1
Climatology and variability of the ice-ocean-atmosphere- terrestrial system on the Yamal Uma Bhatt 1 , Donald A. Walker 2 , Martha K. Raynolds 2 , Howard E. Epstein 4 , Gensuo Jia 5 , Josefino C. Comiso 6 Jorge E Pinzon 6 , Compton J. Tucker 6 1 Geophysical Institute & Dept. Atmospheric Sciences at U. Alaska Fairbanks (UAF), 2 Institute of Arctic Biology at UAF, 4 Dept. of Env. Sci. at University of VA, 5 CAS, Beijing China, 6 NASA Goddard Space Flight Center Poster presented at the LCLUC Science Team Meeting, Bethesda, MD, 20-22 April 2010 Motivation and Methods Goal: Investigate the role of seasonality in current understanding of tundra-climate relationships New GIMMS-NDVI 3g for the Arctic Technical Information • The GIMMS data set uses a histogram matching approach with periodogram parameter regularization • To ensure spatial coherence, temporal consistency among the AVHRR NDVI record and ensure continuity for the coming missions: National Polar Orbiting Operational Environmental Satellite System (NPOESS) and Preparatory project (NPP). • An apparent 72N spatial discontinuity in the previous GIMMS-NDVIg product was fixed by using histogram matching from: • SeaWIFS instead of SPOT which doesn't collect data above 720N • May-September instead of January-December to avoid terminator effects leaking into the parameters. • An integration of NOAA-18 and NOAA-17 NDVI data. • NDVI3g is ready for distribution for all continents (for more information contact [email protected] ). NOAA 7 NOAA 9 NOAA 11 NOAA 14 NOAA 16 NOAA 17/18 NO9 SeaWIFS SPOT Terra/Aqua NPP 81 | 83 | 85 87 | 89 | 91 | | 93 | 95 | | 99 | 01 | 97 | 05 03 07 09 | | | 11 "#$ % & ’$ % & $ % ’$ % ( "#$ % ( ")$ % & *+ % , ’- % , .# % , )" % , /01 ,234 -1 !#$$ $ #$$ +$$ ’$$ )$$ "$$$ "#$ % & ’$ % & $ % ’$ % ( "#$ % ( ")$ % & *+ % , ’- % , .# % , )" % , /01 ,234 1 !#$$ $ #$$ +$$ ’$$ )$$ "$$$ NDVI g NDVI 3g NPOESS Data: Use 25 km resolution SSMI passive microwave Bootstrap Sea Ice Concentration (SIC), AVHRR Surface Temperature (T s ), and new GIMMS NDVI 3g for the Arctic over the 1982-2008 period. Methods: Standard climate trend and correlation analysis techniques applied to regional (Modified Treshnikov basins) time series of Maximum NDVI, Time Integrated NDVI, Summer Warmth Index, and sea ice concentration constructed using data within 50-km of Arctic coastlines (ocean & land) . Map delineating regions used in this research. Color highlights 50-km coastal regions. NDVI = (NIR-R)/(NIR+R) NIR: spectral reflectance in near-infrared band (0.725-1.1 μm) & R: red chlorophyll absorbing portion of spectrum (0.58-0.68 μm) Acknowledgements This study was supported by grants NSF ARC-0531180, NASA NNG6NE00A, NSF ANS-0732885, NSF ARC-0902175, NASA Land Cover Land Use Change on the Yamal Peninsula. References • Pinzon, J. E., E. Pak, C.J. Tucker, 2009 (submitted), A revised AVHRR 8-km NDVI Data Set - Compatibility with MODIS and SPOT Vegetation NDVI Data, American Geophysical Union EOS Transactions. • U.S. Bhatt, D.A. Walker, M.K. Raynolds, J.C. Comiso, H.E. Epstein, G.Jia, R. Gens, J.E. Pinzon, C.J. Tucker, C.E. Tweedie, and P.J. Webber, 2010 (in revision): Circumpolar Arctic tundra vegetation change is linked to sea-ice decline, Earth Interactions. Contrast N. America and Eurasia [Bhatt et al., 2010] -60 -40 -20 0 20 40 60 80 -60 -40 -20 0 20 40 60 80 W. Bering W. Chukchi E. Siberian Laptev E. Kara W. Kara Barents Grnlnd Sea Davis Str. Baffin Bay Can. Arch. Beaufort E. Chukchi E. Bering Eurasia N. America N. Hemis. SWI Trend Sea Ice Trend Percentage Change (1982-2008) Sea Ice Area Summer Warmth Index * * * * * * * * * * * * * * * * * * * * * * -8 -4 0 4 8 12 16 -8 -4 0 4 8 12 16 W. Bering W. Chukchi E. Siberian Laptev E. Kara W. Kara Barents Grnlnd Sea Davis Str. Baffin Bay Can. Arch. Beaufort E. Chukchi E. Bering Eurasia N. America N. Hemis. MaxNDVI TI-NDVI Percentage Change (1982-2008) Time Integrated NDVI Maximum NDVI * * * * * * * * * * * • Trends are larger in N. America • Variability is larger in Eurasia • North Slope: Largest magnitude NDVI & SWI increases • High Arctic: Largest % increases in MaxNDVI and SWI • Bering/W.Chukchi: NDVI decline • Eurasia: small summer warming N. Hemisphere -25% (3.2) 24% (1.4) 2% (0.075) 5% (0.010) Eurasia -28% (5.7) 16% (1.9) 2% (0.102) 2% (0.128) N. America -23% (3.4) 30% (1.6) 3% (0.073) 9% (0.010) 0 0.25 0.5 0.75 0 2 4 6 1985 1990 1995 2000 2005 MaxNDVI TI-NDVI 0 0.25 0.5 0.75 0 2 4 6 1985 1990 1995 2000 2005 MaxNDVI TI-NDVI 0 0.25 0.5 0.75 0 2 4 6 1985 1990 1995 2000 2005 MaxNDVI TI-NDVI 0 10 20 30 40 50 0 20 40 60 80 100 1985 1990 1995 2000 2005 Summer Warmth Index (˚C month) Sea Ice Concentration (%) 0 10 20 30 40 50 0 20 40 60 80 100 1985 1990 1995 2000 2005 Summer Warmth Index (˚C month) Sea Ice Concentration (%) 0 10 20 30 40 50 0 20 40 60 80 100 1985 1990 1995 2000 2005 Summer Warmth Index (˚C month) Sea Ice Concentration (%) 50-km region 0 20 40 60 80 100 1980 1985 1990 1995 2000 2005 2010 14-20May 28May-3Jun 11-17Jun 25Jun-1Jul 9-15Jul 16-22Jul 23-29Jul 30Jul-5Aug 6-12Aug 20-26Aug Year 50-km and full tundra domains similar 0 10 20 30 40 50 0 20 40 60 80 100 1985 1990 1995 2000 2005 Summer Warmth Index (˚C month) Sea Ice Concentration (%) -38% 4.2% 0 0.25 0.5 0.75 0 2 4 6 1985 1990 1995 2000 2005 MaxNDVI TI-NDVI 0.9% 2.1% • Sea ice (blue) shows significant decreases. • SWI (red) and NDVI (greens) display very small positive trend. -60 -50 -40 -30 -20 -10 0 10 20 -6 -4 -2 0 2 4 6 8 10 jan04 jan18 feb01 feb15 mar01 mar15 mar29 apr12 apr26 may10 may24 jun07 jun21 jul05 jul19 aug02 aug16 aug30 sep13 sep27 oct11 oct25 nov08 nov22 dec06 dec20 Ts std trend * 0 20 40 60 80 100 120 140 160 -40 -20 0 20 40 60 80 jan04 jan18 feb01 feb15 mar01 mar15 mar29 apr12 apr26 may10 may24 jun07 jun21 jul05 jul19 aug02 aug16 aug30 sep13 sep27 oct11 oct25 nov08 nov22 dec06 dec20 mean std trend * * * * ** * * * * * MaxNDVI Biweekly April-October Sea Ice weekly Surface Temp weekly -0.4 -0.2 0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 7_apr_a 8_apr_b 9_may_a 10_may_b 11_jun_a 12_jun_b 13_jul_a 14_jul_b 15_aug_a 16_aug_b 17_sep_a 18_sep_b 19_oct_a 20_oct_b Yamal Yamal-std Yamal-trend * * * * * * Weekly Sea Ice Trends & Variability Weekly T s Trends & Variability Bi-weekly NDVI Trends & Variability 0 200 400 600 800 1000 jan18 mar01 apr12 may24 jul05 aug16 sep27 nov08 dec20 80s var 90s var 00s var Yamal 0 20 40 60 80 100 jan18 mar01 apr12 may24 jul05 aug16 sep27 nov08 dec20 80s mean 90s mean 00s mean Yamal • Ice decreasing trends in spring and fall • 90’s spring earlier & 00s spring earlier and fall later • 90’s variability was higher from Mar-Aug & Sep- Nov 0 10 20 30 40 50 60 jan18 mar01 apr12 may24 jul05 aug16 sep27 nov08 dec20 Var 80s Var 90s Var00s Yamal -30 -20 -10 0 10 20 jan18 mar01 apr12 may24 jul05 aug16 sep27 nov08 dec20 80savg 90s avg 00's Yamal • T s is highly variable • 90’s spring was warmer • 90’s variability was higher from Feb-Jul 0 100 200 300 400 500 600 1_jan_a 3_feb_a 5_mar_a 7_apr_a 9_may_a 11_jun_a 13_jul_a 15_aug_a 17_sep_a 19_oct_a 21_nov_a 23_dec_a Var*10^5 80s Var *10^5 90s Var*10^5 00s Yamal 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 1_jan_a 3_feb_a 5_mar_a 7_apr_a 9_may_a 11_jun_a 13_jul_a 15_aug_a 17_sep_a 19_oct_a 21_nov_a 23_dec_a 80s 90s 00s Yamal • NDVI increase primarily in spring. • 00’s spring slightly earlier green up, • 90’s variability was higher from May-Aug Main Findings • Greening has not changed much on Yamal • Land temperatures have warmed slightly in fall and even cooled in spring • Sea ice shows significant declines with earlier melt and later freeze up. • On a weekly (biweekly) time scale the 1990s had the most variability (El Niño and Pinatubo?) Regional Trends Decadal Means Decadal Variance Decadal Means Decadal Variance Decadal Means Decadal Variance
