Search for a command to run...
Summary The north Texas Barnett shale illustrates the successful commercialization of an unconventional reservoir. However, it took 17 years to evolve from pumping crosslinked gel (XLG) carrying more than 1 million lbm of proppant per job to sand waterfracs (SWFs) consisting of large volumes of water with friction reducer and small quantities of sand. This transition to SWF stimulation opened the door for widespread development that has advanced the Newark East (Barnett shale) to the largest producing gas field in Texas. This paper investigates Barnett completion strategy from 1993 to 2002. The 393-well data set includes completion, reservoir, and production data. Unique data-evaluation tools and techniques were used to investigate various completion and reservoir parameters to determine their effects on production (Shelley and Stephenson 2000; Zangl and Hannerer 2003). We found that production results show a broad scattering when crossplotted with various completion and reservoir inputs. This result is not uncommon when analyzing field data. However, general trends were identified through comparisons of large numbers of wells. These trends were confirmed through the use of more-advanced data-mining techniques, which included self-organizing mapping (SOM) of data. The results show that SWF-type stimulation of the Barnett outperformed to varying degrees XLG treatments for the five reservoir types used in this evaluation.