Water that oil and gas operations struggle to dispose of could end up cooling the servers behind the digital economy, according to new research from the University of Texas at Austin.
Engineers found that desalination, the process of separating salt from water, is a strong candidate for treating salty or contaminated water for data center cooling, easing demand on local supplies. The team studied how several data centers could adopt the approach and how the choice of treatment affects cost and environmental impact. The findings appeared in Water Research, and the researchers noted that some data centers are already considering desalination.
“We’re not waiting for a futuristic technology to come along and solve this problem for us,” said Vaibhav Bahadur, a mechanical engineering professor at the university’s Cockrell School of Engineering and a leader of the study. “We can solve parts of this problem with the technology we already have; there’s just a cost to it.”
Membranes or heat
There are two main routes to removing salt. Reverse osmosis pushes water through membranes and is more energy-efficient, but it struggles with water saltier than seawater. Thermal methods boil off the water and condense the vapor. They handle very salty water well but consume far more energy and suffer from scaling on equipment.
Even the most energy-hungry treatment accounts for a small fraction of a data center’s total energy footprint, the researchers said. That means facilities can use water more efficiently without sharply raising costs or overall energy demand.
Permian water has nowhere to go
The study points to a nearby supplier: the oil and gas industry.
“The Permian puts out 1,200 Olympic-sized swimming pools per day,” Bahadur said, noting that the water is transported long distances and injected underground, which carries risks including seismic activity. “We’re also running out of space to bury that water, by as early as 2030.”
Longview scenario scores best
To measure strain on local water sources, the team created a metric called Saline Water Utilization Intensity, or SWUI. Of three data center scenarios studied, a project in Longview, Texas, imposed the least stress by that measure. It would clean brackish groundwater with two-stage reverse osmosis, recovering high levels of freshwater at low cost and with relatively low energy use.
Policy pressure in Texas
The research arrives as the state scrutinizes data centers. In September, Gov. Greg Abbott ordered a pause on new data center permits to evaluate local impacts. The Texas attorney general’s office has also announced a probe into water use at existing facilities.
Bahadur and colleagues in the COMPASS consortium, based at the Bureau of Economic Geology in the UT Jackson School of Geosciences, are building software to analyze individual projects, much as they evaluated the desalination options. The tool aims to inform officials weighing permits. Bahadur and COMPASS colleague Ning Lin recently testified before the Texas Senate Committee on Water, Agriculture and Rural Affairs.
“As they’re making important decisions about data centers, local officials often have no objective analysis of water, power usage and other important factors,” said Lin, the bureau’s chief economist. “We hope to give decision-makers the tools they need to protect and benefit their communities.”