Diurnal Scale Controls on C-Band Microwave Backscatter From Snow-Covered First-Year Sea Ice During the Transition From Late Winter to Early Melt

• Published in: IEEE transactions on geoscience and remote sensing Vol. 55; no. 7; pp. 3860 - 3874
• Main Authors: Yackel, John J., Gill, Jagvijay P. S., Geldsetzer, Torsten, Fuller, Mark Christopher, Nandan, Vishnu
• Format: Journal Article
• Language: English
• Published: New York IEEE 01-07-2017; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Active microwaves; Air temperature; Backscatter; Backscattering; C band; Case studies; Diurnal variations; Ice cover; Incidence angle; Mathematical models; Microwave FET integrated circuits; Microwave integrated circuits; Microwave scattering; Modelling; Ocean temperature; Radar; Radarsat; Regression analysis; SAR (radar); Sea ice; Sea surface; Snow; Snow cover; Snowpack; Surface temperature; Synthetic aperture radar; Temperature effects; Winter
• ISSN: 0196-2892; 1558-0644
• DOI: 10.1109/TGRS.2017.2682859

Diurnal observations of coincident in situ physical-, electrical-, and surface-based C-band microwave scattering properties of a 16-cm saline snow cover on smooth, moderately saline, first-year sea ice are presented for the transition period between late winter and early melt. Statistical regression analysis and backscatter modeling are employed to explore the scattering mechanisms within the snowpack and to assess associations between backscatter and snow properties for both periods. Our results demonstrate substantial variation in both measured snow properties and C-band microwave backscatter over the diurnal cycle during the late winter period when the difference between maximum and minimum air and snow surface temperature was approximately 5 °C. No such variation in snow properties and backscatter occurred for the early melt period when our case study exhibited a small diurnal variation (~1 °C) in air and snow surface temperature. Statistical and modeled results show significant association between the microwave backscatter and snow properties for the top 8 cm of the snowpack. Basal snow properties do not contribute toward total backscatter in either case. As a result, we are certain that the sea ice surface was a negligible scattering interface during both cases. Correlations between backscatter and snow properties are incidence angle dependent, demonstrating the highest association at 50°. Diurnal backscatter from RADARSAT-2 synthetic aperture radar acquisitions support the influence of varying diurnal snow properties on C-band backscatter, showing a difference of ~4 dB for σohhand σ vv o during the late winter period. This difference is reduced to <;1 dB for the early melt period.