Effective predictors of herbaceous plant diversity responses to changes in nutrient availability and herbivory

Biodiversity, the foundation for ecosystem functions and services, is threatened by human activities. In order to protect sensitive biodiversity, a certain goal in ecology is to understand the mechanisms that determine species competition and extinction. In this thesis, I used several methods to investigate and predict the effect of human activities on grassland plant diversity. These approaches take into account the different changes of species in the community and the pivotal traits in the competition process in response to human disturbances, which are most important foundation of management interventions on plant communities, thereby facilitating efforts to maintain and recover plant diversity. In Chapter 2, I developed the concept of space resource utilization and investigated the effect of SRU of community (SRUC) on plant species richness. Height and cover of each species were measured and merged to calculate SRU in a semi-natural grassland experiment with different levels of nitrogen addition. Biomass of each species was also measured to compare with SRU. This study illustrated that SRUC is a better predictor of plant diversity than community productivity under different nutrient conditions. In Chapter 3, I examined how changes in SRU of different abundance groups modify changes in species richness in response to eutrophication and herbivore exclusion. In an alpine grassland, I tested whether abundance groups have different predictions of plant diversity and whether predictions from dominant species are better than those from the full community by a detailed analysis. The study further revealed the broader applicability of our approach across five grassland sites with different habitat types. These results further enhanced the effectivity of changes in SRU of a few dominant species for predicting changes in diversity dynamics in response to eutrophication and herbivore exclusion. In Chapter 4, I investigated the mechanisms of SRUD why it can be better predictions of plant diversity dynamics. I compared bivariate relationships with multivariate partial relationships of structural equation model to examine the extent to which SRUD captures both abundance- and functional-based mechanisms to predict plant diversity dynamics. In this study, we also checked whether non-random loss or random loss of species play a role in diversity dynamics. These outcomes supported the hypotheses that changes in SRUD operate to changes in species richness by representing abundance- to functional-based mechanisms and non-random loss of rarer species is a key reason for the decline in plant diversity. In Chapter 5, I examined the role of early-season growth rate along a manipulated nutrient gradient in a common garden competition experiment and an alpine meadow field experiment. It was shown that early differences in growth rate between species drove short-term competitive dominance under both unproductive and productive conditions and competitive exclusion under productive conditions by eutrophication. It is therefore important that plant species growing faster during the early stage of the growing season gain a competitive advantage over species that initially grow more slowly. Overall, the work presented in this thesis offers several new insights into prediction of plant diversity dynamics and foundational mechanisms of declined plant diversity. My studies substantiated that dominant species and/or fast-growing species always play a vital role in plant diversity dynamics and outcompete other rarer species in a changing world. Therewith, this thesis stressed ‘dominance management’ and ‘faster-growing management’ as effective strategies for conserving species biodiversity and ecosystem functioning in grasslands globally