Snow cover phenology (SCP) is critical to the climate system. China has the most comprehensive snow cover distribution in the middle and low latitudes and has shown dramatic changes over the past few decades. However, the spatiotemporal characteristics of SCP parameters and their sensitivity to meteorological factors (temperature and precipitation) under different conditions (altitude, snow cover classification, or season) in China are insufficiently studied. Therefore, using improved daily MODIS cloud-gap-filled (CGF) snow-cover-extent (SCE) products, the spatiotemporal characteristics (distribution and variation) and respond to climate of snow cover area (SCA), snow cover start (SCS), snow cover melt (SCM), and snow cover days (SCD) are explored from 2000 to 2020. The results show that in the past 20 years, snow cover in China has demonstrated a trend of decreasing SCA, decreasing SCD, advancing SCS, and advancing SCM, with SCM advancing faster than SCS. The greatest snowfall occurs in January, mainly in northeastern China, northern Xinjiang, and the Tibet Plateau. Spatially, the slope of SCP was mainly within ±0.5 day/year (d/y) Statistics indicated that the area proportion where SCD is significantly reduced is greater than increased; SCD, SCS, and SCM are shortened or advanced in three snow-covered area classifications. Moreover, compared with precipitation, the significantly correlated regions (6–47.2% more than precipitation) and correlation degree (1.23–8.33 times precipitation in significantly correlated snow cover classification) between temperature and SCP in different seasons are larger. For stable snow-covered areas (SSA), SCD are mainly affected by spring temperature below 1500 m and mainly by autumn temperature above 1500 m; the precipitation is more affected in autumn. The correlation of SCP with temperature and precipitation has obvious spatial and seasonal differences and shows characteristic variation with altitude. These results can provide important data support for climate prediction, hydrological research, and disaster warning.

High-resolution Synthetic Aperture Radar (SAR), as an efficient Earth observation technology, can be used as a complementary means of observation for snow depth (SD) and can address the spatial heterogeneity of mountain snow. However, there is still uncertainty in the SD retrieval algorithm based on SAR data, due to soil surface scattering. The aim of this study is to quantify the impact of soil signals on the SD retrieval method based on the cross-ratio (CR) of high-spatial resolution SAR images. Utilizing ascending Sentinel-1 observation data during the period from November 2016 to March 2020 and a CR method based on VH- and VV-polarization, we quantitatively analyzed the CR variability characteristics of rock and soil areas within typical thick snow study areas in the Northern Hemisphere from temporal and spatial perspectives. The correlation analysis demonstrated that the CR signal in rock areas at a daily timescale shows a strong correlation (mean value > 0.60) with snow depth. Furthermore, the soil areas are more influenced by freeze-thaw cycles, such that the monthly CR changes showed no or negative trend during the snow accumulation period. This study highlights the complexity of the physical mechanisms of snow scattering during winter processes and the influencing factors that cause uncertainty in the SD retrieval, which help to promote the development of high-spatial resolution C-band data for snow characterization applications.

High spatial resolution snow depth (SD) is crucial for hydrological, ecological, and disaster research. However, passive microwave SD product (10/25 km) is increasingly insufficient to meet contemporary requirements due to its coarse spatial resolution, particularly in heterogeneous alpine areas. In this study, we develop a superior SD downscaling algorithm based on the FT-Transformer (Feature Tokenizer + Transformer) model, termed FTSD. This algorithm fuses the latest calibrated enhanced resolution brightness temperature (CETB) (3.125/6.25 km) with daily cloud-free optical snow data (500 m), including snow cover fraction (SCF) and snow cover days (SCD). Developed and evaluated using 42,692 ground measurements across China from 2000 to 2020, FTSD demonstrated notable improvements in accuracy and spatial resolution of SD retrieval. Specifically, the RMSE of temporal and spatiotemporal independent validation for FTSD is 7.64 cm and 9.74 cm, respectively, indicating reliable generalizability and stability. Compared with the long-term series of SD in China (25 km, RMSE = 10.77 cm), FTSD (500 m, RMSE = 7.67 cm) provides superior accuracy, especially improved by 48% for deep snow (> 40 cm). Moreover, with the higher spatial resolution, FTSD effectively captures the SD’s spatial heterogeneity in the mountainous regions of China. When compared with downscaling algorithms utilizing the raw TB data and the traditional random forest model, the CETB data and FT-Transformer model optimize the RMSE by 10.08% and 4.84%, respectively, which demonstrates the superiority of FTSD regarding data sources and regression methods. Collectively, these results demonstrate that the innovative FTSD algorithm exhibits reliable performance for SD downscaling and has the potential to provide a robust data foundation for meteorological and environmental research.

Deep learning (DL) has been widely applied to enhance the accuracy and reconstruction of snowpack parameters due to its powerful nonlinear mapping ability. Here, we propose a broadband snow albedo (BA) estimation method that infuses the asymptotic radiative transfer (ART) model into the DL model and thus imposes constraints on the input variables of the DL model and compares it to the MOD10A1 BA product and the BA simulated using the ART model alone. The results show that coupling a physical mechanism-based snow radiative transfer model with a DL model can successfully improve the estimation accuracy of BA. Compared to the MOD10A1 BA product and the BA simulated using the ART model alone, the accuracy of the BA simulated by the convolutional neural network model under the constraints of the ART model is improved by 12% and 3%, respectively, and the accuracy of the BA simulated by the long short-term memory model under the constraints of the ART model is improved by 14% and 5%, respectively. The coupling paradigm that integrates the snow radiative transfer model into the DL model effectively balances the advantages of both models. This coupling paradigm creates a complementary association between the snow radiative transfer model and the DL model. This research effectively enhances the simulation and predictive capabilities of snow albedo and is expected to promote the in-depth application of remote sensing big data in combination with physical modeling in the field of cryosphere parameter estimation.

Wetland monitoring is a key means of protecting wetland ecosystems. In order to achieve continuous monitoring of wetlands and predict future patterns, this paper analyzes the spatiotemporal evolution characteristics of wetlands in the upper reaches of the Yellow River from 1990 to 2020, and uses the Patch Generation Land Use Simulation (PLUS) model to simulate the spatial distribution of wetlands from 2040 to 2060 under four scenarios: farmland protection (FPS), wetland protection (WPS), comprehensive protection (CPS) and natural development (NDS). The results show that the total area of wetlands in the upper reaches of the Yellow River is on the rise, increasing by 7.12% in 2020 compared with 1990. The changes in various types of wetlands are different: the areas of river and canals increased by 26.39% and 57.97%, respectively, paddy fields increased by 7.95%, lakes remained basically stable, and tidal flats decreased by 5.67%. The simulation results of the future spatial pattern of wetlands show that: under the FPS scenario, farmland and related land use will expand significantly, mainly through the development of beaches, dry land and unused land, while under the WPS scenario, wetlands will be strictly protected, the area of water resource features such as rivers, lakes and reservoirs will increase significantly, and land use changes will be more ecologically oriented. Compared with the CPS and NDS scenarios, the wetland protection and urbanization process in the upper reaches of the Yellow River can be balanced under the FPS and WPS scenarios. This study has important reference value for the protection and sustainable development of wetland ecosystems in the upper reaches of the Yellow River.