Comparison of isotope methods for partitioning methane production and soil C priming effects during anaerobic decomposition of rice residue in soil

• 13C natural abundance methods overestimate the contribution of plant residue for CH4 production.
• Use of 13C-enriched plant residue was the most reliable approach to estimate sources of soil C for CH4 production.
• Priming effect on CH4 production is regulated by the concentration of electron acceptors.

Management of rice residue in paddies often results in higher methane (CH4) emissions, but the carbon (C) sources contributing to higher emissions are not well characterized. We examined the relative contribution of soil and residue C sources for CH4 production in three different soils using both 13C-labeled or unlabeled (natural abundance) rice and maize residues. The results were compared to a recently proposed natural 13C abundance soil C partitioning method (CNAM) (Conrad et al. 2012b). As anticipated, residue addition increased CH4 production, with the timing and magnitude of emission varying across soils, likely due to differences in soil C and amount of electron acceptors. The CH4 derived from residue-C, when estimated using the typical 13C natural abundance method yielded either negative values or those exceeding 100%. Similarly, the CNAM frequently overestimated the contribution of residue C compared to the 13C enriched residue method (ERM). The over estimation is likely due to C isotopic fractionation during the utilization of rice and maize residues in the different soils. Substantial C isotope fractionation would bias the fundamental assumption of the CNAM. The ERM produced a consistent partitioning result of 50–70% of CH4 derived from residue additions. When typical minimum, average, and maximum values of natural 13C fractionation factors were applied to the ERM the partitioning outcome was unaffected. The priming effect of residue addition on CH4 production was regulated by electron acceptor concentrations in various soils and was positively correlated to the degree of residue decomposition. The priming effect of residue, therefore, depends both on soil and residue properties, with the ERM showing to be more accurate to account for the partitioning of CH4 from various C pools compared to CNAM.