Comparing species-temperature relationships of arctic-alpine plants among three altitudinal transects in Svalbard

• Main Author: Bockmühl, Kathrin
• Format: Master Thesis
• Language: English
• Published: The University of Bergen 2008
• Subjects: Biodiversity; Evolution; Ecology; VDP::Matematikk og Naturvitenskap: 400::Basale biofag: 470; Arctic; Svalbard; Arctic Archipelago; Climate change
• Online Access: http://hdl.handle.net/1956/3262

Broad scale data has normally been used to predict vascular plant species responses to global climate change. The aim of this study is to investigate to what extent a high sampling resolution reveals local variations among species-temperature relationships in arctic regions. I recorded vascular plant communities along three altitudinal transects in the arctic archipelago Svalbard (78- 80 °N). All species in 25 m2 - plots were mapped using a simple abundance measure. I performed detrended correspondence analysis to explore how altitude was related to the variation in the species composition. Afterwards, the altitudinal gradients were converted into temperature gradients. Generalised linear models with a binomial distribution and a logit link were used to generate temperature response curves for 40 species and to detect similarities and differences among temperature relationships of 18 species at different sites. For most of the species clear relationships to temperature were found and many of them exhibited unimodal response curves. The comparison of the temperature response curves showed that only a few species responded consistently to temperature at the different study sites, while the majority revealed different relationships to temperature among the locations. This may indicate that populations of arctic species are locally adapted to environmental and climatic conditions. Further, my results show that altitude is even on low mountains a strong and useful tool to detect local species adaptations within arctic regions, which are characterised by strong climate variations. Hence, this study illustrates that small-scale data reveals local variations, which are overlooked when using broad scale data, but which supply important knowledge on how to optimise predictions for arcticalpine plants under global rising temperatures.