Alkaline and loess grasslands with contrasting richness and biomass patterns are not separated on the CSR strategy spectrum.

Studying the relationship between biodiversity patterns and processes in vegetation has been at the centre of interest in vegetation ecology for several decades. By studying the biomass of loess and alkaline grasslands along a water and salinity gradient, we aimed to analyse species diversity and CSR functional type patterns. We aimed to test the following hypotheses: i) The biomass and species richness scores and the species composition are significantly different between the sampled grassland community types. ii) The sampled communities are well separated based on the CSR strategy spectrum. iii) The amount of green biomass and litter are positively correlated with competitiveness and negatively with stress tolerance. The biomass and species richness scores and the species composition of the sampled communities along the sampled gradients were significantly different; the highest species richness, evenness and Shannon diversity values were found in loess grasslands. The highest level of litter accumulation was found in alkaline meadows. The communities were well separated in the ordinations but surprisingly, calculation of coordinates for CSR strategy types have not shown clear separation of the grassland community types. Based on the results we can conclude that the current CSR classification is too robust to delineate grassland communities in alkaline landscapes which are markedly different in composition. We point out that the CSR classification is based on the magnitude of stress, and it is not able to differentiate between the various forms of stress which might be responsible for the marked compositional differences. Therefore, developing an index focusing specifically on the functional characterisation of stressed communities and able to differentiate between different forms of stress (e.g., water and/or salinity stress) would be highly beneficial for future studies.