--------------------------------------------------------------------------------
COMMENTS RELATING TO THE REFRACTURE FREQUENCY

Comment Number: EPA-HQ-OAR-2010-0505-4246-54
Organization:  Marcellus Shale Coalition (MSC)
Commenter:  Kath Z. Klaber

Comment Excerpt Text:
Issue 13. In the TSD, EPA states that "10 percent of the wells being re-fractured annually (as previously assumed in Subpart W's Technical Supporting Document3). MSC understands that this 10% re-fracture frequency per year was originally developed based on a single paper which evaluated the post re-stimulation performance of five (5) wells in the Barnett shale area which was then incorrectly applied to unconventional wells across the country. MSC requests that EPA fully and completely explain the derivation of the 10% re-stimulation frequency per year, rather than simply referencing the technical support document for Subpart W which simply makes the 10% assumption with no explanation or backup information. As the largest single reduction category in the proposed rule this explanation is necessary to add any credibility to EPA's analysis. It is the contention of MSC members that the re-stimulation rate more realistically falls somewhere between 1 and 2 percent.



Comment Number: EPA-HQ-OAR-2010-0505-4251-24
Organization:  Pioneer Natural Resources USA, Inc.
Commenter:  Gretchen C. Kern

Comment Excerpt Text:
If EPA proceeds with regulating recompletions, it should reevaluate its assumption for the frequency of recompletions. In 2010, Pioneer refractured less than one-percent of its gas wells. This percentage varies significantly from EPA's ten-percent refrac figure. If EPA proceeds with the regulation of recompletions, it should recalculate the cost-effectiveness using updated estimates of emissions and VOC content, as discussed above, as well as recompletion frequency.



Comment Number: EPA-HQ-OAR-2010-0505-4266-218
Organization:  American Petroleum Institute (API)
Commenter:  Howard J. Feldman

Comment Excerpt Text:
15.11. Over Estimate of Number Recompletions per Year 

In the TSD, EPA states that "10 percent of the wells being re-fractured annually (as previously assumed in Subpart W's Technical Supporting Document3). From anecdotal information, API understands that this 10% re-fracture frequency per year was originally from work underlying the National Inventory and was developed based on one paper which evaluated the post re-stimulation performance of five (5) wells in the Barnett shale area which was then extrapolated to the population of unconventional wells nationwide. Obviously relying on extrapolation of somewhat unrelated information on five (5) wells in one area and one formation to the nationwide population of unconventional wells is completely unsupportable. API requests that EPA fully and completely explain the derivation of the 10% re-stimulation frequency per year rather than simply referencing the technical support document for Subpart W which simply makes the 10% assumption with no explanation or backup information. As the largest single reduction category in the proposed rule, this explanation is necessary to reasonably support EPA's analysis. 

According to a 1996 study by the Gas Research Institute (GRI) [Footnote O: Reeves, S.R.: "Assessment of Technology Barriers and Potential Benefits of Restimulation R&D for Natural Gas Wells", Final Report, GRI- 96/0267, July, 1996.], only 2-3% of wells are refractured. This study found that only 15% of the well populations are potentially high potential restimulation candidates which should be subjected to additional analysis. Another study done in 1999 that was presented in a paper at a Society of Petroleum Engineers conference [Footnote P: Reeves, S.R., Hill, D.G., Tiner, R.L, Bastian, P.A, Conway, M.W. and Mohaghegh, S.: "Restimulation of Tight Gas Sand Wells in the Rocky Mountain Region", SPE 55627, Proceedings of the SPE Rocky Mountain Regional Meeting, Gillette, May 15-19, 1999.], showed that further analysis must be done to determine whether these 15% would actually benefit from restimulation. This paper presented an analysis of some 300 tight sand wells which had originally been hydraulic fracture stimulated, high-graded the top 15% of candidate wells for further evaluation and ultimately reduced the wells recommended for restimulation to five (5) of which 2 had been restimulated at the time the paper was written. In other words, only a small number (~2% of the total wells) of the top 15% candidate wells would actually be economic to restimulate anytime in their life. Other developing information suggests the percentage of wells refractured annually is less than 1%. Applying a 1% restimulation rate per year rather than a 10% would reduce the projected number of affected units to ~1,205 rather than 12,050 (90% reduction) and reduce the emissions reductions by the same magnitude.

Comment Number: EPA-HQ-OAR-2010-0505-4241-15
Organization:  America?s Natural Gas Alliance (ANGA), American Exploration and Petroleum Council (AXPC)
Commenter:  Amy Farrell

Comment Excerpt Text:
G. Recompletions

If EPA proceeds with the regulation of recompletions, before doing so it should reevaluate its assumption for the frequency of recompletions. A preliminary review of data collected by industry suggests that the actual frequency of recompletions is significantly lower than the 10 percent assumed by EPA. ANGA and AXPC would like to work with EPA to develop a more appropriate estimate. Before we can do this, however, we need to resolve a discrepancy with the domestic well inventory used by EPA. The data made available by EPA through the rulemaking docket[FN 32: EPA-HQ-OAR-2011-0512-0020.] has significantly fewer wells than EIA's estimate of the number of producing gas wells in the U.S. [FN 33:Energy Information Agency, Number of Producing Gas Wells. September 29, 2011 Release. Available at: http://www.eia.gov/dnav/ng/ng_prod_wells_s1_a.htm.] EPA has indicated that it will not be able to address this discrepancy until after the close of the comment period. We would welcome the opportunity to work with EPA to resolve the discrepancy and develop a more appropriate well recompletion frequency estimate.


