Liquids Rich Unconventional Montney: The Geology and the Forecast

Abstract: The reservoir in the central part of the Alberta portion of the Deep Basin Montney Trend yields the promise of a resource play that may be as extensive, productive and rich in liquids as the best portion of the Eagle Ford Shale in southern Texas. Various geological and petro-physical parameters are compared, along with initial production signatures to show the similarities with the liquids rich, Eagle Ford analog. Liquid and gas analyses are recombined at various, observed, condensate/gas ratios and subsequently input into a compositional simulation model. The extreme representations of these fluid characterizations, varying from rich gas to volatile oil, can be used to achieve comparable history matches of producing wells. The reservoir fluid grades from a rich gas (50 Bbl/MMcf condensate) to a light crude system (3,350 scf/Bbl), from west to east, with a coincident rise in elevation of 100 m. There are no ‘traps" holding the fluid in place other than the very low permeability of the reservoir. Large flow potential gradients exist from the westerly, down-dip, gas rich portion of the reservoir, toward the easterly more liquids rich region. This dynamic liquid on top of gas situation is upside down relative to conventional trapped hydrocarbon deposits. It is the result of dissipation of hydrocarbons away from their point of source and the fact that the catagenesis process converts source materials preferentially to methane with increased depth and temperature. This situation is known to occur in a number of deep basin, over pressured, mixed hydrocarbon deposits in North America including the Montney, the Eagle Ford and the Utica. Reservoir modeling is complicated by the need to initialize a model with lower density fluid underlying more dense fluid and with large potential gradients through the hydrocarbon column. This study illustrates a method used to establish the initial, dynamic as to geologic time, state of fluid distribution. It goes on to illustrate why conventional means of characterizing reservoir fluids are inappropriate due to the nature of the reservoir fluid distribution and, possibly, misleading. Fluids sampled from a point source, be it from a surface location or the bottom of a horizontal well’s vertical section, is not representative of the phase distribution along the entire horizontal well lateral. The 2,430 meter long, Montney horizontal lateral, located at 9–12–64–4W6M, straddles a transition zone that penetrates myriad hydrocarbon phases and compositions, the aggregate of which cannot be represented by a single phase envelope. These types of wells could be classified as either "Gas" or "Oil" if using conventional criteria, depending upon which liquid/gas ratio was used to recombine fluids. The usual, conventional, methods of classifying should therefore be discontinued for wells producing from deep-basin, over-pressured, mixed-hydrocarbon-saturated reservoirs. Data from a detailed core analysis (61m core), and various re-combined fluid analyses, are used to achieve a history match of initial production data from the liquids rich Montney well producing within the Kakwa field. The compositional simulation model is used to run sensitivities on 1) liquid yields and corresponding re-combined reservoir fluid, 2) permeability modifications to the hydraulic fracture and stimulated reservoir volume (affecting fracture conductivity), and 3) quantities of reservoir gas that has migrated up-structure from the source, if necessary, to achieve comparable history matches. The quantity of gas migration is controlled by invoking a miscible flood, 200 years prior to the beginning of well production and varying the permeability within high permeability streaks, or "fractures", to essentially replicate gas migration from the high temperature source in this unconventional reservoir. The "injector" and "producer" used at either end of the reservoir structure, in this model, are used to facilitate the movement of relatively small amounts of gas. The model is initialized, for the purpose of history matching (and forecasting), with a model that not only has a "fingering" distribution of phases and compositions, within the transition zone, but also represents the varying pressure gradients observed along the reservoir dip. Up-dip the pressure gradient approaches 10.5 kPa/m while the down-structure end of this reservoir yields gradients that exceed 13.5 kPa/m. The ultimate purpose of this study is to show what geological conditions prevail within this particular area of the Montney play and why they make this the ideal location for liquids rich gas production. It will show how much detail is required (or not) to generate a representative forecast model or type curve. It will also attempt to quantify error bars associated with parameters typically defined for this purpose, particularly those related to the range of solution gas/oil ratios (or liquid yields) that generate comparable history matches. And finally, this study will show what impact the characterization of reservoir fluids may have on well spacing and reservoir development plans.