Heterogeneous Formation: Permeability Variation, Reservoir Compartmentalization, and WCSB Waterflood Sweep

A heterogeneous formation is a reservoir rock whose petrophysical properties, principally porosity, permeability, fluid saturation, lithology, and pore geometry, change measurably from one location to another within the same producing interval. Almost every commercial reservoir is heterogeneous to some degree; the term is reserved for rock where the variation is large enough to control well placement, recovery factor, and development economics rather than being a minor statistical scatter. Heterogeneity arises from the depositional history of the rock and from everything that happened to it afterward. Fluvial channels, tidal bars, deltaic mouth bars, carbonate reef build-ups, and turbidite lobes each lay down sediment with internal grain-size trends, shale drapes, and bedding contacts that later become permeability barriers or thief zones. Diagenesis then overprints the original fabric through cementation, dissolution, dolomitization, and compaction, while tectonics adds natural fractures and faults. The result is a rock body where permeability can swing across several orders of magnitude over a few metres, from less than 0.1 millidarcy in a cemented baffle to several darcies in an open fracture or a clean channel sand. In the Western Canadian Sedimentary Basin this is the rule, not the exception. The McMurray Formation oil sands carry abrupt inclined heterolithic stratification where mud-draped point bars alternate with high-permeability sand, the Cardium at Pembina mixes conglomerate, sandstone, and tight interbeds, and the Leduc and Nisku carbonate reefs hold vuggy, fractured, and matrix porosity side by side. Engineers quantify heterogeneity with measures such as the Dykstra-Parsons coefficient, which runs from 0 for a perfectly uniform formation to nearly 1 for an extremely variable one, and the Lorenz coefficient derived from a flow-capacity versus storage-capacity plot. These numbers feed directly into recovery forecasts, because a heterogeneous formation channels injected water or gas through the most permeable streaks and bypasses oil held in tighter rock, lowering sweep efficiency. Characterizing that variability through core analysis, well logs, pressure transient testing, and seismic inversion is the central task of reservoir description, and it governs how a naturally fractured reservoir is drained, where infill wells are drilled, and whether enhanced recovery will pay out under AER conservation rules.

Measuring Variability with the Dykstra-Parsons and Lorenz Coefficients

Reservoir engineers convert a cloud of core permeability values into a single heterogeneity index so models stay comparable across fields. The Dykstra-Parsons coefficient is read from a log-probability plot of permeability: V equals (k50 minus k84.1) divided by k50, where k50 is the median and k84.1 is the value one standard deviation below it. A Cardium tight-oil interval at Pembina might log a V of 0.75, flagging a strong contrast between conglomerate streaks and silty interbeds. The Lorenz coefficient instead plots cumulative flow capacity (k times h) against cumulative storage capacity (porosity times h); the further the curve bows from the 45-degree line, the more flow is concentrated in a thin high-perm fraction. Both numbers feed fractional-flow and streamline models that predict how fast injected water in a WCSB waterflood will break through.

Compartmentalization and the Limits of Connected Pore Volume

Heterogeneity at the largest scale produces compartmentalization, where sealing faults, shale drapes, or cemented diagenetic barriers split a reservoir into pressure-isolated blocks that do not communicate over production time. A Montney horizontal well can intersect several such compartments along a 2,500 m lateral, each with its own initial pressure, so early production from one block masks an undrained block next door. Repeat-formation-tester pressure gradients and material-balance analysis expose the disconnect: two zones a few metres apart sitting on different pressure trends prove a barrier between them. Recognizing compartments early changes the completion plan, since cluster spacing and stage placement must target each isolated unit rather than assuming one continuous tank, a distinction worth millions in bypassed Duvernay or Montney pay.

Related Terms

Heterogeneity is best understood alongside permeability, the directional flow property whose spatial variation defines the heterogeneity in the first place, and porosity, which sets storage capacity and often varies independently of permeability in carbonates. A naturally fractured reservoir is a special, extreme case of heterogeneity where fractures dominate flow and matrix holds the oil, while waterflood performance is the clearest economic expression of heterogeneity, since sweep efficiency falls as permeability contrast rises.

Real-World WCSB Scenario: Cardium Waterflood Conformance at Pembina

An operator running a mature Cardium waterflood in the Pembina field, the largest conventional oil pool in Alberta, watched water cut at a row of producers climb past 90 percent while offset producers a few hundred metres away still made dry oil. Tracer surveys and a Dykstra-Parsons coefficient near 0.78 confirmed injected water was channeling through a thin conglomerate streak and bypassing oil in tighter sandstone above and below. Engineers designed a polymer-augmented conformance program, injecting roughly 800,000 CAD of polymer per pattern to raise the displacing-phase viscosity and force water into the bypassed rock.

Within 14 months the treated patterns added incremental oil at a recovery uplift of about 4 percent of original oil in place, and water cut on the channeling producers fell from 92 to 81 percent. The project cleared AER Directive 065 reservoir-management requirements and paid out in under two years at prevailing WTI, a direct return on correctly diagnosing the formation as heterogeneous rather than uniformly swept.