RF PROPERTIES OF ICE AND SNOW-A REVIEW

igs_journal_vol05_issue042_pg773-792.pdf

DIELECTRIC PROPERTIES OF ICE AND SNOW-A REVIEW

By S. EVANs (Scott Polar Research Institute, Cambridge, England)

ABSTRACT. The permittivity and loss tangent of naturally occurring ice and snow are considered. The direct-current conductivity is considered only when it is of importance to the alternating-current and radio­frequency properties. Laboratory measurements on pure ice, and deliberately contaminated ice, are included to help in explaining and extrapolating the behavior of natural ice and snow. The lower band of frequencies from 10 Hz. to 1 MHz. is occupied by a relaxation spectrum in which the relative permittivity falls from approximately 100 to 3. The loss tangent reaches a maximum at a frequency which varies from 50 Hz. to 50 kHz. as the temperature increases from -60°C. to 0°C. We are interested in the effect of snow density, impurities, stress, crystal size, and orientation. For frequencies much greater than 1 MHz., the relative permittivity is 3.17±0.07. The loss tangent reaches a minimum value at approximately 1 GHz beyond which the dominant influence is infra-red absorption. The minimum is 10^-3 at 0°C. or 2 X 10^-5 ; at - 60 ° C. These values are greatly increased by impurities or free water. Some possible applications to glaciological field measurements are mentioned.

 

http://www.mike-willis.com/Tutorial/PF10.htm

Sleet and Snow

The refractive index of water has a strong temperature dependence. The loss of ice is much less than that of water at Radio frequencies, the imaginary component of the refractive index of ice is low and consequently Ice is fairly lossless. Clouds, which contain ice crystals become important for mm-wave satellite links. Although fairly lossless, Needles of ice in clouds can cause large differential phase shifts and hence may degrade the XPD. Sleet consists of wet snow flakes which effectively look like very large raindrops. As a result, sleet can cause very high losses.

As ice absorption is very low the main loss mechanism for ice is scattering, which can also enhance signals beyond the horizon. Below ~100GHz a Rayleigh model can be used as for rain attenuation and rain scatter. Above 100 GHz we need to account for both Rayleigh and Mie scattering.

Frequency (GHz)	A (Mie) dB per km per g/m3	A (Rayleigh) dB per km per g/m3
300	15	15
400	20.5	20.3
500	26	25
800	41	38
1000	50	43

http://seaice.alaska.edu/gi/publications/mahoney/Holt_2009_CRST_SeaIcePennRadar.pdf

Sea ice thickness measurements by ultrawideband penetrating radar: First results

Benjamin Holt a , ! , Pannirselvam Kanagaratnam b , Siva Prasad Gogineni c , Vijaya Chandran Ramasami c , Andy Mahoney d , Victoria Lytle

c aJet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Drive, Pasadena, CA, 91109, USA

b Benaforce Sdn. Bhd., Shah Alam 4000, Malaysia

c CReSIS, University of Kansas, 2335 Irving Hill Road, Lawrence, KS, 66045, USA

d National Snow and Ice Data Center, Boulder CO 80303, USA

A B S T R A C T

This study evaluates the potential of ultra wideband penetrating radar for the measurement of sea ice thickness. Electromagnetic modeling and system simulations were first performed to determine the appropriate radar frequencies needed to simultaneously detect both the top ice surface (snow – ice interface) and to penetrate through the lossy sea ice medium to identify the bottom ice surface (ice – ocean interface). Based on the simulation results, an ultra wideband radar system was built that operated in two modes to capture a broad range of sea ice thickness. The system includes a low-frequency mode that operates from 50 – 250 MHz for measuring sea ice thickness in the range of 1 to 7 m (both first-year and multiyear ice types) and a high-frequency mode that operates from 300 – 1300 MHz to capture a thinner range of thickness between 0.3 and 1 m (primarily first-year ice type). Two field tests of the radar were conducted in 2003, the first off Barrow, Alaska, in May and the second off East Antarctica in October. Overall the radar measurements showed a mean difference of 14 cm and standard deviation of 30 cm compared with in situ measurements over first-year ice that ranged from 0.5 to 4 m in thickness. Based on these initial results, we conclude that ultra wideband penetrating radar is feasible for first-year sea ice thickness measurements. We discuss approaches for further system improvements and implementation of such a system on an airborne platform capable of providing regional sea ice thickness measurements for both ! rst-year and multiyear ice from 0.3 to 10 m thick.