Journal metrics

Journal metrics

  • IF value: 4.524 IF 4.524
  • IF 5-year value: 5.558 IF 5-year 5.558
  • CiteScore value: 4.84 CiteScore 4.84
  • SNIP value: 1.425 SNIP 1.425
  • SJR value: 3.034 SJR 3.034
  • IPP value: 4.65 IPP 4.65
  • h5-index value: 52 h5-index 52
  • Scimago H index value: 55 Scimago H index 55
The Cryosphere, 12, 891-905, 2018
https://doi.org/10.5194/tc-12-891-2018
© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 3.0 License.
Research article
12 Mar 2018
Improving gridded snow water equivalent products in British Columbia, Canada: multi-source data fusion by neural network models
Andrew M. Snauffer1, William W. Hsieh1, Alex J. Cannon2, and Markus A. Schnorbus3 1Department of Earth, Ocean and Atmospheric Sciences, The University of British Columbia, Vancouver, BC V6T 1Z4, Canada
2Climate Research Division, Environment and Climate Change Canada, P.O. Box 1700 STN CSC, Victoria, BC V8W 2Y2, Canada
3Pacific Climate Impacts Consortium, University House 1, 2489 Sinclair Road, University of Victoria, Victoria, BC V8N 6M2, Canada
Abstract. Estimates of surface snow water equivalent (SWE) in mixed alpine environments with seasonal melts are particularly difficult in areas of high vegetation density, topographic relief, and snow accumulations. These three confounding factors dominate much of the province of British Columbia (BC), Canada. An artificial neural network (ANN) was created using as predictors six gridded SWE products previously evaluated for BC. Relevant spatiotemporal covariates were also included as predictors, and observations from manual snow surveys at stations located throughout BC were used as target data. Mean absolute errors (MAEs) and interannual correlations for April surveys were found using cross-validation. The ANN using the three best-performing SWE products (ANN3) had the lowest mean station MAE across the province. ANN3 outperformed each product as well as product means and multiple linear regression (MLR) models in all of BC's five physiographic regions except for the BC Plains. Subsequent comparisons with predictions generated by the Variable Infiltration Capacity (VIC) hydrologic model found ANN3 to better estimate SWE over the VIC domain and within most regions. The superior performance of ANN3 over the individual products, product means, MLR, and VIC was found to be statistically significant across the province.
Citation: Snauffer, A. M., Hsieh, W. W., Cannon, A. J., and Schnorbus, M. A.: Improving gridded snow water equivalent products in British Columbia, Canada: multi-source data fusion by neural network models, The Cryosphere, 12, 891-905, https://doi.org/10.5194/tc-12-891-2018, 2018.
Publications Copernicus
Download
Short summary
Estimating winter snowpack throughout British Columbia is challenging due to the complex terrain, thick forests, and high snow accumulations present. This paper describes a way to make better snow estimates by combining publicly available data using machine learning, a branch of artificial intelligence research. These improved estimates will help water resources managers better plan for changes in rivers and lakes fed by spring snowmelt and will aid other research that supports such planning.
Estimating winter snowpack throughout British Columbia is challenging due to the complex...
Share