Application of conventional material balance methods to evaluate multilayer tight gas reservoirs gives misleading results. This is due to the phenomenon differential depletion, as a result of which very long shut-in times are required to get the correct average reservoir pressure, and pressures from static BHP surveys are generally significantly lower than the average reservoir pressure. This study proposes a new approach which yields a straight line on the p/z plot, without a long shut-in period. This paper proposes to use data from production logging which can be used to perform selective inflow performance (SIP) analysis. From SIP the pressure and productivity of each reservoir can be estimated independently. The average reservoir pressure of the multilayer system can then be calculated from each layer's reservoir pressure weighted by its hydrocarbon pore volume. The method therefore requires estimation of the hydrocarbon pore volume ratio (Fv). A guideline for Fv estimation and the effects of Fv on the p/z plot will be discussed in this study. With this new average reservoir pressure, material balance methods will give a straight line, similar to one for a single layer gas reservoir. Without loss of generality, this study focuses on a 2-layer gas reservoir: a high-permeability layer (layer 1) and a low-permeability layer (layer 2). The results can be applied to multilayer gas reservoirs. Fv is defined here as the ratio of the hydrocarbon pore volume (HCPV) of layer 1 to the HCPV of layer 2. For a given Fv, together with individual layer pressures from SIP analysis, the average reservoir pressure can be calculated. Different values of Fv yield different profiles on the p/z plot. A straight line can be obtained only from the correct value of Fv. If the value of Fv is incorrect, the calculated pressure will be deviating from the theoretical straight line. High and low values of Fv lead to underestimation and overestimation of GIIP and reserves, respectively. During the early production period, most of the production is from layer 1 which dominates the early-time response. The differential pressure between layer 2 and layer 1 (p2 – p1) is increasing during this period. After layer 1 is partially depleted, production from layer 2 starts to increase significantly, and dominates the late-time response. The value of (p2 – p1) reaches a maximum and then decreasing. Beyond this point, all profiles for different values of Fv lead to the correct GIIP and reserve. Then the uncertainty will be reduced dramatically. This study should be beneficial to engineers who wish to estimate GIIP and reserves. It is theoretically sound for multilayer gas reservoirs, both tight and high permeability reservoirs, for both with and without cross-flow. With this new approach, engineers can apply material balance methods to better understand multilayer gas reservoirs and to make better reservoir management decisions.

Applications of the traditional static material-balance method in gas reservoirs become a challenge with production maturity due to variability in aquifer influx, infill drilling, and production-operational changes in offset wells, among others. Besides, some existing modeling approaches involve a trial-and-error method, making the solution outcomes nontrivial. This study proposes a new methodology for analyzing production data involving water-drive gas reservoirs. The main findings of this study include the following: (1) A straight-line plot that yields gas and water in-place volumes, (2) A modified-(pav/z)* plot exhibits a straight-line with an x-intercept of gas initially-in-place, similar to that in a conventional-(pav/z) plot, (3) A new definition of degree of aquifer support that is quantifiable using production data. Synthetic data verified the proposed modeling approach, whereas a field dataset provided validation.

No abstract