Speaker: David Levin National Weather Service – Juneau Alaska
Snow to liquid ratios (SLR) continue to be a large source of error for forecasters in the prediction of snow amounts during the winter months. Current methods of predicting SLR in Alaska range from an empirical method based solely on surface temperatures, to model-derived SLR. Both of these methods have limitations. Thus an SLR climatology was developed yielding a robust data set of snowfall observations. The mean SLR for all Alaska Weather Forecast Offices was found to be much higher than the method based on surface temperatures. Considerable variability was noted in both mean and median SLR values between sites located along the Gulf coast and those in the interior. It is hypothesized that the frequency of events where warmer marine air over- runs cold, dry arctic air from northwest Canada modulates these variations in snow to liquid ratio. An observed sounding climatology was also developed for various low level thermal fields and was matched to observations of SLR at each site. For Southeast Alaska, it was found that the 1000-850mb thickness was a good predictor of SLR with mid level thickness (850-700mb) being the best predictor in general for other areas of Alaska. Finally, the results were then analyzed on a grid using the National Weather Service Graphical Forecast Editor (GFE) and a smart tool was developed which would allow operational forecasters to use this climatology as a starting point when making a prediction of snow to liquid ratio. This presentation will details the work described above.