New method has promise for accurate, efficient soil carbon estimates

To monitor and sustainably manage SOC stocks under agricultural land use, an accurate way to measure SOC is essential. However, current methods of accurately estimating SOC are resource- and cost-intensive. In their new study, published in Geoderma, Agroecosystem Sustainability Center (ASC) researchers tested a new sampling method in hopes of improving the ability to estimate SOC stocks.

The team's previous research suggested that readily available spatial information in public databases could improve the efficiency of SOC sampling in agricultural fields. This study, led by ASC's Eric Potash, a Research Scientist in the Department of Natural Resource & Environmental Sciences (NRES) at the University of Illinois Urbana-Champaign, tested that hypothesis in eight fields across Illinois and Nebraska.

Measuring SOC is challenging due to its variability. The SOC stock at two locations just a few feet apart can differ significantly. This means that many locations need to be sampled to estimate the total SOC stock, which translates to a lot of work in the lab and in the field.

"Past studies, including one that we did one year ago, proposed ways of reducing the number of samples needed," Potash said. "But it was unknown just how much more efficient those methods were. We put those methods to the test using a new high-quality dataset our research team put together."

The team found that SOC stocks in agricultural fields can be more efficiently measured by using a method called doubly balanced sampling, which accounts for auxiliary information available in elevation maps, satellite images, and previous surveys. Doubly balanced sampling is a modern strategy that improves on the classic method of stratified sampling by selecting locations that are more representative of the field in terms of this auxiliary information.

"Quantifying soil carbon stock through soil sampling is a hard and expensive task, but our approach was found to reduce the number of soil samples needed by a very promising 30 percent," said Kaiyu Guan, project lead and coauthor, Founding Director of ASC, and NRES Associate Professor. "We believe this is a significant advancement for improving soil sampling efficiency and should be promoted in future practices by carbon project developers or researchers."

The work is made possible by unique field-level, high-resolution soil samples collected by scientists from different projects.

"I am glad that our hard work and collected soil sampling data enables the development of this approach," said DoKyoung Lee, another coauthor and a Professor of Crop Sciences at the U of I.

The team has made its methods and data publicly available so that the scientific community can benefit from, and collaborate on, further improving the understanding of SOC.

"I am especially excited that we are publicly sharing the data for this study," Potash said. "I hope that this will foster increased collaboration to accelerate progress on soil carbon research."

In addition to Potash, Guan, and Lee, co-authors on this publication include Andrew Margenot, Crop Sciences Associate Professor and ASC Associate Director; Arvid Boe, Professor of Agronomy, Horticulture & Plant Science at South Dakota State University; Michael Douglass, ASC and Crop Sciences Research Technician; Emily Heaton, Professor of Crop Sciences; Chunhwa Jang, Crop Sciences Postdoctoral Researcher; Virginia Jin, USDA-ARS Research Soil Scientist at University of Nebraska; Nan Li, ASC and Crop Sciences Postdoctoral Research Associate; Rob Mitchell, USDA Research Agronomist and Adjunct Professor of Agronomy at University of Nebraska; Nictor Namoi, ASC and Crop Sciences Graduate Research Assistant ; Marty Schmer, USDA-ARS Research Agronomist at University of Nebraska; Sheng Wang, ASC and NRES Research Assistant Professor; and Colleen Zumpf, Bioenergy and Ecosystem Services Specialist at Argonne National Laboratory.