The diversity-productivity hypothesis predicts greater primary productivity in more diverse plant communities due to a more complete utilization of limiting resources by greater niche complementarity.  We hypothesized that grasslands planted to and managed for high plant diversity will promote greater recovery of ecosystem properties (i.e., productivity and carbon accrual) relative to low plant diversity restorations dominated by grasses.  We identified two chronosequences of restored grasslands in northwestern Illinois that contained multiple fields restored for 0-22 years.  The low diversity chronosequence (LDC) was comprised of fields planted with 1-5 species of grasses.  The high diversity chronosequence (HDC) was sown with over 30 species of prairie grasses and forbs.  In each field we measured plant species composition, aboveground net primary productivity (ANPP), total carbon (TC) and microbial biomass C (MBC).

Results/Conclusions

Across all restoration ages, plant species richness was greater in the HDC (15.4 ± 0.4 species/m²) than in the LDC (4.6 ± 0.4 species/m²) (P<0.0001).  Total ANPP was similar between the high (853 ± 75 g/m²) and low diversity (994 ± 74 g/m²) restorations (P = 0.222).  However, grass ANPP was greater in the LDC (869 ± 93 g/m²) than the HDC (609 ± 75 g/m²) (P = 0.021).   Total soil C and MBC increased with time since restoration in both chronosequences, but at a higher rate in the LDC (TC = 0.276 and MBC = 17.7 µg C·g^-1·year^-1) than in the HDC (TC = 0.216 and MBC = 7.2 µg C·g^-1·year^-1).  Contrary to our hypothesis, these results indicate soil C recovery occurs at a faster rate in restorations with lower diversity dominated by grasses than in high diversity restorations.