Heterogeneity in carbonate reservoirs contributes to large ranges in hydrocarbon production rates.  Reservoir heterogeneity is generated by processes related to deposition, diagenesis and structuring.

For example, initial production rates up to 200K BOPD per well were recorded from Yates Field, West Texas, US. Higher rates were enabled by deposition of a coarse-grained shelf margin shoaling system that was intermittently overprinted by intense meteoric karsting.  Poorer producers were developed in inboard muddier facies with little or no karstic overprint. Another example of extremely heterogeneous production patterns comes from Tengiz Field in Kazakhstan. Higher production rates come from wells drilled in the so-called “raised rim” — a culmination feature produced by preferential compaction of adjacent platform interior and outboard deeper-water slope sediments.  Raised-rim sediments were more protected from compaction by microbially-related cementation during deposition. These lower-porosity rocks were more mechanically rigid in the subsurface and subsequently fractured to produce high-permeability reservoir rock.  Best producing wells mimic the ring-shaped distribution of highly fractured rocks in the raised rim. Lower-producing wells are generally distributed within the platform interior and in deeper slope settings where fracturing is not as well developed.

In both preceding examples, dense pattern of development wells confirmed that highest producing wells in the fields correlated with 3D spatial overlaps of beneficial processes — in these particular examples a) deposition of coarse-grained sediment, b) early diagenetic leaching, and c) burial fracturing. Many other combinations of beneficial processes are evident in global production analogues.

One large challenge in predicting patterns of hydrocarbon production in carbonate fields is deconvolving positive and negative effects of multiple input processes. The deconvolution workflow presented in this talk involves a) identification of dominant processes both beneficial and destructive, b) description of associated geobody distributions in three dimensions — sometimes spatially overlapping, and c) characterization of reservoir quality and distributions within individual geobodies. Predictive calibration comes from local field data as well as appropriate production analogues, which should be selected on the basis of fundamentals in physics and chemistry — not necessarily based on age or proximity of the producing formation. Important also is to recognize that dominant reservoir-modifying processes may change over the geological life of the field, modifying the patterns of best productivity.