A comprehensive evaluation of utilizing BeiDou data to estimate snow depths from two ground-based stations.

The BeiDou Navigation Satellite System (BDS) of China started to provide global services in 2020, which provides a new data source for snow depth sensing through the Global Navigation Satellite System Interferometric Reflectometry (GNSS-IR) technique. This study, taking advantage of two stations in the snow season of 2020, first evaluates the performance of the five state-of-the-art snow depth models on BDS data and provides a comprehensive analysis of the characteristics of BDS data toward snow depth estimations. The results show that the snow depth estimations using the SNR, triple-frequency SNR combination (SNR_COM), and triple-frequency phase combination (F3) models are comparable, with high correlation coefficient R² (root-mean-square error, RMSE) values of 0.93 (3.17 cm), 0.93 (3.12 cm), and 0.93 (3.27 cm), respectively. The models of geometry-free linear combinations of the phase measurements (L4) and combination of pseudorange and carrier phase of dual-frequency signals (F2C) are slightly poorer, with mean R² (RMSE) values of 0.69 (7.18 cm) and 0.82 (4.68 cm), respectively. The snow depth retrieval results show similar accuracy using data from the BDS-3 and BDS-2 (RMSE = 3.63 cm vs. 3.72 cm). The results from the IGSO satellites are more reliable than those from the MEO satellites. The effects of satellite elevation angles and sampling rates are also analyzed to determine the optimal parameter selections (i.e., 5°–25° or 5°–30° and sampling rates < 120 s). The findings of this study can provide supporting information to determine the strategy for using BDS signals to retrieve snow depth. [ABSTRACT FROM AUTHOR]