Wade, R.H., 1993, Studies of the geophysics of sea ice: University of Alaska Fairbanks, Ph.D. dissertation, 135 p., illust., maps.
A non-linear growth model that solves the surface energy balance and heat conduction equations was developed to estimate thermal and physical properties of sea ice. The model incorporates several mechanisms that affect the salinity profile, including initial brine entrapment, brine expulsion, and gravity drainage, and is a non-linear extension of the model initially developed by Cox and Weeks (1988). Simulations were run to investigate the effects of the non-linear feedbacks which exist between the ice growth velocity and the thermal properties of the resulting ice. A comparison of the growth rate versus accumulated freeze-days was performed on the linear model, the non-linear model, and empirical formulas based on field observations. Allowing the model to run through the summer months with retarded ice growth and making an attempt at modelling summer desalinization processes produced second and third-year ice with proper temperature and salinity profiles. The ice growth model was then coupled to a Lambertian surface backscattering model for radar. By calculating the average dielectric constant of the penetration depth and using this value in the backscattering model, a comparison of the predicted signature variations in first-year sea ice was performed against observed backscattering values from ERS-1 SAR images of Dease Inlet, Alaska. The agreement between calculated and observed backscatter was surprisingly good considering that other factors may also influence radar returns. However, the more surprising result was the rescaling of the predicted first year ice signature by +6 dB produced a remarkable fit to observed backscattering values of multiyear ice. The predicted backscatter values and ice thicknesses were then used in conjunction with ERS-1 SAR imagery of the high Arctic to estimate areal coverage of the three major ice types in a 100 x 100 km2 area. Heat and mass flux calculations were then performed to produce daily estimates of energy loss and salt infusion for the winter months of October 1991 through March 1992.
Theses and Dissertations
