(715e) Refracturing in the Bakken – an Analysis of Data from across North Dakota

The Energy & Environmental Research Center (EERC), in close coordination with several of Williston Basin’s premier operating companies, is engaged in a research program called the “Bakken Production and Optimization Program,” or “BPOP,” with the goal of simultaneously improving oil recovery in the North Dakota portion of the Bakken petroleum system (Bakken) while reducing its environmental footprint. BPOP is a consortium-funded program, with funding provided from a combination of the North Dakota Industrial Commission (NDIC) Oil and Gas Research Program (OGRP), several operating companies, and the U.S. Department of Energy.

One of the activities conducted by the EERC through BPOP is an evaluation of refracturing performance in the North Dakota portion of the Bakken. In this manuscript, refracturing refers to an operation to restimulate a well after an initial period of production, with the intent of either restoring or increasing well productivity and extending the productive life of a well. More than 90% of the estimated oil-in-place in unconventional reservoirs is not produced via primary production. Refracturing provides one technique for improving oil recovery by counteracting the effects of suboptimal completions of the original well such as undersized fracture treatments, suboptimal spacing of fracture intervals, inoperable completion failures, and damage mechanisms that cause a loss of connectivity to the reservoir.

Refracturing is a recent development for the Bakken but is not a new technique, as refractured wells, or refracs, were first implemented in the 1970s. One of the earliest refracturing operations in the Williston Basin was successfully performed in the Elm Coulee Field of Montana, where single-stage horizontal completions were restimulated. In North Dakota, Marathon Oil Company (Marathon) had an active refracturing program targeting older, openhole (barefoot) wells that were originally completed from 2006 to 2009 (classified as Generation 0 wells) in the Bakken. Marathon noted success with refracturing the Generation 0 wells; however, it also noted technical and operational challenges associated with later-generation wells having more than one completion stage. Whiting Oil and Gas Corporation reported the successful refracturing of a multistage horizontal well in the Bakken. The North Dakota Pipeline Authority (NDPA) identified and compiled all of the Bakken wells that were potentially refractured in North Dakota and evaluated production performance of those wells. The NDPA findings indicated that there were approximately 185 refracs across the Bakken (less than 2% of the total wells) and incremental oil production following refracturing was approximately 200,000 stock tank barrels (stb) per well. These estimates provide an example of the potential increased oil production associated with refracturing in the Bakken.

However, the oil and gas companies operating in the Bakken are cautious about broadly implementing refracturing because of the technical and operational challenges associated with the refracturing process. These challenges create uncertainty in postrefrac performance, which translates to elevated financial risk that limits access to investment capital, despite other economic incentives such as existing infrastructure (pipeline/facilities), a stable environment because the well already exists, known well details and intelligence (understanding of well operation best practices), and lower associated cost.

The goal of this BPOP study was to quantify uncertainty in refrac performance by investigating a large data set of refracs in the Bakken; 272 wells have been refractured in the North Dakota portion of the Bakken as of May 2019, which constitutes the data set used in the present study. The data set was compiled from various data sources including the North Dakota Geological Survey (NDGS), NDIC and FracFocus, the national hydraulic fracturing chemical registry. The data set includes the monthly oil production data as well as information specific to the original and refrac treatments for each of the 272 wells. In addition, decline-curve analysis was used to supplement these data and generate additional metrics for the economic analysis. The data set inventory revealed that the number of refracs increased considerably over the 2017–2018 timeframe, with 104 refracs conducted in 2017 and 2018, as compared with 168 refracs from 2009 through 2016, indicating the increased interest of oil companies in refracturing operations in the Bakken.

This manuscript discusses the results of 1) an evaluation of production performance of wells that have been refractured in the North Dakota portion of the Bakken; 2) an analysis of the impact of refracturing on offset wells (secondhand refracturing); 3) an assessment of refrac economics using a decision-tree modeling approach; and 4) an assessment of the potential economic impact of a broad refracturing program applied throughout the Bakken.

The production analysis of the refracs included several metrics to evaluate refracturing performance, including 1) a comparison of pre- and postrefracturing peak oil rates, 2) a change in daily production data following the refracturing operation, 3) a change in gas-oil ratio (GOR), and 4) incremental estimated ultimate recoveries (EURs) (i.e., the change in EUR between pre- and postrefracturing forecasts). In addition to summarizing the results across all refractured wells, the refracturing production performance was compared between wells that were originally completed as barefoot vs. nonbarefoot wells. Barefoot completions are believed to be understimulated during the original treatment and to have a greater undrained area and larger remaining hydrocarbon volume, resulting in commensurately better post-refracturing production performance. Furthermore, refracturing conducted on wells with barefoot completions are technologically easier and less costly to implement, which makes this group of wells potentially the best candidate for refracturing. The results showed that refracturing performed on barefoot wells outperformed those performed on nonbarefoot wells. For example, the median incremental EUR for the barefoot and nonbarefoot wells under the expected scenario was 340 and 175 thousand stb (Mstb), respectively. The postrefracturing uplift in daily oil production was 421 stb in barefoot completions as compared to 280 stb in nonbarefoot completions. The average GOR of the barefoot wells decreased by 10% on 90-day average daily production data, which provides supporting evidence that the refracturing provided access to previously undrained portions of the reservoir. Conversely, the average GOR of the nonbarefoot wells remained the same after refracturing.

Refracturing can not only restore or improve production of the well that is refractured but also improve production of offset wells located near the refractured well. An analysis of secondhand refracturing was performed to evaluate the production of wells that were offset from the refractured wells. The result showed that refracturing significantly impacted offset wells, and an average increase of 65 stb/day in daily oil production rate was observed in more than half of the offset wells analyzed after nearby refracturing.

An economic analysis was performed that focused on discounted net oil revenue, defined as the oil revenue after deducting royalties (16.67% applied pretax) and state tax (10% applied after deducting royalties), assuming an annual discount rate of 10% per year. A decision-tree modeling approach was used to evaluate the overall economic performance given a set of probabilities assigned to each potential outcome. Comparing the economic performance of wells that were originally completed as barefoot wells to nonbarefoot wells showed that barefoot wells had greater discounted net oil revenue. For example, the median discounted net oil revenue for barefoot wells was $4.9MM (million), while the median for nonbarefoot wells was $1.7MM, or approximately 65% lower.

An initial screening of existing wells in the Bakken identified 400 wells as potential candidates for refracturing using a simple set of criteria, specifically wells that had single-stage completions, older completion dates, and barefoot completions. The potential economic impact from refracturing all of these candidate wells resulted in an estimated discounted net oil revenue of approximately $2 billion, which reflects the median outcome after deducting the refracturing cost, taxes, and royalties. These results show that there is significant production potential of refracturing in the Bakken, particularly if single-stage, older barefoot wells are targeted.