The use of common correlations for the cementation factor, m, has led to discrepancies between log interpretation and actual test results in several tight carbonate formations offshore Abu Dhabi. While logs indicate high water saturations, production tests resulted in unexpected dry hydrocarbons.
The existing common correlations for carbonates yield constant m (Archie) or an increasing m with a decrease in porosity (Shell formula). In the carbonates found offshore Abu Dhabi, m decreases with decreasing porosity. Based on core and log data from five formations in 14 structures, a more representative correlation for m has been derived. This new correlation successfully eliminates the discrepancies between log and test results. Examples confirming the validity of the new formula are included.
Recent interpretation of openhole logs offshore Abu Dhabi has indicated high water saturations across some zones. Hydrocarbon indications during drilling and correlation with nearby wells have shown these zones to be hydrocarbon-bearing. Testing of these zones produced dry hydrocarbons, thus casting doubts on log produced dry hydrocarbons, thus casting doubts on log interpretation results.
Because of the discrepancies between logs and test results, the validity of the different parameters used in log interpretation was investigated. The results indicate that the main reason for these discrepancies is the use of an improper cementation factor, m. This paper reviews logs and core measurements that led to a new m correlation and discusses examples where discrepancies between log and production test results have been reconciled.
In 1942, Archie showed experimentally that the resistivity of a rock completely saturated with a conductive fluid, Ro, is related to the resistivity of the conductive fluid, R.., by
where FR is the formation resistivity factor. When Archie plotted FR vs. rock porosity, o, for various clean formations, he found that
FR =o-m,...................... ............. (2)
where m represents the slope of the linear trend of FR vs. o when plotted on logarithmic scales. Archie stated that m will vary plotted on logarithmic scales. Archie stated that m will vary according to the degree of cementation of the rock. Therefore, m is generally referred to as the "cementation factor."
Archie's relationship was later modified by Winsauer et al. to the general form
FR =ao-m..................................................... (3)
where the constant a was introduced to account for the presence of solid conductors and/or clay in the formation.
Successive work by other authors has confirmed these two relations for different rock types leading to different values of the constants a and m. However, the values generally used for carbon atrocks are a = 1 and m = 2.
Another relationship recommended for use in evaluating low-porosity carbonates is the Shell formula
This formula increases m values with a decrease in porosity.
The following values of m were used in our log interpretation:
(1) Archie's formula with m=2 for medium- to high-porosity formations, and (2) the Shell formula for low-porosity formations of less than 10%.
The study was initiated when intervals showing high water-saturation values in log data processing were tested and produced virtually dry hydrocarbons. The study covered the concession area offshore Abu Dhabi of about 3,860 sq miles [ 10,000 km2] (Fig.1).
Both log and core data from the main carbonate reservoirs encountered offshore Abu Dhabi were used. These reservoirs lie within a sedimentary succession of more than 17,000 ft [5182 m] (Fig. 2). The lithology of these formations is mainly limestone; however, dolomite and dolomitic limestones commonly occur, The reservoirs are essentially clean.
Log data were obtained from the water-bearing formations in 12 different fields (Table 1). Care was taken to use wells where good hole conditions and completeness of logging suites allowed reliable estimation of porosities and resistivities. A cutoff on the gamma ray was used to retain the clean intervals only.
The formation resistivity factor, FR, was calculated as Rt/Rw. Rt was calculated from induction and dual laterologs, while Rw was obtained from laboratory analysis of formation water samples.
Porosities were derived from a combination of density and neutron logs and use of a dual mineral model for a limestone/dolomite matrix.
The data were presented as crossplots of the formation resistivity factor, FR, vs. porosity, 0. on logarithmic scales. Fig. 3a through 3d represents typical examples of FR-vs.-O crossplots obtained for different types of lithologies.
Fig. 3a through 3c demonstrates that the variation of FR with porosity was consistent for all the carbonate formations within the porosity was consistent for all the carbonate formations within the concession area. The presence of scattered anhydrite nodules in some carbonate formations in limited areas was noted. These nodules, however, did not alter the shape of the curve of FR vs. porosity, but raised the FR values above the average values noted with porosity, but raised the FR values above the average values noted with clean carbonates (Fig. 3d).
The findings from logs of limestone. dolomitic limestone, and dolomite formations were confirmed by core measurements.
Core Measurements. Formation resistivity was laboratory-measured in 64 core plug samples 1 or 1 1/2 in. [2.54 or 3.81 cm] in diameter covering limestone, dolomitic limestone, and dolomite formations. The plugs were chosen from cores from six different fields (Table 1).