Opportunities and limitations of the use of functional traits to understand plant invasions

Abstract

Biological invasions comprise the human-mediated introduction, establishment and dispersal of species beyond their natural geographic range. Species movement around the globe can be seen as a large scale community assembly experiment, in which species overcome a series of ecological barriers to become invasive. Plant species ability to become invasive (i.e. species invasiveness) can be explored through trait-based, multi-species approaches, which assess the functional traits associated with the species' ability to overcome these barriers. Importantly however, findings from trait-based studies of plant invasiveness are diverse and not always consistent on the role of traits in facilitating invasion processes.

The overarching aim of this thesis is to investigate the potential of functional traits to act as surrogates for the mechanisms driving plant invasions. I begin by asking whether floras around the world experience similar patterns of urbanization-driven functional change, and suggest general mechanisms that enable persistence and colonization processes in highly disturbed environments. Then, I investigate several aspects that could potentially result in misleading findings from trait-based studies of plant invasiveness, particularly focusing on (1) the existence of several demographic processes underlying invasion success; (2) how the use of different methods to build trait datasets can influence correlations between invasiveness and traits, and (3) the degree by which trait-invasiveness relationships can be misleading as the result of the preferential and often unknown introduction of species with particular combinations of traits.

I found substantial evidence of the role of functional traits in facilitating invasions. My research at both global and regional (Victoria, Australia) scales found that traits such as height, seed mass, specific leaf area and nutrient strategy correlated with plant invasiveness. In both cases, my findings suggest that invasive species are not a homogeneous group and several ecological strategies may lead to the success of invasive plants. One common assumption of invasion studies using binary classifications of invasiveness (i.e. invasive vs non-invasive) is that all invasive plants rely on the same traits to become successful. My findings however suggest that the use of continuous, demography-based invasiveness metrics should help untangle complex relationships between traits and different aspects of plant invasion. Furthermore, my findings highlight that observed trait-invasiveness correlations may be affected by methodological decisions made to build trait datasets - e.g. imputation techniques - and unknown biases on introduced species' traits. For example, overlooking that introduced as well as invasive species show particular values of a given trait may result in the overestimation of that trait's effect in invasiveness.

Based on my findings, future invasiveness studies using a functional approach may benefit from: (1) critical thinking on the links between invasiveness, plant demography and functional traits; (2) further investigation of the limitations posed by trait sampling effort, imputation techniques and the study of different species groups to make confident generalizations; and (3) better efforts to record current introductions and the use of creative ways to account for past introduction biases. I strongly believe these can contribute to increase knowledge of the mechanisms driving plant invasions.