Early-Time Pressure Buildup Analysis for Prudhoe Bay Wells

Abstract

Summary A method of pressure buildup analysis is presented that distinguishes between two possible reasons for the decline in a well's productivity-near-wellbore damage and permeability loss in the well's drainage area. An extensive application is discussed for wells in the Prudhoe Bay field. Heterogeneity of this reservoir is such that semilog analysis methods are not reliable for identifying wells in need of remedial workover. Introduction The production rate of individual wells in the Prudhoe Bay field can decline for a variety of reasons, including workover damage, formation-water precipitate, fines migration, limited drainage area, and gas saturation. Calcium chloride brine was used originally as a workover and packer fluid. This brine, when mixed with formation water, precipitates calcium carbonate, which can practically plug the perforations and near-wellbore formation. Matrix acidization with hydrochloric acid gives only temporary relief. Carbonate scale forms again when the residual workover fluid contacts formation water or CO2 gas from crude-oil production. Fines also migrate to the wellbore. These particles of clay and/or microporous chert plug the perforations, which apparently leads to an insidious type of pressure buildup behavior. The perforation tunnels, perhaps, tend to collapse and thereby compress any fines and/or scale present. This compaction would increase as the amount of drawdown increases. When such a well is shut in and the pressure recovers, this compaction reverses to give a pressure-sensitive near-wellbore damage that can dominate a 24-hour buildup test entirely. Well 6 in Ref. 1 appears to be an example of such a situation at Prudhoe Bay. Factors other than true damage can cause production falloff. Prudhoe Bay field contains a number of mapped faults that limit the drainage area of nearby wells. Relative permeability can also reduce oil production. Since the field has an associated gas cap, the crude oil is at or near its bubblepoint pressure. Consequently, most wells flow at bottomhole pressures (BHP's) below the bubblepoint. These wells will have a near-wellbore region with gas saturation. This reduces relative permeability to oil in the near-wellbore region. Measured relative-permeability curves show that at the minimum mobile gas saturation of 5%, relative permeability to oil is decreased to about two-thirds of its value at zero gas saturation. With continued production, the zone of mobile gas saturation will expand sufficiently to give a measurable drop in productivity accompanied by an increase in GOR. Almost 700 buildup tests have been run in the Prudhoe Bay field. High-precision gauges are used consistently. The ambiguous interpretations of these tests, however, have generated confusion. One interpreter may claim a well is undamaged while another maintains that the well is damaged. Supposedly undamaged wells have been acidized to give threefold rate increases. We believe this confusion arises from attempts to apply Homer-type buildup analysis methods-or slight modifications thereof-to data from a reservoir that departs significantly from the model implicit in the Homer method. The Prudhoe Bay field has both a gas cap and an underlying aquifer. To provide distance from these contacts, the oil zones are usually perforated in only a portion of the net pay. This complicates a Homer-type analysis. Additionally, the formation is a complex sequence of sand-shales resulting from either braided stream or mouth-bar depositions. The sands are separated by shale stringers of widely varied thickness and of generally unknown permeability and lateral extent. As discussed in Ref. 1, gas-cap pressure support can interact with reservoir heterogeneities to give a bewildering and nonunique variety of distortions of the middle-time region so essential to a Homer-type analysis. No attempt is made to identify a middle-time region in the method described in this paper. Instead, early-time, afterflow-dominated buildup data are analyzed to determine whether productivity decline can be attributed to wellbore damage alone. In this paper we first give some comments on the method. This is followed by examples from buildups taken in the Prudhoe Bay field. The method consistently accounts for well productivity behavior. For some examples, the Homer-type analysis is completely satisfactory. In other examples, the Homer approach gives confusing and inconsistent results. Finally, there are examples from a sequence of tests on the same well that produce impossible results by Homer analysis. It is this very lack of consistency that we hope to eliminate. While we concentrate on tests from Prudhoe Bay, the general problem is not unique to this field. This is illustrated by our first example. JPT P. 311^