The effects of long‐term experimental warming on the structure of three High Arctic plant communities

Abstract

AbstractQuestionsAre warming trends in High Arctic tundra consistent across plant communities within a given site, or is there evidence for fine‐scale heterogeneity in the response of plant community structure to experimental warming? How has the composition, cover and canopy height of three High Arctic plant communities responded to 18 yr of experimental warming? To what extent can hand‐held remote sensing methods serve as a proxy for community structure in tundra ecosystems?LocationAlexandra Fiord lowland, Ellesmere Island, Canada.MethodsWe experimentally warmed patches of tundra using the International Tundra Experiment (ITEX) open‐top chambers to simulate climate change. We measured differences in community structure after 18 yr of experimental warming in three distinct plant communities: a wet sedge meadow, a mesic evergreen shrub‐dominated community and a dry deciduous shrub‐dominated community. We measured plant cover, height and growth form composition using a point‐intercept method and investigated the use of hand‐held NDVI at the plot scale as a proxy for plant community structure.ResultsAll three plant communities showed unique differences after long‐term experimental warming. Both the meadow and evergreen shrub communities had significant differences in community composition between control and warmed plots, but none of the communities showed differences in plant cover or height. Deciduous shrubs and bryophytes were more abundant in warmed over control plots at both the evergreen shrub and meadow communities. Warmed plots in the meadow community had fewer graminoids and lower height values. Lichen was consistently less abundant in warmed plots across all communities where it occurred. NDVI values were driven by deciduous shrubs, and graminoids and showed community‐specific correlations with plant cover.ConclusionsOur results suggest community‐level resistance to warming in terms of plant cover and height, but compositional differences support the hypothesis that climate change responses will be community‐specific. Plot‐scale studies such as this one can reveal important details that contribute to the understanding and validation of coarser‐scale measurements from satellite data.