Oil recovery by the injection of water alternately with gas (WAG flooding) is often limited by gravity segregation, which causes the injected gas to rise to the top of the formation and water to migrate to the bottom. This segregation results in a miscible flood in only a thin layer at the top of the formation, whereas the remainder is waterflooded. However, vertical conformance in a WAG flood can be greatly improved by use of injection rates that are high relative to well spacing. The degree of improvement and, hence, the resultant oil recovery is primarily a function of a single dimensionless parameter, which is one form of the viscous-gravity ratio.


A number of processes in which miscible gas was used to displace oil were patented around mid-century. Tn them, complete displacement of oil was claimed. However, research conducted during the next few years showed that the less viscous miscible gas would "finger" through the oil, giving poor conformance and, hence, low recovery.

In 1957, Caudle and Dyes proposed the control of this fingering and Door conformance by injecting water along with the gas. The presence of the water lowers the relative permeability to gas, thereby decreasing its mobility. Field tests of this concept showed that it was most feasible to inject water alternately with gas, and this process became known as WAG flooding.

In 1959, Blackwell et al. reported an investigation of the WAG process. They observed that gravity caused the gas and water to segregate rapidly. As a result, the mobility of the water-gas zone was less favorable than it would have been without segregation. An even more important aspect of this segregation was that the gas occupied a much smaller fraction of the vertical cross section than did water. Only this thin top layer was miscibly flooded by gas, whereas the bottom portion was waterflooded.

Gravity segregation requires some time to occur, so there is a region around the injection well where vertical conformance is good. The size of this region is determined principally by the injection rate, the vertical permeability, and the density difference between water and gas. In some reservoirs it is possible to design a WAG flood so that this region is large relative to the reservoir volume to be flooded by that well.

This paper summarizes the results of a reservoir simulation study investigating the dominant factors controlling the water-gas segregation zone around a wellbore. The objective is to determine the fluid and reservoir properties and operating conditions that determine the size of the zone and, hence, the vertical sweep efficiency of a WAG flood.

An analytical equation for predicting the size of this zone in a homogeneous reservoir will be presented and validated by comparing its predictions with reservoir simulation results. This predictions with reservoir simulation results. This equation, supplemented by scaling theory, is used to define the variables governing the segregation process. These variables are then used to process. These variables are then used to correlate the results of a series of simulations of both homogeneous and heterogeneous reservoirs.


Although in practice alternate injection is used, it is expedient to analyze simultaneous injection. This is justified because simulation has revealed little difference in calculated results as long as injection cycles are kept below one to two months, at least for the reservoir descriptions and conditions simulated in this study.

To derive the analytical equations, we consider the steady state saturation distributions resulting from relatively long-term simultaneous injection of gas and water into a well.

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