Field development and well placement in tight gas fields pose several challenges. First, the reservoirs are highly heterogeneous: permeability varies over several orders of magnitude and the correlation length of the geologic features is relatively small. Second, well rates are low, which imply that hydraulic fracturing will be used to increase well rates, as with other unconventional plays. Third, there is sufficient large scale pressure communication in these fields so that an infill drilling sequence needs to examine the interaction between wells when defining an optimal well spacing. Because of the high degree of heterogeneity, we are led to the use of high resolution (multi-million cell) geologic models to represent the reservoir. Because of the degree of variability, we are led to the use of multiple realizations of these models to represent subsurface uncertainty and our lack of knowledge of the reservoir. Finally, development of an infill program is based on the evaluation of the performance of multiple sequences of wells. Together, the dynamic evaluation of multiple sequences of infill wells in multiple realizations of high resolution multi-million cell geologic models is both a conceptual and computational challenge. To address these challenges, this paper introduces a rapid simulation-free workflow, which allows us to explore the impact of subsurface uncertainty, and the choice of a development strategy on a well placement drilling sequence in tight gas reservoirs.
In this paper, we present a novel simulation-free workflow based on the combination of a reservoir quality map and an uncertainty map. The quality map is produced from a deterministic set of parameters, including reservoir static and dynamic properties. This quality map may be calibrated through the use of reservoir simulation. However, we have found that a sufficiently good evaluation of the depletion can be obtained by a rapid "geometric" pressure calculation, instead of using full flow simulation. The uncertainty map is built to capture the geologic uncertainty, and replaces the use of multiple realizations. We have the most confidence about the reservoir data at the existing well locations, and as the distance from the existing wells increases, our confidence about the data at this location decreases, and the risk correspondingly increases. Good infill well locations and a drilling sequence are then determined through a combination of the quality map and uncertainty map, combined according to a development strategy.
This field development study is performed on a U.S. on-shore tight gas field. We first study a section of this tight gas field to validate the quality maps by comparison with flow simulation results, and then apply this workflow to the full field to generate a proposed infill drilling sequence. We also contrast our results with a risk assessment based upon multiple realizations to further show the benefits of our proposed workflow. This approach does not require flow simulation or multiple realizations; for a multi-million cell full field model the approach can determine a sequence of infill well locations within minutes instead of days. The efficiency of this workflow makes it feasible for large scale field applications, providing good opportunities for improved tight gas reservoir management.