| In this thesis, the spatial distribution and temporal evolution of snow depth over sub- Arctic landfast first-year sea ice in Hudson Bay, having smooth and rough microscale surface texture forms, was characterized and modeled for the late winter-to-early spring transition period prior to sea ice melt. Quantitative and qualitative in situ meteorological, snow depth, and sea ice surface roughness data were acquired and analyzed with descriptive and nonparametric statistics, bivariate linear regression, h-scatterplots, experimental omnidirectional semivariograms, and kriging. Snow cover was characterized as complexly layered. Thin/thick snow was observed over smooth/rough microscale sea ice surfaces, but were not directly correlated. Snow depth was highly variable over spatial and temporal scales, as were spatial autocorrelation decay distances. Snow depth predictions through semivariogram and kriging models were accurate for areas exhibiting gradual snow thickness fluctuations and that were densely sampled. Suggestions for future snow depth modeling were presented. |
