In-situ measured spectral directional emissivity of snow and ice in the 8–14 μm atmospheric window

Dependence of thermal infrared emissive behaviors of snow cover on the surface snow type

“…The effects of the different surface types on BTs were examined by exchanging ground reflectances (i.e., 1-emissivity) which were used as a boundary condition in this RT simulation. The directional emissivity spectra of snow as the boundary condition were employed from those measured by Hori et al (2006), the spectral data of which were taken from the Japan Aerospace Exploration Agency (JAXA) in situ data archive for the GCOM mission (2012), and covered types of fine snow (in situ measured median diameter of snow grains by Hori et al (2006) of d 70 m), medium melt forms (d 600 m), coarse melt forms (d 800 m), sun crust (d 1100 m), and bare smooth ice. A measured emissivity spectrum of surface hoar is available at only the near-nadir exitance angle of 10 in the ASTER spectral library (1999); thus, BT signals of surface hoar were calculated under the assumption that the measured emissivity at the near-nadir exitance angle was valid at all exitance angles of 0-60 .…”

“…Wald (1994) proposed several theoretical approaches for simulating snow-type dependent snow emissivity and pointed out that cementation effects, that is, whether snow is welded or disaggregated, are important for modeling the emissivity of snow. Hori et al (2006) demonstrated the angular and snow-type dependences of snow emissivity by in situ observations. Recently, Hori et al (2013) succeeded in modeling the measured in situ emissivity spectra of five different snow types with various grain sizes using a semi-empirical approximation approach and found that areal fraction of snow grains with specular facets being exposed to the whole sky are key to characterize the spectral snow emissivity.…”

“…For example, snow emissivity has been assumed to be represented by Article 34 a fixed-spectrum model among various snow types and not to exhibit angular dependence in the retrieval of surface temperature (e.g., Hall et al, 2008). However, as Hori et al (2006) demonstrated by in situ observations, snow emissivity varies depending on wavelength, snow type and exitance angle. Thus, the dependence of snow emissivity on snow type and exitance angle has to be taken into account in an analysis of the Earthʼs radiative balances over the polar region (Dybkjaer et al, 2012).…”

“…Hence, snow-type dependent emissivity produces a brightness temperature difference (BTD), for example, between MODIS channels 31 ( 10.8 m) and 32 (12.0 m), that depends on the surface snow type. Thus, snow-type dependent emissivity is potentially usable not only for enhancing the accuracy of surface temperature retrieval but also for discriminating snow types (e.g., Tonooka and Watanabe, 2005;Hori et al, 2006;Hori et al, 2013). As mentioned above, shortwave reflectances at nearinfrared (NIR) to shortwave infrared (SWIR) wavelengths around 0.8-1.6 m have been primarily used for discrimination of snow types with various snow grain sizes (e.g., Stamnes et al 2007;Painter et al, 2009).…”

“…The spectral contrast is enhanced further as the exitance angle increases (not shown in the figure). Finally, bare specular ice has extremely low emissivities, which are the lowest among surfaces made by ice material and can be completely predicted by the Fresnel reflectance formula (Hori et al, 2006). Surface hoar consists of fine acicular ice particles and has high emissivity (even higher than that of fine snow) with a flat spectrum at 11-12 m. Therefore, the spectral features of surface hoar with the lower spectral contrast are also expected to be usable for discriminating it from other types of snow cover.…”