A recent analysis reveals the daunting number of variables that need to be considered when attempting to pair agricultural production and solar generation.
Federal researchers know that solar panels and crops can coexist and provide mutual benefits in certain scenarios. A recent study by the National Renewable Energy Laboratory (NREL) confirms this but also shows that such co-location can lead to crop or financial losses, including from complications like mold-causing dew accumulation and soil damage from construction equipment.
Advocates who see the concept as a potential solution to land-use constraints are now pushing for more funding and collaboration with farmers to test and document outcomes in as many different settings as possible. The hope is that they can prove benefits in enough scenarios to help the solution scale beyond the handful of small farms that have currently implemented it.
“We know we can grow food under solar projects,” said the NREL paper’s lead author, Jordan Macknick. “What remains to be seen is if we can scale up agrivoltaics in a way that meaningfully improves local food production and farmers’ bottom lines while also aligning with the realities of solar development costs, timelines, and practices.”
Moisture and soil
NREL defines agrivoltaics as the “sharing of sunlight between the two energy conversion systems: photovoltaics and photosynthesis,” and notes that “the solar and agricultural activities [must] have an influence on each other.”
Agrivoltaics includes planting pollinator habitat in and around solar panels, and allowing animals to graze around panels. But the sector with the most variables to study is arguably the growing of crops under and between solar panels.
In 2015, the U.S. Department of Energy began researching agrivoltaics through the InSPIRE (Innovative Solar Practices Integrated with Rural Economies and Ecosystems) program. The August NREL paper compiles results from InSPIRE sites with university and other partners in states including Arizona, Georgia, New York, Minnesota, Colorado, Idaho, Oregon, California, Pennsylvania, Massachusetts, and Washington, D.C.
The analysis confirmed that agrivoltaics can help in water-stressed areas, since the shade from panels reduces evaporation due to sun and wind, and water from precipitation or even water used to clean panels can be collected and funneled to crops. Electricity generated can also be used on-site to power pumps for irrigation. At a “dry farming” test site in Oregon’s Willamette Valley, researchers are exploring whether agrivoltaics minimize a condition in tomatoes known as blossom end rot exacerbated by drought.
Increased moisture retention from solar panels can also create complications, however. Rodents and insects may be attracted by humidity and moisture, the study notes; rodents can hurt crops and also chew through electric wires. At Jack’s Solar Garden in Colorado, fungus grew where runoff from dew on the panels collected. Researchers noted that problems could be averted by moving beds away from the drip, or otherwise managing dew collection.
Greg Barron-Gafford, a geography professor and director of food, water and energy research at the University of Arizona’s Biosphere 2, noted that despite some of the challenges, increased moisture under panels is generally a boon for plants.
“If a plant is in a more humid environment, it is less stressed about conserving water, and it can do more photosynthesis,” he said. That means more growth of leafy greens like kale and lettuce, and more resources drawn from the leaves for fruiting plants like peppers.
Construction of agrivoltaics — with heavy equipment — can compact soil, making it harder for it to hold water and nutrients. InSPIRE research in Colorado showed soil at agrivoltaic sites still compacted a decade after construction. But using certain types of equipment and construction processes can reduce the impact on soil. University of Maine researchers are studying whether lower-impact “careful” or “mindful” construction practices can improve agrivoltaic blueberry yields.
Understanding the local soil variation and quality can help minimize harm. Investigating previous uses of the land, including herbicide and pesticide use and types of crops grown, also helps in designing successful agrivoltaic projects.
Mixed results
Performance of certain crops can be counterintuitive, underscoring the need for evidence-based research, the NREL study notes. For example in Colorado, “sun-loving” grass performed better than “shade-loving” grass in the shade of solar panels, surprisingly. Additionally, some crops must be rotated after several seasons because of their effect on soil, so an agrivoltaic array must be planned for multiple seasons.
The study aggregates the effect of agrivoltaics on crop yields at different sites. Tomatoes saw up to double yield with agrivoltaics, while wheat, cucumbers, potatoes and lettuce showed significant negative impacts and corn and grapes showed minimal impact.
In areas that don’t have extreme sun or heat, reduced yields can be due to reducing the amount of sunlight that plants get under panels. These realities must be considered, but there are also ways to moderate the sun-blocking influence, researchers say.
Tracking solar panels that move with the sun only shade plants for part of the day. And Barron-Gafford noted advancing technology includes solar panels might allow the wavelengths of light that plants need most to pass through, while blocking and generating energy from light rays that are less helpful to plants.
Pollinator habitat planted beneath solar panels is perhaps the most widely used type of agrivoltaics currently, with some states including Illinois and Minnesota offering incentives for pollinator habitat solar projects. Research is progressing on this front as well, with more research needed to evaluate the effectiveness of pollinator habitat, and to what extent it is maintained over multiple years.
InSPIRE sites including in Minnesota are exploring the performance of different seed mixtures, hydrologies and soil conditions. Planting pollinator habitat under and around solar panels is meant to benefit nearby farmland. But the study notes there can be diminishing returns with larger pollinator-habitat projects, as pollinators in the middle of the site may not travel out to surrounding fields.
Research challenges
While continued research is crucial to scaling up and expanding agrivoltaics successfully, the research itself presents many challenges.
Many crops grow in cycles of three years or more, so a long time is needed to measure yields. And soil and water conditions can vary greatly even in one plot, not to mention between plots, making standardized comparisons difficult. Since research will generally be happening on active farms, the scientific process needs to accommodate the needs of farmers.
“Often, research activities must accommodate the realities of farm operations,” the study says. “This might mean harvesting more frequently or on different days than planned based on when crops are ripe, or adjusting activities in anticipation of a coming frost.”
Byron Kominek moved back to his family farm in Colorado after working in international development in Africa. His farm became the site of Jack’s Solar Garden, the nation’s largest agrivoltaic research site. He’s collaborated with researchers and seen both the promise and the greatly varying results of agrivoltaics.
“I’m not a professional researcher — my level of research will be a bit messier than the folks from academia who are far more rigid with what they do,” he noted. “It’s fun to learn from them, and I share back what I’ve heard from other researchers in the field or observations I’ve been making as I spend a considerable amount of time here myself.”
Kominek has been especially focused on herbs like lemon balm, peppermint and sweet grass.
“There’s a drastic difference in the quality of the lemon balm, and how much we can pull off each plant” in rows that get different amounts of sun based on their placement. “It dries out and is smaller, with not as good flavor” growing outside the solar array as opposed to under the panels.
“These [herbs] are perennial plants, so it will take some years to get to their maximal extent,” Kominek said. “We’re playing with getting that anecdotal evidence, but if we had someone who wanted to actually fund us, and spend more time figuring it out — that would be helpful.”
The life of a solar installation is 20 to 30 years, longer than most studies continue, yet conditions and performance could change significantly over this time frame. NREL notes that both field research and modeling should be used to predict agrivoltaic outcomes.
NREL also notes that “projects change developers or operators throughout the solar development and permitting process, meaning that decisions can be made by entities that are not going to be operating the site over the long term. If agrivoltaics projects cannot be built and operated successfully over the long term, then the validity of the research on these systems will be jeopardized and/or made less relevant.”
Researchers emphasize that agrivoltaics may not ultimately be well suited to all agricultural situations, but the potential especially for vulnerable and diverse communities in the U.S. and worldwide should be prioritized, in research and deployment.
“Some people who are more glass-half-empty will say, ‘Isn’t this a limited solution,’” said Barron-Gafford. “But in places like Kenya, Israel, Northern Arizona where tribal nations are, people would greatly benefit from being able to grow food by dialing back that water stress and allowing them to have renewable energy for water pumping. That’s the scale that will help rural and tribal communities; we don’t need to focus all our efforts on large projects.”