Species richness impacts total soil carbon more than 19 years of CO2 enrichment or soil nitrogen addition

Background/Question/Methods

Whether the terrestrial biosphere will continue to act as a net carbon (C) sink in the face of multiple global changes is questionable. Part of this uncertainty is related to whether increases in plant C fixation commonly observed under elevated CO₂ (eCO₂) in short-term studies will translate into long-term (i.e., decades or more) C storage and whether this depends on soil nitrogen (N) supply and/or species richness. In this study, we investigated how manipulations of CO₂ (ambient and +180 ppm), soil N supply (ambient and +4 g N m^-2 y^-1), and plant species richness (1, 4, 9, or 16 species) have influenced total ecosystem (soil + plant) C storage over 19 years (1998-2016) in a Minnesota grassland with sandy, nutrient-poor soils. To do so, we used long-term biomass, root ingrowth, and plant tissue chemistry data coupled with measurements of total soil C (to a depth of 60 cm) in a free-air CO₂ enrichment experiment (BioCON).

Results/Conclusions

Species richness impacted total ecosystem C more than CO₂ or N treatments over 19 years. On average, increasing species richness from 1 to 16 enhanced total ecosystem C storage by 1111 g C m^-2 y^-1 (+24%), whereas enriched N modestly increased total ecosystem C by 181 g C m^-2 y^-1 (+4%) and eCO₂ tended to increase total ecosystem C by 129 g C m^-2 y^-1 (+3%). CO₂ effects on total ecosystem C were greater for particular combinations of species richness with year or N level, but were neither consistent nor as strong as the main effect of species richness. Patterns in total ecosystem C were driven by the soil C pool, which made up 90% of total ecosystem C, whereas total plant C contributed the remaining 10%. Though eCO₂ stimulated total net primary productivity by 16% on average, enhanced C losses via soil respiration under eCO₂ likely offset any increases in plant-derived soil C inputs. Our results suggest that increases in plant productivity under eCO₂ may not translate into long-term C storage if the additional C released into soils is labile and decomposes quickly. This calls into question whether eCO₂ will increase the soil C sink in these grassland ecosystems and thereby help to slow climate change. Overall, our results suggest that species richness may influence grassland C storage more than rising CO₂ or soil N supply.