Exploration of the Lower Cretaceous Sands in Leland Area, Alberta, Using Seismically Derived Rock Properties

摘要

This paper provides a case study of a 3D seismic survey in the Leland area of the Deep Basin of Alberta, Canada, where seismically derived rock properties were used for exploration. In this case study, identifying the gas sands within the lower Cretaceous was the primary interpretive focus. Conventional interpretation of Lower Cretaceous sands on normal migrated seismic has typically presented a number of difficulties. First, we ran AVO analysis and deterministic inversions of the AVO attributes. The P-impedance and S-impedance volumes were used to estimate rigidity and incompressibility that are very good indicators for lithology and fluids in the target formations of our study. Finally, neural network analysis was performed in order to estimate reservoir properties (e.g. P-impedance, S-impedance and density) and has proven effective in significantly improving accuracy and vertical resolution in the interpretation of the reservoir. In addition to the rigidity and incompressibility maps, we derived porosity maps calculated from density, in an effort to delimit the reservoir and find new opportunities for field development. This methodology helped in discriminating gas intervals and in drilling new locations, that encountered gas charged reservoir.This paper provides a case study of a 3D seismic survey in the Leland area of the Deep Basin of Alberta, Canada, where seismically derived rock properties were used for exploration. In this case study, identifying the gas sands within the Bluesky, Gething and Cadomin Formation of the lower Cretaceous was the primary interpretive focus. Conventional interpretation of Lower Cretaceous sands of the Bluesky/Gething/Cadomin formations on normal migrated seismic has typically presented a number of difficulties. These include poor well ties to stacked data, lack of a distinct seismic signature for productive zones, poor ties between 2-D and 3-D data and unexplained variations in seismic waveform. Data for the project consist of 47 wells and one 3D seismic survey. First, petrophysical analysis of the well logs was performed in order to provide a trustworthy set of logs that could be used for inversions and multi-attribute analysis and to determine petrophysical relationships that can be useful on seismic data interpretation. Secondly, we ran AVO analysis and deterministic inversions of the AVO attributes. The Pimpedance and S-impedance volumes were used to estimate rigidity and incompressibility (Goodway et al., 1997) that are very good indicators for lithology and fluids in the target formations of our study. Finally, neural network analysis was performed on logs and pre- and post-stack seismic attributes. Neural network estimation of reservoir properties (e.g. P-impedance, S-impedance and density) has proven effective in significantly improving accuracy and vertical resolution in the interpretation of the reservoir. In addition to the rigidity and incompressibility maps, we derived porosity maps calculated from density, in an effort to delimit the reservoir and find new opportunities for field development. This methodology helped in discriminating gas intervals and in drilling new locations, that encountered gas charged reservoir.