Accurately determining the freeze/thaw state (FT) is crucial for understanding land-atmosphere interactions, with significant implications for climate change, ecological systems, agriculture, and water resource management. This article introduces a novel approach to assess FT dynamics by comparing the new diurnal amplitude variations (DAV) algorithm with the traditional seasonal threshold algorithm (STA) based on the soil moisture active passive (SMAP) brightness temperature data. Utilizing soil temperature profiles from 44 sites recorded by the National Ecological Observatory Network between July 2019 and June 2022. The results reveal that the DAV algorithm demonstrates a remarkable potential for capturing FT signals, achieving an average accuracy of 0.82 (0.89 for the SMAP-FT product) across all sites and a median accuracy of 0.94 (0.92 for the SMAP-FT product) referring to soil temperature at 0.02 m. Notably, the DAV algorithm outperforms the SMAP-adopted STA in 25 out of 44 sites. The accuracy of the DAV algorithm is affected by daily temperature fluctuations and geographical latitudes, while the STA exhibits limitations in certain regions, particularly those with complex terrains or variable climatic patterns. This article's innovative contribution lies in systematically comparing the performance of the DAV and STA algorithms, providing valuable insights into their respective strengths and weaknesses.

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).