Effects of climate warming on the distributions of invasive Eurasian annual grasses: a South African perspective

Threats posed by Eurasian annual grasses to ecosystem function have received little attention. Therefore, protocols for prioritising these alien annual species and likely future dimensions of their spread are urgently required. Here we modelled these grasses potential distribution and shifts in distribution ranges in South Africa under current and future climate scenarios. We applied a modelling framework (BIOMOD), which integrated a variety of parametric statistical and non-parametric rule based models to point distribution records of 29 invasive grass species. Correspondence between modelled and recorded distributions was calculated using the model accuracy criteria called the AUC (Area under the Curve). Based on this criteria 12 C₃ species were excellently modelled (AUC = 0.9-1), 11 C₃ species had good model accuracy (AUC = 0.7-0.8) and four C₃ and four C₄ species fell into the fair (AUC = 0.6-0.7) model accuracy class. Mean temperature of the coldest month was the strongest environmental parameter, for most of the alien grass distributions. Modelled distributions of the alien annual grasses projected into the future indicated range contractions in all C₃ species, except Briza minor, which were accompanied by shifts in species distribution ranges into higher altitudes. All C₄ species displayed habitat loss of relatively similar magnitude with climate warming and shifts in their distribution ranges also into higher elevations. These findings conclude that climate change will hinder the spread of European annual grasses in southern Africa. However, shifts in their distributions into pristine areas at higher elevations could pose a threat to the natural vegetation by altering fire regimes.