Experimental and model based investigation of the links between snow bidirectional reflectance and snow microstructure
Abstract
Snow optical properties are unique among Earth surface and crucial for a wide range of applications. The bi-directional reflectance, hereafter BRDF, of snow is sensible to snow microstructure. However the complex interplays between different parameters of snow microstructure namely size parameters and shape parameters on reflectance are challenging to disentangle both theoretically and experimentally. An accurate understanding and modelling of snow BRDF is required to correctly process satellite data. BRDF measurements might also provide means of characterizing snow morphology. This study presents one of the very few dataset that combined bi-directional reflectance measurements over 500-2500 nm and X-ray tomography of the snow microstructure for three different snow samples and two snow types. The dataset is used to evaluate the approach from Malinka, 2014 that relates snow optical properties to the chord length distribution in the snow microstructure. For low and medium absorption, the model accurately reproduces the measurements but tends to slightly overestimate the anisotropy of the reflectance. The model indicates that the deviation of the ice chord length distribution from an exponential distribution, that can be understood as a characterization of snow types, does not impact the reflectance for such absorptions. The simulations are also impacted by the uncertainties in the ice refractive index values. At high absorption and high viewing/incident zenith angle, the simulations and the measurements disagree indicating that some of the assumptions made in the model are not met anymore. The study also indicates that crystal habits might play a significant role for the reflectance under such geometries and wavelengths. However quantitative relationship between crystal habits and reflectance alongside with potential optical methodologies to classify snow morphology would require an extended dataset over more snow types. This extended dataset can likely be obtained thanks to the use of ray tracing models on tomography images of the snow microstructure.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2017
- Bibcode:
- 2017AGUFM.C13C0980D
- Keywords:
-
- 0736 Snow;
- CRYOSPHERE;
- 0740 Snowmelt;
- CRYOSPHERE;
- 0798 Modeling;
- CRYOSPHERE