EconoProp and CarboProp provided higher conductivity and longer effective fractures than a competing broadly sieved ceramic, sand or resin-coated sand. Even in this tight reservoir, proppant quality affects production by a greater degree than predicted with lab data and model forecasts.
A field trial was conducted in the Pinedale Anticline of Wyoming to compare well productivity as a function of proppant type and geologic interval. This study compares production from 446 fracture treatments in 30 new wells completed by Shell between September 2001 and September 2004. Great effort was put into the design of the field trial, the collection of data in a systematic method, and careful analyses of the data. Data and analyses from this field trial have been presented in SPE 96559, SPE 100574, and SPE 106151. This current paper summarizes the techniques utilized in each paper to analyze the field results, compares the relative confidence in conclusions reached with each technique, and highlights lessons learned in the process of conducting and evaluating this field trial.
In this field, the authors conclude with greater than 99.9% certainty that production is affected by proppant selection. Even in this extremely tight formation (~0.002 mD or 2 microdarcies) the fractures remain conductivity-limited, and productivity is greatly influenced by proppant selection. The extremely high level of statistical significance has been achieved by the following:
- A sizeable population of wells were completed by a single operator during a short time interval;
- Drilling, completion, and fracturing stimulation variables were minimized;
- Proppant types were compared in the same geologic intervals from wells in a small geographic area;
- Production logs were conducted and analyzed in a systematic fashion, calibrated and corrected to an equivalent logging date;
- Production data were analyzed with and without normalization to log-derived petrophysical information;
- Prior to initiation of the trial, 22 geologic subintervals and four separate study areas were defined to investigate subgroups of data in the most homogenous conditions;
- Statistical analyses were performed on the full dataset, with smaller subgroups of data, and with numerous normalization techniques to determine whether random reservoir variation could account for the results.
The compilation of these results should be of value to those considering a field trial in a complex geologic setting or comparing the results of propped fracture stimulated wells. Optimizing proppant selection has been shown to significantly increase the economic potential of tight gas wells.
Author(s): M.C. Vincent, Carbo Ceramics Inc., Shell E&P Co., and Paul Huckabee
Paper Number: SPE 108991