| Summary: | Thesis (Master, Geography) -- Queen's University, 2015-09-21 12:00:18.185 Recent climate warming has resulted in substantial changes in vegetation production at high northern latitudes (above 60°N). Understanding the direction, magnitude, and rate of these changes warrants investigation given that tundra and boreal biomes play important roles in global energy partitioning and carbon sequestration. Satellite-based measures of aboveground production have indicated both positive and negative trends across tundra and boreal biomes over the last 30 years, respectively. The purpose of this research was to examine the relationship between the remotely-sensed normalized difference vegetation index (NDVI), a widely used indicator of primary production, and interannual variation in ring-widths of dominant coniferous and deciduous species situated in a boreal mountainous region of southwest Yukon, Canada. Trends in maximum annual NDVI from 1990 to 2013 were generated from two Advanced Very High Resolution Radiometer (AVHRR) datasets: The Alaska Composite NDVI (1km), and the third generation Global Inventory Modeling and Mapping Studies (GIMMS3g) NDVI (8km). Results from the NDVI trend analysis indicated a dominant low-level greening trend across the study region. Twenty 1km2 sites, divided equally between those undergoing “high greening” (NDVI/yr > 0.004) and “low greening” (NDVI/yr 0 – 0.003) trends were selected for dendrochronological sampling. During the summer of 2014, the dominant coniferous and deciduous species were sampled at each site. Annual rings were measured and standardized, yielding two chronologies (coniferous and deciduous) for each site. Ring-width trends varied in direction and magnitude for both coniferous and deciduous species. Agreement between NDVI trends and ring-width trends was low; however, eight coniferous site chronologies showed significant positive correlation (p < 0.05) with GIMMS3g NDVI during the growing season. Our findings suggest that while trends in maximum annual NDVI do not ...
|
|---|