Formation Evaluation and Permeability Prediction in a Highly Heterogeneous Reservoir - The Kuparuk C-Sand

摘要

The Kuparuk River Field on the North Slope of Alaska is one of the largest oil accumulations in North America. Approximately one-third of the original oil-in-place (OOIP) is contained in the C-sands, which are shallow marine sandstones characterized by intense bioturbation and complex diagenesis. Siderite content is extremely variable and an important control on permeability, porosity, and capillarity on the sub-foot scale. Interpretation of wireline logs in the C-sands for mineralogy, porosity, and water saturation is relatively straightforward, provided that proper interpretation parameters for clay, siderite, and glauconite content and core heterogeneity are considered. Computing a realistic permeability log is more difficult due to extreme scatter in cross-plots of core permeability and log porosity that lim-its the effectiveness of defining a trend. Deterministic porosity-permeability transforms, where permeability is predicted from log porosity, are poor predictors since the results do not replicate the extreme variability present in the core data. Recent field description efforts at Kuparak required re-evaluation of the permeability model, with a need for predicted properties that could be scaled up in a straightforward fashion for use in populating a geo-cellu-lar model. A new method was developed for the prediction of permeability. The method is based on random selection of values of core bulk density from sub-groups based on log bulk density, RHOB, and petrofacies. For each half-foot log depth point, core bulk density values are selected at random, repeatedly, until the density averaged over a sliding window matches the RHOB log to within a pre-set tolerance, nominally 0.05 g/cm~3. The values of core porosity and permeability that correspond to the selected value of core bulk density are then selected as the final result at each depth point. The method duplicates the statistical distributions of the core porosity and permeability values, with values obtained every half-foot. Results were converted from measured depth reference to true vertical depth below mean sea level, SSTVD and up-scaled from 0.5 ft sample increment to 1 and 2 ft increments. The up-scaled permeability values match total permeability-height, kH, based on core-plug data, on a well-by-well basis. The values are also consistent with kH determined from maximum flow rates observed in a number of wells. Given the match with other measures of permeability, the up-scaled permeability values were deemed satisfactory for use in the geo-cellular model.