Temperature and salinity distribution of sea ice cover according to experimental and model data (case study of Novik Bay of the Sea of Japan)

• Published in: Ice and Snow
• Main Authors: A. N. Chetyrbotsky, A. U. Lazaryuk
• Format: Article in Journal/Newspaper
• Language: Russian
• Published: Nauka 2018
• Subjects: модель распределения солёности; модель распределения температуры; морской ледяной покров; оценка адекватности моделей; Science; Q; Sea ice
• Online Access: https://doi.org/10.15356/2076-6734-2018-4-559-568; https://doaj.org/article/348c2fb1b7554b88a3d7b734c806a6b2

Analysis of sample distributions of temperature and salinity within depths of sea ice allowed revealing a high negative correlation between temperature of the surface air layer and the salinity vertical distribution across the ice thickness. Tis situation is explained by the fact that when temperature inside the ice rises the vertically oriented pores flled with brine, and this causes increased flow of brine. But when the temperature of the thickness drops, volumes of these pores signifcantly decrease, and as a result of that stresses near the pores grow and brine is squeezed out to both above and under the ice. Comparison of individual cases of the sample salinity distributions made possible to determine that the temperature of the surface air layer signifcantly influences the freezing intensity. When developing a model of spatiotemporal dynamics of the temperature, the diffusion mechanism of its vertical distribution is adopted, where the thermal conductivity coefcient is a linear function of temperature. A computational scheme for solving the model equations had been developed. Te procedure to estimate the model parameters is given. Te results of the parameter estimations had proved the adequacy of both the sample and the model distributions. A degree of the adequacy is the correlation coefcient between the above distributions. It is shown that the numerical simulation of the spatiotemporal salinity dynamics can be performed in framework of the diffusion mechanism of the vertical distribution, where the diffusion coefcient is a linear function of temperature. Te results of the parameter estimations did also show the adequacy of both the sample and the model distributions.