SPECIES AND COMMUNITY CONTROLS ON ECOSYSTEM CARBON CYCLING: DO PLANT INTERACTION TRAITS INFLUENCE ECOSYSTEM PRODUCTIVITY?
As biodiversity is heavily altered by human influence, the need to understand how biodiversity impacts ecosystem productivity is increasingly important. Host plant and arbuscular mycorrhizal fungi (AMF) interaction specificity is a relatively new topic that may facilitate the understanding of the mechanisms behind carbon cycling. AMF interaction traits have been assigned to host plants based on the number of AMF taxa they associate with. Interaction specialists are AMF host plants that associate with few AMF taxa while interaction generalists associate with many taxa. In this study I hypothesised that interaction specialists increase ecosystem productivity due to niche complementarity of AMF interaction partners relative to an interaction generalist especially at high species richness. My study investigates the addition of a late arrival with a defined interaction trait into plant communities with three species richness levels. Soil sterilisation and reinoculation treatment are also included to control for niche partitioning based on AMF interaction partners. I also hypothesised that increasing species richness has a positive effect on productivity while soil sterilisation reduces productivity and soil reinoculation partially restores this lost productivity. The community responses of primary productivity (GPP), ecosystem respiration (ER), net ecosystem exchange (NEE), aboveground plant biomass, belowground plant biomass, leaf area index (LAI), and specific leaf area (SLA) were measured. Community productivity responses to interaction traits did not covary with species richness or soil treatment, so the effect of species richness, soil treatment, and interaction trait were analysed separately. Increasing species richness resulted in increasing productivity as expected, but the pattern of carbon cycling shifted from carbon storage as plant biomass to soil microbial processes. While the rate of carbon drawdown or potential productivity increased, biomass decreased at high species richness. Soil sterilisation likewise resulted in greater carbon drawdown which was opposite to what was expected but decreased ER and plant biomass. Reinoculation of soil restored plant biomass, but negatively affected both carbon flux (GPP & ER) and plant traits (LAI & SLA). The addition of one interaction generalist (Plantago lanceolata) decreased total community aboveground biomass which indicates that interaction trait can be used to predict the direction of community response. The direction of the productivity responses can be partially explained by the P limitation of the mesocosms and the role of the soil microbial community in carbon cycling. By quantifying a diverse range of carbon pools, my study contributes to the understanding of complex carbon allocation processes that control ecosystem productivity responses and provides important insights for the context-dependence of selecting and analysing of productivity measures of whole ecosystems.