Evaluation of high-latitude boreal forest growth using satellite-derived vegetation indices

• Main Author: Berner, Logan T.
• Format: Text
• Language: unknown
• Published: Western Washington University 2010
• Subjects: Arctic; Canada
• Online Access: https://dx.doi.org/10.25710/94sh-ky60; https://cedar.wwu.edu/wwuet/42

Vegetation in northern high-latitudes plays an important role in energy exchange and carbon dynamics, thereby influencing regional and global climate. Vegetation indices derived from the space-borne Advanced Very High Resolution Radiometers (AVHRR) have suggested decreased photosynthetic activity during recent decades within some continental regions of the pan-arctic boreal forests. The purpose of this research was to determine associations between the normalized difference vegetation index (NDVI), as derived by both AVHRR and Moderate Resolution Imaging Spectroradiometers (MODIS), and inter-annual variations in radial stem growth in high-latitude coniferous forests. During 2008 and 2009, tree core samples were collected at 12 sites in northeast Russia and at 10 sites in northwest Canada. Ring-width indices (RWI; n = 27) were generated for larch, spruce, and pine genera and these were correlated with summer NDVI derived from the AVHRR sensors over the 1982 to 2008 period. The correlations between NDVI and RWI were then examined between 2000 and 2008 using both MODIS and AVHRR. The sensors showed similar abilities to proxy radial growth and NDVI-RWI correlations appeared mostly insensitive to changes in MODIS grain sizes between 250 m and 24 km. Over the 27 year period RWI and NDVI showed positive, though variable, correlations (r = 0.43 ± 0.19, n = 27). For pine and spruce, both evergreen conifers, the annual rate of radial growth was significantly correlated with growth during previous years, as was canopy development, as proxied by NDVI. Larch, however, did not show year to year persistence in either radial growth or canopy development, a finding that points to differences in growth patterns between functionally-distinct tree genera. These findings suggest that negative trends in NDVI may reflect decreased radial growth at some locations and that attempts to model tree growth and carbon uptake using NDVI need to take into account multi-year persistence in tree growth. Additionally, the work shows similarities between AVHRR and MODIS, suggesting potential to bridge the historical AVHRR record with the newer and finer resolution MODIS record.