Cross-scale regulation of seasonal microclimate by vegetation and snow in the Arctic tundra

Climate warming is inducing widespread vegetation changes in Arctic tundra ecosystems, with the potential to alter carbon and nutrient dynamics between vegetation and soils. Yet, we lack a detailed understanding of how variation in vegetation and topography influences fine-scale temperatures (‘micro...

Full description

Bibliographic Details
Published in:Global Change Biology
Main Authors: Von Oppen, Jonathan, Assmann, Jakob Johann, Bjorkman, Anne D., Treier, Urs, Elberling, Bo, Nabe-Nielsen, Jacob, Normand, Signe
Format: Article in Journal/Newspaper
Language:English
Published: 2022
Subjects:
Online Access:https://pure.au.dk/portal/en/publications/ed103487-90d3-4eba-9046-6bb86c57155a
https://doi.org/10.1111/gcb.16426
https://pure.au.dk/ws/files/334712809/Global_Change_Biology_2022_Oppen.pdf
Description
Summary:Climate warming is inducing widespread vegetation changes in Arctic tundra ecosystems, with the potential to alter carbon and nutrient dynamics between vegetation and soils. Yet, we lack a detailed understanding of how variation in vegetation and topography influences fine-scale temperatures (‘microclimate’) that mediate these dynamics, and at what resolution vegetation needs to be sampled to capture these effects. We monitored microclimate at 90 plots across a tundra landscape in western Greenland. Our stratified random study design covered gradients of topography and vegetation, while nested plots (0.8 to 100 m2) enabled comparison across different sampling resolutions. We used Bayesian mixed-effect models to quantify the direct influence of plot-level topography, moisture and vegetation on soil, near-surface and canopy-level temperatures (-6; 2; and 15 cm). During the growing season, colder soils were predicted by shrub cover (-0.24 °C per 10% increase), bryophyte cover (-0.35 °C per 10% increase) and vegetation height (-0.17 °C per 1 cm increase). The same three factors also predicted the magnitude of differences between soil and above-ground temperatures, indicating warmer soils at low cover/height, but colder soils under closed/taller canopies. These findings were consistent across plot sizes, suggesting that spatial predictions of microclimate may be possible at the operational scales of satellite products. During winter, snow cover (+0.75 °C per 10 snow-covered days) was the key predictor of soil microclimate. Topography and moisture explained little variation in the measured temperatures. Our results underline the close connection of vegetation and snow with microclimate in the Arctic tundra, but also point to the need for more studies disentangling their complex interplay across tundra environments and seasons. Future shifts in vegetation cover and height will likely mediate the impact of atmospheric warming on the tundra soil environment, with potential implications for below-ground organisms and ...