After fracture stimulation, production often rises slowly, instead of showing an early transient. This indicates either severe reduction in fracture conductivity or reservoir damage. There is still no agreement in the industry about the most important damage mechanisms. Work done by Pratts and Holditch in the 1970’s showed that fracture face damage from filtrate invasion is unimportant, unless there is permeability damage in the invaded zone of at least 99%. New ideas have been proposed which may better explain the behavior commonly seen in actual production data. These ideas include relative permeability with water and gas both immobile at a given saturation ("permeability jail"), small scale reservoir heterogeneity and stress sensitive matrix permeability at high drawdown.
Experience from the field has been contradictory. Sometimes no water is produced back, but gas production does not appear to suffer. In other cases, gas rate is significantly lower than expected, but significant fracture fluid is recovered. With the inherent coupling of fracture length and permeability in welltest interpretation, and the practical impossibility of achieving radial flow in tight-gas reservoirs with large fracture lengths, it has been difficult to prove any theory about the cause of poor performance from tight-gas fracture treatments. This paper shows simulation results of these effects on post-fracture production from an unconventional (0.001 mD) gas well. Furthermore, realistic assumptions about proppant pack cleanup show a connection not only between poor cleanup and short effective fracture length, but also reduction in contacted kh and connected reservoir volume. New reservoir simulation results are presented showing 50% reduction of production in the first years due to these effects in unconventional reservoirs.