Identifying the contribution of deep roots to subsoil organic carbon stocks under long-term switchgrass (Panicum virgatum) cultivation

Background/Question/Methods

Subsoils hold a globally significant quantity of organic C that exchanges relatively slowly with the atmosphere. Atmospheric C can be introduced to subsoils via deep root systems, offsetting anthropogenic C emissions. Quantifying increases in soil organic carbon following establishment of deeply rooted crops is challenging because initial soil C data are often unavailable. In these cases, changes must be inferred by comparing paired plots, which may vary in initial C content. We hypothesized that increased root inputs under switchgrass drive proportionate increases in soil organic C. These changes might be detectable in paired plots over relatively short timescales (10 years) using sensitive isotopic tracers—particularly ¹⁴C—to supplement observations of C stocks. We sampled replicate soil profiles to a depth of 1-3 m at three sites in Oklahoma where switchgrass (Panicum virgatum) had been grown for 10 years in marginal soils. We obtained paired samples from nearby fields that had been left fallow or cultivated with shallow rooted annual crops. We measured root biomass, total C, ¹⁴C, ¹³C, and other soil properties in three profiles in each field and then used a simple one-pool model of ¹⁴C incorporation to estimate the proportion of recently fixed C under switchgrass.

Results/Conclusions

The subsoil (30-120 cm deep) C inventory under switchgrass exceeded the inventory in neighboring fields by 1.6 kg C m^-2 (SD = 0.1) at a site with sandy loam soils and 0.6 kg C m^-2 (SD = 0.3) at a site featuring loam soils. At a third site with clay soils we did not observe a significant difference. Using ¹⁴C we estimated that C introduced after switchgrass establishment comprised 30% of the C stock under switchgrass at the sandy loam site, 21% at the loam site, and 6% at the clay site. These values imply net differences of 0.9 kg C m^-2 (SD = 0.2), 0.6 kg C m^-2 (SD = 0.2), and 0.2 kg C m^-2 (SD = 0.2) at each site respectively. The radiocarbon data thus confirmed that recent plant inputs were at least partly responsible for differences in C stocks at two sites. These results show that ¹⁴C can be used to interpret soil C stock differences between paired plots when initial data are lacking. The rates of C increase we observed suggest that deeply rooted crops can significantly increase subsoil organic C—but also illustrate that this effect may vary with soil texture.