Aim To evaluate the effect of post-glacial migration lags on the current distribution of Alpine plants and the factors responsible for possible range-filling differences among species.
Location Austrian Alps.
Methods We used species distribution models to predict environmentally suitable sites for 183 Alpine plants at a fine spatial resolution (100 × 100 m2). We overlaid these predictions with independent mapping data (3′× 5′) and calculated the extent to which species fill their potential ranges at this coarser grain based on several different approaches. Moreover, we correlated range-filling estimates with the magnitude of improvement of distribution models when using the distance to putative glacial refugia as an additional independent variable. Finally, we compared species-specific range-filling estimates with traits related to dispersal capacity and competitive ability of these species as well as with characteristics of their habitats.
Results Even under a conservative approach, incomplete range filling appears common, with 46% and 31% of the species studied occurring in less than 75% and 50% of their predicted suitable ranges, respectively. Proximity to glacial refugia generally accounts for a lower percentage of the deviance in species distribution data (0–20%, mean 4%) than environmental variables (9–57%, mean 27%). However, its importance correlates closely and negatively with the calculated range-filling estimates. Range filling significantly increases with the dispersal capacity of a species' propagules and the breadth of its altitudinal niche. Calcicolous species have lower range filling than silicicolous plants and substrate generalists.
Conclusions Our results suggest that the current ranges of many Alpine plants are still shaped by delayed Holocene recolonization of suitable sites. Hence, long-term migration lags also affect plant distribution in mountainous areas, at least on regional scales. These findings question whether high mountain floras will be able to track the expected rapid, climate change driven shifts in habitat.