Carbon dioxide evolution in runoff from simulated rainfall on long-term no-till and plowed soils in southwestern Ohio

Water erosion results in the mobilization and depletion of soil organic carbon (SOC), but studies providing direct experimental evidence of eroded C mineralization and its linkage to the global C cycle are lacking. A study was conducted to determine the mineralization of SOC in runoff from a southwestern Ohio Crosby soil (fine, mixed, mesic Aeric Ochraqualf) that had been under no-till (NT), chisel till (CT) and moldboard plow (MP) for 38 years. To simulate present and future soil erosion conditions, the 0-3 and 5-8 cm soil layers from triplicate soil blocks extracted from each tillage practice were used. Soil layers were transferred to runoff trays and simulated rainfall (30±5 mm h-1) was applied for 1 h. Runoff was collected at 20, 40 and 60 min following initiation of rainfall event. Carbon dioxide production was monitored for 100 days in runoff samples incubated without and with N (0.15-0.20 g NH4-N kg-1 sediment) amendment to simulate situations where water erosion coincides with high soil mineral N availability. The data show that soil layer depth and time of runoff sampling had no significant effect on any of the parameters considered. Sediment delivery (g sediment m-2 h-1) was significantly lower in the NT (17.4) than in the other tillage (CT: 48.9, MP: 34.1) practices; it was inversely related to degree of soil aggregation and sediment C content. The average of C export ranged between 0.6 and 1.1 g C m-2 h-1, and was highest in the CT soil. Mineralization of runoff C followed a first-order kinetics and proceeded at rates significantly higher in NT and N-amended runoff. Nearly half of the total runoff C mineralization recorded during the entire incubation occurred during the first 20 days suggesting that under field conditions, most of the erosion-induced CO2 production will occur in the days immediately following rainfall events. Potentially mineralizable carbon (C₀) in runoff was significantly greater with NT (13.5 g C kg-1 sediment) than with the other tillage practices, and was positively correlated with sediment C content (R2=0.76) and cumulative C mineralized (R2=0.77-0.97). The data presented showed that 29-35 and 33-46% of the C exported in runoff was mineralized in 100 days without and with N amendment, respectively. These results are the first quantitative evidence that a significant fraction of the SOC released through water erosion is mineralizable and, therefore represents an additional source of atmospheric CO2.