Elevated CO2 stimulates grassland soil respiration by increasing carbon inputs rather than by enhancing soil moisture

It is not clear whether the consistent positive effect of elevated CO2 on soil respiration (soil carbon flux, SCF) results from increased plant and microbial activity due to (i) greater C availability through CO2-induced increases in C inputs or (ii) enhanced soil moisture via CO2-induced declines in stomatal conductance and plant water use. Global changes such as biodiversity loss or nitrogen (N) deposition may also affect these drivers, interacting with CO2 to affect SCF. To determine the effects of these factors on SCF and elucidate the mechanism(s) behind the effect of elevated CO2 on SCF, we measured SCF and soil moisture throughout a growing season in the Biodiversity, CO2, and N (BioCON) experiment. Increasing diversity and N caused small declines in soil moisture. Diversity had inconsistent small effects on SCF through its effects on abiotic conditions, while N had a small positive effect that was unrelated to soil moisture. Elevated CO2 had large consistent effects, increasing soil moisture by 26% and SCF by 45%. However, CO2-induced changes in soil moisture were weak drivers of SCF: CO2 effects on SCF and soil moisture were uncorrelated, CO2 effect size did not change with soil moisture, within-day CO2 effects via soil moisture were neutral or weakly negative, and the estimated effect of increased C availability was 14 times larger than that of increased soil moisture. Combined with previous BioCON results indicating elevated CO2 increases C availability to plants and microbes, our results suggest that increased SCF is driven by CO2-induced increases in substrate availability. Our results provide further support for increased rates of belowground C cycling at elevated CO2 and evidence that, unlike the response of productivity to elevated CO2 in BioCON, the response of SCF is not strongly N limited. Thus, N limited grasslands are unlikely to act as a N sink under elevated CO2.