Predicting Snow-to-Liquid Ratio over the Western United States

The snow-to-liquid-ratio (SLR) and its inverse, snow density, are crucial for forecasting snowfall in numerical weather prediction models and for estimating snow-water-equivalent (SWE) using remote sensing. SLR also varies widely in space and time, making it challenging to forecast accurately, particularly in the extremely heterogenous geography and climate of the mountains of the Western US. This study utilizes high quality, manually-collected measurements of new snowfall and new SWE from 14 mountainous sites across the region to build an algorithm to predict SLR as a function of atmospheric variables.

When a Multiple Linear Regression (MLR) algorithm is trained on a simple combination of wind speed and temperature at 400 m and 2400 m above ground level from either the ERA5 reanalysis, the GFS, or the HRRR, it predicts SLR with significantly more skill than existing SLR prediction methods. When a more extensive set of variables is considered, skill can be further improved.

The variables used to achieve the most skillful prediction of SLR are (1) temperature, wind speed, relative humidity, and specific humidity, all at 400 m and 2400 m AGL, (2) 500 hPa geopotential height, (3) model QPF, (4) maximum solar angle, and (5) cloud top temperature. When an algorithm is trained using these variables, it can predict SLR with R²=0.44 and MAE=2.91. For the existing SLR prediction techniques currently used in operations, R² ranges from 0.04 to 0.25 and MAE ranges from to 4.30 to 9.27. Therefore the algorithms built in this paper can drastically improve SLR prediction over the mountains of the Western US.

- Indicates paper has been withdrawn from meeting

- Indicates an Award Winner

See more of: Winter Weather Forecasting: Snow-Liquid-Ratio and Freezing Rain
See more of: 33rd Conference on Weather Analysis and Forecasting (WAF)/29th Conference on Numerical Weather Prediction (NWP)

Previous Abstract | Next Abstract >>