2.1 Study species and experimental design
To increase the generality of our results, we selected six common grass species with different functional traits (different photosynthetic pathways (C3 vs C4), life forms (annual vs perennial), and root system characteristics (rhizomevs bunch)) that grow in temperate grasslands in China. These species are Chenopodium glaucum , Cleistogenes squarrosa ,Leymus chinensis , Medicago sativa , Setaria viridis , and Stipa grandis . Among these species, M. sativa is a high-quality forage species and is widely grown in artificial grasslands, while the other five species are common in natural grasslands (Figure 1).
The pot experiment was conducted in the greenhouse of the Institute of Botany, Chinese Academy of Sciences, Beijing, China (39°9′ N, 116°49′ E). A homogenous substrate was prepared by mixing sand, vermiculite (3-5 mm diameter to prevent the roots from sticking to the substrate), and peat soil (Jiffy, Netherlands) at a ratio of 3:1:1 by volume. The planting pots (length: 40 cm × width: 40 cm × height: 30 cm; volume: 48 L) were filled with ~16 kg of the mixed substrate. In this experiment, five densities were established for each species: 25 (D1), 100 (D2), 225 (D3), 400 (D4), and 625 (D5) plants m-2. There were three replicates for each density level and a total of 90 pots for all the species and density levels. The number of plants sown per pot was determined by the planting density and the pot area (0.16 m2); in total, there were 3,960 individuals. The experiment was conducted using a randomized block design with one block per species for a total of six plots (Figure 2).