Ecology of peatland testate amoebae in Svalbard and the development of transfer functions for reconstructing past water-table depth and pH

• Main Authors: Sim, TG, Swindles, GT, Morris, PJ, Baird, AJ, Charman, DJ, Amesbury, MJ, Beilman, D, Channon, A, Gallego-Sala, AV
• Format: Article in Journal/Newspaper
• Language: English
• Published: Elsevier 2021
• Subjects: Arctic; Svalbard; Svalbard Archipelago; Climate change; permafrost; Spitsbergen
• Online Access: https://eprints.whiterose.ac.uk/177418/; https://eprints.whiterose.ac.uk/177418/7/1-s2.0-S1470160X21007871-main.pdf

Peatlands are valuable archives of information about past environmental conditions and represent a globally-important carbon store. Robust proxy methods are required to reconstruct past ecohydrological dynamics in high-latitude peatlands to improve our understanding of change in these carbon-rich ecosystems. The High Arctic peatlands in Svalbard are at the northern limit of current peatland distribution and have experienced rapidly rising temperatures of 0.81 °C per decade since 1958. We examine the ecology of peatland testate amoebae in surface vegetation samples from permafrost peatlands on Spitsbergen, the largest island of the Svalbard archipelago, and develop new transfer functions to reconstruct water-table depth (WTD) and pH that can be applied to understand past peatland ecosystem dynamics in response to climate change. These transfer functions are the first of their kind for peatlands in Svalbard and the northernmost developed to date. Multivariate statistical analysis shows that WTD and pore water pH are the dominant controls on testate amoeba species distribution. This finding is consistent with results from peatlands in lower latitudes with regard to WTD and supports work showing that when samples are taken across a long enough trophic gradient, peatland trophic status is an important control on the distribution of testate amoebae. No differences were found between transfer functions including and excluding the taxa with weak idiosomic tests (WISTs) that are most susceptible to decay. The final models for application to fossil samples therefore excluded these taxa. The WTD transfer function demonstrates the best performance (R2LOO = 0.719, RMSEPLOO = 3.2 cm), but the pH transfer function also performs well (R2LOO = 0.690, RMSEPLOO = 0.320). The transfer functions were applied to a core from western Spitsbergen and suggest drying conditions ~1750 CE, followed by a trend of recent wetting and increasing pH from ~1920 CE. These new transfer functions allow the reconstruction of past peatland WTD and pH ...