Map Shows Where Data Centers Are Being Built in Drought-Hit Areas

A wave of new artificial intelligence (AI) data centers is being built across drought-stricken parts of the U.S., raising concerns about water supplies as the country faces its driest start to a year since 1910.

More than 60 percent of the U.S. is currently experiencing drought conditions, according to the U.S. Drought Monitor, while unusually warm weather and growing water demand are straining supplies nationwide. At the same time, the rapid expansion of AI infrastructure is driving a surge in new data center construction-facilities that can consume millions of gallons of water every day.

There are now more than 5,000 data centers across the U.S., according to the Environmental and Energy Study Institute, with dozens more under construction as major technology companies race to expand AI capabilities.

AI Data Centers Rising Across Drought Zones

Information from a new database launched by environmental activist Erin Brockovich shows more than 50 data centers are currently under construction across the country. A significant number are located in the drought-prone South.

Newsweek has mapped the projects listed in the Brockovich database against U.S. Drought Monitor data to identify which facilities are being developed in areas already experiencing drought conditions.

The findings show eight new data centers under construction in Texas, alongside projects in Louisiana and Mississippi, with several located in areas currently facing severe to extreme drought.

New facilities are also planned across Tennessee, Georgia, South Carolina, North Carolina and Virginia-states where large areas are experiencing severe or exceptional drought conditions.

There are also multiple projects underway in the Northeast and Midwest, including in New Jersey, New York, Pennsylvania, Ohio, Michigan, Indiana and Illinois, though many of those regions currently face fewer drought pressures.

Ayse Coskun, director of the Center for Information and Systems Engineering at Boston University, told Newsweek that “many Southern states are attractive for data center development because of land availability, tax incentives, and growing power infrastructure.”

However, she warned those same regions are increasingly vulnerable to heat stress and water shortages.

“Higher ambient temperatures increase cooling requirements, which can further raise both water and electricity demand simultaneously,” she said. “This could create complex tensions around resource allocation, grid reliability, and long-term sustainability planning, especially in fast-growing regions already facing infrastructure strain.”

Why AI Infrastructure Requires Massive Water Use

Large data centers can use up to 5 million gallons of water in a single day, which is equivalent to the water use of a town populated by between 10,000 to 50,000 people, according to the Environmental and Energy Study Institute.

One report estimated that U.S. data centers collectively consume nearly 450 million gallons of water daily and more than 160 billion gallons annually.

Researchers at the University of California, Riverside, have also estimated that generating a 100-word AI response can consume roughly one bottle of water through cooling and electricity use.

Data centers require vast amounts of water because the powerful computer systems inside generate enormous amounts of heat.

"This power dissipation generates heat that must be extracted from the system," Benjamin Lee, a professor of computer and information science at the University of Pennsylvania, told Newsweek.

"In such places, data centers would compete with other residential and commercial uses of limited water supplies."

Lee added that drought conditions combined with extreme heat can increase cooling needs even further, potentially driving water consumption even higher.

Coskun warned that data center growth could also place additional strain on groundwater systems and municipal infrastructure.

"As climate variability increases, water availability itself could become a constraint on future data center operations," she said.

Environmental Justice Concerns Growing

Researchers have also raised concerns that the environmental burden of data centers is not distributed evenly.

According to the Harvard Science Review, data center development disproportionately affects working-class, Black and Latino communities.

In Bessemer, Alabama, community opposition temporarily halted construction of a data center that was projected to require 2 million gallons of water per day-roughly enough to supply two-thirds of the city's population.

In California, the same report found that roughly 82 percent of data centers are located in communities already suffering from poor air quality, while many are situated in neighborhoods with particularly high levels of diesel pollution.

Can AI Growth Become More Sustainable?

Experts say the challenge facing the industry is balancing water use with energy consumption.

Lee said data centers can be cooled using systems that consume less water, but those approaches often require more electricity.

"Optimizing the data center design for geographic constraints is important," he said.

As AI demand continues to surge, researchers warn that data centers could place increasing pressure on already strained water systems unless companies adopt more sustainable cooling technologies and carefully consider where future facilities are built.

Without stronger oversight and long-term planning, experts say the rapid expansion of AI infrastructure could deepen water shortages and environmental inequalities in the communities hosting the facilities.

2026 NEWSWEEK DIGITAL LLC.

This story was originally published May 29, 2026 at 5:00 AM.