Proximity to forest plantations is associated with presence and abundance of invasive plants in landscapes of south-central Chile

Abstract

Invasive plant species (IPs) are widespread in forests and cause substantial environmental, economic and social impacts. They occupy native ecological niches, causing local extinctions to the detriment of native biodiversity and disrupting ecosystem services provision. How landscape characteristics may determine the success of IPs remains unclear and, more importantly, how land-use and land-cover changes may result in spatial shifts in the invasion risk. Furthermore, the study of how landscape factors may influence biological invasions has focused on particular species, but not the IPs’ community. In this study, we identify and assess landscape variables that influence the presence and distribution of the IPs’ community in temperate forests of a global biodiversity hotspot in south-central Chile. We fitted spatially explicit models, combining field-sampling information and landscape variables related to land-use/land-cover, topography, climate, soil characteristics and anthropogenic factors to explain and predict the presence and distribution of the IPs’ community. From the whole sampling of plant species, we identified eight plant species classified as IPs: three trees and five shrubs. We used field data from 125 500 × 2 m-transects, in which we registered species richness, abundance and basal area of IPs’ community. Distance to forest plantations was the landscape variable with the most substantial influence on IPs’ presence and distribution. Richness, abundance and basal area of IPs’ trees were higher at shorter distances from forest plantations. The basal area of IPs’ trees was the best model explaining the relationship between IPs’ community and landscape variables. All descriptors of the IPs’ community showed similar spatial patterns: species richness, abundance and tree basal area are higher in more disturbed areas. Our findings contribute to increasing our understanding of the distribution patterns of IPs in forest landscapes. Our models can be suitable tools for designing strategies to prevent, mitigate or make integrated control of the impacts of invasive species in forest landscapes.