Nitrogen-fixing tropical trees sequester soil carbon

Abstract

The objective of my dissertation research was to determine how nitrogen-fixing tree species (N-fixers) increase total soil carbon storage relative to non-N-fixers (Eucalyptus) by examining the effects of N-fixers on the decomposition of old soil carbon and the accumulation of new soil carbon. Using stable carbon isotope techniques, I tracked the loss of the old soil organic carbon from the previous C4 land use (SOC4) and the gain of new soil organic carbon from the C3, N-fixer and Eucalyptus plantations (SOC3). I tested four hypotheses. H1: Increased soil N from N-fixation will increase the decomposition of older SOC4 under the N-fixers relative to the non-N-fixers. H2: Greater total soil carbon accretion under N-fixers will result from greater accretion of N-fixer SOC3 relative to the non-N-fixers. H3: Greater accretion of N-fixer SOC3 will result from increased litter inputs rather than increased recalcitrance of N-fixer litter. H4: Soil carbon pools under mixtures of N-fixers and non-N-fixers will follow linear trends with increasing proportions of the N-fixer. I found greater retention of the old soil carbon and greater accretion of new soil carbon under the N-fixers relative to Eucalyptus. The N-fixers sequestered 0.11 kg m-2 yr-1 more total soil carbon with 0.06 kg m-2 yr-1 of that from greater retention of old carbon and 0.05 kg m-2 yr-1 from greater accretion of new carbon relative to Eucalyptus. The annual differences between the N-fixers and Eucalyptus for the total, old, and new soil carbon pools corresponded with nitrogen accretion differences between the N-fixers and Eucalyptus. Eight-month soil incubations of the surface soils from all species at three of the sites supported the hypothesis that greater N-fixer litter inputs are necessary for the greater SOC3 accretion under the N-fixers compared with the non-N-fixers. The results from the incubations cannot rule out the possibility of greater recalcitrance of some portion of the N-fixer litter inputs. Mixed species plots of Albizia and Eucalyptus showed that there is a neutral interaction for species mixtures on soil organic carbon pools. That is, knowing the pure species endpoints of old soil carbon retention and new soil carbon accretion is enough information to extrapolate linearly to mixtures.