Impact of ice temperature on microwave emissivity of thin newly formed sea ice

Abstract

[1] This study examines the impact of ice temperature on microwave emissivity over thin, newly formed sea ice at 6, 19, and 37 GHz during October 2003 in the southern Beaufort Sea, where the physical properties of newly formed sea ice were coincidently measured with microwave emissions. Six ice stations with distinct properties were selected and divided according to ice surface temperature into warm (above -3°C) or cold (below -3°C) stations. The warm stations had a lower emissivity at the vertical polarization by 0.1 than the cold stations and a corresponding difference in brine volume and dielectric properties. Significant correlations were observed between brine volume and ice emissivity (R 2 = 0.8, p value < 0.05). A sensitivity study showed that decreasing ice temperatures from À2.1°to À5.0°C explained the observed difference of 0.1 in ice emissivity between warm and cold stations. The results suggest that the temperature of thin bare ice could be the critical factor in determining ice emissivity near the melting point (about À2°C). Furthermore, a slight decrease in ice temperature (i.e., from À2°to À5°C) significantly reduces the brine volume, thus resulting in high ice emissivity. Finally, we demonstrate the potential of newly formed ice to cause errors in estimating sea ice concentrations using Advanced Microwave Scanning Radiometer-E data.

Helicopter-borne observations with portable microwave radiometer in the Southern Ocean and the Sea of Okhotsk

“…The mean values and their deviations from the helicopter-borne PR values and surface temperature values for these areas are consistent with those shown in past studies (e.g. Scambos and others, 2006; Hwang and others, 2008). Conversely, the MODIS results differ from the helicopter-borne TIR results in the thin-ice and open-water areas.…”

“…The PR 36 for thin ice of the Dalton Polynya with quite uniform surface temperature, and hence presumably ice thickness, is estimated to be 0.137 in October. This value agrees with previous in situ observations and theoretical modeling (Hwang and others, 2008), and again agrees with the AMSR-E thin-ice thickness algorithm for the Sea of Okhotsk of Nihashi and others (2009). Over the DIT, the helicopter-borne 36 GHz TBs are substantially different from those of the other surface types encountered, suggesting that a combination of vertically and horizontally polarized 36 GHz TBs can be used to distinguish ice typical of an Antarctic iceberg tongue from other surface types.…”

“…With these factors in mind, an algorithm was developed to give large-scale information on thin-ice thickness using satellite PMW data from the Special Sensor Microwave/ Imager (SSM/I: 1992–present) and Advanced Microwave Scanning Radiometer–Earth Observing System (AMSR-E: 2002–11; Tamura and others, 2007; Nihashi and others, 2009; Tamura and Ohshima, 2011; Iwamoto and others, 2013). This algorithm is based on the fact that although the microwave penetration depth of bare (thin) sea ice is in the order of 0.1 m at most (Ulaby and others, 1982), microwave brightness temperatures (TBs) at the frequencies of these sensors correlate with the surface salinity (brine volume fraction; Vant and others, 1978), which is sensitive to thin-ice thickness (Cox and Weeks, 1974; Kovacs, 1996; Toyota and others, 2007, Hwang and others, 2008). From in situ observation, Toyota and others (2007) showed that the brine volume fraction at the surface correlates well with ice thickness (root-mean-square error of brine volume fraction: 1.39%), especially for thin ice with thickness <0.5 m.…”

Differentiating between first‐year and multiyear sea ice in the Arctic using microwave‐retrieved ice emissivities

“…However, a certain degree of mismatch is found in the marginal ice zones such as the Bering Strait and the south coast of Greenland near the Denmark Strait, where positive signals exist on 1 January 2010 in spite of the first‐year sea ice shown in the NSIDC data. In those marginal ice zones, EVD is likely contaminated by newly formed thin ice and/or open water, in particular during the winter [ Hwang et al , ]. Since the AMSR‐E footprints for 10.65 GHz and 18.7 GHz are 29 km × 51 km and 16 km × 27 km, respectively, sea ice retrievals such as sea ice concentration in the marginal ice zones are likely to be subject to errors if there is pond‐like water area or open water within the footprint.…”

Retrieving the refractive index, emissivity, and surface temperature of polar sea ice from 6.9 GHz microwave measurements: A theoretical development

“…This is the area showing AMSR‐E‐derived sea ice concentration >95% and where the AMSR‐E product might be less accurate in this region [ Cavalieri et al ., ]. Furthermore, error may be due to the temperature dependence of emissivity and its impact on the sea ice determination over this newly formed thin sea ice area [ Hwang et al ., ]. The overall mean value of N r in January over the entire ice area over the Arctic is about 1.39.…”