Dual-Permeability Reservoir: Matrix-Fracture Flow, Shape Factors, and Fractured Reservoir Simulation

A dual-permeability reservoir is a numerical representation of a naturally fractured formation in which fluid can flow to the wellbore through both the fracture network and the rock matrix, and in which the matrix blocks also exchange fluid directly with neighboring matrix blocks rather than only feeding the fractures. It is a refinement of the more common dual-porosity reservoir idealization, where the matrix stores hydrocarbons and slowly bleeds them into the fractures but is assumed to have no continuous flow path of its own. In the dual-porosity picture the fractures carry essentially all the flow to the well while the matrix acts purely as a storage source term. The dual-permeability model relaxes that assumption: mass and energy balance equations are solved separately for the matrix continuum and the fracture continuum, and both continua are given their own transmissibility so that matrix-to-matrix flow contributes to well deliverability. This matters when the matrix permeability is not negligible relative to the fractures, when gravity drainage moves fluid vertically block to block, or when a displacement process such as waterflooding or gas injection depends on how efficiently the injected phase reaches and sweeps individual matrix blocks. The physical system is idealized as a stack of matrix blocks separated by a fracture network, and the rate of transfer between the two continua is governed by a matrix-fracture transfer function scaled by a shape factor that captures block geometry, size, and the number of fracture sets. Getting that shape factor right, along with the block dimension it embeds, is the central difficulty of fractured-reservoir simulation, because a wrong transfer rate produces the correct total volumes but the wrong recovery timing. In the Western Canadian Sedimentary Basin the approach is applied to fractured carbonates such as the Nisku and Leduc reef complexes and to fractured intervals of the Slave Point and Cardium, where fracture spacing, aperture, and matrix quality all vary and single-continuum models overpredict or underpredict recovery. Properties are reported in dual units, matrix and fracture permeability in millidarcies, block size in m alongside ft, and pressures in kPa alongside psi, and the resulting simulations underpin reserve bookings filed under AER and CER disclosure rules and National Instrument 51-101.

Key Takeaways

  • Both continua flow to the well: Unlike dual-porosity, where only the fracture network delivers to the wellbore and the matrix merely feeds fractures, a dual-permeability model gives the matrix its own transmissibility so matrix-to-matrix flow contributes to deliverability. This captures cases where matrix permeability is not negligible and where gravity moves fluid vertically block to block during depletion or displacement.
  • Two coupled continua, two grids: The simulator solves mass and energy balances separately for a matrix continuum and an overlapping fracture continuum occupying the same volume, coupled by a matrix-fracture transfer term. Each grid block carries both a matrix cell and a fracture cell, roughly doubling the equation count versus a single-porosity model but far cheaper than resolving every discrete fracture.
  • Shape factor sets transfer timing: The rate of fluid exchange between matrix and fracture is a transfer function scaled by a shape factor that embeds block dimension, geometry, and the number of fracture sets. An incorrect shape factor yields correct total volumes but wrong recovery timing, so calibrating it against core, well test, and production history is essential for a trustworthy forecast.
  • Critical for displacement processes: Waterflood, gas injection, and EOR in fractured reservoirs depend on how fast and how completely the injected phase reaches individual matrix blocks through the fast fracture network. Dual-permeability modeling captures matrix-to-matrix imbibition and gravity drainage that dual-porosity omits, giving more realistic sweep and breakthrough predictions in fractured WCSB carbonates.
  • Chosen for computational economy: Fully discrete fracture network (DFN) models resolve every fracture but are computationally prohibitive at field scale. Dual-porosity and dual-permeability continua remain the workhorse of field-scale fractured-reservoir simulation because they are far cheaper while still honoring the first-order effect of fractures on flow, storage, and recovery.

When Dual-Permeability Beats Dual-Porosity

The choice between the two models turns on matrix connectivity. If matrix blocks are effectively isolated and can only communicate through fractures, dual-porosity is adequate and cheaper. Dual-permeability becomes necessary when matrix permeability is high enough that vertical block-to-block flow matters, typically in gravity-dominated processes such as gas-cap expansion or gas injection into a fractured reef, where oil drains downward through stacked matrix blocks rather than only outward into fractures. In WCSB Nisku and Leduc reefs with reasonable matrix quality, ignoring matrix-to-matrix flow can materially underestimate late-life gravity drainage recovery, so simulators select dual-permeability for those intervals while reserving dual-porosity for tight, poorly connected matrix.

Calibrating the Matrix-Fracture Transfer Function

The transfer function converts a pressure or saturation difference between matrix and fracture into a flow rate, scaled by the shape factor and by matrix permeability. The shape factor depends on block size and the number of active fracture sets; smaller blocks and more sets transfer faster. Analysts estimate block dimension from image logs, core fracture spacing, and seismic-derived fracture density, then history-match the transfer parameters against production and pressure data. Because the same recoverable volume can drain quickly or slowly depending on this calibration, transfer-function tuning, not total porosity, usually controls whether a fractured-reservoir forecast matches observed decline and water breakthrough.

Fast Facts

The dual-porosity concept traces to a 1963 paper by Warren and Root, who idealized a fractured reservoir as a regular array of identical sugar-cube matrix blocks separated by fractures and introduced the shape factor still used today. The dual-permeability extension came later specifically to handle gravity-drainage and matrix-to-matrix flow that the original storage-only matrix could not represent, and both formulations remain embedded in every major commercial reservoir simulator more than sixty years after that founding idealization.

A dual-permeability reservoir is the fuller sibling of the dual-porosity reservoir, differing only in whether the matrix is allowed continuous flow. Both describe a naturally fractured reservoir, where tectonic or diagenetic fractures dominate deliverability. The behavior is quantified through permeability assigned separately to matrix and fracture continua, and the fracture side is often characterized with a discrete fracture network before being upscaled into the dual-continuum grid used for full-field simulation.

Real-World WCSB Scenario: A Fractured Nisku Reef in Central Alberta

An operator evaluating miscible gas injection into a fractured Nisku pinnacle reef in central Alberta first history-matched primary depletion with a dual-porosity model, which reproduced early rates but badly underpredicted late-life oil recovery from the reef flanks. Suspecting gravity drainage through connected matrix, the team rebuilt the model as dual-permeability, giving the matrix continuum its own vertical transmissibility. Fracture spacing from core and image logs set a mean block height near 3 m, and the shape factor was tuned against a decade of production. The dual-permeability model cost roughly CAD 250,000 in additional simulation and characterization work.

The refined model showed that block-to-block gravity drainage added several percentage points of recoverable oil under a proposed solvent flood, changing the project from marginal to economic. The reserves were re-booked under NI 51-101 with the dual-permeability forecast, and the gravity-assisted injection scheme was sanctioned on the strength of the corrected recovery timing.