PubAg

Main content area

Invasive acacias differ from native dune species in the hyperspectral/biochemical trait space

Author:
Große‐Stoltenberg, André, Hellmann, Christine, Thiele, Jan, Oldeland, Jens, Werner, Christiane
Source:
Journal of vegetation science 2018 v.29 no.2 pp. 325-335
ISSN:
1100-9233
Subject:
Acacia, biochemistry, biomass, canopy, carbon, coasts, ecosystems, indigenous species, invasive species, least squares, leaves, lignin, models, nitrogen, nitrogen content, plant communities, prediction, shrubs, spectroscopy, stable isotopes, Portugal
Abstract:
AIM: The impact of invasive species may depend on dissimilarity of their functional traits relative to the native community. Therefore, comparing species traits in a multidimensional space can help to better understand invader impacts, but novel methods are needed to effectively measure multiple traits across diverse plant communities. The main aim was to assess biochemical leaf parameters based on field spectra in a whole‐community approach. Our specific objectives were to assess (1) biochemical differentiation within the plant community, (2) accuracy of spectroscopic prediction models of biochemical parameters, and (3) ability to depict the multivariate differentiation using field spectroscopy. LOCATION: Mediterranean dune ecosystems, Atlantic coast, southwest Portugal. METHODS: We analysed leaf biomass of 18 species, including two invasive acacias, for C, δ¹³C, N, δ¹⁵N, lignin, fibre and tannin. Additionally, we collected leaf and canopy field spectra of each sampled plant. We used partial least squares (PLS) regression to predict biochemical parameters from field spectra. Further, we assessed the biochemical differentiation of the species using PCA based on wet chemically determined as well as spectroscopically predicted values. RESULTS: We found high biochemical variation among species and, in particular, marked trait dissimilarity between invasive Acacia spp. and native species, primarily with respect to N content. Biochemical parameters were predicted successfully based on field spectra. Prediction accuracies were particularly high with C, δ¹³C, N and tannin. A PCA of biochemical parameters showed that invasive Acacia spp. were distinct from native species of the same life form, but grouped with native dwarf shrubs. This pattern was accurately reproduced by a PCA using spectroscopically predicted values. CONCLUSIONS: Invasive Acacia spp. have different leaf traits compared to native species of similar growth form. Their trait dissimilarity likely exacerbates their impacts on the ecosystem. This trait dissimilarity in leaf biochemistry can be accurately predicted with hyperspectral whole‐community models. Thus, field spectroscopy can substantially increase the spatial and temporal resolution of measurements and hence facilitate assessments of ecosystem functioning and invader impacts at ecosystem scale.
Agid:
5939571