Soil organic matter turnover over decadal to millennial timescales

Abstract

The sequestration of carbon dioxide (CO2) into soils and subsequent stabilisation of soil organic matter (SOM) represents a potential mechanism for climate change mitigation, while benefiting soil function, ecosystem services and agriculture. However, forecasting carbon (C) fluxes between atmosphere and pedosphere is limited by a lack of certainty around temporal and interlinked biogeochemical processes of SOM stabilisation and destabilisation (i.e. turnover). Improving mechanistic hypotheses that underpin conceptual models is essential to improve SOM model accuracy, especially new-generation models that incorporate soil microbial processes. This thesis thus focusses on SOM turnover rates and processes across timescales and ecosystems to address questions relating to mechanisms and scale. The overarching objective of this thesis was to gain new insights using archived soil samples previously used to establish concepts of SOM turnover with focus on C and nitrogen (N), complemented by recently collected soil samples and modern analytics. The research uses decadal to millennial timescales of soil disturbance and soil development to i) inform mechanistic understanding of the timescales at which SOM stabilisation and destabilisation occur, and ii) improve the ability to measure long-term processes by validating SOM stabilisation theory with short-term observations.Four chapters investigate temporal aspects of SOM processes in contrasting subtropical soil chronosequences, the decadal “Dalal series” under agriculture, and the millennial “Cooloola series” under natural vegetation. The first chapter developed the use of 15N natural abundance to trace SOM mineralisation and N loss in the three-decade old “Dalal series” of land use change from natural forest to cropping; showing that nitrification and subsequent plant uptake is the main contributor to nutrient losses alongside C loss during long-term cropping. The second chapter explored long-term archived soils as a tool to study short-term SOM turnover processes; discovering contrasting results depending on soil texture, with SOM mineralisation functions being less preserved during storage in coarse-textured soils than heavy-clay soils, which were unaffected. The third chapter identified that, based on natural 14C and ramped-pyrolysis techniques, long-term SOM stabilisation mechanisms along a 500,000-year sand dune chronosequence were attributed to a significant contribution of fire-derived carbon in deep subsoils. Using short-term incubations with 13C-labelled glucose addition, the fourth chapter tested for the vulnerability of this ancient SOM to destabilisation by fresh substrate supply, but identified physio-spatial limitations on microbial movement and access to substrate in the sand that limited the capacity for “priming effects” to occur. Combined, this thesis provides new insights to areas of SOM turnover required for developing a new generation of SOM models that would help better prediction of SOM dynamics into the future.