Integrating geophysics, geology and petrophysics: A 3D seismic AVO and borehole/logging case study

Abstract

Introduction Upper Mannville aged strata in southern Alberta consist of a complex lithologic assemblage of narrow shoestring like “Glauconite” and “Lithic” valley-systems trending roughly north-south and variably cutting down into older regional Glauconite, Ostracod and Basal Quartz strata. These valley systems are commonly filled with hydrocarbon-bearing sandstones with varying reservoir quality. In the study area, the Glauconite valley systems are up to 35 meters deep, 2 kilometres wide and several tens of kilometres long. The younger “Lithic” valleysystems generally follow the same trend as the Glauconite valleys and are up to 40 meters deep, 3 to 4 kilometres wide and also several tens of kilometres long. In 1996, a 3D seismic survey was acquired over the valley trends to image reservoir quality sands. It is often difficult to seismically distinguish between Glauconite sand and Ostracod shale by using poststack amplitudes. Improved techniques from recent prestack seismic work suggest that AVO (Amplitude Versus Offset) might be an effective way to extract rock properties and detect gas (Smith and Gidlow, 1987; Gidlow et al., 1992; Fatti et al., 1994, Goodway et al., 1997). Smith and Gidlow (1987), Gidlow et al(1992) and Fatti et al. (1994) proposed a method to create a display which highlights the Vp/Vs ratio anomaly (often caused by the presence of gas) by using the so called “fluid factor stack”. Goodway et al. (1997) proposed a new improved fluid detection and lithology discrimination indicator using petrophysical parameters and where and are Lame’s constants (incompressibility and rigidity) and is density. The geological and geophysical effort in this area is to determine where the two valleys are present, and to differentiate sands from shale. Geology and Reservoir Quality Log analysis and core data, where available, indicate that the Glauconite sandstones generally have good to excellent reservoir quality whereas the Lithic sandstones are generally moderate to poor quality rocks. For example, the Glauconite sandstones in well B (Figures 1 & 2) have core porosities ranging from 18 to 26% and permeabilities ranging from 948 to 4900 md.. In contrast, the Lithic sandstones in well C (Figure 1) have core porosities ranging from 10 to 14% and permeabilities from 1 to 40 md.. Glauconite channels in the study area produce both oil and gas whereas Lithic channels produce primarily gas. The differences in reservoir quality between the Glauconite and Lithic channels are related to major differences in sandstone framework grain composition. Sedimentological interpretations from available core indicate that the Glauconite and Lithic channels consist predominantly of fine to medium grained fluvial sandstone at the base and grade upwards into finer grained marine influenced (estuarine) deposits at the top. Typical regional, non-channel, deposits in this area can be seen in wells A & D in Figure 1. The Ostracod interval, which consists of shale and limestone is a significant regionally extensive stratigraphic marker. Absence of the Ostracod is one of the key correlation indicators for downwards incision by overlying Glauc and Lithic channels. Petrophysical Analysis and Lame Parameters Petrophysical parameters for “complex” lithology, porosity, and fluid saturations were modeled. A rather complete log-data set including density, neutron, sonic, gamma ray and magnetic resonance logs permitted a comprehensive model to be developed and correlated to core lithology, porosity and saturation data. This “complex” lithology model is required to account for bitumen, gas, irreducible water and intraparticle porosity evident in the log, core and thin section data. Magnetic resonance data enables meaningful evaluation of bitumen and irreducible water in this lithology. Model parameter selection is constrained by comparison of model results to core lithology, porosity and fluid saturations. The petrophysical model illustrated by figure 2 for well B shows a clean quartz rich sand lithology having a maximum effective porosity of 0.26 (core porosity overlaid in red). The porosity match deviates in intervals where intraparticle porosity is not included in the core porosity. Effective porosity is reduced by shaliness at the top and by bitumen at the base of the sand. Modeled hydrocarbon saturation is overlain by core oil saturation (in red). The saturation curves correspond in intervals containing residual oil and deviate in gas bearing intervals.