Phylogenetic relatedness and plant invader success across two spatial scales

Aim  Successful invaders often possess similar ecological traits that contribute to success in new regions, and thus under niche conservatism, invader success should be phylogenetically clustered. We asked if the degree to which non-native plant species are phylogenetically related is a predictor of invasion success at two spatial scales.
Location  Australia – the whole continent and Royal National Park (south-eastern Australia).
Methods  We used non-native plant species occupancy in Royal National Park, as well as estimated continental occupancy of these species from herbarium records. We then estimated phylogenetic relationships using molecular data from three gene sequences available on GenBank ( matK , rbcL and ITS1 ). We tested for phylogenetic signals in occupancy using Blomberg's K .
Results  Whereas most non-native plants were relatively scarce, there was a strong phylogenetic signal for continental occupancy, driven by the clustering of successful species in Asteraceae, Caryophyllaceae, Poaceae and Solanaceae. However, we failed to detect a phylogenetic signal at the park scale.
Main Conclusions  Our results reveal that at a large spatial scale, invader success is phylogenetically clustered where ecological traits promoting success appear to be shared among close relatives, indicating that phylogenetic relationships can be useful predictors of invasion success at large spatial scales. At a smaller, landscape scale, there was no evidence of phylogenetic clustering of invasion success, and thus, relatedness plays a much reduced role in determining the relative success of invaders.