Estimating the landscape and soil freeze-thaw (FT) dynamics in the Northern Hemisphere (NH) is crucial for understanding permafrost response to global warming and changes in regional and global carbon budgets. A new framework for surface FT-cycle retrievals using L-band microwave radiometry based on a deep convolutional autoencoder neural network is presented. This framework defines the landscape FT-cycle retrieval as a time-series anomaly detection problem, considering the frozen states as normal and the thawed states as anomalies. The autoencoder retrieves the FT-cycle probabilistically through supervised reconstruction of the brightness temperature (TB) time series using a contrastive loss function that minimizes (maximizes) the reconstruction error for the peak winter (summer). Using the data provided by the Soil Moisture Active Passive (SMAP) satellite, it is demonstrated that the framework learns to isolate the landscape FT states over different land surface types with varying complexities related to the radiometric characteristics of snow cover, lake-ice phenology, and vegetation canopy. The consistency of the retrievals is assessed over Alaska using in situ observations, demonstrating an 11% improvement in accuracy and reduced uncertainties compared to traditional methods that rely on thresholding the normalized polarization ratio (NPR).

This paper presents a convolutional autoencoder deep learning framework for probabilistic characterization of the ground freeze-thaw (FT) dynamics in the Northern Hemisphere to enhance our understanding of permafrost response to global warming and shifts in the high-latitude carbon cycle, using Soil Moisture Active Passive (SMAP) satellite brightness temperatures (TB) observations. The autoencoder recasts the FT-cycle retrieval as an anomaly detection problem in which the peak winter (summer) represents the normal (anomaly) segments of the TB time series. The results demonstrate that the new framework outperforms the widely used fixed-thresholding of the Normalized Polarization Ratio (NPR) by learning the land surface structural and radiometric complexities that might arise in TB times series due to snow cover and vegetation. Validation against ground-based measurements over Alaska shows that the accuracy of the FT-cycle retrievals can be improved by 12%, primarily due to a marked reduction in false detection of short snowmelt episodes as ground thawing by the NPR thresholding approach.

The process of soil freezing and thawing refers to the alternating phase change of liquid water and solid water in the soil, accompanied by a large amount of latent heat exchange. It plays a vital role in the land water process and is an important indicator of climate change. The Tibetan Plateau in China is known as the roof of the world, and it is one of the most prominent physical characteristics is the freezing and thawing process of the soil. For the first time, this paper utilizes the spaceborne GNSS-R mission, i.e., CYGNSS (Cyclone Global Navigation Satellite System), to study the feasibility of monitoring the soil freeze-thaw (FT) cycles on the Tibetan Plateau. In the theoretical analysis part, model simulations show that there are abrupt changes in soil permittivities and surface reflectivities as the soil FT occurs. The CYGNSS reflectivities from January 2018 to January 2020 are compared with the SMAP FT state. The relationship between CYGNSS reflectivity and SMAP soil moisture within this time series is analyzed and compared. The results show that the effect of soil moisture on reflectivity is very small and can be ignored. The periodic oscillation change of CYGNSS reflectivity is almost the same as the changes in SMAP FT data. Freeze-thaw conversion is the main factor affecting CYGNSS reflectivity. The periodical change of CYGNSS reflectivity in the 2 years indicates that it is mainly caused by soil FT cycles. It is feasible to use CYGNSS to monitor the soil FT cycles in the Tibetan Plateau. This research expands the current application field of CYGNSS and opens a new chapter in the study of cryosphere using spaceborne GNSS-R with high spatial-temporal resolution.