Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →Data centers use electricity directly to run servers and cooling equipment; some cooling systems also consume water on site. Electricity generation can consume additional water and produce emissions elsewhere. These are separate impacts with different locations and accounting boundaries, so there is no universal water-per-query figure—and operational totals are not a complete lifecycle footprint.
How much electricity do data centers use?
Data-center electricity demand is rising, but forward-looking figures are model estimates, not measurements of future consumption. The U.S. and global estimates below come from different models and scopes; they should not be treated as one continuous series.
| Geography and source | Period | Estimate |
|---|---|---|
| United States; Lawrence Berkeley National Laboratory (LBNL), United States Data Center Energy Usage Report: 2025 Update, published June 2026 | 2030 | 649 TWh in the Reference Case; uncertainty bounds of 521–843 TWh. The report estimates data centers could represent 9.5%–15.3% of total U.S. electricity use. |
| Global; International Energy Agency (IEA), 2025 assessment | 2024 | 415 TWh, about 1.5% of global electricity consumption. |
| Global; IEA, 2025 assessment | 2030 | About 945 TWh in the main outlook. |
| Global; IEA, 2025 assessment | 2035 | About 1,200 TWh in the Base Case. |
LBNL’s U.S. forecast uses a bottom-up model involving planned IT-equipment shipments, per-device electricity use, cooling simulations, facility types, and locations. IEA’s global assessment uses a separate scope and modeling approach. IEA estimates global data-center electricity consumption grew by around 12% annually from 2017 through 2024; that growth rate is historical, not a forecast that the same rate will continue.
These totals cover data centers and their workloads broadly, not AI alone. AI is one contributor within the overall estimates. IEA notes that the worldwide share is modest relative to all electricity use, while local effects can be more pronounced: nearly half of U.S. data-center capacity is in five regional clusters, according to its 2025 assessment.
How much water do data centers use?
“Water use” can mean different things. Direct water consumption occurs at the facility, including water lost through evaporation in some cooling systems. Indirect water consumption occurs at power plants generating electricity used by the facility. It is not water delivered to the data-center site. LBNL defines consumption as water removed from the immediate water cycle through evaporation or other irreversible processes; withdrawals and consumption are not interchangeable.
| Impact and geography | Period | Estimate and boundary |
|---|---|---|
| U.S. data-center electricity use; LBNL 2024 report | 2023 | About 176 TWh of electricity and approximately 66 billion liters of direct on-site water consumption. |
| U.S. direct on-site water consumption; LBNL 2024 report | 2024 projection | 60–124 billion liters. |
| U.S. direct on-site water consumption; LBNL 2024 report | 2028 projection | 145–275 billion liters. |
| Water associated with electricity generation for U.S. data centers; LBNL 2024 report | 2023 | Nearly 800 billion liters of indirect water consumption, estimated using regional grid water factors. The national average factor was 4.52 liters per kWh. |
The 2023 direct-water figure and the 2024 and 2028 figures are estimates or projections in LBNL’s 2024 report, not measurements that apply to every facility. The report’s indirect-water calculation did not incorporate individual facilities’ power-purchase agreements or behind-the-meter generation. The direct and indirect figures therefore describe different parts of the system and should not be mistaken for a site-level water bill.
Why do water and electricity figures vary so much?
A workload’s footprint depends on the computing performed and the conditions under which it runs. A 2025 LBNL-authored review found more than 10,000-fold variation among modeled workload-level water-use estimates. That spread reflects variation across modeled conditions, not a measured difference between every pair of real-world tasks.
The review ranks the main determinants as server efficiency, the water-consumption factor of the electricity grid, server utilization, cooling-system type, infrastructure efficiency, climate zone, the share of inactive servers, and server refresh cycle. In practical terms, a workload on efficient, well-used servers in a low-water-intensity grid may have a different footprint from the same workload on less efficient equipment or a more water-intensive grid. Cooling choices can also shift impacts between on-site water and electricity demand.
Rank #3
There is no single cooling or siting choice that minimizes water use everywhere. LBNL’s 2021 spatial study found that one-fifth of the direct water footprint of U.S. data-center servers fell in moderately to highly water-stressed watersheds, and nearly half of servers were fully or partly powered by plants in water-stressed regions. Those are findings from that study’s methods and period, not current proportions for all facilities.
To compare facilities or proposals meaningfully, assess at least:
- On-site water consumption, and the local watershed’s water stress.
- Electricity demand for computing and cooling, alongside the water intensity and emissions of the local grid.
- Server efficiency and utilization, including how much equipment is inactive.
- Climate, cooling-system design, and reliability or operational requirements.
What emissions are included in data-center environmental footprint estimates?
Electricity-related emissions are an operational impact that occurs mainly where power is generated, not necessarily at the data center. IEA’s 2025 assessment estimates that data centers cause around 180 million metric tons (Mt) of indirect CO2 emissions from electricity consumption today, excluding backup-power emissions. Its scenarios put electricity-related emissions at 300 Mt by 2035 in the Base Case and 500 Mt in the Lift-Off Case. These are scenario estimates for all data-center workloads, with AI as one component.
LBNL’s 2024 estimate for U.S. data-center electricity use in 2023 attributed 61 billion kilograms of CO2-equivalent emissions to that electricity, alongside nearly 800 billion liters of indirect water consumption. Its national 2023 average was 0.34 kg CO2e per kWh. The calculation used grid factors and did not incorporate individual facilities’ power-purchase agreements or behind-the-meter generation; it is not a complete accounting of every facility’s emissions.
A facility’s operational footprint may include electricity for IT equipment and cooling, direct cooling-water consumption, electricity-related water and emissions, and backup generation where it is measured. A full lifecycle footprint is broader: it can also include land and construction, materials, semiconductor and server manufacturing, and end-of-life. The quantitative estimates above do not provide a complete inventory across those lifecycle stages, so they should not be presented as the entire environmental footprint of a data center.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What can reduce or manage these impacts?
Measures need to be assessed against the local grid, watershed, weather, and reliability needs. Improving server efficiency and utilization can reduce the electricity needed for a given computing workload; cooling and infrastructure choices affect both facility energy and water. The LBNL review cautions against a universal recipe because trade-offs depend on site conditions.
The U.S. Department of Energy identifies options for meeting and managing fast-growing, geographically uneven, often continuous data-center loads: clean generation, storage, existing nuclear and hydropower, grid expansion, efficiency, demand resources, and planning. These are options for local system analysis, not guaranteed impact reductions. For example, an electricity contract does not by itself establish that a facility has eliminated physical grid impacts; its actual effects depend on the power system and accounting boundary.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
