Background

As interest in the voluntary soil carbon market surges, carbon registries have been developing new soil carbon measurement, reporting, and verification (MRV) protocols. These protocols are inconsistent in their approaches to measuring soil organic carbon (SOC). Two areas of concern include the type of SOC stock accounting method (fixed-depth (FD) vs. equivalent soil mass (ESM)) and sampling depth requirement. Despite evidence that fixed-depth measurements can result in error because of changes in soil bulk density and that sampling to 30 cm neglects a significant portion of the soil profile’s SOC stock, most MRV protocols do not specify which sampling method to use and only require sampling to 30 cm. Using data from UC Davis’s Century Experiment (“Century”) and UW Madison’s Wisconsin Integrated Cropping Systems Trial (WICST), we quantify differences in SOC stock changes estimated by FD and ESM over 20 years, investigate how sampling at-depth (> 30 cm) affects SOC stock change estimates, and estimate how crediting outcomes taking an empirical sampling-only crediting approach differ when stocks are calculated using ESM or FD at different depths.

Results

We find that FD and ESM estimates of stock change can differ by over 100 percent and that, as expected, much of this difference is associated with changes in bulk density in surface soils (e.g., r = 0.90 for Century maize treatments). This led to substantial differences in crediting outcomes between ESM and FD-based stocks, although many treatments did not receive credits due to declines in SOC stocks over time. While increased variability of soils at depth makes it challenging to accurately quantify stocks across the profile, sampling to 60 cm can capture changes in bulk density, potential SOC redistribution, and a larger proportion of the overall SOC stock.

Conclusions

ESM accounting and sampling to 60 cm (using multiple depth increments) should be considered best practice when quantifying change in SOC stocks in annual, row crop agroecosystems. For carbon markets, the cost of achieving an accurate estimate of SOC stocks that reflect management impacts on soils at-depth should be reflected in the price of carbon credits.

This data is comprised of two different datasets. The first dataset comes from University of California, Davis’ Century Experiment at the Russel Ranch Sustainable Agriculture Facility (38°32′24″N, 121°52′12″W). The second dataset comes from UW Madison’s Wisconsin Integrated Cropping Systems Trial (WICST) at the UW Madison Agricultural Research Station in Arlington, WI (43°18″N, 89°20″W). Data were previously published in Sanford et al. (2012) (https://doi.org/10.1016/j.agee.2012.08.011) and Tautges et al. (2019) (https://doi.org/10.1111/gcb.14762). To explore management impacts on soil carbon accrual over time. See these publications for details on data collection.

All datasets were cleaned to remove any missing values and organized for input into the SimpleESM R function (https://github.com/fabienferchaud/SimpleESM). SimpleESM requires an excel input document with bulk density (g cm^-3) and carbon concentration (g kg^-1) data. Bulk density ("BD")and carbon concentration ("Concentrations") are recorded on separate sheets. Each sheet contains a campaign variable (year in which the sample was taken), treatment (ID of experimental treatment), block (ID of experimental block),  plot (ID of sampled plot), point (ID of sampled point), and the lower depth and upper depth (in centimeters) of the sample.