Access to sub-surface cores for running multiple experiments is gradually becoming a rarity, chiefly for two reasons; it is hard on time and on money. Additionally, the alternative provided by the digital methods is becoming increasingly popular not only for the better economics but also for the repeatability of results and the flexibility in modeling multiple scenarios on the same digital volume. Digital Rock Physics (DRP) offers to model the static and dynamic reservoir properties with better control on subjective biases of the experimentation and is non-destructive in nature. DRP involves imaging the formation samples and simulating the field performance to account for various non-homogeneities in the reservoir formation. DRP has proved to be highly successful in clastic reservoirs but in case of complex reservoirs such as carbonates and unconventional resources like shales, it is still at the feasibility stage only. The reasons are plenty ranging from method of imaging, availability of calibration libraries, transition space error, and its quantification.
In this paper, we look at a number of CT and micro CT images from carbonate formations - exhibiting range of heterogeneity - to understand the detectability of pore volume, the transition boundaries between pore and grains and the pore types -w.r.t aspect ratio and size-present in the formation. Given the set of challenges, this work had two fold objectives;
to estimate the impact of heterogeneity on successful determination of the pore volume in carbonate rocks and
to segregate the pores into their types; by size (micro, meso, and macro) and by aspect ratio (AR) which intuitively determines the flow regime in the formations.
Whereas, size of pores controls the schemes of fluid flow in formations, their aspect ratio controls the elastic property behavior in stressed rocks. Because the processes are coupled, it becomes all the more important to calibrate DRP modeling results with laboratory measurements for larger understanding and for creating rock physics forward models that combines these processes. In this work, we explain the approach of testing and filtering to reach a successful segmentation algorithm, its comparison with the laboratory measurements and also its limitation. We will also discuss the effect of wettability on the fluid flow, especially in the scenario where oil gets trapped in the pores (in the case of a water wet or oil wet reservoir).
