Implications of mountain shading on calculating energy for snowmelt using unstructured triangular meshes

Abstract

In many parts of the world, the snowmelt energy balance is dominated by net solar shortwave radiation. This is the case in the Canadian Rocky Mountains, where clear skies dominate the winter and spring. In mountainous regions, solar irradiance at the snow surface is not only affected by solar angles, atmospheric transmittance, and the slope and aspect of immediate topography, but also by shadows from surrounding terrain. Many hydrological models do not consider such horizon-shadows. The accumulation of errors in estimating solar irradiance by neglecting horizon-shadows can lead to significant errors in calculating the timing and rate of snowmelt due to the seasonal storage of internal energy in the snowpack.

A common approach to representing the landscape is through structured meshes. However, such representations introduce errors due to the rigid nature of the mesh, creating artefacts and other constraints. Unstructured triangular meshes are more efficient in their representation of the terrain by allowing for a variable resolution. These meshes do not suffer from the artefact problems of a structured mesh.

This thesis demonstrates the increased accuracy of using a horizon-shading model with an unstructured mesh versus standard self-shading algorithms in Marmot Creek Research Basin (MCRB), Alberta, Canada. A systematic basin-wide over-prediction (basin mean expressed as phase change mass: 14 mm, maximum: 200 mm) in net shortwave is observed when only self-shadows are considered. The horizon-shadow model was run at a point scale at three sites throughout MCRB to investigate the effects of scale on the model results. It was found that small triangles were best suited for this topographic region and that shadow patterns were captured accurately. Large triangles were found to be too easily shaded by the model, created many disjointed regions. As well, model results were compared to measurements of mountain shadows by timelapse digital cameras. These images were orthorectified and the shadow regions extracted allowing for a quantitative comparison. It was found that the horizon-model produced results within 10 m of the measured shadows, and properly captured shadow transits.

A point-scale energy balance model SNOBAL was run via The Cold Regions Hydrological Model, an HRU based hydrologic model. It was found that in the highly shaded valleys, snowpack ablation could be incorrect by approximately 4 days. Although MCRB was generally not significantly impacted by the over-estimation in irradiance in this study, insight into the horizon-shadowing process was possible as a result of the existing network of radiometers and other meteorological stations at MCRB. Because down-stream processes such as flooding depend on correct headwater snowmelt predictions, quantitative results demonstrating inaccuracies in a modelled component of the surface energy balance can help improve snowmelt modelling.