Invasive plant species effects on carbon and nitrogen cycling in temperate wetlands

Background/Question/Methods

Plant functional traits, such as leaf tissue chemistry and biomass production, are useful because they allow us to simplify complex controls on ecological/biogeochemical processes to a few plant traits of major influence.  Here, we investigate how dominant plant species influence ecosystem carbon (C) and nitrogen (N) stocks, litter quantity and quality, and soil C and N transformation rates in Michigan wetlands.  We hypothesized that invasive grasses (such as Phalaris arundinacea and Phragmites australis) have greater nitrogen use efficiency (NUE) compared to the native species they often replace, resulting in lower quality litter inputs and increased C and N storage.  We also hypothesized that soil C and N mineralization rates are influenced by the dominant vegetation due to the quantity and quality of litter inputs, and that these litter characteristics differ among invasive species.  We conducted a regional survey of wetlands near the Kellogg Biological Station in SW Michigan, where we characterized plant community composition, key traits such as productivity and tissue chemistry, and chemical and physical properties of litter and soil.  Also, we performed laboratory incubations to characterize soil C and N mineralization rates at a subset of sites characterized by monocultures of invasive species.

Results/Conclusions

The degree of P. arundinacea invasion is positively correlated with litter C and N stocks, and total invasive species biomass and litter chemistry are significant predictors of soil C and N stocks.  Total ecosystem C and N stocks are positively correlated with invasive species dominance.  There is no relationship between NUE and invasion, mainly due to variable NUE for P. arundinacea, though P. australis did have the highest NUE.  Incubation results show significant effects of dominant species and incubation temperature (7 and 21^oC) with the highest soil C mineralization rates occurring in P. arundinacea sites and the lowest mineralization rates occurring at P. australis sites.  C mineralization rates were strongly controlled by soil C/N ratio, which was influenced by species identity.  Q₁₀ values for C mineralization varied by species: 2.2 for Typha latifolia, 2.4 for P. arundinacea, and 2.6 for P. australis.  Net N transformation rates show significant differences among monocultures of invasive species, with the highest rates at sites dominated by P. arundinacea.  Our results suggest organic matter inputs differ among dominant species, with P. arundinacea inputs more labile and P. australis inputs more recalcitrant.   Our continuing work utilizes an experimental framework for investigating linkages and feedbacks between plant traits, species composition, and ecosystem processes.