Monitoring Snow Depth by GNSS Reflectometry in Built-up Areas: A Case Study for Wettzell, Germany

摘要

Snow storage dynamics is essential to predict floods, to quantify water resources for human use and irrigation, and to assess the risk of avalanches. Recently, Global Navigation Satellite System (GNSS) ground stations have been successfully used to continuously estimate snow depth at an intermediate scale of about 1.000 m2 around the stations. In this study, GNSS signal-to-noise ratio (SNR) data at the station Wettzell, Germany, are used to estimate snow depth variations from 2012 to 2015. The station Wettzell is located in a built-up area. The most challenging task at this site is to separate the GNSS reflections from the ground and from surrounding buildings. We modified the interference approach previously used for snow depth estimation using the phase of the multipath interference pattern instead of their frequency. Additionally, we complemented the analysis by including satellites transmitting the L2P signal into the processing. We studied the performance of the L1 signal. The derived GNSS snow depth ranges between 3 and 25 cm and corresponds well to in situ observations by an ultrasonic sensor, with a correlation of 0.8 for daily time series. The residuals of GNSS snow depths compared to the ultrasonic sensor reveal a root-mean-squared error (RMSE) of 4.3 cm for the L2 and 5.9 cm for the L1 signal with a small bias of 1 cm. The results show that the existing data of the global GNSS tracking network promise to provide valuable complementary snow depth observations to the existing sensors at several hundred sites worldwide, including urban areas.