Actual and potential distribution of non-native plants in the Central-Southern Andes of Chile

Ondertitel:the role of anthropization at multiple spatial scales

Several factors have been identified as relevant for the distribution of non-native plants in mountain ecosystems. These factors depend mainly on the environmental and vegetational characteristics of each ecosystem (invasibility) and the traits of each invasive species (invasiveness). Knowing the importance of each factor that determines the distribution of non- native plants at multiple scales, is critical to advance the knowledge of the invasion process and efficient protocols for the control and management of non-native in the Andes, all this considering that these ecosystems are a hotspot of biodiversity. Therefore, the general objective of this doctoral thesis is to deepen the existing knowledge on the distribution of non-native plants in the Andes. More specifically, I aim to evaluate how abiotic, biotic, and anthropogenic factors determine the distribution patterns, richness, and abundance of non-native plants along elevational gradients at multiple scales in the Andes. Complementarily, using the current distribution patterns of the most abundant non-native lowland plants, we sought to project their potential future distribution using anthropogenic and microclimatic variables, to understand how the inclusion of these variables can improve the performance of species distribution models and would thus allow us to assess more accurately the redistribution of invasive plants in the Chilean Andes. In this thesis, the following hypotheses were evaluated: (1) The richness and abundance of non- native plants at the regional scale in the Andes Mountains will be determined by abiotic variables such as soil temperature and nitrogen content, while at the local scale anthropogenic variables, such as the presence of roads and human settlements, will be the main drivers. Due to the importance of anthropogenic and microclimatic factors for non-native plant distributions in mountain ecosystems (2), the inclusion of microclimatic and anthropogenic data should improve the performance of models of these distributions in the Andes. Finally, because of the importance of roads as defining factor for non-native plant species distributions at multiple scales in mountain ecosystems worldwide, we propose that (3) roads will similarly be the main drivers of the distribution of non-native plants at the continental scale in the Andes Mountains. To answer these hypotheses, three methodological approaches will be used: (1) First, through a literature review we seek to identify the main drivers of non-native plant richness patterns in the Andes, main impacts and management strategies (Objectives 1, 4), (2) using observational studies and local and regional measurements of abiotic, biotic and anthropogenic factors to assess the drivers of non-native plant richness and abundance (Objective 1,2). Finally, (3) using species distribution models (SDM), we modelled the distribution of five invasive plants in the south-central Andes, evaluating the importance of incorporating anthropogenic and microclimatic variables (objective 3). The literature review (Chapter 1) showed that most studies conducted in the Andes in relation to plant invasions are aimed at understanding distribution patterns, often with case studies focused on specific species. In addition, we found that issues such as the impact of climate change, the impact of anthropogenic disturbances on distribution, and the impacts of non-native plants on ecosystems are the least addressed when studying invasive plants in the Andes. In Chapter 2, we analysed the main factors that explain the richness and abundance of non-native plants at multiple scales. We determined that at the local scale, the presence of roads, the distance to human settlements and the presence of agricultural activities were in all but one case the most important factors explaining the richness and abundance of non-native plants. Only in Malalcahuello National Reserve, biotic factors (the abundance of native species) more strongly related to the abundance of non-native plants. At the regional scale, the richness of non-native plants was explained best by soil ammonium levels and pH, while the abundance of non-native plants was explained best by the presence of agricultural activities. In chapter 3, I show that the inclusion of anthropogenic variables (road presence and distance to human settlements) and microclimatic variables (soil temperature) significantly improved the performance of SDMs, especially for herbs and shrubs. It is however important to note that in all models, precipitation was a key macroclimate variable explaining the distribution of non- native plants in the Andes. Additionally, the results highlight that it is possible that non-native plants expand to higher elevations in the Andes, with significant invasion potential in protected areas. Finally, in Chapter 4 we demonstrate the importance of anthropogenic variables for explaining the distribution of non-native plants along continental-scale elevation gradients in the Andes. These results are fundamental for Andean biodiversity conservation, as they allow us to focus on these anthropogenic variables (e.g., presence of human settlements, agricultural activities, cattle ranching, etc.) when developing biosecurity protocols and management strategies to avoid further expansion of non-native plants in protected areas in particular and the Chilean Andes as a whole.

Jaar van publicatie:2022

Trefwoorden:Doctoral thesis

Toegankelijkheid:Open