Elevational species shifts in a warmer climate are overestimated when based on weather station data

TLDR: The results demonstrate that weather station data are unable to reflect the complex thermal patterns of aerodynamically decoupled alpine vegetation at the investigated scales, which might lead to misinterpretation and inaccurate prediction.

Abstract: Strong topographic variation interacting with low stature alpine vegetation creates a multitude of micro-habitats poorly represented by common 2 m above the ground meteorological measurements (weather station data). However, the extent to which the actual habitat temperatures in alpine landscapes deviate from meteorological data at different spatial scales has rarely been quantified. In this study, we assessed thermal surface and soil conditions across topographically rich alpine landscapes by thermal imagery and miniature data loggers from regional (2-km2) to plot (1-m2) scale. The data were used to quantify the effects of spatial sampling resolution on current micro-habitat distributions and habitat loss due to climate warming scenarios. Soil temperatures showed substantial variation among slopes (2–3 K) dependent on slope exposure, within slopes (3–4 K) due to micro-topography and within 1-m2 plots (1 K) as a result of plant cover effects. A reduction of spatial sampling resolution from 1 × 1 m to 100 × 100 m leads to an underestimation of current habitat diversity by 25% and predicts a six-times higher habitat loss in a 2-K warming scenario. Our results demonstrate that weather station data are unable to reflect the complex thermal patterns of aerodynamically decoupled alpine vegetation at the investigated scales. Thus, the use of interpolated weather station data to describe alpine life conditions without considering the micro-topographically induced thermal mosaic might lead to misinterpretation and inaccurate prediction.