Monitoring snow variations can know how fast the snow is melted and how much water is frozen on the Earth. However, traditional ground techniques have some limitations to monitor snow variations due to high cost. Recently, global navigation satellite systems-reflectometry (GNSS-R) has been developed and used to sense snow variations, while previous studies mostly used global positioning system (GPS) observations to estimate snow depth. In this paper, snow depth is estimated from GLONASS signal-to-noise ratio (SNR) and phase-based multipath reflectometry at one IGS GANP station in Slovakia, which is compared with in situ observations. For two snow seasons, snow depth estimations from GLONASS SNR1 and SNR2 data have a good agreement with in situ results with correlations of 0.94 and 0.94 in 2012 and 0.92 and 0.89 in 2013, respectively. Compared to GPS results, snow depth results from GLONASS are almost similar, but have some differences due to low coverage of less GLONASS satellites. However, results from GLONASS geometry-free linear combination (L4) are not pretty much, which needs more works to improve it in the future. Combined GPS and GLONASS observations have no significant improvement on the precision, but improve the spatial resolution because of more satellites.