1. How does snow metamorphism affect surface albedo?
Snow metamorphism leads to spherical and larger snow grains, resulting in a decrease in surface albedo. This process is supported by research from Warren (1982), Colbeck (1983), and Gubler (1985). The decrease in surface albedo is due to the change in snow grain shape and size, which affects the reflection of sunlight. As the snow grains become more spherical, they reflect less sunlight, leading to a decrease in surface albedo. This phenomenon is significant in understanding the impact of snow metamorphism on the Earth's climate and energy balance.
read more
2. What instruments were used for airborne measurements during the ACLOUD campaign?
During the ACLOUD campaign, the Polar 5 aircraft was equipped with remote sensing instruments measuring solar spectral radiation. These instruments included the Spectral Modular Airborne Radiation measurement system (SMART) albedometer, which measured the solar spectral downward and upward irradiance with 2 Hz temporal resolution. Additionally, two imaging spectrometers, AisaEagle and AisaHawk, were used to measure upward radiance with 20 Hz temporal resolution. AisaEagle had a field of view of 36 * and 1024 pixels, while AisaHawk had a field of view of 36 * and 384 pixels. Both instruments covered a wavelength range from 400 nm to 2500 nm. The instruments were calibrated using a certified diffuse radiation source, ensuring accurate measurements of solar spectral radiation and radiance.
read more
3. How are radiative transfer simulations applied to model snow reflectance spectra?
Radiative transfer simulations for snow reflectance spectra utilize the library of radiative transfer routines and programs (libRadtran). This library, as mentioned by Emde et al. (2016) and Mayer et al. (2019), is used to model the radiative transfer in a dense medium like snow layers. The far field assumption and multiple scattering assumption are crucial in these simulations. The far field assumption presumes that particles are at a distance, allowing scattering waves to be considered planar. The multiple scattering assumption defines particles by their single-scattering properties and assumes no interaction between particles. However, these assumptions may be violated when simulating a snow layer with a density increase of over a hundredfold. Pohl et al. (2020) addressed this issue by showing that the effects can be neglected. To simulate effective particle radii larger than 25 um, the database of optical properties in libRadtran was expanded. Single scattering properties and Legendre moments representing the scattering phase function of ice crystals were taken from an external database (Yang et al., 2000). The 'smooth droxtal' shape was chosen to account for the rounding of ice crystals during snow aging. For liquid water spheres, the Mie-tool provided by libRadtran was used to derive single scattering properties. The d-M-approach (Wiscombe, 1977) was applied to reduce the number of Legendre moments needed for an adequate representation of the scattering phase function. The bulk optical properties were scaled accordingly. Detailed information on the simulation setup can be found in Appendix A.
read more
4. What are the spatial maps of r eff and f LW for eleven selected flight sections?
The spatial maps of r eff and f LW for eleven selected flight sections were derived using retrieval methods. The maps show the derived properties for thirteen observation angles between -15 * and +15 * converted to distance from nadir on the y-axis and the along track distance on the x-axis. The r eff -frequency distribution in Fig. 5c shows effective radius values between 100 um and 350 um with occurrence of generally higher values towards the south-west (negative distances from nadir) as depicted in the map (Fig. 5b). This r eff gradient was visible on all south-east or north-west heading flight sections. The reference curves for the r eff -retrieval show a dependence on observation angle with deviations up to 100 um between an observation angle of +15 * and -15 * , indicating a sensitivity to surface inhomogeneities and roughness in the vicinity of melt ponds and pressure ridges. The filtered out sections are indicated as dark blue areas of f LW up to 100 % in the f LW -map. Overall, a variation of f LW is visible, with a particularly homogeneous area highlighted in the maps by the right red box. The retrieved liquid water fractions were between 8.7 % and 15.6 %, corresponding to very wet (8 - 15 %) and soaked (> 15 %) snow layers according to the international classification for seasonal snow on the ground (Fierz et al., 2009). The overall high liquid water fraction indicates the runoff phase of snow melting for all cases but flight section 2017/06/08 (I). The f LW -distributions of flight sections 2017/06/25 (IV) and 2017/06/25 (I) show some geographical variability, with the f LW -distribution of section 2017/06/25 (IV) being narrower than that of section 2017/06/25 (I). The maximum depth of 0.33 m was derived for the pond center, with pond parts to the right between 30 m and 45 m along track distance mostly shallower with depths varying around 0.2 m. The magnitude of the here retrieved depths can be assumed to be quite reasonable at the start of the pond evolution.
read more