Predicting Snow Depth in a Forest–Tundra Landscape using a Conceptual Model Allowing for Snow Redistribution and Constrained by Observations from a Digital Camera

Abstract

Estimation of snow depth in the forest–tundra landscape remains a challenge because of a lack of reliable and frequent observations on precipitation and snow depth. Snow models forced by gridded meteorological datasets are often the only option available for assessing snow depth at the local scale. Unfortunately, these models generally do not take into account the snow redistribution process between open and forested areas which frequently occurs in the forest–tundra landscape. A simple modification to an existing snow accumulation and melt model is proposed in order to allow for snow redistribution. Along with a technique for taking advantage of snow depth observations obtained from a digital camera, the model is shown to provide accurate predictions of snow depth at the local scale when forced with precipitation data from Environment Canada's Canadian Precipitation Analysis. Results from this study suggest that instrumenting automated weather stations with a digital camera, together with small modifications to an existing model used operationally for snow depth prediction, could result in significant improvements to snow depth prediction and analysis in this environment. Further testing at sites where snow water equivalent of the snowpack is available should, however, be performed to fully validate the method.