Biogeochemical coupling and microbial regulation of soil carbon and nitrogen cycles in grasslands : A thesis submitted in partial fulfilment of the requirements for the Degree of Doctor of Philosophy at Lincoln University

In New Zealand, a dramatic change to grasslands throughout the last two decades has been widespread conversion from dryland sheep farming to irrigated dairy farming. The detrimental impact of intensive dairy farming on groundwater quality requires new strategies to reduce carbon (C) and nitrogen (N) losses from grazed grasslands. Developing these strategies requires thorough understanding of the processes that regulate the coupling of the biogeochemical cycles of soil C and N. The objective of this work was to investigate the biogeochemical coupling of C and N cycles in grassland soils. To address this objective, two microcosm experiments in controlled conditions were undertaken and the findings were used to interpret observations made in a third experiment at a long-term field site. Carbon and nitrogen inputs to soil were manipulated in the microcosm experiments using different plant species and the addition of N, and long-term treatments at the field site consisted of mowing frequency and biomass retention or removal. The objective of the first microcosm experiment was to determine the role of enhanced root-derived C availability on soil nitrification activity for five different grassland species: Cichorium intybus (chicory), Lolium perenne (perennial ryegrass), Plantago lanceolata (ribwort plantain), Raphanus raphanistrum (wild radish), and R. sativus (cultivated radish). These species were grown under controlled conditions for nine weeks and N was added at a low (no urea-N) or a high rate (550 kg urea-N ha-1). Compared to the soils with low N addition, the high N addition rate resulted in an increase in water-extractable C concentrations and a decrease in potential nitrification activity. This suggests that increased C availability for microbial uptake may have stimulated microbial N immobilisation, resulting in reduced nitrification. In the second microcosm study, the objective was to investigate the effects of increasing amounts of N addition on ecosystem C balance, C rhizodeposition, and the regulation of soil functional processes by changes in soil microbial community composition. Lolium perenne and P. lanceolata were grown for seven to eight weeks under controlled conditions and treated with increasing amounts of N (220, 300, 450, and 750 kg N ha-1). A 13CO2 pulse-labelling approach was used to trace photo-assimilated C through the plant-soil-microbe system. Plant C and N uptake and C rhizodeposition increased with increasing N addition, with the greater amounts observed for P. lanceolata than those for L. perenne. There were also plant species-specific differences in the soil microbial community composition and microbial uptake of rhizodeposited C. Plant species-specific variation in microbial uptake of rhizodeposited C changed with increasing N addition, suggesting that microbial processing of rhizodeposited C from different plant species depends on N availability. Although the microbial community composition and the uptake of rhizodeposited C were closely related with soil respiration rates, there was no significant effect on soil N transformation rates. These findings suggest a decoupling of soil C and N cycles when N availability exceeds plant N uptake and highlights the important role of the soil microbial community composition in regulating soil C cycling. Using an established long-term (>25 years) field experiment, the concept of ecological stoichiometry was used to link soil biogeochemical processes with microbial cycling of C, N, and P. The objective was to investigate relationships between microbial elemental limitation and soil organic matter concentrations, soil respiration and N mineralisation and nitrification rates. The long-term experiment is composed of 32 plots with eight different treatments: never mown, frequently and infrequently mown with clippings retained, and infrequently mown with clippings removed, all with and without N addition (50 kg N ha-1). The C:N and C:P ratios were both greater for the soil microbial biomass than those ratios for the available soil substrates across all treatments, suggesting that the soil microbial community was primarily C-limited. The stoichiometric imbalance between available substrates and microbial elemental requirements were associated with changes in the soil microbial community composition and metabolic enzyme production. Significant relationships between the microbial community composition and SOM fractions, soil respiration rate, and C-acquiring enzyme activity highlighted the dependence of the soil microbial community on C. This may indicate that each microbial community has a specific C demand. The strong C limitation of the soil microbial community may explain the marginal effect of microbial and stoichiometric indices on soil N transformation rates. This work has provided evidence that the composition and the stoichiometric elemental demand of the soil microbial community are key regulators for the biogeochemical processes that couple C and N cycles in grassland soils and that these processes can be influenced by grassland management practices. The findings demonstrate that root-derived soil C availability can be manipulated by supplying N for enhanced plant growth. Because the soil microbial community was shown to be primarily limited by C, increasing soil C availability could increase the stoichiometric N demand of the soil microbial community and this could lead to increases in microbial N immobilisation. Both, soil N status and plant species were shown to interactively affect the allocation of rhizodeposited C to different microbial groups. This may determine the fate of rhizodeposited C in the soil due to the critical role of soil microbial community composition in regulating soil C cycling. These findings can help with identifying and developing management practices that avoid uncontrolled decoupling of elemental cycles and C and N losses from grassland systems