Numerical Simulation of Thick, Tight Fluvial Sands

Abstract This paper presents several workflows for constructing adequate flow models of the Jonah Field, located in Wyoming. The numerical flow models were built by integrating seismic, petrophysical, geological and engineering data, including hydraulic fracture data. The reservoirs consist of several sand units over a gross thickness of 4000ft in a fluvial depositional environment. Reservoir rock permeabilities are in the micro-darcy range. The over-pressured reservoirs become economically viable only by hydraulic fracturing. Two major challenges of modeling the field are reservoir upscaling and appropriate representation of the hydraulic fractures. A streamline based flow model was utilized to upscale geological features. Some practical assumptions were made to apply this technology to the Jonah Field study. Multiple models were generated using different upscaling scenarios and techniques. The models were setup with the same boundary conditions (injector-producer pairs, injection-production rates, etc), and their results were compared with the fine-grid geocellular model results. Pseudo fluid properties (low viscosity) and very long time scale had to be used because of the low permeability of the sands. The fluid recovery and injected fluid breakthrough times for the flow models and the geocellular model were then compared. The flow model with the most reasonable volumetrics and flow characteristics was chosen for the numerical simulation of the Jonah Field. Jonah wells are hydraulically fractured with multi-stages. Single well and sector models were utilized to determine the ultimate fracture properties that were used in the final simulation model. First, local grid refinement was used to represent the fracture properties. Then, a parametric study was conducted to establish the effective global cell properties that are required to simulate the flow of hydrocarbons along the hydraulic fracture without using the local grid refinements. Production and pressure performance were compared over a long period of time. Effective permeability and pore volume calculation yielded the best results, and they were used during the history matching of the wells' performances. Introduction The Jonah field is located in the northwestern Green River Basin in Sublette County, Wyoming. It produces gas from the fluvial channel sandstones of the Upper Cretaceous Lance Formation at drilling depths ranging from 11,000ft to 12,500ft. The Lance Formation consists of several hundred feet of stacked lenticular sands and siltstones, floodplain shales, and minor coals that were deposited in a broad alluvial plain. The reservoirs consist of several sand units over a gross thickness of 4000ft in a fluvial depositional environment. The thickness of individual sand units ranges from 5ft to 50ft. These geological heterogeneous reservoirs are poorly correlatable over a large area. The sandstones are fine to medium grained with porosity ranging from 5 to 14% with permeabilities in the range of 0.001 to 0.03md. Water saturation varies from about 30% updip to about 60% downdip. The study area situated updip has better reservoir quality, and produces very small amount of water, which is probably water of condensation and rock compaction. The reservoirs are overpressured with the top of the overpressure zone corresponding approximately with the Cretaceous-Tertiary boundary. Based on stimulation breakdown pressures, pressure gradient of up to 0.7psi/ft has been recorded. To honor the pressure gradient, infinitissimal vertical premeabilities, which prevent gravity equilibrium, were specified in the simulation models. The models were were initialized by enumeration of reservoir pressure. Wells are stimulated using multiple limited entry hydraulic fracturing to attain economical gas production rates. Operators of the Jonah Field use different completion strategies, which have evolved over the years. Three to six individual sands were completed per stage with four to six stages per well. The number of stages has increased over the years and six to ten stages are now common.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.