Do plant functional traits determine spatial pattern? A test on species‐rich shrublands, Western Australia. (27th September 2012)

• Record Type: Journal Article
• Authors: Perry, George L.W.; Enright, Neal J.; Miller, Ben P.; Lamont, Byron B.; Kühn, Ingolf
• Is Part Of: Journal of vegetation science. Volume 24:Number 3(2013:May)
• Page Start: 441
• Page End: 452
• Publication Date: 2012-09-27
• Subjects: Plant ecology -- Periodicals; Plant communities -- Periodicals; Plant populations -- Periodicals; 581.7
• DOI: 10.1111/j.1654-1103.2012.01476.x
• Languages: English
• ISSNs: 1100-9233

Abstract

<abstract abstract-type="main" id="jvs1476-abs-0001"> <title>Abstract</title> <sec id="jvs1476-sec-0001" sec-type="section"> <title>Questions</title> <p>A species' spatial pattern is the outcome of a series of filters: demographic, disturbance, environmental and functional, all varying over space and time. To evaluate the importance of function as a filter we ask: (1) do a species' functional traits allow prediction of its fine‐scale spatial patterning, and (2) how consistent is the fine‐scale spatial pattern shown by the same species across multiple sites where the properties of the other filters may vary?</p> </sec> <sec id="jvs1476-sec-0002" sec-type="section"> <title>Location</title> <p>Species‐rich Mediterranean climate shrublands of southwestern Australia's northern sandplains.</p> </sec> <sec id="jvs1476-sec-0003" sec-type="section"> <title>Methods</title> <p>Using fully mapped plots at four sites (each >10 000 individuals, 74–112 species), we characterized individual species spatial patterning using point pattern analyses. We classified species spatial patterns in three ways: (1) whether they departed from a null model controlling for first‐order effects (i.e. are they aggregated?), (2) parameterization of Thomas cluster processes (what form does any clustering take?), and (3) their position in a multivariate 'pattern space' (do species show different types of pattern?). We then explored the extent to which a species' functional traits, abundance and/or the site at which it occurred predicted these three facets of its spatial pattern.</p> </sec> <sec id="jvs1476-sec-0004" sec-type="section"> <title>Results</title> <p>Although at all sites most species were aggregated, site was consistently important in predicting a species' spatial pattern. Regenerative response to fire – whether a species is killed by fire and recruits solely via seeds, or survives and resprouts vegetatively after fire – was the functional trait most consistently useful in predicting a species' spatial pattern. Fire‐killed species tended to show more aggregated distributions than resprouters. Species present at multiple sites did not show consistency in their spatial patterns across those sites more than expected by chance alone.</p> </sec> <sec id="jvs1476-sec-0005" sec-type="section"> <title>Conclusions</title> <p>Although functional traits relating to fire responses and water use predict species spatial distributions at the landscape level, at the fine scales we considered site effects were as important as functional traits in explaining spatial pattern. Within and between site heterogeneity, some of which can be generated by the stochastic properties of fire, may mask the deterministic effects of species functional traits.</p> </sec> </abstract>