AZORES · Community phylogenetics and conservation on an oceanic archipelago: the flora of the Azores as a case study

Community phylogenetics and conservation on an oceanic archipelago: the flora of the Azores as a case study

In our project, we investigated the biology of invasion in plants, and show that evolutionary metrics can predict invasiveness and are likely to be more reliable than the currently used ecological metrics. Our work builds on 10 years of fieldwork in the remote Azores archipelago and large genetic and ecological datasets. We reconstructed the phylogeny for the entire archipelago's flora (796 species), and tested various possible correlates of invasion, including Darwin's 'naturalisation hypothesis'. Having found a global predictor of invasiveness has tremendous importance. Invasive species are a major threat to ecosystems worldwide. This work is also likely to be receptive to the general public. In the past 20 years, many campaigns were launched to raise the awareness of the general public towards invasive species. As a result, the topic is now frequently discussed in newspapers and on popular science websites. Quarantine regulations aimed at preventing spread of potential invaders affect everybody travelling between different continents and invasive species like Japanese knotweed or tansy ragwort regularly produce headlines. Invasive species are a major threat to ecosystems worldwide, and especially on oceanic islands like the Azores. An estimated half a million species are classified as 'invasive aliens' causing costs of over 1 trillion per year. So far, it has not been possible to predict whether an introduced species will become invasive, and it is unclear whether certain traits promote invasiveness. Our results show that introduced plant species are likely to become invasive when there are no closely related species in the area, verifying an hypothesis published by Charles Darwin in 1859 (Darwin's 'naturalisation hypothesis'). Using phylogenetic information based on the first complete DNA dataset for the angiosperm flora of an entire archipelago (the Azores - a group of nine volcanic islands in the middle-Atlantic), we found that the phylogenetic nearest neighbour distance (a measure of evolutionary relatedness) is a reliable predictor of invasiveness. Using analyses at different geographic scales, we argue that enemy release rather than competition plays an important role in invasiveness. Some ecological traits, such as life form and seed size, also predict invasion success in the Azores, but not in other systems. We suggest that plant invasions are therefore best forecasted using evolutionary metrics, and our analyses have important implications for solving this problem worldwide. Plant invasions are now predictable, and our results can help solving this problem across the globe. With the rapid advance of next generation DNA sequencing technologies, we can expect to recover species-level phylogenies for an increasing number of ecosystems worldwide and evolutionary metrics will help predict which of those taxa are the potential invaders. This may well be the first tangible contribution to global conservation of current efforts for assembling the 'tree-of-life'.

Data: CORDIS, © European Union