The ground beneath a solar farm is easy to forget.
It sits in moving shade, crossed by wires, and for much of the industry’s short history it has been mowed short or left nearly bare.
At two small solar sites in southern Minnesota, someone tried something different.
Crews seeded the floor between the panel rows with native prairie grasses and wildflowers.
Then counters walked it, summer after summer, to see what moved in.
What a bee finds that a mowed field does not
A native bee needs three things that a field of row crops rarely offers at once: flowers through the warm season, undisturbed ground to nest in, and shelter from wind. A prairie planting threaded between panel rows can supply all three on ground no plow touches. Most native bees in the Upper Midwest nest in soil or stems rather than in hives, so stable, undisturbed ground matters as much as blossoms do.
Because the floor is not tilled once the panels go in, ground nesting bees can hold their tunnels from one season to the next. And because a prairie mix blooms in sequence rather than all at once, a forager finds something open across much of the summer instead of a single flush and then nothing.
Shade complicates the picture rather than simply helping. Companion fieldwork at these same Minnesota sites found that native plant cover under the arrays was in places lower than in adjacent full sun plots, so the panels act as a constraint on the planting as well as a shelter for it.
What the restored floor ultimately restores is continuity of bloom, the thing continuous row cropping stopped providing.
What showed up between the rows
The headline result was the bees. Native bee abundance rose roughly twentyfold across the monitoring period, the steepest change anywhere in the dataset, and total insect abundance tripled. Every habitat and biodiversity metric the team tracked moved in the same direction.
Flower abundance, the number of flowering plant species and the diversity of insect groups all increased as the planting matured. Beetles, flies, moths, wasps and bees went into the ledger together.
Yet the most numerous groups were not bees at all. Goldenrod soldier beetles and hoverflies dominated the counts, which is roughly what a young prairie looks like when it is still finding its footing.
Two sites, one slow count, and a published result
The Atwater and Eastwood solar sites sit on retired farmland in southern Minnesota, and construction at both was finished in 2017. Atwater covers about 35 acres and carries 4 megawatts; Eastwood covers about 42 acres and carries 5.5 megawatts, for a combined footprint near 77 acres. Native grasses and flowering plants went into the ground between the rows in early 2018.
From August 2018 through August 2022, researchers from two national laboratories working with a native landscape firm and university collaborators logged 358 transect observations at the two sites, returning four times each summer. They detected about 10,900 beneficial insects spanning four orders, and published the work in Environmental Research Letters.
Lead author Lee Walston, an Argonne landscape ecologist, said the study “highlights the relatively rapid insect community responses to habitat restoration at solar energy sites,” and that “if properly sited, habitat friendly solar energy can be a feasible way to safeguard insect populations and can improve the pollination services in adjacent agricultural fields.”
The restoration was young, and the insects did not wait for it to mature.
The fence the bees crossed
Spillover into the neighboring crop was one of the questions the team set out to test, not a surprise. Bee visitation to soybean flowers beside the solar plantings was comparable to visitation in soybeans beside grassland enrolled in the Conservation Reserve Program, and greater than at soybean field interiors or roadsides. The boundary fence turned out not to be much of a boundary for a foraging bee.
The same pattern shows up where other infrastructure meets living systems. Ecologists studying sugar kelp grown between offshore wind turbines have examined whether such structures can become productive ground when something living is deliberately introduced around them.
Even passive wild colonization turns up tenants nobody planned for. Water sampling near North Sea turbine foundations detected five species of shark and ray, including a basking shark in winter, around structures designed for none of them.
What the panels still cannot fix, and what comes next
The Minnesota result carries real limits. It covers two sites and about 77 acres, with a record just long enough to catch a planting establishing itself, and the gains depend on continued management rather than on the panels alone.
Solar leases commonly run for decades, and a prairie community that took years to knit together can be turned under in an afternoon if the land returns to row crops. That fragility sits at the center of any serious policy conversation about what solar ground should look like.
Later work from the same group reports that ecovoltaic solar development can also promote grassland bird communities, which suggests the effect does not stop at insects. The open question now is scale: a national solar buildout will eventually cover far more ground than these two fields. What lives under that canopy is still largely unwritten.
Read the whole thing?
Get the week's signal, not the noise
Our sharpest reporting on energy, climate and nature — free, once a week.
