Producing-Gas/Oil-Ratio Behavior of Multifractured Horizontal Wells in Tight Oil Reservoirs

Abstract

Summary Horizontal wells with hydraulic fractures in tight oil reservoirs show producing-gas/oil-ratio (GOR) behavior that is very different from conventional, higher-permeability reservoirs. This paper explains the reasons for the observed behavior by use of reservoir simulation, with field examples from the STACK and SCOOP plays of the Anadarko Basin in central Oklahoma. A framework for interpreting observed GOR behavior in tight black-oil reservoirs is modeled after the following stages in a well's history. Some stages may not be visible because of the degree of undersaturation, flowing-bottomhole-pressure schedule, finite-conductivity fractures, and duration of the transient-flow period. Stage 1: Early GOR is constant at the initial solution GOR (Rsi) while bottomhole flowing pressure is above the bubblepoint. Stage 2: A rise in GOR as bottomhole flowing pressure declines to less than the bubblepoint. Stage 3: The transient GOR “plateau”, which is characteristic of transient linear flow. Stage 4: A continuous rise in GOR during boundary-dominated flow. Fundamental differences between linear and radial flow, which cause the dependence of GOR on flowing bottomhole pressure, are explained by use of simulation. During transient linear flow, the GOR response to changes in flowing bottomhole pressure is independent of permeability for infinite-conductivity fractures, but not for finite-conductivity fractures. Several practical observations are made. Knowing Rsi and the transient-GOR-plateau level in an area can help one interpret where a well is in its GOR history. Rate-transient-analysis (RTA) diagnostic plots are altered by rising GOR, and sometimes show an early unit slope. During boundary-dominated flow, GOR is more a function of cumulative production than of time; wells with closer fracture spacing have a faster GOR rise with time, but also recover oil more quickly. If compound linear flow develops, GOR can decline late in the well life. The Meramec and Woodford formations in STACK can be history matched without invoking a suppressed bubblepoint caused by pore-proximity effects. The critical gas saturation in the Meramec appears to be in the range of 0–5%. Technical contributions include a framework for interpreting GOR behavior over well life; the effect of changing bottomhole flowing pressure on GOR; the effect of fracture spacing, conductivity, and half-length on GOR; and the effect of GOR on RTA diagnostic plots.