5.5 Seasonality of climate parameters that characterize Arctic tundra vegetation

Wednesday, 4 May 2011: 9:30 AM
Rooftop Ballroom (15th Floor) (Omni Parker House )
Uma Bhatt, University of Alaska, Fairbanks, AK; and D. Walker, M. Raynolds, P. Bieniek, J. Comiso, J. Pinzon, and C. J. Tucker

An increase of Arctic tundra vegetation greenness has been documented using the remotely sensed Normalized Difference Vegetation Index (NDVI) dataset (a measure of vegetation photosynthetic capacity). Previous work has shown coherent variability between NDVI, coastal sea ice cover and land surface temperatures (defined as the sum of the degree months above freezing during May-August or the summer warmth index, SWI). The goal of this paper is to understand the forcing factors of this change and variability better through an analysis of the seasonality of sea ice, summer tundra land-surface temperatures, and NDVI.

This study uses weekly 25-km resolution sea ice concentration from Special Sensor Microwave Imager (SSM/I) data and Advanced Very High Resolution Radiometer (AVHRR) radiometric surface temperature from 1982 to 2010. The bi-weekly NDVI data set (1982-2010) has been recently corrected (Pinzon et al. 2011, in progress) for biases that were revealed by our seasonality analysis. In addition, standard climate data (station, reanalysis, and model data) and ground observations of arctic vegetation are used in this study to synthesize results.

For overall trends, we find that during the period of satellite observations (1982-2010) summer time open water area has increased most in the Beaufort, Chukchi, Laptev, and Kara Seas. The largest increases in SWI are found over land areas adjacent to the Beaufort and Chukchi and in the High Canadian Arctic. NDVI increases have been largest on the North Slope of Alaska and along the Laptev Sea. The Taimyr and Yamal display NDVI decreases consistent with SWI declines over these areas. Caution must be exercised when analyzing trends in these parameters as they are characterized by large interannual to decadal variability. Time series analysis of NDVI and SWI show a notable dip during 2009 over the tundra from Northern Canada to the Laptev Sea. While still speculative, this may be linked to volcanic activity at high latitudes during spring of 2009. The analysis of the weekly and bi-weekly data suggests that the largest declines in near coastal (within 100-km) sea ice are in the spring and the fall (earlier breakup and later freeze-up). NDVI has increased most in spring in Eurasia and more during the peak vegetation activity (July) over North America.

The trends, variability and seasonality findings all suggest that the role of local atmospheric circulation as well as other local factors (i.e. soil types) likely plays an important role in vegetation productivity.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner