Assessing range limits and niche shifts in invasive weeds : A thesis submitted in partial fulfilment of the requirements for the Degree of Doctor of Philosophy at Lincoln University

Abstract

Invasive species pose global threats to the environment, human health, and endemic species. Being able to predict which species will become invasive, and which countries are vulnerable, helps prevent species invasions before they occur. Current methods consider which environments a species occupies in the native range and extrapolates this onto new regions. However, when species undergo niche shifts they occupy different environments in the introduced range than those they occupied in the native range. In this thesis I will combine statistical and empirical approaches to uncover what processes lead to niche shifts, and provide a better understanding of how to predict species’ distributions in the presence of niche shifts. To accomplish this I consider three globally invasive ruderal weed species: Rumex obtusifolius Linnaeus, R. crispus L., and R. conglomeratus Murray. I begin by assessing where Rumex species undergo niche shifts globally. I found that Rumex species undergo different niche dynamics in each of the three non-native regions considered. In New Zealand, the climatic niche expanded towards warmer, wetter climates; in Australia, the niche was largely conserved; and in North America the niche predominantly expanded into drier, less stable climates. From here, I wanted to understand which variables were best at predicting species’ distributions both in the native and introduced range where the species were shown to have undergone a niche shift. Here I considered land use, human-related, and hydrological non-climatic variables. Whilst climatic variables are the most commonly used variables in predicting species’ distributions, I found that climate models often underperformed when compared to non-climatic models. In particular, when projecting models from the native range onto the introduced ranges the inclusion of climatic variables was detrimental to model performance. Utilising non-climatic variables, such as human impact, resulted in more accurate model projections. Finally, I conducted a large-scale common garden experiment in the introduced range to ascertain whether individuals from the introduced range had evolved to occupy new niche space beyond what individuals from the native range tolerate. Contrary to my expectations, I found that climatic niche shifts in Rumex species are the result of preadaptation rather than rapid evolution. This study is the first of its kind to consider niche shifts across multiple introduced regions and provide experimental evidence of how species are capable of shifting their niches into non-analogue climates. In this thesis I will provide new insights into how we assess niche shifts, and open new avenues for future research. I suggest that we should first focus on better understanding the species’ fundamental niche, before assessing whether species have undergone rapid adaptation in new regions.