Spatial and temporal patterns of net carbon exchange in the polar semi-desert vegetation type on Melville Island, Nunavut

Abstract

While studies of mesic tundra vegetation types across latitudinal gradients have shown decreasing levels of net ecosystem exchange (NEE) of carbon dioxide at more northern sites, little work has explored the factors regulating NEE in the polar semi-desert, a vegetation type which is widely distributed across the High Arctic. In 2013, eight ADC Automated Carbon Exchange (ACE) systems were deployed at the Cape Bounty Arctic Watershed Observatory (CBAWO) to quantify the contribution of the polar semi-desert vegetation type to landscape-scale NEE. As polar semi-desert vegetation cover varies at relatively small spatial scales, the chambers were distributed between vegetated areas (18-51% cover) and bare soil. Measurements were made every 30 minutes from late May to late July. Air temperature, soil temperature, and soil moisture measurements were collected in conjunction with NEE and ecosystem respiration (Re). In July 2013, Normalized Difference Vegetation Index (NDVI) data were collected to quantify variability in vegetation cover within the polar semi-desert. NDVI varied from -0.12 to 0.31, with the highest values occurring at vegetated sites and low values occurring on bare soil. Percent vegetation cover and NDVI correlated well at peak biomass (R2 = 0.96). NEE is driven by variability of several biophysical factors, and the factors that best predict NEE vary throughout the season. In the early season, respiration drives NEE, and air temperature is the strongest predictor (R2 = 0.23 to 0.55). During the warmer part of the season, photosynthesis is the dominant component of NEE, and photosynthetically active radiation becomes the best predictor. Results suggest that once a threshold temperature is reached photosynthesis will dominate NEE in polar semi-desert vegetation types. Therefore, longer growing seasons, if associated with higher temperatures, would enhance NEE. These relationships may be useful for quantifying NEE in polar semi-deserts using remotely-sensed data in future studies.