Northern Virginia Wants to Turn Data Center Waste Heat into a Community Resource
Turning data center waste heat into a regional energy asset

The numbers are hard to ignore. In Northern Virginia – already home to the largest concentration of data centers in the world – peak energy usage from those facilities hit an estimated 2.8 gigawatts in 2022. By 2038, the regional utility is forecasting that number climbs to roughly 14 gigawatts. With approximately 265 data centers currently operating and another 110 planned before 2035, the region is staring down an infrastructure challenge that no single solution can fix on its own.
But a growing coalition of planners, engineers, and energy companies thinks district energy might be a bigger part of the answer than most people realize – and not just for the obvious reasons.
"The practical reality is both are not going to succeed without working together," said Michael Ahern, who leads the system development group at Ever-Green Energy, a 225-person district energy company headquartered in Minnesota. He was referring to the data center industry and the district energy sector – two worlds that have historically operated in parallel without much coordination. "It's the right thing to do, but it's also the only way that we're going to achieve the decarbonization goals."
Heat as a Resource, Not a Problem
Every kilowatt-hour of electricity that flows into a data center eventually leaves as heat. That's not a design flaw – it's thermodynamics. What district energy advocates argue is that this heat, which data centers currently reject into the atmosphere through cooling towers, is an enormous untapped resource sitting in the middle of communities that increasingly need affordable, low-carbon heating.
Luke Gaalswyk, president and CEO of Ever-Green Energy, framed it in straightforward terms. "Taking every kilowatt-hour of electricity that you put into a data center as far as possible is enabled through district energy integration," he said. Connecting data centers to district heating networks reduces cooling tower electricity usage, can eliminate active cooling tower water consumption entirely, and provides a far more efficient heat source for buildings than individual electric heat pumps or resistance heating.
Water is a particularly acute issue. Data centers are notoriously thirsty, and communities near major campuses have raised concerns about the draw on local water supplies. Eliminating evaporative cooling through heat recovery integration directly addresses that pressure.
The efficiency case extends to the broader grid, too. As utilities struggle to keep up with surging data center load – and as data center developers increasingly turn to on-site generation to work around interconnection delays – district energy integration offers a way to extract more value from every unit of electricity already being consumed, rather than simply adding more supply.
The Temperature Differential Question
For HVAC engineers, the conversation gets more technical quickly. One of the key design variables in any heat recovery scheme is the temperature differential – the delta T – between the heat source and the district heating network's supply temperature.
"The closer that source of heat is to the supply temperature on the hot water network, the more efficient your system is going to be," Gaalswyk explained, "because less electricity you need to use with your heat pump to lift that temperature up from the heat source to the hot water network."
That relationship directly determines heat pump coefficient of performance, which in turn drives project economics. The other major economic lever, he noted, is the spark spread – the price differential between a kilowatt-hour of electricity and a kilowatt-hour of natural gas – which shapes the financial case for displacing fossil fuel heating with electrified, heat-pump-driven district systems.
There's a convergence happening on both sides of that equation that's making the math more attractive. On the data center side, the shift toward direct liquid cooling and advances in chip thermal tolerances are pushing waste heat temperatures higher – meaning the heat coming off servers is increasingly closer to useful district heating supply temperatures without additional lifting. On the district energy side, the industry has been moving away from legacy steam networks and high-temperature hot water systems toward lower supply temperatures that improve overall system efficiency.
"That really provides for some great opportunities for integrating data center waste heat," Gaalswyk said.
Ahern added that designing for the right temperature differential is already standard practice in district energy. "That's already inherent as a design standard within the district energy industry – we need our customers to be operating at that temperature differential to optimize efficiency," he said. "It's already programmatic with most district energy systems."
Geothermal as a Storage Bridge
One of the more intriguing technical wrinkles in Ever-Green's current project portfolio involves pairing data center heat recovery with geothermal storage – essentially using the ground as a thermal battery.
The concept addresses a fundamental timing mismatch: data centers produce waste heat continuously, including in summer months and overnight when there's little or no demand for building heat. Rather than rejecting that heat or trying to find an immediate use for it, the system injects it into the ground, where it can be retrieved weeks or months later when heating demand peaks.
"Regardless of the temperature coming off the data center – whether it's 100°F, 90°F, 110°F – you can put that heat right into the ground without having to do anything to it, and then utilize that heat at a later date when it's valuable," Ahern said. "It's BTUs getting put in the ground to be used whenever it is needed."
The economic appeal is significant. In a conventional heat pump system, the cost and energy penalty of lifting low-grade waste heat to district heating supply temperatures can be a barrier, particularly when the heat is available at inconvenient times. Geothermal storage sidesteps that constraint – the heat goes in whenever it's produced and comes out whenever it's needed, without requiring active thermal management in between.
Northern Virginia's Particular Urgency
Dale Medearis, senior regional planner with the Northern Virginia Regional Commission – joined the conversation representing and laid out why the stakes feel especially high in this corner of the country.
The economic dependency is stark. Data centers generate roughly $1 billion in local tax revenue annually for jurisdictions like Loudoun County alone, with Prince William County – the second-largest concentration – collecting somewhere in the range of $150 to $200 million. At the same time, the region is absorbing significant federal job losses and the contracting ripple effects that come with them. Local governments that have come to rely on data center tax revenue can't afford to see that industry constrained or relocated.
But the industry's growth is generating its own friction. Residents near data center campuses have raised persistent complaints about noise – much of it from cooling systems – and about the visual and aesthetic impact on communities that want green space and neighborhood character, not acres of mechanical equipment. Grid congestion is forcing difficult conversations about new high-voltage transmission infrastructure through Virginia's hunt country and Blue Ridge foothills.
"It's the union of not just the energy efficiencies that we see to be realized through district energy," Medearis said. "It's also the ability for district energy to address some really critical social and economic problems."
He pointed to European precedents – Frankfurt, Copenhagen, Paris – as validation that the model works at scale. Northern Virginia has exactly one district energy system currently operating, a legacy district cooling installation dating to the 1970s that, by Medearis's own admission, hasn't been maintained to the standard the technology deserves. The goal now is to change that trajectory.
"Success for us looks like a district energy project in the next 24 to 36 months," he said, "where we're able to pull together an operator, a collection of planners and designers from local government, and create this nice synergy between supply and demand – and then build it out and replicate it across our region over the next five to ten years."
The Replication Problem
That word – replicate – came up repeatedly. The people in this conversation aren't just trying to build one project. They're trying to demonstrate something that other communities can copy.
Gaalswyk made the point that once a handful of these projects are operating and producing real performance data, the calculus changes for everyone. "Communities across the United States are going to see the value, and data center developers are going to see the value too," he said. Data center operators, he noted, are acutely sensitive to speed-to-market. Anything that helps them navigate power constraints, water constraints, and community opposition faster is worth serious attention.
The pitch to data center developers isn't purely altruistic, either. Offering waste heat to a neighboring college campus, healthcare system, or downtown district gives a data center project something it currently lacks in most communities: a tangible local benefit beyond digital infrastructure. In an environment where local governments are increasingly scrutinizing data center approvals, that community integration story matters.
For the engineering community, the near-term opportunity is in the design work – mapping data center campuses against existing and planned district energy infrastructure, modeling temperature differentials and heat pump performance, and identifying the sites where the economics pencil out first. Northern Virginia's planners have already begun that mapping exercise, overlaying data center locations against proximity to university campuses, corporate facilities, and residential density.
The pilot project may still be a year or two away. But the technical case is solid, the economic pressure is intense, and the precedents from Europe are clear. The question now is mostly one of execution – finding the right combination of stakeholders, the right site, and the right moment to put a shovel in the ground.
"We're so ravenous for a successful pilot project," Medearis concluded. "Any help, any support, any examples of best practices that can feed us with the technical and economic data is going to be so valuable."
Looking for a reprint of this article?
From high-res PDFs to custom plaques, order your copy today!






