Can Data Centers Run on Renewable Energy?

For the next four years the amount of electricity needed to power new data centers will double. The growing AI boom will need twice the amount of electricity than is available today to run their data. Tapping into renewable energy sources such as wind, solar and hydro could see companies lessen their reliance on fossil fuels—the fuels that produce greenhouse gas emissions with deleterious climate change impacts.

Last month United Nations Secretary-General António Guterres called on major AI companies to power all data centers with renewable energy by 2030.

Can it be done?

Although renewable energy use is growing, so is the use of fossil fuels. The International Energy Agency (IEA) projects growth in electricity generation for data centers worldwide for the next five years, saying by 2030 renewables will produce by 210.03 TWH (terawatt-hours), equal to 210,030,000,000 kilowatt hours followed by Natural Gas at 192.54 TWh, Coal at 77 TWh and finally nuclear at 42 TWh.

IEA also finds that using renewables to generate electricity will surpass coal-fired power this year (2026), growing “by more than 8% in 2026. Its share in the electricity generation mix is set to rise from 33% in 2025 to 37% by 2027.”

Integral to using renewable energy is modernizing electricity grids to become more efficient in utilizing more and different energy sources and to optimally lower the cost of electricity. It’s a process that could take years.

Michael Thomas, founder of the market intelligence site Cleanview, says right now existing data centers use a variety of ways to generate power. Among alternate sources of energy at data centers, companies use on-site mobile gas generators, aero-derivative turbines designed for aircraft and warships, reciprocating engines and refurbished turbines used in industrial operations. Thomas notes that “power generation efficiency is out….and that AI data center can generate $10-12 million per mega-watt annually.” That means that data center developers are willing to have a center up and running as quickly as possible, as opposed to waiting for years to bring it online via the grid, even if it means with inefficient systems using fossil fuels.

Environmental Concerns

https://www.maps.com/data-center-map-shows-data-centers-and-energy-infrastructure/

But it’s not only about the carbon footprint set by data centers using various types of energy, it’s other uses that impact the overall environment. A new report by United Nations University, Institute of Water, Environment and Health, (UNU-INWEH), reveals that “…every kilowatt-hour used by AI carries carbon, water, and land implications, and these footprints do not always move in the same direction: low-carbon electricity is not automatically low-water or low-land.”

That the AI ecosystem uses carbon-only metrics are no longer sufficient for AI is clear as stated in the report. AI’s environmental impact should require transparency to “show carbon, water and land footprints so regulators and investors can compare like with like.”

One of the most startling points of the report is that the AI program ChatGPT alone processes around 2.5 billion prompts per day, translating to roughly 383 GWh of electricity per year for a single product. The report says “Offsetting associated carbon emissions would require 2.6 million tree seedlings grown for 10 years, enough trees to cover a land area the size of Manhattan. The water footprint is equivalent to the minimum annual domestic water needs of roughly 500,000 people in Sub-Saharan Africa, and the land footprint is equal to over 800 football fields.”

image: https://usdatamap.com/