Total soil C and N sequestration in a grassland following 10 years of free air CO2 enrichment

Soil C sequestration may mitigate rising levels of atmospheric CO2. However, it has yet to be determined whether net soil C sequestration occurs in N-rich grasslands exposed to long-term elevated CO2. This study examined whether N-fertilized grasslands exposed to elevated CO2 sequestered additional C. For 10 years, Lolium perenne, Trifolium repens, and the mixture of L. perenne/T. repens grasslands were exposed to ambient and elevated CO2 concentrations (35 and 60 Pa pCO2). The applied CO2 was depleted in d C and the grasslands received low (140 kg ha ) and high (560 kg ha ) rates of N-labeled fertilizer. Annually collected soil samples from the top 10 cm of the grassland soils allowed us to follow the sequestration of new C in the surface soil layer. For the first time, we were able to collect dual-labeled soil samples to a depth of 75 cm after 10 years of elevated CO2 and determine the total amount of new soil C and N sequestered in the whole soil profile. Elevated CO2, N-fertilization rate, and species had no significant effect on total soil C. On average 9.4 Mg new C ha 1 was sequestered, which corresponds to 26.5% of the total C. The mean residence time of the C present in the 0–10 cm soil depth was calculated at 4.6 1.5 and 3.1 1.1 years for L. perenne and T. repens soil, respectively. After 10 years, total soil N and C in the 0–75 cm soil depth was unaffected by CO2 concentration, Nfertilization rate and plant species. The total amount of N-fertilizer sequestered in the 0–75 cm soil depth was also unaffected by CO2 concentration, but significantly more N was sequestered in the L. perenne compared with the T. repens swards: 620 vs. 452 kg ha 1 at the high rate and 234 vs. 133 kg ha 1 at the low rate of N fertilization. Intermediate values of N recovery were found in the mixture. The fertilizer derived N amounted to 2.8% of total N for the low rate and increased to 8.6% for the high rate of N application. On average, 13.9% of the applied N-fertilizer was recovered in the 0–75 cm soil depth in soil organic matter in the L. perenne sward, whereas 8.8% was recovered under the T. repens swards, indicating that the N2-fixing T. repens system was less effective in sequestering applied N than the non-N2-fixing L. perenne system. Prolonged elevated CO2 did not lead to an increase in whole soil profile C and N in these fertilized pastures. The potential use of fertilized and regular cut pastures as a net soil C sink under longterm elevated CO2 appears to be limited and will likely not significantly contribute to the mitigation of anthropogenic C emissions.