Future distribution of invasive weed species across the major road network in the state of Montana, USA

Invasive plant species are a significant global problem, with the potential to alter structure and function of ecosystems and cause economic damage to managed landscapes. An effective course of action to reduce the spread of invasive plant species is to identify potential habitat incorporating changing climate scenarios. In this study, we used a suite of species distribution models (SDMs) to project habitat suitability of the eleven most abundant invasive weed species across road networks of Montana, USA, under current (2005) conditions and future (2040) projected climates. We found high agreement between different model predictions for most species. Among the environmental predictors, February minimum temperature, monthly precipitation, solar radiation, and December vapor pressure deficit accounted for the most variation in projecting habitat suitability for most of the invasive weed species. The model projected that habitat suitability along roadsides would expand for seven species ranging from + 5 to + 647% and decline for four species ranging from − 11 to − 88% under high representative concentration pathway (RCP 8.5) greenhouse gas (GHG) trajectory. When compared with current distribution, the ensemble model projected the highest expansion habitat suitability with six-fold increase for St. John's Wort (Hypericum perforatum), whereas habitat suitability of leafy spurge (Euphorbia esula) was reduced by − 88%. Our study highlights the roadside areas that are currently most invaded by our eleven target species across 55 counties of Montana, and how this will change with climate. We conclude that the projected range shift of invasive weeds challenges the status quo, and requires greater investment in detection and monitoring to prevent expansion. Though our study focuses across road networks of a specific region, we expect our approach will be globally applicable as the predictions reflect fundamental ecological processes.