Can we predict which species win when new habitat becomes available?

Land cover change is a key component of anthropogenic global environmental change, contributing to changes in environmental conditions of habitats. Deforestation is globally the most widespread and anthropogenically driven land cover change leading to conversion from closed forest to open non-forest habitat. This study investigates the relative roles of geographic features, characteristics of species climatic niche and species traits in determining the ability of open-habitat plant species to take advantage of recently opened habitats. We use current occurrence records of 18 herbaceous, predominantly open-habitat species of the genus Acaena (Rosaceae) to determine their prevalence in recently opened habitat. We tested correlation of species prevalence in anthropogenically opened habitat with (i) geographic features of the spatial distribution of open habitat, (ii) characteristics of species climatic niche, and (iii) species traits related to dispersal. While primary open habitat (naturally open) was characterised by cold climates, secondary open habitat (naturally closed but anthropogenically opened) is characterised by warmer and wetter conditions. We found high levels of variation in the species prevalence in secondary open habitat indicating species differences in their ability to colonise newly opened habitat. For the species investigated, geographical features of habitat and climatic niche factors showed generally stronger relationships with species prevalence in secondary open habitat than functional traits. Therefore, for small herbaceous species, geographical features of habitat and environmental factors appear to be more important than species functional traits for facilitating expansion into secondary open habitats. Our results suggested that the land cover change might have triggered the shifts of factors controlling open-habitat plant distributions from the competition with forest trees to current environmental constraints.

A novel supermatrix approach improves resolution of phylogenetic relationships in a comprehensive sample of danthonioid grasses.

Phylogeny reconstruction is challenging when branch lengths vary and when different genetic loci show conflicting signals. The number of DNA sequence characters required to obtain robust support for all the nodes in a phylogeny becomes greater with denser taxon sampling. We test the usefulness of an approach mixing densely sampled, variable non-coding sequences (trnL-F; rpl16; atpB-rbcL; ITS) with sparsely sampled, more conservative protein coding and ribosomal sequences (matK; ndhF; rbcL; 26S), for the grass subfamily Danthonioideae. Previous phylogenetic studies of Danthonioideae revealed extensive generic paraphyly, but were often impeded by insufficient character and taxon sampling and apparent inter-gene conflict. Our variably-sampled supermatrix approach allowed us to represent 79% of the species with up to c. 9900 base pairs for taxa representing the major clades. A 'taxon duplication' approach for taxa with conflicting phylogenetic signals allowed us to combine the data whilst representing the differences between chloroplast and nuclear encoded gene trees. This approach efficiently improves resolution and support whilst maximising representation of taxa and their sometimes composite evolutionary histories, resulting in a phylogeny of the Danthonioideae that will be useful both for a wide range of evolutionary studies and to inform forthcoming realignment of generic delimitations in the subfamily.