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).

The vertical temperature distribution in the permanently shaded region (PSR) has a significant impact on the temporal and spatial distribution of the cold trap. To obtain the vertical temperature profile of the PSR, an inversion method that fuses microwave and infrared brightness temperature (TB) data is proposed. In the inversion process, the infrared data were initially used to derive the optimal value of the H-parameter that controls the density profile. Subsequently, high-frequency (37 and 19.35 GHz) microwave TB data were used to ascertain the range of surface density, whereas low-frequency (3 GHz) microwave TB data were used to determine the range of bottom density. A fixed correction was applied to the 3-GHz brightness temperature data to account for the calibration error. Due to the inherent uncertainties associated with the thermal model, both the Hayne and Woods' models were used in the inversion process, yielding disparate results. The PSR in the Haworth impact crater was selected as a case study for the inversion. The Woods' model was found to provide a superior explanation for the microwave observation. The optimal surface density of the PSR of the Haworth crater was determined to be within the range of 1200-1300 kg m(-3), while the bottom density was within the range of 2100-2200 kg m(-3). The inverted vertical temperature distribution in the PSR of Haworth crater indicates that the depth of the cold trap can reach approximately 8.5 m. In addition, the impact of heat flow on microwave TB is discussed.

Near-surface temperatures of permanently shadowed regions (PSRs) on the Moon provide fundamental information for water ice exploration. Seasonal temperature variations of PSRs are found in both Chang'E-2 microwave radiometer data and Diviner Lunar radiometer observations. Furthermore, unusual microwave brightness temperature variations between February 2011 and May 2011 of double-shaded PSRs are shown in the Chang'E-2 observational data, i.e., that the minimum microwave brightness temperature occurs before the time when the infrared brightness temperature reaches the minimum in double-shaded PSRs. To interpret this phenomenon, the 1-D thermal model and the microwave radiation transfer model are used. In the thermal model, the reradiation energy from the illuminated area is estimated by effective solar irradiance, which is an analytic solution for the radiative equilibrium temperature in the shadowed area of a spherical bowl-shaped crater. In the simulation, an assumed internal 0.4 W/m(2) heat flow beneath the lunar surface made a plausible fit to the unusual variations during some lunations. However, this is a huge value compared with the well-known heat flow value of about 0.018 W/m(2). Furthermore, it is difficult to obtain this extra heat energy by lateral conduction below the surface in a large impact crater due to the small thermal conductivity of the lunar regolith. Finally, the unusual microwave brightness temperature (TB) changes are concluded to be caused by a calibration problem after excluding other possible reasons. In addition, a statistical correction method is applied to revise the problematic TB data to obtain the proper variation trend of the brightness temperature.