The snow/snow water equivalent ratio and its predictability across Canada | Posted on:2007-12-21 | Degree:M.Sc | Type:Thesis | University:McGill University (Canada) | Candidate:Cox, Jessica | Full Text:PDF | GTID:2440390005473617 | Subject:Atmospheric Sciences | Abstract/Summary: | | The current practice of snowfall forecasting in Canada is to determine the snow water equivalent (SWE) expected to precipitate using a numerical weather predication model and then multiplying this amount by a snow/SWE ratio to determine the forecast snow depth. The 10:1 "rule of thumb" is still widely used operationally as this ratio, even though it is well-known to introduce error in the forecasts because the density of snow is highly variable. In 2003 Ivan DubE developed a decision tree type algorithm to find the snow/SWE ratio which has subsequently been automated in 2004 by the Meteorological Service of Canada (MSC). The objectives of this study are to explore the behaviour of snow/SWE ratio by developing a Canada-wide climatology of this quantity and examining performance of the MSC algorithm over the winter 2004-2005 using several verification techniques. We found that the mean annual snow/SWE ratio across Canada is 13:1 with large variations temporally and spatially and that the MSC algorithm performed with equal or better skill than the 10:1 algorithm in 84% of the events. | Keywords/Search Tags: | Ratio, Canada, Snow, MSC, Algorithm | | Related items |
| |
|