Carefully planning where to build solar and wind power facilities can maximize the amount of renewable power available while minimizing costs by reducing the need for battery storage, according to a new analysis. The approach takes advantage of a characteristic of renewable energy sources that is usually thought of as a liability: their variability.

Skeptics of renewable energy development often point out that the wind doesn’t always blow and the sun doesn’t always shine. True enough, but when the sun isn’t shining in one place, the wind is often blowing there, or somewhere else that’s not too far away.

These patterns open up the possibility of complementarity: using different renewable energy sources to balance each other out across time and space.

“We want the best alignment of energy supply with demand which will improve efficiency and reliability of the energy system overall,” says study team member Liying Qiu, a postdoctoral researcher at the Massachusetts Institute of Technology.

Right now, renewable energy planning is mostly done with a very broad brush: build this many wind farms in a country, or aim to get a certain percentage of power from solar. Meanwhile, decisions about where to site specific facilities are made by individual developers or utilities, who tend to put renewable energy installations where there is the most sun or wind overall.

But this can result in excess power generation (if it’s especially windy when electricity demand is relatively low), or shortfalls (if there’s plenty of sunshine during the day, but peak electricity demand happens after sunset). Resolving these challenges often requires building costly battery storage, or turning to fossil fuel-based power at times of peak demand.

In the new study, Qiu and her colleagues used high-resolution meteorological data from the National Renewable Energy Laboratory and a high-resolution energy system model that they developed in order to determine the most cost-effective places to build solar and wind farms in three regions of the country: California, Texas, and New England.

The study is the first to make use of both weather and electricity demand data at a resolution of a few kilometers. “This approach enables more precise siting of renewable installations to find the best match with time-varying energy demand and reduces overall system costs,” Qiu says.

The locations of renewable energy facilities in a lowest-cost energy system are different than those identified by the typical approach of maximizing annual energy production, the researchers report in the journal Cell Reports Sustainability.

The model suggests that in New England, wind power facilities should be located not necessarily in the windiest places overall, but in places that are reliably windy at night, when solar power is unavailable.

In Texas, wind power facilities should be sited in the western part of the state, which is windiest in the morning, as well as along the Gulf Coast, where it is windiest in the afternoon.

“Only high-resolution data can accurately represent the variability of renewable resources, which is essential for optimizing resource complementarity,” says Qiu. At a more typical, coarser planning scale, the possibilities for identifying such complementarities are lost, the researchers found.

Balancing fluctuations in renewable energy supply and power demand over the course of a 24-hour period “had a much larger impact than expected” in terms of lowering overall energy system costs, she adds.

The analysis does not yet account for climate change, notes study team member Michael Howland, assistant professor of civil and environmental engineering at MIT. “Climate change may impact energy generation and also energy demand,” he says. “So our next step is to analyze whether we should design decarbonized energy systems differently in a changing climate to make sure we minimize cost and also maximize resilience.”

Source: Qiu L. et al. “Decarbonized energy system planning with high-resolution spatial representation of renewables lowers cost.” Cell Reports Sustainability 2024.

Image: Wayne S. Grazio via Flickr.