Changing Seasonality of Tundra Vegetation and Associated Climatic Variables

Uma Suren Bhatt1, Donald A Walker2, Martha K Raynolds3, Peter Bieniek1, Howard E Epstein4, Josefino C Comiso5, Jorge E Pinzon6, Compton J Tucker7, Michael Steele8, Wendy S Ermold9 and Jinlun Zhang10, (1)University of Alaska Fairbanks, Fairbanks, AK, United States, (2)University of Alaska Fairbanks, Alaska Geobotany Center, Fairbanks, AK, United States, (3)University of Alaska Fairbanks, Institute of Arctic Biology, Fairbanks, AK, United States, (4)University of Virginia, Department of Environmental Sciences, Charlottesville, VA, United States, (5)NASA Goddard Space Flight Center, Cryospheric Sciences Laboratory, Greenbelt, MD, United States, (6)NASA Goddard Space Flight Center, Greenbelt, MD, United States, (7)NASA Goddard Space Flight Center, Earth Science Division, Greenbelt, MD, United States, (8)Univ Washington, Seattle, United States, (9)Applied Physics Laboratory University of Washington, Kenmore, WA, United States, (10)University of Washington, Seattle, WA, United States
Abstract:
This study documents changes in the seasonality of tundra vegetation productivity and its associated climate variables using long-term data sets. An overall increase of Pan-Arctic tundra greenness potential corresponds to increased land surface temperatures and declining sea ice concentrations. While sea ice has continued to decline, summer land surface temperature and vegetation productivity increases have stalled during the last decade in parts of the Arctic. To understand the processes behind these features we investigate additional climate parameters.

This study employs remotely sensed weekly 25-km sea ice concentration, weekly surface temperature, and bi-weekly NDVI from 1982 to 2013. Maximum NDVI (MaxNDVI, Maximum Normalized Difference Vegetation Index), Time Integrated NDVI (TI-NDVI), Summer Warmth Index (SWI, sum of degree months above freezing during May-August), ocean heat content (PIOMAS, model incorporating ocean data assimilation), and snow water equivalent (GlobSnow, assimilated snow data set) are explored. We analyzed the data for the full period (1982-2013) and for two sub-periods (1982-1998 and 1999-2013), which were chosen based on the declining Pan-Arctic SWI since 1998.

MaxNDVI has increased from 1982-2013 over most of the Arctic but has declined from 1999 to 2013 over western Eurasia, northern Canada, and southwest Alaska. TI-NDVI has trends that are similar to those for MaxNDVI for the full period but displays widespread declines over the 1999-2013 period. Therefore, as the MaxNDVI has continued to increase overall for the Arctic, TI-NDVI has been declining since 1999. SWI has large relative increases over the 1982-2013 period in eastern Canada and Greenland and strong declines in western Eurasia and southern Canadian tundra. Weekly Pan-Arctic tundra land surface temperatures warmed throughout the summer during the 1982-1998 period but display midsummer declines from 1999-2013. Weekly snow water equivalent over Arctic tundra has declined over most seasons but shows slight increases in spring in North America and during fall over Eurasia. Later spring or earlier fall snow cover can both lead to reductions in TI-NDVI. The time-varying spatial patterns of NDVI trends can be largely explained using either snow cover or land surface temperature trends.

Back to: Changing Ecosystems of the Arctic and Antarctic Posters II