Bureau of Economic Geology, The University of Texas at Austin (www.beg.utexas.edu).
AAPG Hedburg Conference, Adelaide, Australia, Nov. 29Dec 6, 2002
Paul R. Knox
Risk assessment for petroleum prospects, aquifer contamination, and carbon dioxide sequestration relies partly on successful prediction of either capillary characteristics or permeability of known shales that form either a potential reservoir seal or a confining aquitard. For these purposes, the ability of the shale to impede flow is frequently considered to be a consequence of its thickness. The distribution of hydrocarbons in Frio sandstone intervals of West Fulton Beach field in Copano Bay, Texas, suggests that, instead, this ability is a function of the quality of the shale (fabric and ductility), which can be predicted by the shale's relationship to a hierarchy of flooding surfaces and its position within a depositional cycle. Furthermore, a possible interpretation of reservoir-filling history indicates that the hierarchical system of shales partitioned hydrocarbon charge, both volume and type (gas vs. oil), through time as oil and gas moved from the migration path (a major growth fault) into reservoir sandstones.
Subregional correlation of SP and resistivity logs in the area of West Fulton Beach field identified 3rd-, 4th-, and 5th-order maximum flooding surfaces, and associated shales, on the basis of stacking pattern of barrier-strandplain position. A repetitive pattern was observed at each scale of flooding-surface-bounded depositional cycle of larger, gas-dominated reservoirs transitioning downward to smaller, oil-dominated reservoirs, then to wet sandstones. This pattern is independent of the thickness of the shale above each reservoir. In fact, many of the shales above the largest reservoirs are thinner than most others, suggesting that shale quality (low permeability/high capillary entry pressure), not shale thickness, probably governs seal capacity. Thinner, possibly higher quality shales in the transgressive limb of a depositional cycle, and at lower order (e.g., 3rd-order) cycle boundaries, may be thinner because their distal setting results in less clastic input, as well as less silt content and higher organic content. This combination results in a more aligned clay platelet fabric, increased ductility, and smaller pore throats, which together yield lower vertical permeability (hydraulic conductivity) for wetting phases (usually water) and greater capillary entry pressures in nonwetting phases (usually hydrocarbons). In contrast, higher frequency flooding shales in progradational limbs may be thicker but siltier, having higher permeability (conductivity) and lower entry pressures.
Another, potentially complementary, explanation of observed reservoir patterns is that hydrocarbon charge, rising along the bounding fault of West Fulton Beach field, is preferentially partitioned into reservoir sandstones beneath a lower order flooding shale. This partitioning might occur because the presumed finer grained and more ductile composition of such shales forms a better gouge within the fault plane, preventing fluids of a given buoyancy pressure from hydraulically opening the fault plane to allow further vertical migration. Fluids are then diverted first into the highest sandstone under a major flooding-event shale until the reservoir pressure increases to a point at which the next underlying sandstone presents a lower pressure, thus lower energy, pathway for the migrating fluids. Under this scenario, the higher gas percentages in higher reservoirs might be the result of later upward remigration of the more mobile gas portion of underlying reservoirs through higher order shale seals of lesser seal capacity (in effect, baffles). An alternate explanation of gas distribution is that multiple sands are receiving charge simultaneously and that the higher mobility and buoyancy of the gas result in a partitioning into higher reservoirs under a major seal.
Interestingly, in third-order reservoir packages, total reservoired hydrocarbons decrease downward from cycle to cycle, as they do at other scales, but each cycle becomes slightly more gas rich, in contrast to patterns in fourth- and fifth-order packages, where lower reservoirs are more oil rich. One possible explanation is that these packages filled separately over a longer period of time, during which source maturity increased, progressively producing a greater percentage of gas. This observation, combined with those described earlier, suggests that these stacked sandstones filled more as partly partitioned hydraulic systems than as individual reservoirs.
Further evaluation of these observations, and those in other fields, is required to more fully evaluate these hypotheses. Such a study should relate measured shale petrophysical properties to the shale microfabric, depositional setting, and position within a stratigraphic hierarchy and should also correlate known hydrocarbon column heights with that framework. It may thus be possible to improve reservoir charge/seal prediction, contaminant transport risk assessment, and long-term leakage risk for underground storage projects.