Layer | Fill | Outline |
---|
Map layers
Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|
Fill | Stroke |
---|---|
Collaborating Authors
Results
The most common technique used for modeling and estimating the elastic properties of rocks under various fluid scenarios is Gassmann's method (Gassmann 1951). However there is always a risk of making a human error when applying this method, especially on heterogeneous media. One example for such media is a thinly bedded sand-shale sequence, in which laminations are often if not always ignored, especially in industry software. In this paper we will discuss three different fluid substitution approaches based on Gassmann's method performed on well log data from a deep water turbidite offshore Angola and how each of them yields a different outcome despite the fact that the underlying relationships are the same. They key message here highlights the significant difference we observed on the Gassmann fluid substitution results when sand-shale laminations are ignored compared to when they are properly handled. Introduction Gassmann's equation which relates the bulk modulus of a rock to its pore frame and fluid properties assumes a homogeneous mineral modulus and statistical isotropy of the pore space but free of assumptions about the pore geometry.
Summary Sub-resolution thin shaly-sand reservoirs have subtle seismic signatures requiring a probabilistic interpretation of their seismic attributes. We model thin sand-shale sequences and their corresponding synthetic seismograms to generate a training set. The set is used to extract attributes based on kernel principal components for characterizing sub-resolution reservoir properties such as net-to-gross ratio. The training set is created by varying several important parameters. Even though some of these parameters are not reservoir properties of interest, they should be included to yield more accurate seismic responses. We find that the impact of the non-reservoir parameters can overshadow that of reservoir parameters. This observation is confirmed by a sensitivity analysis in which we rank all parameters according to their impact on seismic signatures. To reduce the impact of the non-reservoir properties, we perform a cascaded interpretation scheme by first limiting the ranges of the non-reservoir parameters and generating a new training set. This new training set which has a better relation between seismic signatures and reservoir properties can then be applied to real seismic data for property estimation (e.g., net-to-gross ratio).
- Geology > Rock Type > Sedimentary Rock > Clastic Rock > Sandstone (0.87)
- Geology > Rock Type > Sedimentary Rock > Clastic Rock > Mudrock > Shale (0.71)