HOlistic Management of Emerging forest pests and Diseases

Funder

Organizational Unit

Publications

Modelling the dispersal of invasive species using landscape variables

Publication . Nunes, Pedro Mourato Catela; Simões, Manuela Rodrigues Branco; Silva, José Carlos Franco Santos; Jactel, Hervé

Biological invasions are a great threat for ecosystems functioning and human welfare. In many cases, the more realistic strategy in the combat of invasive species is controlling their spatial expansion. Our work focused on studying the role of the landscape structure towards invasive species dispersal. For this we developed three different dispersal models using landscape factors and ecology concepts. Models were applied to two highly relevant invasive species for Europe, the pine wood nematode (PWN) and the african citrus psyllid. To understand the role of landscape heterogeneity towards the spread of the PWN, we studied the dispersal trajectories of the vector, M. galloprovincialis. We used least-cost path analysis to estimate the beetle’s behaviour response to the different land uses from a mark-release-recapture essay conducted in a heterogenous forest. The importance of isolated host trees for invasive species spread was tested in two spatio-temporal models for the dispersal of T. erytreae at large and local scales. At large scale, the role of human activities for the invasion patterns of T. erytreae in Portugal was also addressed. The inclusion of human-mediated spread and isolated host trees improved the F1-Scores of the model validation with the official reports as observations. For the local scale model, in an heterogenous agriculture region, we developed a spatio-temporal model, using an epidemiological approach for the chain of tree infections, to explain the dispersal of the species across the landscape. This model further allowed incorporating different management strategies. Altogether, our findings highlighted the importance that landscape structure, and its elements, can have for invasive species dispersal, making their inclusion invaluable for invasive species modelling. Additionally, we were able to contribute with promising novel methodologies and modelling approaches for invasive biology, including a tool for pinpointing the origin of captured individuals in trapping grids.

2023-12-07Doctoral thesis

Open access