Water management

Saltwater Disposal Optimization Drives Water Midstream Sector

Operators of unconventional plays face a conundrum—how to dispose of produced water economically without risking seismicity or aquifer contamination. A recent paper and virtual forum offer ideas for optimizing saltwater disposal.

Source: Getty Images.

Salt water, produced water, waste water, oilfield brine—regardless of what you call it, large volumes have been coproduced with oil in the US for decades. But the volumes have surged in the past few years and doubled since 2009, along with widespread seismicity in some regions, most notably Oklahoma and, more recently, the Permian Basin. The increase in produced water and concerns about its effects have recently spawned a new business sector known as “the water midstream.”

An estimated $9 billion to $11 billion of private capital has been committed to the oilfield water midstream business to date, and a further $16 billion is projected to be required. The value proposition for this business is optimizing the treatment and disposal of produced water, currently at water/oil ratios of approximately 4:1 for unconventional wells and 13:1 for conventional, at scale. A key performance indicator of that value is the ability to provide reliable underground water-storage ­capacity while avoiding seismic events and potential aquifer contamination, both of which have been linked to produced water.

Managing Unconventional Water Production, Unconventionally

Just as the hydrocarbon production mechanisms differ between conventional and unconventional reservoirs, so does managing produced water. In conventional wells, produced water is managed primarily by injecting it into the producing horizon, often to maintain pressure or for enhanced oil recovery (EOR). The rapid increase in unconventional oil production is associated with an increase in coproduced water that cannot be reinjected into the low-permeability tight-oil reservoirs. This produced water is managed primarily by subsurface injection into nonproducing geologic intervals through saltwater disposal (SWD) wells.

A 2013 National Research Council (NRC) report emphasized the impact of the net fluid balance—fluid injection minus extraction (or production)—in controlling subsurface pressure changes and induced seismicity. Because injection and extraction are generally balanced in conventional reservoirs, net pore-fluid pressure changes tend to be minimal, reducing the risk of seismicity, according to the NRC. Produced water from unconventional reservoirs, on the other hand, is injected into non-oil-producing intervals using SWD wells, because the low-permeability reservoirs are not suitable for produced-water reinjection. The consensus is that the combination of injection into nonproducing horizons and increased volumes has likely led to larger net-positive reservoir-pressure changes at the regional scale.

Three entities—the Texas Railroad Commission (RRC), TexNet, and the Center for Integrated Seismicity Research (CISR)—are involved in managing the potential risk associated with SWD in Texas. The RRC has been responsible for regulating SWD wells in Texas since 1982. In 2014, the commission adopted rule amendments designed to address disposal-well operations in areas of historic seismic activity, including taking preventive action when necessary. TexNet and CISR are both housed at The University of Texas at Austin and managed by the Bureau of Economic Geology, a research institution at the university. TexNet, which is funded by the state, collects and catalogs data on earthquakes in Texas. CISR is funded by the petroleum industry. TexNet and CISR collaborate with the help of numerous industry partners to conduct research to better understand naturally occurring and potentially induced earthquakes and their associated risks. Seismicity monitoring began statewide in 2017 using TexNet seismic stations in areas of anomalous seismicity such as Dallas-Fort Worth, northeast of Snyder, Texas, the Permian Basin, and the Eagle Ford operating area of south Texas.

The Seismicity vs. Aquifer Conundrum

According to the Ground Water Protection Council, increasing pore-fluid pressure reduces effective stress on faults, making fault slip more likely. The pore-pressure buildup results from produced-water injection that is not offset by production. Critical factors to consider for induced seismicity include the following:

  • Pore-pressure buildup from injection
  • Presence of an optimally oriented fault for movement in a critically stressed region
  • A pathway connecting the pressure increase with the fault

The time period of injection is also important.
Bridget Scanlon, senior research scientist with the Bureau of Economic Geology, Jackson School of Geosciences, The University of Texas at Austin, wrote with her coauthors in a paper (Scanlon et al. 2019) that the factors listed above result from produced-water injection rates, cumulative injection volume, and proximity to the basement (the older, deformed igneous or metamorphic rock layer below which sedimentary rocks are not common). Of these, Scanlon said, proximity to the basement is the most important. Among the reasons is that large faults are more prevalent at greater depth, particularly in old, brittle, basement rocks that have been subjected to different stresses over long periods of time.

Historical data from plays in the US suggest that shallow disposal may help reduce seismicity; however, tradeoffs between shallow vs. deep disposal need to be considered. Deep disposal wells that extend below oil reservoirs may cost 2–3 times more than shallow wells. However, produced water disposed into shallow intervals has a higher likelihood of contaminating overlying aquifers. With more than half a million oil wells drilled in the Permian during the past century, faults, fractures, and abandoned wells that were not properly plugged can provide pathways for overpressured fluids to migrate upward and potentially affect aquifers.

Optimizing the Process

“Despite the downturn, we produce a lot of water that needs to be managed,” said Scanlon. “Among the biggest issues,” she continued, “is disposal capacity. How can we accommodate and manage historical and projected produced-water volumes? How much capacity is left, and where is it? And how do we manage seismicity, contamination, and blowouts?”

Joey Ndu, vice president of strategy, Enhanced Energetics, has similar ­questions and concerns. To address them, he organized a virtual SWD optimization summit in September that brought together representatives from several water midstream companies.

“The pains and challenges water midstream is facing today are not the same as they were 12 months ago,” he said. “Today we are looking not so much for answers as for how to ask better and more intelligent questions. How do we learn how to think long term; to get more efficient; and to create disposal capacity when and where we need it, with capex, headcount, rig count, and production down, and under regulatory limitations?”

“The looming question,” he continued, “is whether there is a link between injection and seismicity, and if so, how and why. The other challenges stem from that.”

“There is pressure on operators to get volume allowances to make SWD economics feasible,” he said. “For many, the only real option is to go deep. But the rock isn’t homogeneous, so we end up with $8-million gambles. And, in some situations, no remediation is possible. This is very difficult for operators to overcome.”

One path to optimization, according to Ndu, is leveraging data. “Our sector is reaching the point where we have real, meaningful, actionable data about old wells, about the reservoirs and infrastructure around current wells, and about the produced-water supply chain.”

Data-Driven Evaluation of SWD and Seismicity

Nate Alleman, HSE and water infrastructure specialist with ALL Consulting, illustrated Ndu’s point in his presentation of a recent data-driven evaluation of west Texas seismicity correlation with SWD and hydraulic fracturing in west Texas, the RRC response to seismicity, and how to use seismic monitoring to increase the injection rate allowed by the RRC.

The evaluation focused on Culberson and Reeves counties. Four areas of interest were selected based on concentration of seismic events. Two radii were used for each area—one 9.08-km radius (the standard radius the RRC uses for areas of interest within 100 square miles of permit applications) and one 25-km radius, based on a study indicating the potential of SWD wells to affect seismicity up to 25 km away from the well.

Area 2, near the Culberson-Reeves county line, centered on the location of a 4.6-magnitude seismic event that occurred 26 March 2020, at a depth of 23,293 ft. Data from the 9.08-km radius showed little correlation between SWD and seismicity. Extending to the 25-km radius revealed a growing increase in SWD volumes and seismicity in the area, but still no major correlations (Fig. 1).

Fig. 1—A data-driven evaluation of seismicity-to-SWD correlation in Culberson and Reeves counties in west Texas revealed no major correlation in either a 9.08- or 25-km radius from a 4.6-magnitude seismic event.


“Within the 25-km radius, seismicity is occurring in the west-southwest area, and injection in the north-northeast. Just because disposal and seismicity are both increasing doesn’t mean that one is causing the other. SWD alone doesn’t necessarily determine seismicity. Oil and gas production also influences it,” said Alleman.

When filing an application, the RRC uses a grading system that is triggered by seismic events within 100 square miles. Grading criteria include seismic event and fault proximity and characteristics, and depth to the Precambrian basement. A score of B or C indicates that the RRC is somewhat concerned and may want to reduce injection rates.

“What they are really looking for,” said Alleman, “is additional data and more seismic monitors for better depth accuracy, which is something lacking with the current distance between seismic monitors.”

For this reason, the applicant in some instances may be granted an option to increase the injection rate by an additional 10,000 gallons per day by implementing a seismic-monitoring plan and seismic-response plans. Alleman said data-driven evaluations such as the one presented for Culberson and Reeves counties can help operators avoid restrictions by identifying and avoiding sensitive areas, and can also help the RRC in its mission to look at all possible ways to reduce or eliminate seismicity from SWD and hydraulic fracturing.

A Vision of Optimization

When asked what she thought SWD optimization would look like, Scanlon suggested reusing produced water to hydraulically fracture new producing wells as an alternative to subsurface disposal. “This would accomplish a number of goals,” she said, “including managing the net-fluid balance, reducing produced-water disposal, and reducing water demand for hydraulic fracturing from other water sources.”

Scanlon pointed out that it wasn’t so long ago that producers believed they needed fresh water for hydraulic fracturing. “Now we can almost use produced water with very little treatment; just remove solids, kill biologic components, and generate clean brine. Treatment costs are minimal and require minimal effort,” she said, noting that some operators are applying this method now.

Processes based on the application of chlorine dioxide chemistry have proven effective to provide slickwater frac-compatible water at an all-in treatment cost of less than half the equivalent cost of using fresh water (SPE 174956).
Table 1 shows examples of cost considerations that are ­included in the overall water cost for any water-
management plan.



Disposal generally contributes the largest portion of overall cost of using fresh or brackish water, but even when disposal is neglected, produced water can be treated for less than the equivalent purchase-and-handling costs alone. The breakeven point for each treatment system is defined as the volume necessary to recycle before recovering the initial capital investment. SPE 174956 is based on an extensive case study in which the installations were designed to handle more than 25,000 B/D of water. Breakeven points for batch and inline treatment systems were approximately 4.5 million and 3.5 million barrels, respectively.

The inline process can treat produced water continuously, on the fly, and on demand. According to the authors, this process further reduces cost vs. batch treating by efficiently using chemicals and reducing maintenance and cleaning labor. The process is scalable and can be modified to treat single-well or full-field production.

“To make reuse of produced water achievable, the industry needs to conduct more detailed geologic studies to understand the units being targeted for disposal and what has happened to them in the past. Monitoring reservoirs and downhole pressures will improve understanding of ­changes associated with injection volumes,” said Scanlon.

Scanlon cautioned that the potential for this approach to work depends in part on the ability to match produced-water supplies with hydraulic-fracturing-water demand both spatially and temporally. TexNet-CISR data show that cumulative produced-water volumes match hydraulic-fracturing-water demands better in the Bakken and Permian Basin plays than in others. Although the quality of some of the produced water in the Bakken is extremely saline—greater than 10 times that of seawater—studies suggest that advances in fluid chemistry can accommodate such saline water with minimal treatment.

When asked about potential links in some areas between seismicity and water injection during hydraulic fracturing, Scanlon noted that hydraulic-fracturing operations usually take place over a period of days, and any impacts are generally mitigated within a short time period.

Ndu said SWD optimization will depend on the ability to leverage data to do more with less. “Producers need to be technologically aware of how to optimize the assets they currently hold. They don’t want to send their produced water off-lease because of the huge costs,” he said, “but they don’t all track metrics well.

“They need to understand flow volumes of water during different stages of production and how that impacts disposal ability,” he continued. “They need to be able to evaluate temporary storage rather than moving the water off-site. For this, they need to work with data companies to see what is happening off-site so they can renegotiate third-party contracts if that makes sense.

“The industry, including the water midstream sector, is becoming very smart and focusing on using intelligence rather than brute force,” Ndu said. “Service providers need to do a better job of communicating the synergistic value of their offerings.”

SWD in the Energy Transition

The role of produced-water management, including SWD, is not confined to conventional and unconventional oil and gas development. “Produced-water management in unconventional or dewatering reservoirs is similar to other energy technologies that inject or extract large fluid volumes over long periods of time and may also modify pressures that could induce seismicity,” Scanlon said, citing carbon capture and storage and some geothermal systems as examples.

For Further Reading

Managing Basin-Scale Fluid Budgets to Reduce Injection-Induced Seismicity from the Recent U.S. Shale Oil Revolution. B. Scanlon, R. Reedy, University of Texas-Austin, et al.; M. Weingarten, Stanford; K. Murray, San Diego State University, 2019.

SPE 174956 The New Reality of Hydraulic Fracturing: Treating Produced Water Is Cheaper Than Using Fresh by C. Barnes, R. Marshall, et al., Apache Corporation; and J. Mason et al., Sabre

Induced Seismicity Potential in Energy Technologies, National Research Council, Washington, DC. 2013.