Evidence for scale‐ and disturbance‐dependent trait assembly patterns in dry semi‐natural grasslands

Abstract

Summary1. The mechanisms driving nonrandom assembly patterns in plant communities have long been of interest in ecological research. Competing ecological theories predict that coexisting species may either be more functionally dissimilar than expected by chance (with functional 'divergence' mainly reflecting niche differentiation) or be functionally more similar than expected (with functional 'convergence' reflecting either the outcome of environmental filtering or weaker-competitor exclusion effects). Assembly patterns are usually assessed at a single scale and disturbance regime, whereas considering different spatial scales and disturbance regimes may clarify the underlying assembly mechanisms. 2. We tested the prediction that convergence and divergence are scale-and disturbance-dependent in grazed and abandoned species-rich dry grasslands within a 22 km 2 landscape in south-eastern Sweden. Convergence and divergence were tested for plant species' traits and phylogenetic relationships at three nested spatial scales: within 412 plots (50 9 50 cm, divided into 10 9 10 cm subplots), within 117 grassland patches (from 0.02 to 11.63 ha) and within the whole landscape (across patches). 3. At the finest scale (10 9 10 cm subplots within plots), coexisting species were more different than expected by chance (divergence), both functionally and phylogenetically, suggesting niche differentiation. At the intermediate scale (50 9 50 cm plots within patches), coexisting species showed convergence, suggesting environmental filtering. No significant deviations from random expectations were detected at the broadest scale (patches within the 22 km 2 landscape) -suggesting the prevalence of dispersal limitation at this scale. The fact that nonrandom patterns were particularly evident under grazed conditions is consistent with the prediction that assembly patterns are disturbance dependent. 4. Synthesis. This study shows that multiple trait-based assembly processes operate simultaneously in species-rich communities, across spatial scales and disturbance regimes. The results support earlier theoretical predictions that divergence between coexisting species may be an important driver of community assembly, particularly at finer spatial scales, where species compete for the same local resources. In contrast, environmental filtering is expected at broader spatial scales, where species growing in particular environmental conditions share traits that are adaptive under those conditions. Within given habitat types, dispersal limitation may, however, override environmental filtering at increasing spatial scales of observation.

Changes in local‐scale intraspecific trait variability of dominant species across contrasting island ecosystems

“…However, consideration of ITV at the local (within-plot) scale yields additional information that may help understand how specific environmental factors drive community trait assembly (Albert et al 2010a, Jung et al 2010, Paine et al 2011, Siefert 2012. For instance, contrary to other studies on the role of spatial heterogeneity in promoting community trait divergence (e.g., de Bello et al 2013), our result shows that spatial heterogeneity of soil resources was not important in driving greater ITV among dominant species on the most productive islands. Further, ITV at local scales v www.esajournals.org may be an important (though largely neglected) element of community-level functional composition and diversity, and one that could potentially be an important driver of ecosystem processes such as primary productivity and nutrient cycling in the manner that has been shown for intraspecific variation at landscape scale (Garnier et al 2004).…”

Grass strategies and grassland community responses to environmental drivers: a review

“…dispersal limitation) at large spatial scale (landscape, km 2 >), while at more local scale (m 2 -km 2 ) deterministic processes are ruling species assemblage (Lortie et al 2004;Gravel et al 2006). In dry semi-natural grassland in south-western Sweden, this pattern was effectively observed among a series of leaf and whole-plant traits comparing landscape scales with patch scales (de Bello et al 2013). In addition, these authors observed for this series of traits clear patterns of convergence at a patch scale and divergence at a smaller spatial scale (neighborhood, cm 2 -m 2 ), Importance of a functional trait and b functional dissimilarity values to explain species abundance in three communities of six species.…”

Disturbance by an endemic rodent in an arid shrubland is a habitat filter: effects on plant invasion and taxonomical, functional and phylogenetic community structure

“…However, both assembly processes can operate in a given community (Kraft et al, 2015), but their predominance is likely to vary with spatial scale and stress level, among other factors (Cavender-Bares et al, 2009;Spasojevic and Suding, 2012;de Bello et al, 2013;Laliberté et al, 2013). In our system, habitat filtering may have restricted species composition and functional divergence of communities, by species sorting, before interspecific interactions could limit the phylogenetic similarity between co-occurring species.…”

“…We propose two nonmutually exclusive explanations for this pattern. First, greater environmental heterogeneity generated by disturbance may allow functionally different species to occupy different microhabitats within a site, thus increasing overall functional divergence (de Bello et al, 2013;Kraft et al, 2015) and decreasing the degree of phylogenetic clustering, provided that life history traits show phylogenetic signal (Pausas and Verd u, 2010;de Bello et al, 2013). Actually, it has been shown that increased disturbance by fossorial mammals causes soil heterogeneity, because inter-mound microsites have higher cover of dominant species and different microclimatic conditions compared with mound microsites (Whitford and Kay, 1999).…”