A new study lays out a wide range of options available to cost-effectively eliminate greenhouse gas production from the energy system in the United States by 2050. The findings give policymakers and industry leaders valuable insights on how to chart a path forward to address climate change.

The paper, "Diverse Decarbonization Pathways Under Near Cost-Optimal Futures," is published in the open-access journal Nature Communications.

"There isn't just one way to cost-effectively decarbonize our energy system," says Jeremiah Johnson, co-author of the study and a professor of civil, construction and environmental engineering at North Carolina State University.

"In fact, we have many technologies to choose from. Our study helps people understand exactly what those options are, and how we may want to prioritize them."

"There are a range of models out there that are designed to find the least expensive path forward to decarbonize our energy system—essentially identifying the optimal approach to eliminating greenhouse gas production in everything from electric power production to transportation and industry," says Aditya Sinha, corresponding author of the study and a research scholar at NC State.

"The problem is that it is difficult for these models to fully capture uncertainty in such a complex system," Sinha says. "There are a lot of different technologies that can help us decarbonize, and it's difficult to determine how much flexibility we have in identifying which of these tools can be used to reach an optimal outcome.

"One way to address this challenge is to stop trying to identify the path that gets you to a perfect solution and instead identify alternative options that get us very close to the least expensive path forward."

For this study, the researchers defined "very close" as coming within 1% of the optimal cost for decarbonizing the entire energy system.

Specifically, the researchers used an existing model, called Temoa, that was originally designed to determine the least expensive pathway to achieve decarbonization. They ran that model to identify what the optimal cost would be. They then added 1% to the optimal cost and modified the model using that number as a constraint.

"The model then has thousands of decisions to make," Johnson says. "How much solar should be built? Should homeowners swap natural gas heat for electric heat pumps? And so on.

"We ran our modified version of Temoa 1,100 times, each time telling the model to favor—or disfavor—any given technology. In part, this reflects the fact that humans make all sorts of decisions that are not driven solely by what makes economic sense, which we wanted to account for."

"This approach gave us a clearly defined range of technologies that would allow us to eliminate greenhouse gas production from the energy system and still stay within 1% of the optimal cost," says Sinha.

The findings can be broken down into four categories:

• Category 1 consists of technologies that were broadly adopted in all 1,100 solutions the model identified. This includes expansion of both solar and wind energy generation, as well as expansion of energy storage capacity on the power grid.
• Category 2 includes technologies that were either eliminated or greatly reduced. This includes greatly reducing reliance on petroleum in the transportation sector and eliminating coal power generation that wasn't mitigated by carbon capture and sequestration.
• Category 3 consists of emerging technologies with a wide range of possible outcomes, meaning that some of the model's scenarios found the technologies receiving widespread use, while other scenarios didn't include these technologies at all. These technologies include things like direct air capture—which pulls carbon dioxide out of the air—or the use of hydrogen in transportation and industry.
• Category 4 covers technologies the model generally didn't use at all—but when it did make use of these technologies, it relied on them heavily. These include synthetic fuels produced from carbon dioxide and coal power plants that incorporate carbon capture and sequestration.

"Running the model 1,100 times produced an enormous range of potential outcomes, to the point where it was difficult to know where to start," says Sinha. "It was only after an in-depth analysis of these outcomes that we were able to identify these categories, which provide a good way of understanding what our options are and how we may want to prioritize them."

"From a practical standpoint, these findings tell us a few things," says Johnson. "First, we need to figure out how to facilitate the more widespread adoption of the technologies in Category 1.

"Second, we need to figure out how to plan for an orderly and just—but timely—transition away from the technologies in Category 2," says Johnson. "Third, we won't need all of the technologies in Category 3, but we'll need some of them. That means we need to invest in research and development to determine which technologies we should prioritize and how to deploy them. Lastly, we also need to invest in research and development to determine if any of the technologies in Category 4 are truly worthwhile and, if so, how to capitalize on those technologies."