CO2, nitrogen, and diversity differentially affect seed production of prairie plants

Plant species composition and diversity is often influenced by early life history stages; thus, global change could dramatically affect plant community structure by altering seed production. Unfortunately, plant reproductive responses to global change are rarely studied in field settings, making it difficult to assess this possibility. To address this issue, we quantified the effects of elevated CO2, nitrogen deposition, and declining diversity on inflorescence production and inflorescence mass of 11 perennial grassland species in central Minnesota, USA. We analyzed these data to ask whether (1) global change differentially affects seed production of co-occurring species; (2) seed production responses to global change are similar for species within the same functional group (defined by ecophysiology and growth form); and (3) seed production responses to global change match productivity responses. We found that, on average, allocation to seed production decreased under elevated CO2, although individual species responses were rarely significant due to low power (CO2 treatment df = 2). The effects of nitrogen deposition on seed production were similar within functional groups: C-4 grasses tended to increase while C-3 grasses tended to decrease allocation to seed production. Responses to nitrogen deposition were negatively correlated to productivity responses, suggesting a trade-off. Allocation to seed production of some species responded to a diversity gradient, but responses were uncorrelated to productivity responses and not similar within functional groups. Presumably, species richness has complex effects on the biotic and abiotic variables that influence seed production. In total, our results suggest that seed production of co-occurring species will be altered by global change, which may affect plant communities in unpredictable ways. Although functional groups could be used to generalize seed production responses to nitrogen deposition in Minnesota prairies, we caution against relying on them for predictive purposes without a mechanistic understanding of how resource availability and biotic interactions affect seed production.