Print PageThe most appropriate source of propagules used in ecological restoration remains controversial largely due to unknown hierarchical consequences (e.g., community assembly and ecosystem functioning) of different population sources in multi-species assemblages, over time, and across a broad geographic region. Distance and connectivity between restored and remnant populations, as well as the degree and size of disturbance, are important considerations in the selection of local genotypes, mixed genotypes, or cultivars for propagule sources in restoration. Cultivars of the dominant prairie grasses have been selected for agronomic plant traits and are widely used in grassland restorations, but there is limited information on the ecological consequences of using these sources when aiming to restore diverse prairie communities. We established an experimental prairie restoration in southern Illinois (SIL) using a split plot design to test whether cultivar or locally collected sources of the dominant grasses differentially shape community structure that scales to affect ecosystem functioning. Local and regional generalizations regarding ecological consequences of these source populations in restoration were justified by using three pools of other species (with equivalent representation of life forms) seeded within each grass source treatment in SIL, and through establishment of a similar field study in Kansas.
Results/Conclusions
Initial comparisons of plant traits between the dominant grass sources demonstrated similar or enhanced root characteristics and physiological performance of cultivars relative to non-cultivar sources of the dominant grasses. In SIL, community composition was distinct among the 3 species pools of non-focal grasses (P<0.05), but there was no effect of grass source on diversity. Total ANPP of planted species and focal grasses increased over the first 4 years of restoration in SIL (P<0.001) and total planted ANPP varied among species pools (P=0.006). The only focal grass to exhibit a difference in ANPP among sources was Schizachyrium scoparium, where ANPP of the non-cultivar source was moderately lower than the cultivar source over all years in SIL (P=0.089) due to lower ANPP in one species pool (P=0.035). There was no effect of grass source on diversity or ANPP in KS plots established using the same seeding rate and species pool as SIL. Thus, we have no evidence that cultivar and locally-collected propagules of dominant grasses differentially affect community assembly or ecosystem functioning in prairie restored with a 60:40 ratio of grass:forb seeds and moderate richness of seeded species.
