A practical formulation of snow surface diffusion by wind for watershed‐scale applications

Noriaki Ohara

doi:10.1002/2013wr014744

Abstract

AbstractPrediction of snow drift is of importance for structure design and traffic management on snowy and windy prairie landscapes. The snow redistribution by wind is also regarded as one of the largest sources of error in hydrologic snowmelt models. In this study, a snow movement equation was generalized and customized for horizontal two‐dimensional watershed‐scale applications by incorporating snow transport, wind snow diffusion, and snow gravitational movement. Then, the snow surface diffusion process by wind turbulence was formulated in terms of the autocorrelation functions of the measurable wind velocity field using G.I. Taylor's theorem. However, analysis of the example wind data suggested a delta correlation in wind turbulent component that resulted from subtracting their moving average values from the original wind speed data. The dynamic model based on the proposed formulation was able to effectively reproduce the observed equilibrium snow profiles affected by wind drifting. A two‐dimensional model simulation using a 10 m digital elevation model in Muddy Gap, Wyoming was also presented for qualitative validation of the model in the watershed‐scale applications. Additionally, the theoretical extension for preferential snow accumulation process was presented in Appendix . These modeling results together with the observations on the prairie suggested the importance of the snow surface diffusion process in addition to the snow transport.

Highlights

Understanding the snow redistribution process is crucial for accurate snowmelt estimation using processbased snow modeling because that is considered to be a major source of uncertainty in snowmelt simulations, especially on windy prairie landscape [Pomeroy and Gray, 1990].Snow redistribution is mainly driven by the wind near the snow surface, which may be significantly affected by the ground surface conditions, such as topography, vegetation, etc
The snow transport formula by Tabler et al [1990a, 1990b] in Table 1 is used for the snow drift component because this was developed in the same Mountain States region of the USA
The model based on this formulation is named the SMOOTH (Snow Movement Over Open Terrain for Hydrology) model in this study

Summary

Introduction

Understanding the snow redistribution process is crucial for accurate snowmelt estimation using processbased snow modeling because that is considered to be a major source of uncertainty in snowmelt simulations, especially on windy prairie landscape [Pomeroy and Gray, 1990].Snow redistribution is mainly driven by the wind near the snow surface, which may be significantly affected by the ground surface conditions, such as topography, vegetation, etc. Quantification of snow redistribution and preferential snow accumulation still remains a challenge since it is difficult to estimate the local wind field and the variability in snow physical characteristics. This snow movement may be a prime driver of the subgrid-scale variability in watershed-scale modeling (several-meter to a few-kilometer grid size), which is yet to be modeled effectively. The north facing or forested slopes have more snow storage than south facing or clear-cut slopes in the late spring in the northern hemisphere [e.g., Ohara and Kavvas, 2006].The snowmelt process can be modeled well by an energy balance framework under a known atmospheric forcing. Snow is often redistributed by five primary factors: (1) snow transport by wind, (2) preferential snow accumulation, (3) snow surface diffusion by wind, (4)

Objectives

Results

Conclusion