Effects of Snow Grain Shape and Mixing State of Snow Impurity on Retrieval of Snow Physical Parameters From Ground‐Based Optical Instrument

Abstract

We proposed new snow grain model and snow impurity mixture models for the purpose of accurate retrievals of snow grain size and concentration of light‐absorbing particles (LAP) in snow from the optical remote sensing data. Two kinds of ice crystal models, irregularly shaped Voronoi columns, and Voronoi aggregates were employed. LAP can be captured by the snow through two processes: dry and wet deposition. Two different snow impurity mixture models were proposed. One is an external mixture model. We employed a coated sphere model in which soot were hydrophilic particles to represent a hygroscopic property of aerosol and assumed the hydrophilic soot particles to be externally mixed with snow particles. The other is an internal mixture model. We employed a dynamic effective medium approximation method in which soot particles were randomly located within snow particle with any size distribution and number concentration. Validation of these models is conducted using a ground‐based spectral radiometer system with in situ measurement data. For snow grain size retrievals, a Voronoi mixture model seamlessly representing a geometrical shape and an optical properties of various snow types can provide accurate retrievals. For the retrieval of LAP concentration in snow, employing different mixing state models depending on season and measurement site gives accurate results. We also discussed the uncertainty of retrieved snow parameters on the surface slope involved in the retrieval accuracy. These models are expected to be useful for advanced airborne/satellite remote sensing and climate studies via radiative transfer modeling in the atmosphere‐snow system.

Quantifying Relative Contributions of Light‐Absorbing Particles From Domestic and Foreign Sources on Snow Melt at Sapporo, Japan During the 2011–2012 Winter

“…This implies that, in terms of RMSD, only 40% of the total error of NHM‐Chem‐SMAP in simulating shortwave snow albedo can be attributed to the atmospheric chemical process despite the relatively large uncertainties discussed above. Recently, it has been recognized that mixing states of LAPs within the snowpack (Flanner et al., 2012; He et al., 2014, 2019; Liou et al., 2014; Tanikawa et al., 2020) and optically equivalent snow grain shapes (He et al., 2014; Libois et al., 2013; Tanikawa et al., 2020) can control the snow albedos. Hence, it is necessary to consider these processes explicitly in the SMAP model to improve the model performance in the future.…”

The Determination of the Snow Optical Grain Diameter and Snowmelt Area on the Greenland Ice Sheet Using Spaceborne Optical Observations

“…The spherical grain assumption was motivated by the successful estimation of spectral hemispherical reflectances of snow with nonspherical ice particles when representing them as spherical grains of a similar volume-to-surface-area [37]. While this technique has been widely applied, it has been noted that the spherical assumption was limited in accounting for the directional variation of snow reflectance [38][39][40][41][42][43][44][45] and therefore would lead to errors when used on remotely sensed directional reflectance. Several models using the nonspherical grains assumption have been applied to snow characteristics retrievals, for example on data from the Sea and Land Surface Temperature Radiometer (SLSTR) onboard the Sentinel-3 satellites [43,44] or on data from the Moderate Resolution Imaging Spectroradiometer (MODIS) onboard the Terra and Aqua satellites [29,38].…”

The retrieval of snow properties from SLSTR Sentinel-3 – Part 2: Results and validation

“…However, there is no publication with respect to the retrieval of the ice crystal shape in the snow layer using passive multi-spectrum satellite observations. Although habit mixture models are preferable for the description of snow grain shapes (Saito et al, 2019;Tanikawa et al, 2020;Pohl et al, 2020), the information content from satellite observation is limited compared to fieldbased measurements. Thus, an optimal single shape, which provides the best agreement between simulation and satellite observation (e.g., top-of-atmosphere (TOA) reflectance), is also needed.…”