| dc.description.abstract | Shallow gas and overpressure are potential drilling hazards, and detailed mapping of such risks must be done prior to drilling to prevent loss of money, damage to equipment, and in the worst case, accidents on personnel. In the shallow part of the sediments of the North Sea glacial deposits create networks of possible porous and permeable rock. These rocks can act as reservoirs for shallow gas, trapped in the subglacial forms, overlaid by impermeable till and glacial and marine clay. Amplitude variation with offset (AVO) and amplitude variation with offset and azimuth (AVAZ) measurements have been performed on a tunnel valley, an esker, a channel system and other amplitude anomalies in an area above the Edvard Grieg Field in the Central North Sea. High Resolution 3D (HR3D) data have been used. Detailed interpretations have revealed several subglacial forms not detectable on conventional seismic, including narrow tributary tunnel valleys, eskers and glacial lakes. Geophysical methods, combined with AVO modelling using Smith and Gidlows approximation, have been used as tools to detect if these deposits contain shallow gas and represent drilling hazards. The results also contribute to the geological understanding of subglacial deposition.The high quality HR3D site survey has been compared to a conventional near stacked cube and a BroadSeis survey. The comparison shows the significance of the HR3D cube, but also its limitations. The importance of using high quality data will be discussed, as it may not be a necessary solution compared to the advantages of other data types. The results indicate gas and/or overpressure in the glacial channel network. The lithology variations related to the esker show an anomaly which can be related to a minor gas saturation. The upper tunnel valley sediments does not show an AVO response related to fluid changes. However, variations in lithology support the theory of tunnel valleys being formed by several episodes of erosion and deposition, creating composite tunnel valleys. AVAZ and wide azimuth measurements on the top tunnel valley interface support this, as anisotropy is indicated in the upper part of the tunnel valley infill. However, AVAZ modelling does not support this affirmation.The measurements are contaminated by several sources of uncertainty, making the results questionable. Both statistical data processing effects, related to channel balancing, and deterministic modelling effects like mode conversions, focusing effects, transmission losses, geometrical spreading and tuning, implicate the response. Due to the size of the measured anomalies, the Fresnel zone plays an important part in the apprehension of the measurements. | nb_NO |
