Search

Wind‐driven redistribution of surface snow is one of the key factors leading to heterogeneous accumulation of snow at small scales. Understanding the processes that lead to this heterogeneous accumulation is, therefore, of great importance to many glaciological and hydrological questions. High‐quality information on the wind field is necessary to realistically represent drifting snow. Here, we introduce a novel, intermediate‐complexity drifting and blowing snow module for the Weather Research and Forecasting (WRF) model that integrates seamlessly into the standard WRF infrastructure. The module also accounts for snow particle sublimation and considers the thermodynamic feedback on the atmospheric fields. In an idealized model environment the module was tested for physical consistency. Sensitivity experiments showed that drifting snow sublimation has on the one hand local effects on the erosion and deposition patterns and on the other hand non‐local effects on the larger‐scale atmosphere. The simple and computationally efficient implementation allows this module to be used for small‐scale and large‐scale simulations in polar and glaciated regions. Plain Language Summary: Transport of snow by the wind can have high impact on local glacier mass changes as it leads to non‐uniform amounts of snow on the ground. In order to simulate and better understand this process we introduce a new modeling framework that is included into a widely used atmospheric model. Test simulations and sensitivity experiments show the physical consistency of the model. Complex interactions between different processes like snow erosion, drifting snow sublimation and the wind field show the necessity of coupling the snow and atmospheric models. Key Points: We introduce a new module to simulate drifting snow within the framework of the Weather Research and Forecasting (WRF) modelThe module is tested in an idealized setting for physical consistencyWe show that local sublimational cooling has larger‐scale non‐local effects on the wind field [ABSTRACT FROM AUTHOR]

Abstract Wind‐driven redistribution of surface snow is one of the key factors leading to heterogeneous accumulation of snow at small scales. Understanding the processes that lead to this heterogeneous accumulation is, therefore, of great importance to many glaciological and hydrological questions. High‐quality information on the wind field is necessary to realistically represent drifting snow. Here, we introduce a novel, intermediate‐complexity drifting and blowing snow module for the Weather Research and Forecasting (WRF) model that integrates seamlessly into the standard WRF infrastructure. The module also accounts for snow particle sublimation and considers the thermodynamic feedback on the atmospheric fields. In an idealized model environment the module was tested for physical consistency. Sensitivity experiments showed that drifting snow sublimation has on the one hand local effects on the erosion and deposition patterns and on the other hand non‐local effects on the larger‐scale atmosphere. The simple and computationally efficient implementation allows this module to be used for small‐scale and large‐scale simulations in polar and glaciated regions.

The temporal evolution of seasonal snow cover and its spatial variability in environments such as mountains, prairies or polar regions is strongly influenced by the interactions between the atmospheric boundary layer and the snow cover. Wind-driven coupling processes affect both mass and energy fluxes at the snow surface with consequences on snow hydrology, avalanche hazard, and ecosystem development. This paper proposes a review on these processes and combines the more recent findings obtained from observations and modeling. The spatial variability of snow accumulation across multiple scales can be associated to wind-driven processes ranging from orographic precipitation at large scale to preferential-deposition of snowfall and wind-induced transport of snow on the ground at smaller scales. An overview of models of varying complexity developed to simulate these processes is proposed in this paper. Snow ablation is also affected by wind-driven processes. Over continuous snow covers, turbulent fluxes of latent and sensible heat, as well as blowing snow sublimation, modify the mass, and energy balance of the snowpack and their representation in numerical models is associated with many uncertainties. As soon as the snow cover becomes patchy in spring local heat advection induces the development of stable internal boundary layers changing heat exchange toward the snow. Overall, wind-driven processes play a key role in all the different stages of the evolution of seasonal snow. Improvements in process understanding particularly at the mountain-ridge and the slope scale, and processes representations in models at scales up to the mountain range scale, will be the basis for improved short-term forecast and climate projections in snow-covered regions.

In order to predict snow distribution on a roof caused by snow drift, a simulation based on coupled CFD (computational fluid dynamics)and DEM (discrete element method)methods was employed. The numerical simulation method includes motion equations of fluid and snow particle, fluid-snow particle coupling control eqnation and snow particle collision model on the coupling condition. Coupled computing method, material parameters of snow particle, the parameters of the collision and border control conditions was built based on the measured results.A simulation of the snow drift adopting this method on a typical Tsuchiya stepped roof was done. Compared the simulated results of snow distribution on the lower roof with previous data from field measurement and wind tunnel test, the feasibility of using coupled CFD and DEM methods to study snow drift was demonstrated. Numerical simulation shows that backflow whirlpool decreases obviously above the low roof due to the existence of snow, the nondimensional horizontal wind velocity on the end of the lower roof is-0.4 when there is no snow distribution, but it increases to 0 with snow distribution. [ABSTRACT FROM AUTHOR]

Copyright of Journal of the Japanese Society of Snow & Ice / Seppyo is the property of Japanese Society of Snow & Ice and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The planning of snow fence constructions in alpine terrain is site-specific and lacks published guidelines. A methodology of snow drift analysis for the evaluation of snow drift constructions is presented. The background to the methodology are recent findings of time series of spatial snow depth maps obtained by laser scanning over a period of several years, where snow accumulation patterns over a whole winter season were similar in different years of observation. It could be concluded that the knowledge of spatial snow distribution at annual peak accumulation provides a reliable basis for the planning of drift constructions. It was hypothesized that snow accumulation patterns at a new project site could be assessed using a combination of snow mapping via terrestrial laser scanning and high resolution numerical wind field modelling. Modelling results localized areas of large vertical and horizontal wind speeds, which coincide with extrema of wind stress on the snowpack surface. TLS surface maps provided the spatial snow depth distribution, and erosional and depositional zones were apparent. Insufficiencies in fence functioning at the study site were determined and adaptations to fence constructions using existing recommendations could be applied. The functionality of the fences was significantly improved using the presented methodology, so it is recommended that future construction projects consider a similar site evaluation. [ABSTRACT FROM AUTHOR]

The blizzard phenomenon is the climate risk that has the greatest impact on the road-ways in the South-Eastern part of Romania. The E85 section between Urziceni and Buzău has a great vulnerability to blizzard. Its vulnerability was proved during the 26th of January - the 14th of February 2012, when the road-way was blocked for 10 days. In order to establish the vulnerability level for E85 road, the section between Urziceni and Buzău, the Roads Vulnerability to Blizzard Index (RVBI) has been calculated and spatially modelled by using the GIS techniques. This index integrates several factors, such as: road sections orientation, buildings density in the vicinity of the road, width of the villages that are crossed by E85, the snow drift potential, within built-up area's orientation to E85, the negative relief forms along the road section. The calculation of the RVBI points out that 74% of total length of the road present high and very high vulnerability during blizzard phenomenon. The road sections with high and very high vulnerability need protective measures, such as: forest belts plantation, using bigger snow fences, increasing the snow fences density and achieving more equipment for snow moving off by the local authorities. [ABSTRACT FROM AUTHOR]

Previous wind tunnel experiments on snow drift usually used artificial snow as a substitute of natural snow, which can not fully reflect the motion characteristics of snow drift in natural environments. In this paper we conducted a series of experiments in wind tunnel to investigate the motion of natural snow (fresh snow and old snow), which was collected out-door without destroying the surface structure. The results indicated that the threshold velocity of fresh snow is less than that of old snow, and that the mass flux rates of the two kinds of snow exponentially decrease with height whereas the snow transport rate increases exponentially with wind velocity. Based on the PIV measuring of the velocities of snow particles over two kinds of terrains (flat ground and roadbed), we found that the particle velocity obeys a Gaussian distribution for flat ground, top surface and leeward slope of the roadbed. However, for the windward slope of roadbed the particle velocity distribution displays poor correlation with a Gaussian function due to the acceleration of snow particles when saltating across the slop. Statistical analysis showed that impacting and liftoff velocities of snow particles also obey Gaussian distributions. [ABSTRACT FROM AUTHOR]

Experiments in a cold wind tunnel were used to verify drifting snow sublimation models. A layer of drifting snow particles was formed over a sintered snow surface. Sublimation and drifting snow flux were estimated from two vertically resolved profile measurements separated along the flow path and were compared to a simple, one-dimensional diffusion model of drift and drifting snow sublimation. The experiments show an increase in water vapor content of the air from drifting snow sublimation. The measured drifting snow sublimation appeared to be consistent with albeit somewhat larger than theoretical values found in the model study. Under wind tunnel conditions, particle number density appears to be the most important controlling factor on the sublimation rate. For experiments with external solar radiative forcing, the increase of the sublimation rate was also larger than theoretical predictions. The experiments suggest that irregular snow crystals and solar radiation might increase sublimation rates more than described by many drifting snow models. [ABSTRACT FROM AUTHOR]

Abstract: The interactions between atmosphere and snowpack are important for modelling the boundary layer (BL) over snow surfaces. To study these interactions, the mesoscale local model (LM) of the German Meteorological Service is applied for simulations over Greenland at a horizontal resolution of about 14 km and a high vertical resolution of the BL. Since usage of the snow/soil model as implemented in LM leads to unrealistic representation of the snowpack, adjustments of the representation of snow properties in LM have been done for the simulation over the Greenland ice sheet. Simulations using LM are performed for a ten-day period in July 2002, during which a field experiment took place at station Summit. To validate the model LM, the model results are compared to these measurements. The comparison shows some deficies for the near-surface values of temperature as well as turbulence quantities. The turbulence parameterisation in the LM has deficies for very stable stratification. Therefore, a local mixing length approach and the parameterisation of a scalar roughness length over ice and snow are implemented in the LM and their influence on the model performance is tested. These sensitivity studies show an improvement of turbulent kinetic energy and heat flux with usage of scalar roughness length as well as local mixing length approach. For a first estimation of the influence of snow drift on the snow accumulation, the LM results are used to drive the one-dimensional snow model SNOWPACK in an off-line mode. SNOWPACK has a much better representation of snow than the LM snow/soil model and allows for simulating snow microphysical properties. While over large parts of the plateau area of the ice sheet the influence of snow drift on the snow accumulation is much smaller than the evaporation/sublimation, snow drift influence is of at least the same magnitude in regions of the ice sheet margins, associated with large divergence/convergence. [Copyright &y& Elsevier]

Abstract: For about 10 years, Météo-France has developed and operated a real time operational suite aimed at snowpack simulation and avalanche risk forecasting: the numerical model chain Safran-Crocus-Mepra (SCM). It presently includes only a crude formulation of snowdrift effects at large spatial scales (massif). This paper presents the improvements to a snow drift module suited to the SCM chain. The new version called SYTRON3 which aims at an improved simulation of snow drift effects at smaller spatial scales. It is coupled with the SCM environment, which provides hourly meteorological conditions and snow forcing. The main modifications concern the new parameterisation schemes, which are more realistic and more physically based. The increased number of vertical layers allows now to represent explicitly the three modes of movement during snow drift: creep, saltation and diffusion. The validation of this new version as well as comparisons with the previous version are performed at the Col du Lac Blanc test site (2700 m a.s.l., French Alps) by the use of digital photographs and field observations. [Copyright &y& Elsevier]

Field measurements were carried out to calculate the threshold friction velocity for snow saltation, and mass fluxes during snow drift. The wind was measured in three components by an ultrasonic anemometer, and the mass fluxes were determined using an optical sensor (‘snow particle counter’), acoustic sensors (‘Flowcapt’) and mechanical traps. The threshold friction velocity was found to be correlated to the grain size (R2=0.75). The mass flux measurements were compared with numerical simulations of snow drift, and it was demonstrated that the maximum snow transport takes place at shear stress values of roughly two times the average shear stress over 20 min. By implementing a probability distribution for the shear stress the mass flux was simulated with only the mean measured value of the shear stress as input. This procedure enables the future use of the numerical model for operational applications. [ABSTRACT FROM AUTHOR]

This paper introduces a computational fluid dynamic (CFD) model for two and three-dimensional simulation of wind blown particles such as sand, soil or snow. The model is based on the homogenous two-phase flow theory, where the flow field is predicted by solving Navier–Stokes equations for transient, incompressible viscous flow. The particle volume fraction is predicted by solving the transport convection/diffusion equation. Particles transported by suspension and by saltation modes are modeled separately and added to the transport equation as extra source terms. A new solid interface boundary is introduced to the flow computational domain as particles accumulate to form deposition regions. The model treats control volumes fully blocked by particles as solid surfaces whenever the deposition conditions are satisfied. The transport equations are discretized in Eulerian reference frame using finite volume method. The model is used to simulate the flow field around fences for two applications, snow drift at single row fence and sand drift at double row fence. In both cases, the model shows good agreement with observations and a realistic behavior of the snow and sand particles that deposited at porous fences as compared with both field and wind tunnel measurements. [Copyright &y& Elsevier]

Formed metal deck roofs play an important role in steel building construction. There are many design variables in deck design to make it important for designers to put the right combination together to achieve a better design. In this paper, a procedure for the design of steel roofs subjected to non-uniform loads such as drifted snow is presented. The presented approach uses genetic algorithms to achieve optimal or near-optimal designs. The advantages of the model stem from two main points: (1) it searches for the optimum or near-optimum number of purlins for the roof; and (2) it determines the near-optimum spacing between purlins to minimize the structure weight. Details of model development are described and example applications are presented to demonstrate the model capabilities. [Copyright &y& Elsevier]

Blowing snow is a process, which can be observed during the entire winter season in Arctic catchments. On Spitsbergen at 78° North, the ground is bare of tall vegetation and the snow is easily moved by the wind. During frequent storms, large masses of snow are relocated from erosion to deposition areas. In this study, the mass of snow, transported out of an Arctic valley to the open sea is estimated via direct measurements and model calculations. The study area is a valley on Spitsbergen in the high Arctic. The valley is approximately 4 km wide and 10 km long and ends in a fjord arm. The wind direction in the valley is very uniform. During the winter season, the wind is blowing out of the valley 80% of the time. There is one permanent automatic weather station located in the valley. In addition, three automatic weather stations were installed in the valley during a study period of 2 months in February and March 2000. These stations measure, in addition to wind speed and direction, the snow drift flux with acoustic snow drift sensors. The results of the study period are related to those of the permanent station to quantify the accumulated snow mass passing from the valley to the open sea for the entire season. The results show that the transport snow mass equals 0.2% of the annual precipitation in the catchment. Thus, the losses due to snow drift to the open sea are of little importance to the valleys water balance. [Copyright &y& Elsevier]

This paper presents a source term model to predict the drifting snow near the ground, which is based on the assumption of local equilibrium saltation. The features of the one way coupled conservation equation for the snow transport in the Eulerian reference frame are examined through calibrating the input sources versus output snow concentrations by using a numerical experiment method. The target source term for the snow concentration compensation due to the surface erosion is derived with the help of a semi-empirical mass flow rate of the equilibrium saltation and the calibration results under different friction velocities, threshold friction velocities and settling velocities. With the validation on the equilibrium saltation above an open and flat snow bed, rational saltation mass concentration is solved. Using this source term model, good adaptabilities for various wind and snow conditions are confirmed. With the application to the non-equilibrium saltation around a cube, intuitively plausible distributions of the snow concentration near the ground and the transient topography changes in the snow surface are predicted through a time dependent dynamic mesh solution. The non-equilibrium properties of the drifting snow revealed in space and time are further evidenced by the spatial distributions and time histories of the friction velocity and the saltation snow concentration • The source term model is built on the assumption of local equilibrium saltation and derived from the numerical experiments. • Using this model, good adaptabilities for various wind and snow conditions above an open and flat snow bed are confirmed. • The source term model can predict the non-equilibrium saltation and the transient topography changes in the snow surface. [ABSTRACT FROM AUTHOR]