The fate of alpine plant diversity under climate warming is currently contentious. Species distribution models (SDMs) predict major upslope shifts, habitat loss and extinction risk of plants above the treeline. However, these models have been criticised for working on a too coarse spatial resolution. In fact, microclimatic variation can be pronounced in alpine terrain. This variation might allow species to follow a changing climate via retraction into spatially close, colder microsites instead of migrating upslope. However, whether accounting for this variation would reduce predicted range loss of alpine species is unclear as modelling habitat on a finer grain does not necessarily imply that projected habitat area or its change would differ.

A further disadvantage of SDMs, independent of the spatial resolution, is the that they are sensitive to a disequilibrium between species distribution and climatic conditions. Such a disequilibrium has likely emerged over the past four decades of rapid warming. As a consequence, using recent distribution data for model fitting might result in overestimation of warmth tolerance and hence in underestimation of future range loss.

In REFINE-Alps, we want to tackle both of these issues in a modelling study focused on ca. 100 alpine plant species in the central parts of the Austrian Alps. We will model changes of potential ranges of these species under three scenarios of climate change until the end of the century at a spatial resolution of 1 m2 across an area of about 500 km2 of above-treeline landscapes. We will therefore use both species distribution models and demographic distribution models (DDMs). The latter models are based on correlations of climatic conditions with vital rates which are likely responding to climatic changes much more rapidly than spatial distribution patterns. They should hence not be affected by disequilibrium problems.

The applicants have already collected part of the necessary species distribution, demographic and environmental data in other projects and will expand and adapt this large dataset to the needs of this project during the runtime of REFINE-Alps. Modelling results will be compared among finescale SDMs and SDMs fit on a coarser resolution (100 x 100m), and between fine-scale SDMs and DDMs. We will thereby specifically tackle the following questions: (1) How much of their currently suitable range will alpine plants lose under moderate, intermediate and strong climate change until the end of the century according to SDMs fitted and projected on a 1m2-resolution? (2) Does the alpine flora appear significantly more vulnerable when models fit and projected at coarser spatial resolutions (100 x 100 m)? (3) Do DDMs predict more pronounced losses than SDMs? (4) Do differences between SDM and DDM projections depend on species traits especially those related to mobility and persistence, and thus implicitly to the degree of inertia in current species distributions?

Finally, we will overlay predictions from both SDMs and DDMs to identify major refugia of the modelled species under climatic scenarios. The resulting map shall support ‘climate-smart’ conservation-related land management in the area. To guarantee that project products are useful for that purpose workshops with relevant regional stakeholder shall allow for project co-design, discussion and transfer of results.