Fluid flow in light, tight oil (LTO) reservoirs is by nature capillary-force dominated, as opposed to viscous-force dominated. However, most LTO reservoir models rely solely on well-log data to initialize volumes of original hydrocarbons in place, ignoring the physics of capillarity. This reliance on well logs owes to the difficulty of generating accurate, core-derived saturation height function models (SHFM). Moreover, when initializing reservoir saturation distributions to drainage capillary pressure curves, the resulting dynamic model must then obey matrix fluid transport, which is governed by oil-water capillary pressure relationships. This often creates difficulty in the history matching process for numerical simulation models that must honor production data. A contradictory approximation of "zero capillary pressure" is often made for simplification. We successfully constructed a core-derived SHFM and matrix permeability model for the Lower Etchegoin/Williamson reservoir at the Lost Hills field, California. This unconventional reservoir contains light oil (20 to 25° API) with low matrix permeability (0.05 to 10 md). The SHFM and matrix permeability models were constructed using routine and special core analyses from three wells that were cored and logged. High pressure mercury injection (HPMI) capillary pressure measurements were acquired on 16 core-plugs taken across the three main reservoir zones. From capillary pressure analyses, a Williamson SHFM was generated based on best-fits of irreducible water saturation, capillary entry-pressure, Brooks-Corey exponent, and permeability/porosity combinations. A single free water level (FWL) parameter was tuned in the SHFM at each of the well locations to force a match with the resistivity-derived saturation profiles. An excellent match of the saturation profile for the three wells was achieved. Unexpectedly, the Williamson SHFM has also proven useful in identifying the presence of geologic faults, independent of logs and seismic. When the capillary pressure model was applied to step-out areas, requiring a ‘depth offset’ correction to maintain a constant FWL across fault blocks (relative to the main reservoir block FWL), it was successfully hypothesized that faulting was present. The ‘depth offset’ correction represents the overall vertical fault displacement between reservoir blocks.
It sounds like a contradiction -assessing waterflood oil recovery by layer in a reservoir without knowing the original saturation distribution. However, we have developed a new method for estimating layer oil recovery efficiency using nuclear magnetic resonance (NMR) and dielectric logs run in wells within a mature, onshore waterflood. We use post-waterflood NMR T2 signals to give the reservoir rock pore-body size distribution at depths along the wellbore. This depth-dependent distribution can be converted into a series of oil-water capillary-pressure curves. The capillary pressure curves can then be modeled into the original, pre-waterflood saturation height model (SHM), which gives original oil in place (OOIP). The dielectric log, also acquired postwaterflood, is used to measure the remaining oil distribution in the reservoir. The breakthrough concept is that the saturation difference between the NMR-derived saturation height model and the post-waterflood dielectric log, which corresponds to the amount of oil swept by the flood, can be measured post-waterflood "all in one go."Conventionally, pre-primary recovery well logs are needed to delineate native-reservoir saturation distributions and provide a baseline for OOIP. Although ~1500 resistivity well logs were acquired in the diatomite reservoirs of the South Belridge field in the San Joaquin Valley, California, between 1979 and 1987, our understanding of localized original saturation variations is complicated by many factors. These include primary depletion and the injection of variable salinity water over time. For these reasons, this "all in one go" alternative method was developed to estimate OOIP and waterflood sweep efficiency in the diatomite.The biggest test of this new technique came when comparing the OOIP calculated from the NMR-inferred capillary pressure model against the existing geologic model (localized averages of early well logs) used for the South Belridge field. We found that the OOIP calculated by the new technique agreed favorably with that calculated by the existing geologic model. As a result, a re-assessment of sweep efficiency in the multi-layered South Belridge waterflood is being undertaken on a layerby-layer basis. This new, post-waterflood NMR and dielectric logging assessment of sweep has potential applications to other reservoirs.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.