Abstract: Sea ice is melting rapidly in the Arctic region, and further monitoring of the ice situation in the polar region is critical. The dielectric constant is an important parameter for remote sensing monitoring of sea ice. This study utilizes a cryogenic laboratory to investigate the relationship between the dielectric and physical properties of sea ice and quantitatively analyze the dominant factors affecting changes in the dielectric constant of sea ice under different conditions. The air temperature of the polar region is simulated to test the dielectric and physical properties of artificially frozen brine ice within a frequency range of 80 kHz~50 MHz. The investigation is focused on changes in the dielectric constant of brine ice with frequency, ice temperature, and salinity with increasing temperature (from −45 ℃ to −15 ℃), and the effects of different factors on the dielectric constant are assessed using correlation analysis. The results show that the dielectric constant of brine ice is positively correlated with ice temperature and salinity, and increases by ~1.5 for every 10 ℃ increase in ice temperature. Furthermore, it is negatively correlated with frequency, with higher salinity lowering the rate of change. Finally, the test results are used to establish a dielectric constant prediction model for pure ice and brine ice, yielding R² values of 0.75 and 0.68, P-values of 0.005 and 0.008, root mean square error values of 1.1 and 3.4, and residual sum of squares values of 40 and 127, respectively. The model fitting effect is improved, the regression effect is significant, and the accuracy requirements are met. This indicates that it is feasible to invert the dielectric constant of polar sea ice through physical properties, and that polar sea ice can be monitored via the relationship between its physical and dielectric properties.

Key words: saline ice, dielectric properties, low and medium frequencies, physical properties, predictive model, Arctic sea ice