Energy use is one of the most common concerns raised by residents when a hyperscale data center is proposed. The amount of electricity a data center uses depends on several factors, including the size of the facility, the cooling system it uses and the type of customer it serves.  Some cooling systems use more water and less electricity, while others do the opposite.  A common way to measure energy efficiency is through a Power Usage Effectiveness (PUE) rating.  This industry standard rating compares the total energy used by the entire facility to the energy used by the computer equipment itself.  A lower PUE indicates that a larger share of the facility’s electricity is being used for computing, rather than supporting functions such as cooling, lighting and other building operations.  A PUE rating will always be above 1.

The amount of water a data center uses depends largely on its cooling system and the overall size of the facility. Some data centers rely on water-based cooling systems, while others use air-cooled systems that require little or no water. A common measure of water efficiency is Water Usage Effectiveness (WUE), which compares the amount of water used to the amount of energy consumed by the data center’s computer equipment. Lower WUE values indicate more efficient water use. Across the industry, many facilities average about 1.8 liters of water per kilowatt-hour of electricity used, while highly efficient facilities may use less than 0.2 liters.

There is often a tradeoff between water use and energy use. Facilities that rely on dry or closed-loop cooling systems typically use less water but may require more electricity for cooling. In contrast, facilities that use water-based evaporative cooling systems often use less electricity for cooling but consume more water.

The most water intensive aspect of a hyperscale data center’s workflow happens offsite with electricity generation, using more than 10 times as much water as cooling. Learn more: UW Milwaukee Data Center Legislative Model Report.

Before a data center is approved, a municipality may require engineering studies to evaluate potential impacts on local water and wastewater systems. These studies can help determine whether existing infrastructure has sufficient capacity to meet the facility’s needs and identify any upgrades that may be necessary.  

Hyperscale data centers have a large physical footprint which in some cases may exceed 1,000 acres. Developers often target farmland or undeveloped open space in urban adjacent rural areas. Public concerns over the impact on local resources is not solely focused on water and electricity, but includes the overall concern of the environmental impacts a large-scale land use development may create.

The Wisconsin Department of Natural Resources oversees environmental analysis when development uses state or federal grant funds, state agency actions are required, or if the proposed development potentially impacts wetlands, navigable waters, stormwater discharges, floodplains, shorelands, high-capacity water withdrawals, or wastewater facilities. County/municipal zoning ordinances may also introduce development standards used to best “fit” these large land use developments into their communities.