Drought tolerance is determined by species identity and functional group diversity rather than by species diversity within multi-species swards

摘要

Low-input grasslands depend on the presence of grasses and legumes for the provision of nitrogen and a sufficient amount and quality of herbage for livestock. Increasing variability in the amount and distribution of rainfall, associated with climate change, is a major factor affecting production from permanent grassland and particularly grass-clover swards. Previous work has shown that introduction of deep-roofing dicotyledonous forbs into grass-legume mixtures increases functional trait diversity (FTD), which facilitates the resistance and resilience of grassland to drought. It also increases species diversity and functional group diversity (FG(D)). However, the role of species diversity at a given level of FG(D) and associated FTD for the resistance and resilience of low-input grassland towards drought remains unclear. This knowledge is required to design forage mixtures adapted to drier climatic conditions. We conducted a drought-stress experiment with 16 different sown sward types, grown in cylindrical mesocosms, combining three levels of species diversity (one, three, five) and three levels of FG(D) (one, two, three) all being associated with FTD in terms of phenology, roofing depth and nitrogen fixation to determine whether maximum levels of species diversity or FG(D) promote adaptation to drought. The study was conducted over three drought-stress cycles in an open-sided greenhouse where the climate followed ambient natural conditions. Over the duration of 685 days, we found no clear species diversity effect but a distinct species identity effect on biomass production and agronomic water use efficiency. A positive species identity effect depended largely on white clover presence and on the FTD within mixtures. The best performing mixtures for resistance and resilience combined FTD and FTD in a complementary way, irrespective of whether the swards contained three or five species, as FG(D) and FTD were already maximized within the best-performing three-species swards. Increasing complementarily of traits by species grown in carefully designed mixtures is one measure to achieve diversity effects and facilitate future sustainable grassland production. We propose to exploit the benefits of species identity and functional group diversity in designing future mixtures to cope with drought.