Maximum residual gas saturation (SgrM) is known to be a key factor to evaluate the gas recovery from a lean gas reservoir invaded by aquifer water. This work focuses on variations of SgrM within heterogeneous gas-bearing sandstone reservoirs. Our aim is to produce a reservoir model based on SgrM values consistent with petrophysical trends. Literature presents no SgrM values below 20 % and reported trends usually indicate that SgrM decreases as porosity increases.
Three hundred measurements of SgrM have been performed by liquid capillary imbibition at laboratory conditions on dry plugs. The following data are also available: lithological description, thin sections, XRD analysis, NMR measurements, porosity, permeability, grain density, formation factor and cementation factor. The core plugs are selected from two Far East and one West Africa sandstone gas reservoirs and from Fontainebleau Sandstone outcrops. Plug porosities and permeabilities, respectively range from 6 % to 25 % and from 0.1 to 3 000 mD, with clay content ranging from 0 to 33 %.
The main results are:SgrM values are very scattered from 5 % to 85 %.SgrM against porosity plots present three major trends: Two very different but clear trends exist in the low to medium porosity range (below 14 %). As porosity decreases, SgrM increases for Fontainebleau sandstone whereas it decreases for other sandstones. Concerning the highest porosity values, i.e. above 14 %, the two above mentioned trends merge to an average of around 25–35 %.No relationship exists between SgrM and either grain density, or formation factor, or cementation factor.The amount of clay controls the SgrM versus porosity relationship. SgrM decreases as the clay content increases. No relationship was found between the type of clay and SgrM.SgrM decreases as microporosity, as represented by the NMR Clay- Bound Water, increases. It suggests that microporosity does not trap gas.
Introduction
During depletion of gas fields, the aquifer often encroaches into the reservoir, and residual gas saturation (Sgr) is used to estimate microscopic recovery. Published values of Sgr vary between 15 and 80%. The economic impact of Sgr on gas reservoirs can be extremely high. This is even more crucial for heterogeneous reservoirs where assessing Sgr for different rock types is the key issue.
Many studies have attempted to understand gas-trapping mechanisms. First, Geffen (ref. 1) established that residual gas saturation measured in the laboratory on core plugs is the same as in a gas reservoir. Crowell (ref. 2) illustrated the effect of initial gas saturation (Sgi) on trapped gas saturation (see also refs 3 and 4). Land (ref. 5) first proposed a characteristic shape for the relationship between Sgi and Sgr:Equation (1).
The effect of water flooding rates on Sgr was found to be negligible (refs 1, 2 and 6). Katz (ref. 7) showed that the residual gas left behind the moving waterfront remains constant and equal to that obtained during the measurement of capillary pressure. Several authors demonstrated that Sgr obtained by water flooding and spontaneous imbibition are very close (refs 1, 2 and 3), provided the reduction in Sgr due to diffusion is disregarded (ref. 6). The type of displacing liquid was also found to be negligible in effect (refs 1, 4 and 8). The same Sgr values were obtained whatever the pressure and temperature prevailing during the core test (refs 1, 3, 6 and 9).
