New research suggests electricity demand from data centers and cryptocurrency mining is likely to increase power costs in some parts of the country by up to 57% by 2030, with a national average increase of 6%-29%. Electricity demand related to data centers is also likely to increase CO2 emissions by up to 28% by 2030, relative to a future with no data center growth, according to the analysis from North Carolina State University, Carnegie Mellon University, the University of Pittsburgh and the University of Toronto.
“Power demand in the U.S. was relatively flat for almost 20 years,” says Jeremiah Johnson, corresponding author of a journal article on the work and an associate professor of civil, construction and environmental engineering at North Carolina State University. “But in the past couple of years we’ve seen a significant increase in power demand, due largely to data centers and – to a lesser extent – cryptocurrency mining.
“We wanted to understand the implications of this increased demand,” Johnson says. “What new power infrastructure will need to be built? Where? How will these systems be operated? What will that mean for the cost of electricity? And what will it mean for carbon emissions?”
The researchers drew on recent research to estimate data center and cryptocurrency power demand through 2030, and then made use of computational modeling tools to forecast what technologies would be used to generate that power.
“Specifically, we used an energy system optimization model,” says Anderson de Queiroz, co-author of the paper and an associate professor of civil, construction and environmental engineering at NC State. “An energy system is the full supply chain that delivers energy to people. And optimization models are tools that can be used to search for the least expensive ways to plan, maintain and operate energy systems in order to meet energy demand while complying with existing laws and regulations.”
“The optimization model we used for this work was designed to focus on electrical power generation,” says Johnson. “We were able to look at energy supply and demand on an hourly level for 26 regions of the power grid, covering the lower 48 United States.”
One key finding from the optimization model is that increased demand will lead to increased carbon dioxide emissions from electricity generation, by up to 28% over the next three and a half years.
“The power sector has made progress in reducing carbon emissions over the past 20 years, but the increased demand will essentially erase a lot of that progress,” says Johnson.
“We also found that electricity costs will increase by an average of 6%-29%, nationally. However, those prices could increase as much as 57%, depending on where you are in the country.”
Those electricity price increases would be most pronounced in Virginia, eastern North Carolina, Pennsylvania, Maryland, Delaware, New Jersey, west Texas, Ohio, West Virginia and New York.
“But those future price increases depend on where new data centers are built,” Johnson says. “For example, price increases in Virginia jump due to substantial expansion of data centers. If the data centers are distributed more broadly across the country, Virginia won’t be hit as hard. Prices will still go up for everyone, but the expense will be spread more evenly across the country.
“There is a great deal of uncertainty regarding the cost of installing new natural gas turbines and the cost of natural gas itself,” Johnson says. “But regardless of fuel cost and the cost to build new natural gas plants, we still see substantial increases in electricity cost and CO2 emissions.
“The public and policymakers need to be aware of these near-term challenges – 2030 is less than four years away,” says Johnson. “Our findings highlight the need for regulators and utilities to make informed decisions about near-term power generation, and for government officials at all levels to make informed decisions related to the construction of data centers.”
The paper, “Power System Costs and Emissions from Data Center and Cryptocurrency Mining Expansion in the United States,” is published open access in the journal Environmental Research Letters. The paper was co-authored by Cameron Wade of Sutubra Research; Michael Blackhurst of the University of Pittsburgh; Joseph DeCarolis and Paulina Jaramillo of Carnegie Mellon University; and Daniel Posen of the University of Toronto.