There are numerous examples of tight naturally-fractured gas reservoirs with active water in the foothills of Alberta and Northeastern British Columbia. Examples include the Pincher Creek field in Alberta and the Bucking Horse, Pocketknife, Sikanni and Grassy fields in British Columbia.
Recovery factor is typically low from this type of reservoir due to water production. The problem is frequently attributed to coning, but coning may actually play a minor role. In fact, the gas/water contact in the fracture system may be relatively flat.
Initial gas recovery comprises gas displaced from fractures plus pressure depletion from the matrix. The amount of pressure depletion in the matrix is a function of structural relief above the original gas/water contact.
Subsequent gas recovery is an imbibition process, which may be very slow. Laboratory work, conducted at the TIPM Laboratory on behalf of Husky, demonstrates that water will continue to imbibe into tight matrix rock submerged under water for months.
This work implies that the best operating strategy maybe to produce the wells at the highest rates possible until water breakthrough, followed by a shut-in period of perhaps several years to allow gas to re-accumulate.
Significant gas reserves are contained in structured Mississippian-age reservoirs along the eastern foothills of the Rocky Mountains. Equivalent formation names include Livingstone, Turner Valley and Debolt (from southern Alberta through northeastern British Columbia).
Matrix permeability in these reservoirs is often insufficient to support commercial production rates. Fortunately, natural-fracturing enhances reservoir permeability, and many foothills gas reservoirs are intensely fractured. Fracturing results from post-depositional thrusting. Prolific wells can be drilled where fracture intensity is greatest, usually along the hinge of a folded structure.
In most cases, these reservoirs overlie inactive aquifers. Recovery is simply a function of abandonment pressure, which is in turn a function of minimum economic production rate. Recovery is typically greater than 80 % of original gas-in-place.
Aquifer influx can have a devastating influence on recovery, however. Recovery from a reservoir overlying an active aquifer may be less than 20 % of the original gas-in-place. Wells may water-out very abruptly.
The problem is often blamed on coning or channeling. Fractures are mistakenly seen as preferential conduits, or "pipelines" into the water leg. With this view, gel treatments, rate restrictions and plug-backs may be implemented to reduce water production; almost always unsuccessful1y.
Our premise is that the problem is not caused by coning or channeling. Fractures are planar; not tubular. Even in a vertical fracture, horizontal permeability along the plane of the fracture is approximately equal to vertical permeability. Fractures are not preferential conduits into the water leg.
Furthermore, viscosity of gas is typically two orders of magnitude less than viscosity of water, and density of gas at reservoir conditions may be an order of magnitude lower than density of water. Therefore, extremely high pressure gradients would be required to generate a small cone in a permeable fracture system.
Rather, in most cases, gas is efficiently displaced from the fractures as the free gas/water contact rises uniformly across the pool.
