Origin and composition of organic N forms and related stabilization mechanisms in a soil chronosequence
Final Report Abstract
Organic N bound to minerals is currently disregarded in models related to N turnover in soils. However, mineral-bound ON might represent an importent reservoir of N being available to microorganisms and plants. This project, therefore, aimed at the mechanisms of ON stabilization in native rainforest soils with special emphasis on the mineralogical control. It bases on the “long substrate aging gradient” chronosequence at Hawaiian Islands that covers substrate ages from 0.3 to 4100 kyrs at otherwise fairly constant environmental variables. This allowed to track changes in ON accumulation and stabilization as related to mineral composition, which changes during soil formation. Using density fractionation, we showed that the majority of OC and ON in the mineralogically different Hawaiian soils is associated with minerals (OC: 79 ± 4%; ON: 88 ± 4% n = 12). Both, the accumulation of OC and ON depended strongly on the prevailing mineralogical regime. More OC and ON was present at sites rich in poorly crystalline minerals (Fe and Al hydroxides, allophane) while less was present in youngest and oldest soils being dominated by crystalline minerals. The differences in OC and ON concentrations among the sites were substantial (e.g., differing by factor 15). This indicates that natural gains and losses of OC and ON in soils due to long-term soil weathering are immense and exceed by far changes due to anthropogenic land use changes. XPS and NEXAFS spectroscopy revealed that, irrespective of the mineralogical composition, peptide N is the dominating form of ON in mineral-organic associations and that aromatic N moieties (considered to be refractory N) is of minor importance. Mineral-associated ON (amino acids, amino sugars) closely matched trends in OC accumulation, suggesting that most ON binds to minerals as part of sorbing OM. Mineral-bound ON in A horizons, being largest at sites with poorly crystalline minerals (20−400 kyrs), is primarily of bacterial origin and turns over quickly irrespective of the mineral assemblage. In deeper soil, the smaller portion of hydrolyzable ON, with an even larger microbial contribution, went along with the presence of older proteins. Subsoil horizons rich in poorly crystalline minerals more effectively stabilize acidic protein structures. The preferential association of acidic amino acids with reactive minerals at pH values typical for the Hawaiian soils has also been confirmed in simple batch experiments. In separate sorption and incubation experiments we showed that even intrinsically labile microbial-derived substances (EPS) can effectively be stabilized by interaction with minerals or multivalent metals. These parallel experiments helped explaining the large amount of microbial derived ON in sorbed OM in the Hawaiian mineral soils. Changes in the mineralogical composition fostered changes in the chemical composition of sorbed OM with preferential enrichment of lignin-derived phenols at sites rich in poorly crystalline minerals and crystalline Fe and Al oxides (>20 kyr). A larger C-normalized fraction of N-related biomarkers was present at the yougest soil (0.3 kyr) composed of primary silicate minerals. The differential accumulation of the tested biomarker suggests that the underlying mineral compositions controls the selective retention of OM (sorptive fractionation). This fractionation was most recognizable in the A horizons where sorption sites at mineral surfaces are already limited. A large number of proteins could be identified by mass-spectrometry based proteomics with the abundance being related to the OM content, i.e., being lowest at sites containing little reactive minerals (primary silicates or gibbsite and kaolinite) and highest in soils containing poorly crystalline minerals (allophane, Fe and Al hydroxides). In summary, ON in mineral–organic associations is the dominating form of N in the investigated tropical rainforest ecosystem. Mineral-bound ON is actively involved in the cycling of N in surface horizons, hence, represents a bioavailable N fraction. The active participation of mineral-bound ON in A horizons suggests that mineral-bound OM (incl. ON) cannot be considered as passive, refractory OM. We therefore propose that the N dynamics at the ecosystem level is not only affected by microbial–plant interactions but includes mineral–organic feedbacks. Future research should be directed to the effect of minerals on the rates of ON transformation and the respective N transfer to microorganisms and plants.
Publications
- 2009. Biogeochemistry of mineral-organic associations across a long-term mineralogical soil gradient (0.3-4100 kyr), Hawaiian Islands. Geochimica et Cosmochimica Acta 73, 234-260
Mikutta R., Schaumann G.E., Gildemeister D., Bonneville S., Kramer M.G., Chadwick O.A., Guggenberger G.
- 2009. Biogeochemistry of mineral-organic associations across a mineralogical soil gradient (0.3-4100 kyr), Hawaii. Soil Science Zvieri 25.2.2009, University of Zürich
Mikutta R.
- 2010. Interaction of organic matter with minerals - ultimate stabilization in soils? Geochimica et Cosmochimica Acta, 74, A489
Kalbitz K., Mikutta R.
- 2010. Mineral control on organic carbon and organic nitrogen biochemistry. Geochimica et Cosmochimica Acta 74, A361
Guggenberger G., Mikutta R., Chadwick O.A., Chorover J., Klaus K., Kramer M.G., Vollmer A.
- 2010. Mineralogical impact on organic nitrogen across a long-term soil chronosequence (0.3–4100 kyr). Geochimica et Cosmochimica Acta 74, 2142–2164
Mikutta R., Chadwick, O.A., Chorover J., Dörr N., Kaiser K., Kramer M.G., Vollmer A., Guggenberger G.
