Peer-reviewed study: Yard Stick's handheld probe predicted 0–30 cm soil carbon stocks with R² of 0.88 on six Illinois farms

January 2025 — A handheld spectral probe from Yard Stick PBC predicted soil organic carbon stocks to 30 centimeters with an R² of 0.88 against lab-analyzed cores on six Illinois farms, according to a paper in Geoderma by Soil Health Institute and Yard Stick scientists. Here's what happened, what it cost, and what went wrong.

Cost and outcome (as of Jan. 29, 2025)

Outcome: 0–30 cm SOC stocks predicted with R² = 0.88, RMSEP 7.8 and bias −0.49 Mg/ha. (Gyawali et al., Geoderma 453, Jan. 2025, via DOAJ record of the paper)
Weak spot: bulk density wasn't predicted accurately (R² = 0.28). (DOAJ record of the paper)
Cost: no per-measurement cost given; the releases say it saves time and money. (Yard Stick and SHI, Jan. 29, 2025)
Who: three of six authors are from Yard Stick; three, including the lead author, from SHI. (Soil Health Institute)

What the study tested

The authors say soil carbon developers need measurements across 100,000 to 1,000,000 hectares that are "accurate, unbiased, inexpensive, and fast." They tested a handheld visible and near-infrared (VisNIR) probe in Midwest Mollisols. (DOAJ record of the paper)

Current methods involve collecting physical samples and shipping them to a lab, the releases note. (Yard Stick and SHI, Jan. 29, 2025) The authors write that VisNIR is widely used for soil properties in labs and in the field, and that recent low-cost spectrometers made handheld devices possible. (DOAJ record of the paper)

They collected probe readings and soil cores from six non-contiguous Illinois farms and used a one-farm hold-out modeling approach, predicting SOC concentration, bulk density, 5-cm SOC stock increments down to 45 cm, and 0–30 cm stocks. (DOAJ record of the paper) The releases describe the probe as recording a "movie" of the soil as it measures reflected light. (Yard Stick and SHI, Jan. 29, 2025)

The releases say the work was supported by ARPA-E, the U.S. Energy Department's advanced research agency. (Yard Stick and SHI, Jan. 29, 2025)

The results

SOC concentration: R² = 0.72, RMSE 0.33%, bias 0.0005%. Five-centimeter stock increments: R² = 0.68. Stocks from 0 to 30 cm: R² = 0.88, bias −0.49 Mg/ha. (DOAJ record of the paper)

Bulk density couldn't be predicted accurately (R² = 0.28). The authors found that, in this loess landscape, stock accuracy was driven by accurate SOC concentration rather than bulk density. (DOAJ record of the paper)

Modeling stocks directly gave lower bias than computing them from concentration and bulk density (0.05 vs. 0.15 Mg/ha). (DOAJ record of the paper) The authors concluded that the probe "shows promise" as a rapid, low-cost tool for Midwestern Mollisols and called for more investment in in-situ spectral libraries. (DOAJ record of the paper)

What the authors said

"This probe presents an inexpensive yet accurate solution for rigorous, measurement-based carbon markets," said lead author Dr. Ayush Joshi Gyawali of SHI. (Yard Stick and SHI, Jan. 29, 2025)

SHI chief scientific officer Cristine Morgan, a co-author, said the data "show that Yard Stick's spectral probe can deliver." (Yard Stick and SHI, Jan. 29, 2025) Co-authors from Yard Stick were Marissa Wiseman, Sarah Coffman and Kevin Meissner; SHI's Jason Ackerson also co-wrote it. (Soil Health Institute)

What went wrong

Bulk density failed. The probe couldn't predict bulk density (R² 0.28), one of the two inputs to a carbon stock. (DOAJ record of the paper)
Narrow test. Six farms in one state and one soil type; the authors limit their conclusion to Midwestern Mollisols. (DOAJ record of the paper)
Commercial ties. Three of the six authors work for Yard Stick, which sells measurement services using the probe. (Soil Health Institute)
Depth wording differs. The releases say the probe accurately measured stocks to 45 cm; the abstract's best headline result is for 0–30 cm. (Yard Stick and SHI, Jan. 29, 2025, DOAJ record of the paper)
What we don't know yet: the cost per sample compared with cores and lab analysis, and performance on other soils.

Our follow-up question for Yard Stick

What does a probe measurement cost per hectare compared with cores, and how does it perform outside Midwestern Mollisols?

Sources

Gyawali, A.J., Wiseman, M., Ackerson, J.P., Coffman, S., Meissner, K., Morgan, C., “Measuring in situ soil carbon stocks: A study using a novel handheld VisNIR probe,” Geoderma 453, January 2025. doi.org
DOAJ record with abstract of the same paper. doaj.org
Yard Stick PBC and Soil Health Institute, “The Soil Health Institute and Yard Stick PBC Jointly Publish On Innovative Soil Carbon Measurement Technology,” Jan. 29, 2025. useyardstick.com
Soil Health Institute, “The Soil Health Institute and Yard Stick PBC Publish on Innovative Soil Carbon Measurement Technology.” soilhealthinstitute.org

Disclosure: [not in copy] Editor placeholder