SEARCH

SEARCH BY CITATION

FilenameFormatSizeDescription
gcb12283-sup-0001-FigS1-S8.pdfapplication/PDF2235K

Figure S1. Comparison of large-scale variations in the timing and length of the photosynthetic growing season showing that the results are not sensitive to the choice of the approach to extract phenology (see Methods section).

Figure S2. Comparison of variations in timing, length and intensity of the photosynthetic growing season between 1982 and 2006 based on the latest version of biweekly AVHRR NDVI data (GIMMS NDVI3g – green) and the earlier version of the dataset (GIMMS NDVIg – gray).

Figure S3. Comparison of variations in timing, length and intensity of the photosynthetic growing season between 2001 and 2011 based on 15-day AVHRR NDVI from the GIMMS NDVI3g data set (green) and on 16-day Terra MODIS NDVI from the MOD13C1 data set (gray).

Figure S4. Comparison of variations in the timing of spring snow melt based on NOAA weekly snow-cover (SMT – purple) with the timing of the SSM/I-based spring thaw (STH – black) and start of the thermal growing season (STS – orange) at latitudes north of 45°N.

Figure S5. Changes in spring and autumn number of nonfrozen days based on satellite microwave (SSM/I – black) and gridded surface air temperature observations (HadGHCND – orange) north of 45°N. Correlations between the series for the period 1988-2007 and linear trends (β) for the HadGHCND series over the 1950-2011 period are given in each panel.

Figure S6. Comparison of 10-year high-pass filtered variations in seasonally integrated NDVI (green) with seasonally integrated temperature (orange), and peak-to-trough amplitude of the annual cycle of atmospheric CO2 at Point Barrow (blue).

Figure S7. Comparison of parameters describing the amplitude and phase of the annual cycle of atmospheric CO2 at northern latitudes based on observations at Point Barrow (71°N – blue) and on estimates from the marine boundary layer matrix of GLOBALVIEW–2011 averaged north of 60°N (red).

Figure S8. Correlation between growing season integrated GIMMS NDVI3g (TI–NDVI) and peak-to-trough amplitude of the annual cycle of atmospheric CO2 at Point Barrow (black dot) from 1982 to 2010.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.