Effects of artificially varying amounts of rainfall on two semi-natural grassland types





Is a lowland grassland community with a lower amount of ambient precipitation and a high abundance of subdominant species more sensitive to altered rainfall input in comparison with mountain grassland with higher ambient precipitation and a low abundance of subdominant species?


(1) Lowland Festuca grassland characterized by dry acidophilous vegetation, low mean annual precipitation and high species richness, and (2) mountain Nardus grassland with high mean annual precipitation and low species richness in the Eastern Czech Republic.


We conducted a manipulative experiment with rainout shelters and gravity irrigation systems. We simulated three levels of precipitation input (50%, 100% and 150%) and examined the treatment effects on above-ground net primary productivity (ANPP) of the main plant functional groups (grasses, forbs, legumes, shrubs and mosses) and on plant species composition over three consecutive years.


Altered rainfall input had a significant effect on total live ANPP in the drier, lowland grassland community, while no pronounced effect was observed in the mountain grassland characterized by higher ambient precipitation. In the lowland grassland, changes in total ANPP resulted primarily from enhanced growth of subdominant non-N-fixing forbs in response to the direct effect of higher artificial rainfall input and an indirect effect of reduced competition with grasses in drier treatments.


Wetter, species-poor mountain grasslands will probably show smaller changes in response to expected climate changes in ANPP, species number and abundance than lowland grasslands, characterized by dry acidophilous vegetation, with higher species richness and higher abundance of subordinate species.