Simulation of soil organic carbon dynamics under different land use and crop management practices

Abstract

Soils are the largest reservoir of terrestrial carbon (C) (~1500 Pg organic C in top one meter) and any change in this large soil organic carbon (SOC) stock due to change in land use, management practice or climate change may have significant and long-lived effects on the global C cycle. The SOC turnover models can project the change in SOC storage and turnover rate under future scenarios of land use, management practice, technological improvement and climate change, and investigate hypotheses that are beyond the feasibility of experimental work. In Australia, there is a greater need for modelling such scenarios across a wider range of agro-ecological regions and land uses. In this thesis, SOC dynamics under different land uses and crop management practices in the northern plains and slopes of New South Wales (NSW) was explored using the Rothamsted carbon model (RothC). The specific objectives were (i) to determine turnover times of SOC fractions, separated by a combination of physical and chemical methods, in contrasting land use systems by natural 14C abundance, (ii) to examine whether a combination of physical and chemical laboratory methods of SOC fractionation has the potential to quantify different SOC pools, as required by RothC, (iii) to evaluate the performance of RothC in simulating the effect of land use change (LUC) on SOC dynamics, particularly following a proposed land use change from native vegetation to cropping, using paired-site data sets, (iv) to examine whether initializing RothC with measured SOC pools, rather than using default model equilibrium pools, improves RothC performance in prediction of LUC effects on SOC dynamics, (v) to evaluate RothC performance in simulation of SOC dynamics under a cotton based cropping system on an irrigated Vertosol using long-term field experiment data, (vi) to explore different scenarios of SOC dynamics under different cotton based cropping systems in irrigated Vertosol, (vii) to test whether RothC projections of grassland SOC under climate change were sensitive to the model initialisation method, and (viii) to projected climate change impacts on grassland SOC with three different global climate models (GCMs) forced with four different climate scenarios for the time period 2008-2100, using RothC in the northern slopes and plains of NSW, Australia.