The Arctic-boreal zone (ABZ) is warming due to climate change. Current spaceborne remote sensing techniques and retrieval methodologies need to be complemented to improve systematic monitoring of the cryosphere. To that end, this article presents a new investigation of the use of the global navigation satellite system reflectometry (GNSS-R) remote sensing technique by a SmallSat constellation. A new freeze/thaw (F/T) seasonal multithresholding algorithm (STA) is developed using high-inclination orbit near-Nadir Spire Global GNSS-R data acquired through the National Aeronautics and Space Administration (NASA) Commercial Smallsat Data Acquisition (CSDA) Program. Five different soil surface reflectivity Gamma models are proposed to account for the impact of vegetation cover and small-scale surface roughness on Earth-reflected GNSS signals. The sensitivity of the Gamma models to F/T surface state transitions is evaluated, and the optimum model is selected to construct a seasonal scale factor. Then, a multithresholding matrix is obtained for F/T classification using a specific threshold for every surface grid cell. Results for the annual frozen soil duration (days yr(-1)) are compared with those by the Soil Moisture Active Passive (SMAP) mission. Additionally, freezing and thawing periods are analyzed to determine when the moisture exchange with the atmosphere is locked, which is an important climatic factor. A novel metric is introduced to characterize the freeze intensity moving beyond classical F/T binary classifications. Results are evaluated using air and soil temperature, snow depth and temperature, and soil moisture content (SMC) provided by the European Centre for Medium-Range Weather Forecasts (ECMWF) ERA5-Land reanalysis product.

The freeze-thaw (F/T) process plays a significant role in climate change and ecological systems. The soil F/T state can now be determined using microwave remote sensing. However, its monitoring capacity is constrained by its low spatial resolution or long revisit intervals. In this study, spaceborne Global Navigation Satellite System-Reflectometry (GNSS-R) data with high temporal and spatial resolutions were used to detect daily soil F/T cycles, including completely frozen (CF), completely thawed (CT), and F/T transition states. First, the calibrated Cyclone Global Navigation Satellite System (CYGNSS) reflectivity was used for soil F/T classification. Compared with those of soil moisture active and passive (SMAP) F/T data and in situ data, the detection accuracies of CYGNSS reach 75.1% and 81.4%, respectively. Subsequently, the changes in spatial characteristics were quantified, including the monthly occurrence days of the soil F/T state. It is found that the CF and CT states have opposite spatial distributions, and the F/T transition states distribute from the east to the west and then back to the east of the Qinghai-Tibet Plateau, which may be due to varying diurnal temperatures in different seasons. Finally, the first day of thawing (FDT), last day of thawing, and thawing period of the F/T year were analyzed in terms of the changes in temporal characteristics. The temporal variation of thawing is mainly different between the western and eastern parts of the Tibetan Plateau, which is in agreement with the spatial variation characteristics. The results demonstrate that the CYGNSS can accurately detect the F/T state of near-surface soil on a daily scale. Moreover, it can complement traditional remote sensing missions to improve the F/T detection capability. It can also expand the applications of GNSS-R technology and provide new avenues for cryosphere research.