Material-Balance Analysis of a Water-Drive Reservoir with An Unusual Development History

Abstract

Abstract Frequently, the analysis of an undersaturated waterdrive reservoir by a material balance is unsatisfactory because reservoir conditions do not reasonably satisfy the assumptions of the material-balance equation. This occurred in the case of a Devonian reservoir in West Texas which proved difficult to analyze because of a rather unusual development history. The standard material-balance analysis technique had to be modified before a satisfactory oil-in-place value could be determined. The modified technique consisted of a material-balance formulation which permitted the analysis of pressure-production data from separate portions of the reservoir with one equation; it also permitted a judicious application of the "least-squares method" to reduce the effect of inherent inaccuracies in the early data. The analysis was accomplished with the aid of an IBM 704 computer and represents a good example of the flexibility in material-balance calculations that may be obtained by computer usage. Introduction The determination of oil in place and water-influx performance of undersaturated water-drive reservoirs, using the material-balance principle, is a well established practice. Frequently, however, the material-balance results are questionable, and the practicing engineer begins to regard such calculations as either totally unreliable or applicable only to a strange breed of reservoirs which apparently are owned only by competitors. Inconclusive or questionable results will occur when reservoir conditions do not reasonably satisfy the assumptions of the material-balance equation and when the true average reservoir pressure is not accurately represented by an average of individual well pressures. These difficulties often can be overcome by modifying the material-balance equation to introduce additional independent parameters which may be assigned various values during the calculations to determine the effects, within certain limits, of possible inaccuracies in basic assumptions. This is exemplified by the reservoir study which is the subject of this paper. The subject reservoir is a structural trap in the Devonian dolomite-limestone formation of West Texas. The initial reservoir pressure of 3,231 psi declined to about 2,250 psi during 12 years of production, but it has remained above the saturation pressure of 1,775 psi. Material-balance calculations, made early in the life of the field, indicated a much larger reservoir than had apparently been defined by the drilling of 10 producing wells and three dry holes. Although a material balance during the early life of an expansion-drive reservoir is notably unreliable, the effort proved rewarding in this case. The consistency of the calculated value of oil in place after nearly eight years of production prompted additional drilling which eventually extended the field to include a volume approximately three times that of the initial discovery and resulted in an additional 28 producing wells. Even with the additional drilling, the amount of reservoir pore volume was still questionable. Volumetric calculations indicated 40 million STB of oil in place, but very few wells had penetrated the full Devonian section; since the top of the Devonian is eroded over most of the structure, the estimated pay thickness may be considerably in error. Material-balance calculations, although not so consistent as the earlier ones, gave an oil-in-place value of about 140 million STB and no definite indication of water influx. Furthermore, data from drill-stem tests of the Devonian formation taken in dry holes in the vicinity of the reservoir indicated the possibility of a very limited aquifer. The pressure-decline curves indicated a recovery above saturation pressure much greater than normal for an expansion drive--if the amount of actual oil in place was reasonably close to the volumetric estimate. JPT P. 37^