Hydraulic fracturing treatment design historically has been deterministic in nature. In a deterministic design, the required design variables are single-valued and can be established with a reasonable degree of certainty. Additionally, their values are assumed to be laterally continuous. Fulfilling these assumptions is complicated for geologic settings in general, but nearly impossible in fluvial geologic environments, which violate the assumption of lateral continuity.

Recent work has defined a methodology for hydraulic fracturing treatment design using statistical methods to account for the complexity and heterogeneity in fluvial environments. The methodology demonstrates how to use statistics to address treatment design questions such as, “What is the vertical fracturing interval needed to create a well-confined, economical and productive fracture treatment in a highly layered, fluvial environment?” and “What is the optimal fracture treatment in such complex geology?”

This paper presents and describes the statistical-based hydraulic fracturing design methodology, including constructing and calibrating a basic geological model, incorporating geostatistics for the selection of alternate geological models, forecasting production for alternate fracturing treatments, and determining an optimum hydraulic fracturing design using all available information. The methodology is illustrated using geological and well data from a producing fluvial tight gas formation, though the approach should apply to other complex geological settings.

This work is valuable as it demonstrates a method for extending the deterministic approach taken with formation characterization and applying it directly to hydraulic fracturing design (thus, using the data where good values are available) to include statistics, thereby providing a basis for selecting an optimum treatment design when a range of possible design parameters exists for less well defined variables.

This content is only available via PDF.
You can access this article if you purchase or spend a download.