Questions:
To what extent is trait variation in grasses and forbs driven by land-use intensity, climate, soil conditions and plant diversity of the local neighbourhood? Do grass and forb species differ in the degree of intraspecific trait variation?

Location:
Managed grasslands in three regions of Germany.

Methods:
Using a phytometer approach, we raised 20 common European grassland species (ten forbs and ten grasses) and planted them into 54 plots of different land-use types (pasture, meadow, mown pasture). After 1 yr in the field, we measured above- and below-ground plant functional traits. Linear mixed effects models (LMEM) were used to identify the most powerful predictors for every trait. Variation partitioning was applied to assess the amount of inter- and intraspecific trait variation in grasses and forbs explained by environmental conditions (land-use intensity, climate and soil conditions) and plant species diversity of the local neighbourhood.

Results:
For 12 out of the 14 traits studied, either land-use intensity or local neighbourhood diversity were predictors in the best LMEM. Land-use intensity had considerably stronger effects than neighbourhood diversity. Root dry matter content and root phosphorus concentration of forbs were more affected by land-use intensity than those of grasses. For almost all traits, intraspecific trait variation of grasses was much higher than that of forbs, while traits of forbs varied more among species. Overall, inter- and intraspecific variation was of the same magnitude.

Conclusion:
The similar magnitude of intra- and interspecific trait variation suggests that both sources should be considered in grassland studies at a scale similar to that of our study. The high amount of intraspecific trait variation that was explained by environmental factors and local neighbourhood diversity clearly demonstrates the high potential of species to adjust to local conditions, which would be ignored when only considering species mean trait values.