Master Thesis

The same, only different: analysing the world’s grassland heterogeneity in terms of productivity and beta diversity

Samuel Hoffmann (06/2025-12/2015)

Support: Carl Beierkuhnlein, Severin D.H. Irl

Ecological studies have been evolving for about a hundred years. So-called Coordinated Distributed Experiments (CDE) are recently discussed in ecology. The Herbaceous Plant Diversity Network (HerbDivNet) applied an ecological CDE in order to address the lack of investigations that consider grassland diversity at the global scale. Grassland is one the Earth’s major ecosystems and is enormously contributing to the human well-being by ecosystem functions and services that depend on biodiversity. Beta diversity is heterogeneity in species composition and can be defined in many ways. It is an essential component of the concept of biodiversity. Discovering causes for differences within and between species assemblages is thus crucial for the understanding and the conservation of biodiversity and ecosystem functions or services, from which we profit. But beta diversity of the world’s grassland is poorly examined. Therefore, I analysed beta diversity of the global grassland ecosystems by using the HerbDivNet data set. Former studies revealed that biomass production determines beta diversity by stochastic or deterministic, ecological processes, whose relative importance alters along the productivity gradient. Stochasticity and determinism are the main concepts of two contrasting, yet fundamental theories in ecology, the neutral theory and the niche theory, respectively. By relating biomass production, heterogeneity in biomass production, species richness and climatic parameters to thirteen beta diversity metrics, which are distinguished into four groups of distinct theoretical computation and practical meaning, I identified drivers and causes of grassland beta diversity worldwide, which can be assigned to one of both theories. Statistical methods included robust linear regression, the Mantel test and variance partitioning. Results showed that both stochastic and deterministic processes influence species composition of the Earth’s grassland ecosystems. However, findings depend on the applied definition of beta diversity. Hence, I recommend taking advantage of the diversity of beta diversity concepts in order to unravel ecological mechanisms generating species assemblages in future studies. Moreover, a large part of heterogeneity in species composition could not be explained at all. I finally suggest other unconsidered factors of grassland beta diversity worldwide to be incorporated into further research.