ECOSYSTEM ECOLOGY Natalie T. Boelman Marc Stieglitz Kevin L. Griffin Gaius R. Shaver Inter-annual variability of NDVI in response to long-term warming and fertilization in wet sedge and tussock tundra Received: 8 September 2004 / Accepted: 11 January 2005 / Published online: 12 April 2005 Ó Springer-Verlag 2005 Abstract This study explores the relationship between the normalized difference vegetation index (NDVI) and aboveground plant biomass for tussock tundra vegeta- tion and compares it to a previously established NDVI– biomass relationship for wet sedge tundra vegetation. In addition, we explore inter-annual variation in NDVI in both these contrasting vegetation communities. All measurements were taken across long-term experimental treatments in wet sedge and tussock tundra communities at the Toolik Lake Long Term Ecological Research (LTER) site, in northern Alaska. Over 15 years (for wet sedge tundra) and 14 years (for tussock tundra), N and P were applied in factorial experiments (N, P and N+P), air temperature was increased using greenhouses with and without N+P fertilizer, and light intensity was re- duced by 50% using shade cloth. during the peak growing seasons of 2001, 2002, and 2003, NDVI mea- surements were made in both the wet sedge and tussock tundra experimental treatment plots, creating a 3-year time series of inter-annual variation in NDVI. We found that: (1) across all tussock experimental tundra treat- ments, NDVI is correlated with aboveground plant biomass (r 2 =0.59); (2) NDVI–biomass relationships for tussock and wet sedge tundra communities are com- munity specific, and; (3) NDVI values for tussock tundra communities are typically, but not always, greater than for wet sedge tundra communities across all experi- mental treatments. We suggest that differences between the response of wet sedge and tussock tundra commu- nities in the same experimental treatments result from the contrasting degree of heterogeneity in species and functional types that characterize each of these Arctic tundra vegetation communities. Keywords Aboveground biomass Arctic tundra Normalized difference vegetation index Introduction Arctic tundra soils currently store 14% of the Earth’s terrestrial carbon (Oberbauer et al. 1996). While net carbon sequestration in tundra regions has been the norm over the last 10,000 years (Billings 1987; Gorham 1991; Oechel and Billings 1992), given the recent and ongoing climate warming (Oechel et al. 1993; Houghton et al. 1995; Keyser et al. 2000; Serreze et al. 2000), the net response of future carbon exchange between the vegetation, soil and atmosphere in Arctic tundra eco- systems is now uncertain. In addition, numerous experimental and modeling studies predict that these high northern latitude ecosystems are particularly sen- sitive to changing climate (Schlesinger and Mitchell 1987; Houghton et al. 1990; Maxwell 1992; Chapman and Walsh 1993; Curry et al. 1996; Myneni et al. 1997; Oechel et al. 1993; McKane et al. 1997; Stieglitz et al. 2000) and there is evidence suggesting that ecosystem form and function have been, and are predicted to continue responding to this warming in a variety of ways. The responses include changes in the direction and magnitude of CO 2 exchange between the soils and Communicated by Jim Ehleringer N. T. Boelman M. Stieglitz K. L. Griffin Lamont-Doherty Earth Observatory, Columbia University, 61 Route 9W, Palisades, NY 10964, USA Present address: N. T. Boelman (&) Department of Global Ecology, Carnegie Institution of Washington, 260 Panama Street, Stanford, CA 94305, USA E-mail: nboelman@globalecology.stanford.edu Tel.: +1-650-4621047 Fax: +1-650-4625968 M. Stieglitz Georgia School of Civil and Environmental Engineering and School of Earth and Atmospheric Sciences, Georgia Institute of Technology, Atlanta, GA 30332, USA G. R. Shaver Marine Biological Laboratory, The Ecosystems Center, Woods Hole, MA 02543, USA Oecologia (2005) 143: 588–597 DOI 10.1007/s00442-005-0012-9