AVO Modeling and Interpretation related to Thermal Effects: Åsgard Field, Norwegian Sea
Abstract
In oil exploration, one of the first steps is to conduct a seismic survey to get an overview of the area of interest. If a possible reservoir is found, the only way to determine reservoir quality is to drill an exploration well and analyze cores and logs extracted from the well. Based on previous project work in collaboration with Statoil s research center, AVO has been studied in terms of thermal effects. Specifically the impact of quartz cementation due to diagenesis and the possible avoidance of this due to chlorite coating. Well data from two different wells in Åsgard Field situated in the Norwegian Sea were used as input data. The presence of chlorite coating, which might preserve porosity, is known to be present in some sand intervals in the wells, and the main goal of this dissertation was to see whether zones affected by chlorite coating could be detected using seismic attributes. This was done by combining various models, including temperature history modeling and rock physics models for shale and sand. The temperature history was found by geological backstripping of present-day values followed by a forward temperature modeling. Based on the temperature history, the elastic properties of shale were found by using Dræge s shale model (Dræge et al., 2006). For sand, quartz cementation and the following porosity were found from Walderhaug s cementation model (Walderhaug, 1996), from which the P- and S-wave velocities were derived using Han s empirical relations (Han, 1986). Based on a simple two-layer model with an upper shale layer and a lower sand layer, the reflection coefficients were calculated using the Aki-Richard s approximation (Aki and Richards, 1980) of the Zoeppritz equation. There has not been any uplift in this area, and the sand and shale are assumed to follow the same temperature history (Kyrkjebø, 2014). The elastic parameters for the sand were derived both with and without the effect of chlorite coating, and for different grain sizes. The resulting AVO responses were plotted for different temperatures, amount of chlorite coating and grain sizes. The AVO responses from the two-layer model were compared to extracted AVO from synthetic seismograms made from well logs. It was shown that it was not straightforward to compare synthetic two-layer models to synthetic seismograms. This was due to both a difference in the sand and shale properties between the model and well, and that real data often have variations in lithology that are not present in a simple two-layer model. Also, for a given temperature, it seems hard to distinguish between the effect of the grain size and chlorite coating based on the AVO response. Small grain size and chlorite coating gave the same response as for larger grain size with no chlorite coating.