Monte Carlo Simulation of Snow Depth in a Forest

Abstract

Snow depth in a forest is highly variable and to reduce the cost of extensive field sampling for obtaining mean depths, a simulation model was developed. First, the location of individual trees in a representative portion of the forest is either surveyed in the field or simulated based on the statistical characteristics pertaining to the distribution of trees. In this forest, a large number of randomly located sample points is generated by Monte Carlo technique. The azimuth and distance from each point to the nearest tree is determined, and a snow depth simulated based on the observed snow depth distribution around individual trees. The model was applied successfully to a northern spruce forest in subarctic Ontario, showing that this simulation provides a useful approach to determine mean snow depth.