Ellie M. Goud, Tim R. Moore and Nigel T. Roulet
Plant functional traits, including morphological, physiological, and phenological characteristics, can influence the ratio between carbon (C) gains and losses by ecosystems, through effects on photosynthesis, respiration and litter decomposition rates. There is growing evidence that rates of C gas (carbon dioxide and methane) exchange (fluxes) can be directly predicted from plant traits such as specific leaf area (leaf area per unit mass) and leaf nitrogen content. These traits require destructive sampling of plant material, which limits their use when traits are to be directly compared to gas fluxes from the same individual or community. Non-destructive traits such as leaf area, leaf longevity or growth form may provide valuable information for field studies requiring repeated measurements of the same sampling area.
In this study, we assessed the ability of four non-destructive plant traits (leaf area, leaf longevity, growth form, tissue containing air spaces (aerenchyma)) to predict rates of carbon dioxide (CO2) exchange and methane (CH4) emissions in a temperate peatland in Ontario, Canada. Peatlands, including bogs and fens, are key features of the northern landscape and support an impressive diversity of plants adapted to the harsh, acidic wetland conditions, such as Sphagnum peat mosses, dwarf blueberry shrubs, sedges and carnivorous pitcher plants. Peatlands are also critical in the global C cycle, storing approximately one-third of the global organic soil C pool. However, the ability of plant traits to predict peatland C cycling is relatively unexplored. We found that CO2 fluxes (rates of net CO2 exchange, ecosystem respiration and gross photosynthesis) were positively related to leaf area and longevity, and negatively related to the proportion of woody species. CH4 fluxes (emission rates) were positively related to aerenchyma tissue and leaf area of sedges and rushes. The significance of trait-flux relationships differed depending on whether data were averaged at the level of plot, species or microsite. We found that leaf area was the strongest predictor of both CO2 exchange and CH4 emissions, and is therefore recommended in other systems where it is not ideal to measure traits destructively.