Effect of landuse on organic matter stabilized in organomineral complexes: A study combining density fractionation, mineralogy and delta C-13

Landuse changes for the purposes of cultivation often destabilise a substantial part of the initially stabilized organic matter (OM) in surface soils. However, the mechanisms of OM destabilisation are poorly documented, particularly with respect to organomineral complexes. The aim of this study was to characterize and quantify the effects of landuse on various OM pools, while focusing particularly on mineral-bound OM in a ferralsol. Four different parameters were assessed: the proportion of mineral-bound OM, the nature of the minerals (poorly crystalline aluminosilicates, gibbsite, halloysite and iron oxides) contributing to organomineral complexes, landuse, and soil depth (surface horizons versus non-tilled deeper horizons). The study site had a field with C4 vegetation that had been cultivated for 186 years and compared to an uncultivated reference plot with C3 vegetation. Organomineral complexes were separated by densimetric fractionation and characterized by mineralogical and carbon isotopic methods. Depending on the considered horizon, 58 to 80% of the OM was stabilized through organomineral complexation. Chemosorption of organic compounds at the surface of mineral phases was thus found to be a major stabilisation process in the ferralsol. Although cultivation significantly affected OM pools that were not bound to minerals (particulate non-occluded and particulate occluded OM), these pools represented a low proportion (similar to 5%) of the carbon budget variations in the profile. Most variations in carbon stocks within the profile were attributed to organic matter located in organomineral complexes. Several potential processes were highlighted on the organomineral complex scale: (i) cultivation may have modified the physicochemical stability of mineral phases (involving dissolution and/or formation); (ii) cultivation could have led to vertical migration of organomineral complexes and their accumulation deeper in the profile; (iii) cultivation systematically modified the amount of organic compounds linked to minerals, with a decrease in topsoil and an increase in deep soil. In further studies to determine the impact of cultivation on soil carbon stocks, it would thus be interesting to continue assessing mechanisms that control the dynamics of mineral-bound OM while also investigating soil layers to a depth of at least 1 m. (C) 2009 Elsevier B.V. All rights reserved