Satellites show Arkansas farmers planted far more cover crops without subsidies than with them, and a U.S. study finds a yield cost

October 2025 — Farmers in eastern Arkansas planted about 11 times as many winter cover-crop acres on their own as government cost-share programs paid for between 2013 and 2019, according to a satellite study in PLOS One. Here's what happened, what it cost, and what went wrong.

Cost and outcome (as of Oct. 21, 2025)

Outcome: 4,486,197 acre-years of voluntary cover crops vs 402,839 cost-shared; that's about 11 to 1 by our arithmetic. (Ahmed et al., PLOS One, Oct. 21, 2025)
Subsidized growth: cost-shared acres rose from 9,617 in 2013 to 201,361 in 2019. (Ahmed et al., PLOS One, Oct. 21, 2025)
Cost to farmers: across more than 100,000 U.S. fields, average yields fell ~3% for corn and ~2% for soybean on fields that adopted cover crops. (Lobell et al., Nature Sustainability, July 28, 2025 (abstract))
Carbon: estimated, not measured: 355,346 to 474,631 tons C a year (+34%). (Ahmed et al., PLOS One, Oct. 21, 2025)

What the Arkansas study found

University of Arkansas researchers trained a random-forest model in Google Earth Engine on Landsat 8 images and NRCS records of where subsidized cover crops were planted. They then mapped cover crops across the Arkansas part of the Mississippi Alluvial Plain, all or parts of 27 counties. (Ahmed et al., PLOS One, Oct. 21, 2025)

Cost-shared cover crop acres were lowest in 2013, at 9,617, and stayed low until 2017, at 24,386. They then jumped to 94,464 in 2018 and 201,361 in 2019. Voluntary cover-crop acres were 667,655 in 2013 and grew 5% (35,608 acres) by 2019. (Ahmed et al., PLOS One, Oct. 21, 2025)

Over all counties and years, voluntary adoption totaled 4,486,197 acre-years, against 402,839 that got government cost-sharing. The two were positively correlated (Pearson's R = 0.46). The authors say this suggests "spillover effects," with farmers in the programs inspiring others to adopt. (Ahmed et al., PLOS One, Oct. 21, 2025)

In 2013, soybean followed by soybean without a cover crop was the most common sequence, covering 21% of cropland. The abstract says soybean/soybean rotations had the strongest link to more winter cover. (Ahmed et al., PLOS One, Oct. 21, 2025)

Estimated carbon

Applying an assumed rate of 0.52 tons of carbon per acre per year, the authors estimated sequestration rose from about 355,346 tons in 2013 to 474,631 tons in 2019, up 34%. Most of the gain came from subsidized acres, which went from 5,046 tons to 105,648. (Ahmed et al., PLOS One, Oct. 21, 2025)

A national view: the yield cost

A Stanford-led team used satellite observations from more than 100,000 U.S. fields, half of which had recently adopted cover crops. They found average yields fell about 3% for corn and 2% for soybean. Planting was delayed about four days for corn and 2.5 days for soybean. (Lobell et al., Nature Sustainability, July 28, 2025 (abstract))

But in the wet spring of 2019, cover-crop fields were only half as likely to have prevented planting. "Cover cropping appears to reduce important aspects of farmer risk in wet conditions but increase them in dry conditions," the authors wrote. Eliminating planting delays would cut the yield penalty by about 50% for corn and 90% for soybean, they found. (Lobell et al., Nature Sustainability, July 28, 2025 (abstract))

The authors are David B. Lobell, Stefania Di Tommaso, Qu Zhou, Yuchi Ma, James Specht and Kaiyu Guan. NASA Acres and the Keck Foundation funded the work. (University of Illinois Experts record)

What went wrong

Modeled carbon. The Arkansas carbon totals assume a fixed sequestration rate and weren't measured in soil. (Ahmed et al., PLOS One, Oct. 21, 2025)
Bad GPS points. Some NRCS records logged the farmer's home or shop rather than the field; the team filtered them out with a threshold rule. (Ahmed et al., PLOS One, Oct. 21, 2025)
Yield penalty. U.S. adopters in the Stanford-led study gave up about 3% of corn yield on average, partly because of late planting. (Lobell et al., Nature Sustainability, July 28, 2025 (abstract))
What we don't know yet: whether Arkansas's voluntary adopters face the same yield cost, and how much soil carbon is really gained.

Our follow-up question for University of Arkansas and Stanford

Have the Arkansas maps been checked against field-measured soil carbon, and do yield penalties show up in the Mississippi Alluvial Plain?

Sources

Ahmed Z., Connor L., Brye K., Green V.S., Popp M., Shew A. et al., “Satellite Remote Sensing Reveals Voluntary Cover-Crop Adoption and Crop-Rotation Hotspots in the Mississippi Alluvial Plain,” PLOS One 20(10), Oct. 21, 2025. journals.plos.org
Lobell D.B. et al., “The mixed effects of recent cover crop adoption on US cropland productivity,” Nature Sustainability 8, 1004–1012, published online July 28, 2025 (abstract). nature.com
University of Illinois Urbana-Champaign, Illinois Experts record for the Lobell et al. paper (authors and funding). experts.illinois.edu

Disclosure: [not in copy] Editor placeholder