Abstract. Accurate knowledge of the reflectance from snow/ice covered surface is of fundamental importance for the investigation and retrieval of snow parameters and atmospheric constituents. This is a prerequisite for identifying and quantifying changes in the environment and climate in Polar Regions. However, the current differences between simulated and measured reflectance in a coupled snow-atmosphere system, leads to systematic errors in the determination of the amount of trace gases, aerosol and cloud parameters from space based and airborne passive remote sensing observations. In this paper, we describe studies of the retrieval of snow grain morphologies, also called habits, and their use to determine reflectance and test the accuracy of our radiative transfer model simulations of reflectance by comparison with measurements. Firstly we report on a sensitivity study. This addresses the requirement for adequate a priori knowledge about snow models and ancillary information about the atmosphere; the objective being to minimize differences between measurements and simulation. For this aim we use the well-validated phenomenological radiative transfer model SCIATRAN. Secondly and more importantly, we present a novel two-stage snow grain morphology (i.e. size and shape of ice crystals in the snow) retrieval algorithm. We then describe the use of this new retrieval to estimate the most representative snow model, using different types of snow morphologies, for the airborne observation conditions, performed by NASA's Cloud Absorption Radiometer (CAR). The results show that the retrieved ice crystal shapes are consistent with the expected snow morphology (estimated from temperature information) in the measurement area over Barrow/Utqiaġvik, Alaska in 2008. Thirdly, we present a comprehensive comparison of the simulated reflectance (using retrieved snow grain size and shape as well as independent atmospheric data) with that from airborne CAR measurements in the visible and NIR wavelength range. The results of this comparison are used to assess the quality and accuracy of the radiative transfer model in the simulation of the reflectance in a coupled snow-atmosphere system. Assuming that that snow layer consists of ice crystals with aggregates of 8 column ice habit having an effective radius ~ 98.83 μm, we find that for a surface covered by old snow, the Pearson correlation coefficient, R, between measurements and simulations to be 0.98 (R2 ~ 0.96). For freshly fallen snow on areas having surface inhomogenity, the correlation is ~ 0.97 (R2 ~ 0.94) in the infrared and 0.88 (R2 ~ 0.77) in the visible wavelengths assuming that snow layer consists of aggregate of 5 plate ice habit with effective radius ~ 83.41 μm. Largest differences between simulated and measured values are observed in glint, i.e. in the angular regions of specular and near-specular reflection, with relative azimuth angles