Estimation of Snow Depth Using L1 GPS Signal-to-Noise Ratio Data

Abstract

Accurate measurements of snowpack properties are needed by scientists to better understand effects of climate variability on water resource availability. Satellite measurements currently assess snow cover rather than snow depth. Many in situ snow sensors/networks lack the necessary spatial and temporal sensitivity needed for such studies. Existing GPS networks are a potential source of new snow data for climate and hydrology studies, but current operational analyses only use signal-to-noise ratio (SNR) data from the new GPS signal centered at 1.2 GHz (L2C). These data are often unavailable in GPS archives. A snow depth algorithm that used the older (less precise) GPS signal centered at 1.5 GHz (L1) would provide longer snow depth time series that are needed by climate scientists. Here, an algorithm is developed to use the L1 SNR data. Snow depth estimates are derived for 23 sites for 5 years. These data are compared with existing snow depth time series derived from the L2C signal. They show an average bias of 1 cm and correlation of 0.95. Some of this disagreement is due to differences in the azimuthal coverage of the two datasets. The L1 snow depth solutions are also compared with in situ measurements, yielding a bias of $- {4}\;\text{cm}$ , comparable to the $- {6}\;\text{cm}$ bias found in a previous study of the L2C retrieval algorithm.