Snow cover thickness estimation using radial basis function networks

Abstract

Abstract. This paper reports an experimental study designed for the in-depth investigation of how the radial basis function network (RBFN) estimates snow cover thickness as a function of climate and topographic parameters. The estimation problem is modeled in terms of both function regression and classification, obtaining continuous and discrete thickness values, respectively. The model is based on a minimal set of climatic and topographic data collected from a limited number of stations located in the Italian Central Alps. Several experiments have been conceived and conducted adopting different evaluation indexes. A comparison analysis was also developed for a quantitative evaluation of the advantages of the RBFN method over to conventional widely used spatial interpolation techniques when dealing with critical situations originated by lack of data and limited n-homogeneously distributed instrumented sites. The RBFN model proved competitive behavior and a valuable tool in critical situations in which conventional techniques suffer from a lack of representative data.

Improving gridded snow water equivalent products in British Columbia, Canada: multi-source data fusion by neural network models

“…More recently Tong et al (2010) compared SWE predictions of ANNs using microwave brightness temperatures (TBs) to those of the SPD algorithm Aschbacher (1989) and other TB difference algorithms (Chang et al, 1987;Derksen, 2008) in the Quesnel River Basin of BC, finding that ANNs which included the most TB channels outperformed other networks and algorithms. Binaghi et al (2013) applied radial basis function networks to estimate snow cover thickness in the Italian Central Alps, finding that this approach outperformed inverse distance weight and spline interpolation methods commonly used in similar contexts with limited numbers of homogeneously distributed measurement sites. These and other studies covered in various review papers (Gan, 1996;Evora and Coulibaly, 2009;Shi et al, 2016) demonstrate the promise of more accurate snow estimates via machine learning methods, but they do not incorporate existing SWE products directly.…”

Using the MODIS Sensor for Snow Cover Modeling and the Assessment of Drought Effects on Snow Cover in a Mountainous Area

“…This model presented a desirable result, with a correlation coefficient of R = 0.90, which is consistent with the results of the present study. Binaghi et al [38] used the RBF neural network to estimate the thickness (depth) of snow in the central Alpine area in Italy. Meteorological inputs (temperature and precipitation) were used in the study and NRMSE ranged from 0.04 to 0.07, indicating a better accuracy than the current study.…”

Improving gridded snow water equivalent products in British Columbia, Canada: multi-source data fusion by neural network models

“…More recently Tong et al (2010) compared SWE predictions of ANNs using microwave brightness temperatures (TBs) to those of the SPD algorithm Aschbacher (1989) and other TB difference algorithms (Chang et al, 1987;Derksen, 2008) in the Quesnel River Basin of BC, finding that ANNs which included the most TB channels outperformed 5 other networks and algorithms. Binaghi et al (2013) applied radial basis function networks to estimate snow cover thickness in the Italian Central Alps, finding that this approach outperformed inverse distance weight and spline interpolation methods commonly used in similar contexts with limited numbers of homogeneously distributed measurement sites. These and other studies covered in various review papers (Gan, 1996;Evora and Coulibaly, 2009;Shi et al, 2016) demonstrate the promise of more accurate snow estimates via machine learning methods, but they do not incorporate existing SWE products directly.…”