Monday, 13 January 2020
When wind blows over dry snow, the snow surface self-organizes. This forms meter-scale roughness elements, known as 'bedforms', which include dunes, ripples, snow waves, and sastrugi. These bedforms cover up to 11% of the surface of the Earth [1], on alpine tundra, ice sheets, and sea ice. When they form, they change the interactions between the surface and the atmosphere: snow surfaces with bedforms are less reflective, aerodynamically rougher, and faster to melt [2,3] than flat snow surfaces. Despite their wide range and atmospheric effects, however, snow bedforms have rarely been quantitatively studied.
Here, we present new quantitative studies of snow bedforms, demonstrating that (1) snow bedforms are ubiquitous in dry, unsheltered snow with winds higher than a 7 m/s threshold speed; (2) the sizes of snow barchans increase non-linearly with the wind speed; and (3) snow bedforms expedite snow accumulation and snow cohesion in windy conditions. These results are obtained through a mix of field studies and numerical methods.
The field results are drawn from a series of detailed observations of snow bedform movement in the Colorado Front Range [4], including the first time-lapse documentation of snow bedform evolution. We show examples of the movement of snow ripples, barchan dunes, and sastrugi, which move through our field site at speeds between 0.1-2 m/h . These observations reveal that the shapes of common snow bedforms result from the interaction of granular processes, turbulent boundary-layer structures, and time-dependent cohesion in the snow. Correlating these observations with weather data shows that snow surfaces in Colorado are flat only when the wind speed is lower than 6.7 m/s, and that most falling snow remains flat for less than a day.
The simulation results were obtained with Rescal-snow [5], a model that we developed to unify our field observations with previous dune modeling work by the geomorphology community. In our field studies, we saw that snow often looks analogous to fields of ripples and dunes in sand, but that it is additionally shaped by snowfall and by time-dependent cohesion. Rescal-snow represents both sand-like and snow-like processes through a cellular automaton. The model is open source and intended to be a community modeling resource for studies of wind-blown snow on a 1-100 m scale.
Figure:
Evolutionary patterns of (a) flat snow surfaces (b) barchan dunes (c) snow-steps and (d) snow-waves.
References:
[1] Filhol and Sturm, 2015. "Snow bedforms: a review, new data, and a formation model". Journal of Geophysical Research: Earth Surfaces, 120 (9) 1645-1669.
[2] Amory, C., Gallée, H., Naaim-Bouvet, F. et al. 2017. "Seasonal variations in drag coefficient over a sastrugi-covered field in East Antarctica". Boundary-Layer Meteorology 164 (107). doi:10.1007/s10546-017-0242-5
[3] Petrick, Eicken, et al., 2012. "Snow dunes: A controlling factor of melt pond distribution on Arctic sea ice". Journal of Geophysical Research: Oceans, 117 (C9).
[4] Kochanski, 2018. "Time-lapse observations of snow bedforms in the Colorado Front Range", Zenodo. doi:10.5281/zenodo.1253725
[4] Rescal-snow: www.github.com/kellykochanski/rescal-snow
Here, we present new quantitative studies of snow bedforms, demonstrating that (1) snow bedforms are ubiquitous in dry, unsheltered snow with winds higher than a 7 m/s threshold speed; (2) the sizes of snow barchans increase non-linearly with the wind speed; and (3) snow bedforms expedite snow accumulation and snow cohesion in windy conditions. These results are obtained through a mix of field studies and numerical methods.
The field results are drawn from a series of detailed observations of snow bedform movement in the Colorado Front Range [4], including the first time-lapse documentation of snow bedform evolution. We show examples of the movement of snow ripples, barchan dunes, and sastrugi, which move through our field site at speeds between 0.1-2 m/h . These observations reveal that the shapes of common snow bedforms result from the interaction of granular processes, turbulent boundary-layer structures, and time-dependent cohesion in the snow. Correlating these observations with weather data shows that snow surfaces in Colorado are flat only when the wind speed is lower than 6.7 m/s, and that most falling snow remains flat for less than a day.
The simulation results were obtained with Rescal-snow [5], a model that we developed to unify our field observations with previous dune modeling work by the geomorphology community. In our field studies, we saw that snow often looks analogous to fields of ripples and dunes in sand, but that it is additionally shaped by snowfall and by time-dependent cohesion. Rescal-snow represents both sand-like and snow-like processes through a cellular automaton. The model is open source and intended to be a community modeling resource for studies of wind-blown snow on a 1-100 m scale.
Figure:
Evolutionary patterns of (a) flat snow surfaces (b) barchan dunes (c) snow-steps and (d) snow-waves.
References:
[1] Filhol and Sturm, 2015. "Snow bedforms: a review, new data, and a formation model". Journal of Geophysical Research: Earth Surfaces, 120 (9) 1645-1669.
[2] Amory, C., Gallée, H., Naaim-Bouvet, F. et al. 2017. "Seasonal variations in drag coefficient over a sastrugi-covered field in East Antarctica". Boundary-Layer Meteorology 164 (107). doi:10.1007/s10546-017-0242-5
[3] Petrick, Eicken, et al., 2012. "Snow dunes: A controlling factor of melt pond distribution on Arctic sea ice". Journal of Geophysical Research: Oceans, 117 (C9).
[4] Kochanski, 2018. "Time-lapse observations of snow bedforms in the Colorado Front Range", Zenodo. doi:10.5281/zenodo.1253725
[4] Rescal-snow: www.github.com/kellykochanski/rescal-snow
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner