Summary: | Thesis (Master, Geography) -- Queen's University, 2016-01-28 16:55:12.415 Wet sedge meadows are the most productive vegetation communities in the High Arctic. Preliminary research suggests that this vegetation type is a net carbon sink, yet the controls – and the scale at which those controls act – are not well understood. If warming of the High Arctic enhances or limits wet sedge growth, we may observe changes in the percentage of land area occupied by these meadows, resulting in significant alterations to the carbon balance of high arctic landscapes. Here, the factors controlling carbon dioxide (CO2) exchange of wet sedge meadows are examined at different spatial and temporal scales and environmental data is used to create predictive models of CO2 exchange. Automated and static CO2 exchange systems recorded CO2 exchange at three wet sedge sites at the Cape Bounty Arctic Watershed Observatory (CBAWO), Melville Island, NU, from June to August, 2014. In conjunction, time-series normalized difference vegetation index (NDVI) data were collected to quantify the phenological stage of the wet sedge vegetation type through the growing season, and soil temperature, air temperature, photosynthetically active radiation (PAR), soil moisture, and active layer depth were measured. Net ecosystem exchange (NEE) measurements indicated dominant plant uptake through photosynthesis, and spectrally separable ‘wet’ and ‘dry’ sedge areas yielded significantly different NEE values at both sampling scales. NDVI measurements indicated that spring greening and peak summer biomass differed between wet and dry areas, but that NDVI was not strongly related to CO2 exchange trends in these systems. Abiotic factors such as air and soil temperature and soil moisture – varying over time and space throughout the season – influenced CO2 exchange to varying degrees at each scale. Predictive models of ecosystem carbon flux were created using NDVI in combination with environmental measurements as predictors. This facilitated an evaluation, at two scales, of the drivers of CO2 exchange in these communities – both spatially and temporally. Static chamber measurements (bi-weekly) were unsuccessful in modelling CO2 exchange, but autochamber measurements (half-hourly) provided reasonable predictions. I suggest, though, that linear multivariate-regression models are insufficient for capturing variation in these systems, and that more complex models may provide greater success in the future. M.Sc.
|