The data, analysis and simulation resulting from a three-well interference test performed in a tight, lenticular naturally fractured reservoir at the Department of Energy (DOE) Multiwell Experimental (MWX) site, near Rifle, Colorado, will be discussed. The test data, that includes bottomhole pressure measurements from a production well and two closely spaced observation wells, provides considerable evidence that reservoir gas production was dominated by flow through a layered, anisotropic naturally fractured system. The interference testing consisted of almost two months of continuous data acquisition that included the introduction of several pressure pulses from the production well and the subsequent observation of those reservoir pressure transients at the remote well locations.
A series of novel completion schemes using high pressure nitrogen gas and explosive fracturing were employed in the completion process of all three wells. The primary intent was to eliminate foreign liquids from entering the reservoir fracture system following perforation in order to minimize external intervention in the natural production process.
When wellbore storage effects were considered to negatively impact the data set, bottomhole shut-in tools were employed and virtually eliminated storage volume. Both observation wells made use of these shut-in tools throughout the entire test yielding an unobstructed pressure interference data set.
The tight Fluvial reservoirs found in the Mesaverde interval at MWX contained some of the most interesting candidates for testing and stimulation experiments. Although in situ matrix properties for these reservoir did not represent a very favorable productive prospect, the inclusion of a rather extensive natural fracture system indicated that if these complex reservoirs could be effectively stimulated, it would substantially increase the likelihood that they would become favorable, economic targets for commercial production.
Previous testing and stimulation experiments in underlying Paludal, and Coastal reservoirs at MWX made clear the extreme sensitivity of the natural fracture system to invading completion and hydraulic stimulation liquids. Although damage to the reservoir matrix was considered to be detrimental, in a relative sense the more serious problem was clearly maintaining the high productive flow paths of the natural fractures in their original state. Unless certain natural fracture properties could be adequately understood and defined, any hydraulic "stimulation" scheme would be more speculative than certain, and in f act, may very well not only prove to be ineffective in enhancing production, but could in f act impart sufficient damage such as to reduce production.
The effects of anisotropic producing natural fractures have been shown to significantly impact post-stimulation production.