Gas Reservoir

A large gas reservoir in North Africa had unusual water production especially in its crestal wells.  There were uncertainties as per the source of this water.  The gas-water contact did not move fast enough to breakthrough at these wells which were completed in the upper layer.  First argument that the water is encroaching from other layers through cracks in the casing was eliminated by careful water fingerprinting from the wells.   Once it was confirmed that the water is indeed reservoir water, attention turned in re-investigating well-related data. 

Detailed analysis of modified isochronal tests revealed systematic repetition of negative skin values.  In addition, the tests of horizontal wells can only be matched when vertical fractures were allowed.  This hypothesis of vertical fractures seemed more feasible when the tectonics of the area was re-visited as it seemed the crestal folds and faults were more conducive to creating and maintaining swarms of vertical fractures.  This finding was back-propagated in to the history match model to successfully converge to an agreeable fluid production scenario.

Midale reservoirs in Williston Basin are primarily fractured carbonates interspersed with anhydrites that appear and disappear in a haphazard manner. 

Matrix is tight and production is possible only because of the natural fractures present.  The wells are routinely acid-washed to keep them flowing.  Early attempts of history matching with single porosity models failed repeatedly.  Subsequent simulations of other fields with similar characteristics consistently suggested dual porosity (MINC) system behavior. 

One major difference between conventional and unconventional reservoirs is that the gas storage in unconventional reservoirs, apart from free gas, may also contain adsorbed gas.  Specialized algorithm incorporating gas, water and condensate flow under the constraint of overall mass balance has been devised in-house.  The deliverability of both vertical and horizontal wells intercepting multi-staged fractures can be computed.  In this algorithm, gas rate, water rate, condensate rate, water saturation in fractures and reservoir pressure, are solved implicitly by simultaneously accounting for material balance and inflow equations for water and gas phases.  The model parameters can be perturbed to match historical production or the model can be used to forecast expected production behavior of unconventional gas reservoirs. 

The model has been validated against production from shale plays across the country.  One perturbation parameter, bulk volume, provides an estimate of Stimulated Rock Volume (SRV).  The model focuses well-by-well and therefore has a faster turnaround time and is repetitive, unlike specialized fine grid-based simulation models.