Inferring temporal shifts in landuse intensity from functional response traits and functional diversity patterns: a study of Scotland's machair grassland


R. J. Lewis, The James Hutton Institute, Craigiebuckler, Aberdeen, AB15 8QH, UK. Applied Vegetation Dynamics Laboratory, School of Environmental Sciences, Univ. of Liverpool, Liverpool, L69 3GP, UK. Present address for RJL: Inst. of Ecology and Earth Sciences, Univ. of Tartu, Lai 40, Tartu, EE-51005, Estonia. E-mail:


Plant functional response traits, which consistently respond to the environment, are useful for identifying drivers of vegetation change, particularly in response to disturbance gradients. Similarly, functional diversity indices have proven useful for investigating processes governing community assembly, particularly patterns of functional convergence/divergence.

This study investigated the functional ecology of biodiverse, seminatural coastal grasslands (Scottish machair) at the national scale. We examined temporal shifts in functional response traits and functional diversity metrics using a series of null model, multivariate and regression analyses. The aim was to link temporal shifts in traits and diversity metrics to environmental variables in which to gauge the contribution of landuse change to plant functional composition and processes governing plant assembly.

We observed significant shifts in the composition of 8 out of 12 functional response traits at the national scale, whereas at the regional scale all traits displayed at least one significant shift. Ordination of response traits found PC axis 1 (accounting for 39% of the variation) to be positively correlated to vegetation height and negatively correlated to specific leaf area, similar to that expected along a disturbance gradient. Significant changes in functional diversity indices were also observed at both national and regional scales, with varying convergence/divergence patterns observed across individual regions. We found functional richness (t = 4.87, p < 0.001) and divergence (t = 9.3, p < 0.001) to increase along PC axis 1, suggesting greater convergence and lower divergence along a disturbance gradient.

This study demonstrates the potential for using functional diversity indices in combination with response traits as a sensitive method for detecting landuse change and its impacts on biodiversity. We conclude that landuse change, particularly management declines and intensification is a major driver governing change among the functional composition and functional diversity for machair grasslands, influencing convergence/divergence patterns, and subsequently community assembly processes.