The effect of dissolved gas in the reservoir on oil production rates in SAGD is investigated by means of theory and simulation. Butler's LinDrain equation for the oil rate during the spreading phase of the steam chamber is generalized to include the effect of gas dissolved in the oil phase. The theory incorporates methods to calculate saturation distributions in the flow region and links the (dead) oil rate to the producing water-oil ratio (WOR) and gas-oil ratio (GOR). For a given GOR in a suitable range, the theory indicates that the oil rate with dissolved gas approaches the oil rate without dissolved gas as the operating pressure increases.

Simulation studies were carried out to evaluate the impact of dissolved gas on oil rates for increasing operating pressure and the conclusion from theory stated above was confirmed in the simulation runs. The results here indicate that the reduction n the oil rate due to dissolved gas may be substantial for operating pressures close to the initial pressure and could influence the selection of a suitable operating pressure for SAGD.


In the early theoretical and experimental work on steamassisted gravity drainage (SAGD), the effect of dissolved gas on the process was ignored (see Chapter 7 of Butler1). Later studies of the effect, based on reservoir simulation, indicated a decrease in the bitumen production rate due to the presence of dissolved gas (Gittins et al. 2). Since this effect was apparently not in agreement with field data, some authors concluded that there was something inherently wrong with the simulation results and claimed that the exclusion of dissolved gas resulted in bitumen rates more in accord with field experience (pp. 583 - 584 of Mukherjee et al. 3).

In this paper we study the issue of dissolved gas in SAGD by means of theory and simulation and show that the effects on bitumen rate predicted by simulation are in accord with theory. owever the effect does decrease as the operating pressure increases and the (dead) bitumen rates with and without dissolved gas become nearly equal.

LinDrain Theory for SAGD with Dissolved Gas

For the case when no dissolved gas is present, Butler (p. 176 of Ref. 4) has developed a 2-D theory for calculating the bitumen rate for SAGD, during the spreading phase of the steam chamber, by assuming a triangular shape for the steam chamber. The reservoir is assumed to be laterally infinite. This theory, referred to by Butler as LinDrain Theory in his writings, is generalized here to the case when dissolved gas is present. The dissolved gas is taken to be methane and it is assumed that methane is soluble in the oil phase but not in the water phase. Dead bitumen is represented by a single pseudo-component. Only steam injection is considered here-there is no injection of methane with steam.

For the case of dissolved gas, the steam chamber of SAGD becomes a vapor chamber containing a mixture of water vapor and methane

This content is only available via PDF.
You can access this article if you purchase or spend a download.