Water Rights and Impacts

Data centers generate massive amounts of heat and need millions of gallons of water annually for cooling. Training AI models can consume hundreds of thousands of liters of water, and even everyday AI queries add up. These facilities also need permits for air emissions (from backup diesel generators), stormwater management, and impacts on wetlands. Some federal and state environmental laws require review processes that developers can’t skip.

Environmental reviews under federal and state law create protected public participation opportunities. Agencies must respond to substantive comments, and weak environmental analysis can be challenged through appeals and lawsuits. The scientific evidence on AI’s water and energy consumption is growing rapidly and communities can provides concrete data to cite when demanding comprehensive Environmental Impact Statements instead of cursory reviews.

Most water issues are addressed at regional, state, or local levels. In general, the federal government sets standards for the amount of pollution allowed in water discharges (“point sources”) and site run-off (“non-point sources”) which are implemented by the states. Access to water from a lake, reservoir, or aquifer is based on water law, which is quite different in the generally wet eastern US and the often arid western United States.

This gateway matters because developers want quick approvals with minimal conditions and limited public disclosure. Communities can file detailed scoping comments demanding thorough review, present expert testimony from hydrologists and scientists, challenge chemical risks from cooling systems, and organize environmental justice coalitions to document cumulative impacts. When EPA regional offices, state environmental agencies, and water authorities see organized community opposition backed by scientific evidence, they’re normally forced to take a harder look.