Gas Reservoir Performance in Abnormally High Pressure Carbonates

Abstract

Abstract Two-dimensional r-z reservoir simulation and tank material balance were used to develop a prediction technique for gas reservoir performance in abnormally high pressure reservoirs. It is shown that p/z versus Gp plot is very sensitive to the variability of the effective compressibility, ce. Nevertheless, our technique gets around this problem and, with availability of reliable reservoir history data, accurately predicts the reservoir performance and original gas in place, especially in single-well reservoirs such as the Cotton Valley reefs in East Texas. The calculations presented in the paper are simple and easy to use. Introduction In abnormally high pressure sandstone gas reservoirs of the Gulf of Mexico, where formation porosity and permeability are relatively large, the early p/z decline rate is smaller than the p/z decline rate at the normal pressure levels. This concept is demonstrated by a field example in a paper by Ramagost and Farshad. This behavior is attributed to the high pore compressibility and/or compaction of the rock frame. Ramagost and Farshad and Fetkovich, et al. derived a material balance equation for volumetric gas reservoirs shown by Eq. (1): (1) where ce is the effective compressibility defined by Ramagost and Farshad below: (2) and (3) Fetkovich, et al. have a similar but a more comprehensive definition for ce to account for shale dewatering, etc. In this earlier work it is assumed that the effective compressibility is essentially constant as the reservoir pressure declines. New Developments Our laboratory measurements of the pore compressibility, cf, under simulated reservoir conditions, however, indicate that cf and, therefore, ce vary significantly with reservoir pressure decline, especially in high permeability sandstone reservoirs. Intact, at pore pressures in the abnormal pressure region these compressibilities could be an order of magnitude greater than in the normal pore pressure region. A recent publication by Harari, et al. provides data (please refer to Fig. 7 of Ref. 3) in support of our experience. To account for variability of ce with pore pressure we propose a simple modification of Eq. (1) as shown (4) P. 213