A New Workflow for Comprehensive Petrophysical Characterization of Carbonate Reservoirs Drilled with Water-Base Muds

Abstract

One of the top concerns for carbonate reservoir evaluation is the effect of rock texture on permeability, capillary pressure and relative permeability. Recent advances in log analysis combined with new logging sensors that are sensitive to carbonate rock texture have led to an improved workflow for petrophysical analysis of carbonates. The authors have earlier described an approach to estimating permeability in carbonates from borehole NMR logs and electrical images, and have earlier studied the relationship between NMR T2 distributions and capillary pressure curves in carbonates. Additional enhancements have been made to this workflow to include estimates of relative permeability by modeling invasion of mud filtrate utilizing a fluid flow model in combination with array resistivity logs. Analyzing relative permeability in conjunction with formation permeability and capillary pressure leads to new insights in log-based rock typing for comparison with Special Core Analysis (SCAL) data. This workflow is presented in the form of a case study of a carbonate well in the UAE. The workflow will be reviewed in its entirety with particular emphasis on the relationship between log rock texture and permeability, capillary pressure and relative permeability. Introduction Many of the giant carbonate fields of the Middle East are undergoing Enhanced Oil Recovery (EOR) via water flood, gas injection and/or combinations of both to improve ultimate oil recovery. The advance of fluid fronts in fields undergoing EOR typically varies within the formation layers, with permeability usually being the controlling factor. Decades of experience with routine and special core analysis, surveillance logging and production logging have confirmed the critical role permeability plays in hydrocarbon recovery and Masalmeh et al. (2003) examine the effect relative permeability and capillary pressure play in EOR. Several authors, Gyllensten et al. (1999) and Haro (2004) have proposed various methods for estimating permeability from wireline logs and one of the most common is to identify rock types, which each have a particular porosity-permeability relationship. The workflow presented here facilitates various forms of rock typing. Marzouk et al. (1997) presented a form of carbonate rock typing based on porosity partitioning that is consistent with the Dunham Classification (Dunham, 1962, Embry and Klovan, 1971). Amaefule et al. (1993) describe a more generalized rock typing method based on the Flow Zone Indicator (FZI) that delineates rock types from poro-perm relationships. Grotsch et al. (1998) have presented a rock typing scheme based on property cut-offs, thin section analysis, and high pressure capillary pressure data and pore-throat size distributions. These three forms of rock typing can be supported by the workflow presented in this paper and, in addition to these methods, there is potential to examine a new form of rock typing based on estimates of relative permeability. Carbonate Analysis Workflow The workflow presented has three basic steps and is described in detail by Ramamoorthy et.al. (2008). At each step of this workflow there are log displays and cross-plots to check results against available core data. These checks will be described for each step. Both routine core analysis (RCAL) data such as porosity, permeability and grain density, and Special Core Analysis (SCAL) data such as mercury injection capillary pressure (MICP), and relative permeability curves can be utilized during the workflow. Even though such core data may be available only on a few key wells in a given field, the fine tuning of parameters based on the detailed core-log analysis can normally be applied on a field wide basis.

Effects of pore size and pore connectivity on trapped gas saturation

“…8 (Lima, et al, 2020). This correlation of the results makes it possible to partition the pores (e.g., Gyllensten et al, 2008).…”

“…As noted earlier reported, only three samples were submitted to MICP tests based on the choice of Sgt values (high, medium, and low value), i.e., samples SD_35_03 (silurian dolomite), MC20P2 (coquina), and IL_09_05 (Indiana limestone). The throat size partitioning scheme described by Gyllensten et al (2008) was used to differentiate between different regions of the pore system: micropores from 0 to 0.5 µm, mesopores from 0.5 to 5 µm, and macropores above 5 µm.…”

Numerical Modeling of In-Situ Relative Permeability and Capillary Pressure from Inversion of Wireline Formation Tester Data

“…It is also possible to have indications of an initial PC curve from the T2 distribution of Nuclear Magnetic Resonance (NMR) logs. The PC curve derived from NMR logs is equivalent of a drainage PC curve as described by Gyllensten et al (2008).…”

Integration of Dielectric Dispersion and 3D NMR Characterizes the Texture and Wettability of a Cretaceous Carbonate Reservoir

“…Ramamoorthy et al (2008) and Gyllensten et al (2008) have demonstrated a workflow applicable to the Cretaceous carbonate reservoirs in the Middle East which exploits the information on rock texture available in borehole NMR and image logs along with conventional triple combo data and array resistivity data.…”

“…Recent field trials of an advanced dielectric logging tool and of a multi-dimensional NMR tool have shown that they provide very accurate saturations in the zone around the well-bore invaded by the mud filtrate Zielinski et al 2010). These tools have also been shown to provide textural information describing pore geometry (Gyllensten et al 2008).…”