Peatlands constitute a significant soil carbon (C) store, yet the C gas flux components show distinct spatial variation both between and within peatlands. Determining the controls on this variability could aid in our understanding of the response of peatlands to global changes. In this study, we assess the usefulness of different vegetation related parameters to explain spatial variation in peatland C gas flux components. We hypothesise that spatial variation is best explained by trait-based indices (similarly to other terrestrial ecosystems), and that the impact of soil physicochemical properties, such as nitrogen (N) content or water level, can be manifested through the traits. Furthermore, we expect that the spatial variability associated with each of the C gas flux components can be explained by a distinct set of traits. To address our aim, we used a successional peatland chronosequence from wet meadows to a bog, along which all variables were recorded with similar methods and under similar climatic conditions. We observed spatial variability with all measured gas fluxes, with carbon dioxide (CO2) fluxes showing significant variability between sites, while within site variability was more important for methane (CH4) fluxes. As expected, our results show that the impacts of physicochemical conditions were directed via vegetation. However, the cover of functional plant types that capture multiple traits proved to be more powerful in explaining gas flux variability compared to functional trait-based indices. Our findings imply that for future gas flux modelling purposes, rather than attempting to use individual traits - as is the ongoing trend in ecology - it might be more useful to refine plant functional groupings and ensure they are based on a set of plant traits relevant for the studied ecosystem process. This could be facilitated by the collation of a large data set of traits measured from peatlands.
Keywords: Carbon fluxes; Functional diversity; Functional trait; Plant diversity; Plant functional type; Structural equation model.
Copyright © 2022 The Authors. Published by Elsevier B.V. All rights reserved.