction" or "phase changes" whenever possible. There is no doubt that in most cases this method fails, because of a lack of understanding of what kind of anomaly one should expect to see, due to too little information. The appearence of features related to gas saturated sands are strongly dependent upon the acoustic impedance contrasts of the sediments themselves as well as the presence of gas. This would imply that a more detailed knowledge of the geology of the area is necessary before a more accurate prediction can be made (D. Milas, pers. com.). The geological information can be derived from the regional integration of well logs, geotechnical boreholes and seismic data. This information should serve two purposes:To investigate the possibility of generation and entrapment of shallow gas.To provide the physical parameters (composition, porosity, consolidation) necessary to evaluate the acoustic properties of the sediments. If a geological model can be defined, then the seismic response of the sediments can be predicted. Also shallow gas indications in seismic profiles can be interpreted at stratigraphical intervals where the depositional environments and sedimentary processes indicate possible shallow gas accumulations.