Topographic heterogeneity explains patterns of vegetation response to climate change (1972–2008) across a mountain landscape, Niwot Ridge, Colorado

ABSTRACT

The distributions of biomes worldwide are predicted to shift as vegetation tracks climate change. Ecologists often use coarse-scale climate models to predict these shifts along broad elevational and latitudinal gradients, but these assessments could fail to capture important dynamics by ignoring fine-scale heterogeneity. We ask how the elevational ranges of vegetation types have changed in a mountainous landscape, and investigate the influence of fine-scale topographic, snowpack, and soil properties on vegetation change. We manually classified vegetation from high-resolution repeat aerial photographs from 1972 and 2008 at Niwot Ridge, Colorado, USA, and generally found that trees and shrubs colonized tundra, while tundra colonized barren soils. Only shrubs expanded their elevational range. Several fine-scale topographic, soil and snow characteristics, including elevation, slope, solar radiation, soil bulk density, and interannual snowpack variability, modulated where plant establishment occurred. Each vegetation type had a unique suite of variables best predicting its establishment in new areas. We suggest that fine-scale heterogeneity may strongly control how plants in mountainous regions respond to climate change, and different vegetation types may be sensitive to different aspects of this heterogeneity. An improved understanding of the factors controlling vegetation change gives us a broader understanding of ecosystem response to climate change, nitrogen deposition, and release from grazing.

KEYWORDS:

• Climate change
• alpine vegetation
• topography
• snowpack
• solar radiation

Acknowledgments

Logistical support was provided by the Niwot Ridge LTER program and the CU Mountain Research Station. We thank Dr. William D. Bowman and Dr. Jane G. Smith for helpful suggestions about our fieldwork, and Dr. Lauren M. Hallett for suggestions on the article.

Additional information

Funding

We thank the Niwot Ridge LTER program (NSF DEB 1027341) for funding CPB, the NSF REU program (NSF DBI 1460906) for funding LST, CDB and KCR, and NSF EAR 1141764 for funding NPM. Publication of this article was funded by the University of Colorado Boulder Libraries Open Access Fund.