Abstract Introduction Geologic Setting Stratigraphic Setting Porosity Determination Data Set Porosity vs. Depth Depth vs. Age Effect of Pore- Water Composition Limestone & Dolomite Porosity Summary References

Table 2 summarizes the wire-line data set used in this study. Initial wire-line coverage totaled 82,638 ft (25,188 m) of section in the South Florida basin. Obvious non-carbonate facies were eliminated by visual inspection of the wire-line logs, leaving 50,002 ft (15,241 m) of wire-line data that were subdivided into 1,695 intervals of variable thickness (5 to 70 ft; 1.5 to 21.3 m) but roughly homogeneous composition and log character. The lithology of each subdivision was quantitatively determined from lithology logs (borehole-gravity data) or the M-N plot (conventional-log data), and intervals with evidence of non-carbonate minerals were rejected. This left as the final composite data set 1,302 intervals from 15 wells representing 41,459 ft (12,637 m) of clean carbonate rocks. Each time-stratigraphic unit used in this report is represented in the composite data set (Table 2).

The rather irregular distribution of interval thicknesses in the composite data set is shown in Figure 4a. The large peak is due to the arbitrary subdividing of uniform lithologies (primarily in the Gulfian) into 50-ft (15.2-m) intervals. Interval thicknesses are roughly comparable to the vertical scale of hydrocarbon reservoirs. A full range of limestone-dolomite percentages, including end members, is represented in the composite data set (Fig. 4b). Figure 4c shows the distribution of interval-porosity values, which range from near zero to 55%. Taken as a whole, the intervals studied represent a very porous assemblage of carbonate rocks.

Fig. 4 - Distributions of interval thickness (A), carbonate composition (B), and porosity (C) for the wire-line data of this study. [larger image]

The 1,302 intervals form a data set from which general porosity relations, representative of the basin as a whole, can be calculated. Such statistical results represent averages for thick carbonate sequences and are not site specific. Individual carbonate strata may show a considerable variation above or below the regionally representative regression curve.

Wells in which wireline data were obtained probably were not drilled at random, but according to criteria that may have biased the sampling of basin facies. Also, it is possible that the methods of wireline analysis used here resulted in a systematic rejection or acceptance of intervals that should or should not have remained in the data set. Having considered these possibilities, we nevertheless believe that the data of this study represent a reasonable sampling of carbonate porosity in the South Florida basin, from which we can draw conclusions with basin-wide applicability.