From the species perspective: the contribution of a fine scale microtopography analysis for the conservation of alpine bryophytes

Abstract

Ecosystems and their biodiversity are under pressure from many threats. One way to predict their impact on species is to use Species Distribution Models (SDM). However, most SDMs use macroclimatic variables measured by weather stations 2 m above the ground and macro-scale variables with a spatial resolution of 1 km, far larger than the area occupied by species. Especially in the mountains, due to the complex topography, and for small species sensitive to microclimatic variations, a large difference exists between the measured temperature and that experienced by the species at their scale, which could lead to erroneous predictions in SDMs. This Master Thesis aims to i) compare the accuracy of the distribution of alpine bryophytes sampled on two summits in Valais (Switzerland), between 3 SDMs based on microtopographic variables: the first model using only 1 cm resolution variables, the second using 2 m resolution variables and the last using both variables at 1 cm and 2 m resolution, ii) predict, for the snow-free period, thanks to microtopographic variables at 1 cm or 2 m resolution, the offset in mean temperature and relative humidity between those measured by ERA-5 (spatial resolution of 31 km) and those measured by 71 loggers placed on the two summits from September 2023 to June 2024. The results show that i) the best model is the one using only microtopographic variables at 2m resolution (AUC = 0.7 and TSS = 0.37), ii) the relative humidity offset, and especially the temperature offset, are well predicted by microtopographic variables at 1 cm resolution (R²=0.23 and R²=0.43, respectively, for the fixed effects). Although this study could be extended in time and space, it supports the idea that fine-scale variables can be used in SDMs, which is useful for identifying microrefugia in mountains and thus could contrast the ability of alpine bryophyte to survive under climate change.