One of the main features of the Asian Water Tower imbalance is the massive melting of snow, necessitating enhanced snow monitoring. However, the sensors aboard polar-orbiting satellites, such as MODIS, yield only one to two valid observations daily. This, coupled with the extensive cloud cover and prolonged duration over the Asian Water Tower, results in a significant number of data gaps. China's new generation of geostationary satellites FY-4A and FY-4B have high frequency observations, making it possible to monitor snow with high precision. In this study, we systematically analyze the image pixel size stretching of FY-4A and FY-4B in the Asian Water Tower region, based on their imaging geometry. This analysis offers a theoretical foundation for fractional snow cover retrieval in the subsequent integration of these two satellites. Concurrently, this study conducts fractional snow cover (FSC) retrieval for FY-4A and FY-4B, utilizing the multiple endmember spectral mixture analysis algorithm with automatic endmember extraction (MESMA-AGE). High spatial resolution Landsat-8 imagery serves as reference data for accuracy assessment. The results indicated that FY-4A's retrieval accuracy remained unaffected by pixel size stretching, achieving an Overall Accuracy (OA) of up to 0.97 and a Root Mean Square Error (RMSE) of less than 0.13. For FY-4B, the retrieval accuracy demonstrated higher snow identification precision, with an OA of up to 0.95. However, the RMSE varied significantly due to pixel size stretching, ranging from 0.12 to 0.21. FY-4A and FY-4B fusion enables high precision and near-real-time snow monitoring.

The normalized difference snow index (NDSI) is a fundamental spectral indicator of snow/ice in visible and shortwave-infrared imagery. The complex grain shapes in nature have well-known significant effects on the single-scattering properties (SSPs) and subsequently the bidirectional reflectance of snow. The shape effects on snow NDSI need to be further characterized as NDSI is a nonlinear combination of two reflectance bands. Considering the common snow grain shapes represented by sphere, spheroid, hexagonal plate, and Koch snowflake, we use the ray-tracing approach to simulate the SSPs of ice particles and the discrete ordinate algorithm to solve the bidirectional reflectance function and calculate NDSI of snow. According to simulating results, the angular pattern of snow NDSI is subject to snow grain shape, whereas the shape effects can be significantly weakened by the increasing surface roughness of ice particles. The shape of Koch snowflake causes an NDSI habit different from other three shapes for large snow grain size. Moreover, snow NDSI also has complex responses to aspect ratio (AR) for spheroid and hexagonal prism. The theoretical characterization of the snow NDSI responses to various grain shapes would enrich the knowledge of NDSI variation mechanism in snow-covered area mapping applications.

The normalized difference snow index (NDSI) plays an important role in mapping snow cover with spaceborne visible and shortwave-infrared imagery. The NDSI variation depends on illuminating-viewing geometry and snow physical properties, including equivalent grain size (EGS), snow depth (SD) and impurity concentration, as well as fractional snow cover (FSC) within a mixed pixel; however, it is still not fully understood. To quantifiably characterize the pattern of snow NDSI variation, we use a light scattering model of snow to calculate bidirectional reflectance and consequent NDSI values for a wide range of illuminating-viewing geometries, SD, and EGS values. In the model designated bicontinuous snow model using Geometric Optics theory and Radiative Transfer Equation (bicontinuous-GO/RTE), snowpack is represented by a bicontinuous microstructure, and bidirectional reflectance is simulated based on geometric optics and vector radiative transfer equation. The discrete ordinates radiative transfer (DISORT) algorithm is used to simulate the soot concentration effect on snow NDSI. A soil spectral reflectance model (SOILSPECT) is utilized with the assumption of the linear spectral mixture of snow and soil to quantify the effect of FSC on NDSI. As for discontinuous forests, an analytical hybrid geometric-optical and radiative transfer (GORT) model in conjunction with the bicontinuous-GO/RTE model and the PROpriétésSPECTrales (PROSPECT) model is used to examine the effect of canopy cover, which is related to the maximal FSC viewable to satellites. Modeling results indicate that: 1) snow NDSI is comparably low at off-nadir viewing angles, and this effect is exacerbated by the decline in solar elevation but limited by large EGS; 2) snow NDSI increases with EGS yet becomes saturated at EGS of 500$\mu \text{m}$; 3) the effect of SD that is as low as 1 cm on NDSI is rarely noticeable; 4) the concentration of internally mixed soot up to 1 ppm has little reducing effect on snow NDSI ($> 20^{\circ }$in forests; and 8) forests complicate the nonlinear relationship between NDSI and canopy cover with fully snow-covered ground beneath canopies. These findings imply important uncertainty sources of binary and FSC mapping with NDSI.

Spatiotemporal continuous fractional snow cover(FSC) dataset is needed as an important input for the study of large-scale hydrological, meteorological and climate research. But optical data often have gap due to cloud cover. The widely used Moderate Resolution Imaging Spectroradiometer (MODIS) snow cover dataset uses an eight-day composite approach to remove cloud effects, but it cannot meet the requirement of monitoring snow cover, a parameter with high temporal and spatial variability. This paper uses multiple endmember spectral mixture analysis (MESMA) algorithm to retrieve FSC from geostationary satellite (FY-4A and Himawari-8) and polar-orbiting satellite (MODIS) data; Then geostationary satellite retrieve results are used to fill the MODIS cloud cover pixels, which can reduce the cloud cover from 50% to 15%; Finally, the daily FSC product is obtained by using Piecewise Cubic Hermite Interpolating Polynomial (PCHIP) algorithm. Landsat-8/OLI data processed by the MESMA algorithm is determined as “ground truth” to validate this product. The result shows that the accuracy of the daily cloud-free FSC product is high with the root mean square error is 0.1-0.15.

Many algorithms and products for snow cover have been developed. We built a high spatial and temporal resolution validation data set, which can be used as a unified standard for the validation of existing snow products. In recent years, Sentinel-2 MSI has been widely used because of its higher spatial and temporal resolution. In this study, we applied the linear spectral mixture analysis to Sentinel-2 MSI to obtain the fractional snow cover products, then we explored whether the Sentinel-2 MSI FSC products can meet the accuracy requirements with GF-2 data. The results show that the accuracy of Sentinel-2 MSI FSC can meet the requirements. Finally, we use Sentinel-2 MSI FSC to validate the MODAGE product.

The inversion of snow properties with optical remote sensing often relies on snow reflectance modelling. In this study, snow reflectance models including Mie-DISORT, Mie-Mishchenko and ART and the model based on bicontinuous microstructure are evaluated and compared with in situ measurements. Results show the bicontinuous microstructure based reflectance model can accurately simulate spectral albedo of clean and sooty snow, and have best performance on directional reflectance of snow surface compared with other three models.

The moderate resolution imaging spectroradiometer (MODIS) snow algorithm has been used to generate global fractional snow cover (FSC) at a pixel size of 500 m using a linear regression relationship (called “FRA6T”) between FSC and the normalized difference snow index (NDSI). However, the linear relationship is problematic because of the considerable NDSI variation in nonsnow conditions. In this letter, we propose a universal ratio snow index (URSI), which is the ratio of the visible reflectance and the sum of the near infrared and shortwave infrared reflectances. It is called “universal” because it has weak sensitivity under snow-free ground conditions and, therefore, can improve the stability of the linear snow index methodology. A comparison between NDSI and URSI with regard to estimate FSC using the linear snow index methodology is carried out for the Tibetan Plateau. The scatter plots of MODIS NDSI/URSI and Landsat-7 Enhanced Thematic Mapper Plus (ETM+) FSC indicate that a linear relationship can be assumed for both NDSI and URSI for barren land conditions and is more appropriate for URSI than it is for NDSI in forested areas. Validation efforts show that the linear relationship using URSI (designated “FracURSI”) achieves fewer errors in FSC estimation compared with the developed NDSI method (“FracNDSI”), particularly for forested areas and for moderate FSC values. Averaged over all comparisons, the root-mean-square error (RMSE) of FSC estimates for FRA6T is 0.13, and for FracNDSI is 0.12, whereas FracURSI RMSE is 0.11.

High-resolution soil moisture dataset is crucial for various application such as meteorology, climatology, hydrology and agriculture. Active microwave remote sensing sensors like radar provide earth observations at high spatial resolutions. This study based on physical model simulations (Advanced Integral Equation Method, AIEM, and Water Cloud Model, WCM) combined with the Artificial Neural Networks to investigate the potential of the ALOS-2 and Sentinel-1 radar images for estimating soil moisture at high spatial resolution. The results shows that the statistical parameters of the relationships between estimated and measured soil moisture, expressed in terms of R, bias, and RMSE, are 0.834~0.878, 1.59~3.65 vol% and 3.36~6.15 vol% for ALOS-2, and 0.722~0.896, 1.75~2.97 vol% and 3.24~6.86 vol%, for Sentinel-1. In densely vegetated area, RMSE significant increases, due to the limited penetration ability of L and C bands in high vegetation areas.

Many algorithms and products for snow cover have been developed. Then a unified set of “ground truth” data is important to validate snow cover products. In this study, Landsat-8/OLI data processed by linear unmixing algorithm was determined as “ground truth” to validate the moderate resolution snow products. In order to evaluate the cloud removing effect of the daily fractional snow cover (FSC) dataset of MODIS over High Asia, we use the MOD10A1 FSC product which is calculated by recommended equations as the before cloud removing data, then the Landsat-8/OLI FSC was used to validate both of the MODIS data. The results show that when the percentage of cloud pixels is less than 10%, the binary accuracy can reach 0.85 or more, the mean absolute error is less than 0.25, and the root mean square error is less than 0.35. These results suggest that the product has high credibility, despite there is still a small amount of cloud in the product.

High-resolution soil moisture dataset is important for studying cold and humid temperate forest climates, estimating forest carbon emissions and storage, and identifying the influence of water circulation and global change in Genhe area. Leaf Area Index (LAI) is an important vegetation biophysical variable and has been widely used for analysis of the vegetation biomass, land-surface process simulation, and many other global change studies. This paper based on LAI from Global LAnd Surface Satellite (GLASS), microwave polarization difference index (MPDI) from Soil Moisture Active Passive (SMAP) L band Brightness Temperature (TB), and synthetic Land Surface Temperature (LST) from combine AMSR-2 TB and MODIS LST data proposed a downscaling method for SMAP L3 soil moisture product. Using the multiple linear regression method, we obtained 1km spatial resolution of soil moisture data in Genhe area. The results showed that downscaling SMAP soil moisture can present more details than before with low errors.

This study investigated a data fusion method based on pixel-based robust stepwise regression technique to retrieve a long-term snow depth dataset from passive microwave observations. The NOAA's Snow Data Assimilation System (SNODAS) snow depth(SD) product covered the United States was selected as standard reference to train the brightness temperature data from the MEaSUREs Calibrated Enhanced-Resolution Passive Microwave Daily EASE-Grid 2.0 Brightness Temperature Earth System Data Record. In order to achieve robustness against the presence of outliers and avoid multicollinearity problem in regression, the robust stepwise regression technique was selected as the training approach. The retrieved snow depth were evaluated against in situ observations and SNODAS SD. The results show that the retrieved SD have a good agreement with both in situ observations and SNODAS SD, and are more consistent with SNODAS SD than in situ data.

Land surface freeze/thaw (FT) state is important for identifying the variable of carbon-nitrogen, water and energy cycling and soil erosion. The Soil Moisture Active Passive (SMAP) mission produces global and northern hemispheric daily landscape FT dataset [1] at a spatial resolution of 36 km from L-band radiometer. Parameterized discriminant function algorithm (PDFA) [2] detect landscape FT state from Advanced Microwave Scanning Radiometer 2 (AMSR2) observations. In this study, we use the in-situ soil temperature to validate the SMAP global FT dataset (36 km) and PDFA-based AMSR2 FT dataset (0.25°). Performance evaluation are obtained from two regions at northern hemispheric located in the China. The evaluation results show that overall accuracies usually greater than 85% for SMAP global FT dataset at Genhe (GH) region, but lower than 70% at Saihanba (SHB) region. AMSR2 FT dataset have accuracies that always higher than 85% at both regions.

China’s new generation of geostationary weather satellite, FengYun-4A (FY-4A) carrying the Advanced Geosynchronous Radiation Imager (AGRI) was launched on December 1 2016 and its data became publicly available on March 12 2018. The availability of high temporal observation at visible, near infrared, short-wave infrared and long-wave infrared bands over stable snow covered areas in China inspired this study to investigate the feasibility of FY-4A AGRI to estimate fractional snow cover. In this preliminary study, we present the multiple end-member spectral mixture analysis on FY-4A AGRI’s 5-band reflectances for snow covered area estimation. The end-members were extracted automatically from multispectral images and typical end-members were selected using the vector length. To account for the reflectance variation due to sun-target-sensor geometry, the end-member extraction and selection and subsequent estimation were implemented by the 5°×5° spatial window. The fractional snow cover estimates were evaluated by using reference data from corresponding Landsat-8 Operational Land Imager (OLI)’s 30-m resolution observations. The outcome indicated that FY-4A AGRI’s fractional snow cover agrees well with Landsat-8 OLI’s 30 m estimates. Comparisons showed that root mean squared error of FY-4A AGRI fractional snow cover ranges from 0.10 to 0.16 with R2exceeding 0.7.

Snow depth (SD) and snow water equivalent (SWE) are significant parameters in climate and hydrologic models. Successful estimation of SD (SWE) can improve the accuracy of snowmelt-runoff predictions and the management of water supplies. Currently, passive microwave (PMW) remote sensing is the most efficient way to monitor SD on global and regional scales; however, there are many challenges for accurate SD estimation. In this study, a new spatial dynamic method is developed by introducing random forest (RF) model, AMSR-2 TB and other auxiliary data. The main objective of this work is to produce long term SD dataset with the dynamic method using the Special Sensor Microwave Imager (SSM/I) and Special Sensor Microwave Imager/Sounder (SSMI/S) which span from 1987 to present. Through evaluation and analysis, the RF method performs better than traditional linear-fitting model. However, it tends to overestimate SD in shallow snow cover areas. Now, a preliminary spatial dynamic method (pixel-based model) is developed. For further work, the evaluation will be conducted to assess the feasibility of SD reconstruction. Moreover, to address overestimation over shallow snow areas, the snow depletion curve (SDC) incorporating SD and fractional snow cover (FSC) is expected to improve SD retrievals.

High resolution and long-term soil moisture products play an important role in estimating forest carbon storage and carbon emissions in Genhe, China. In order to obtain the high spatial and temporal resolution of soil moisture datasets in China, this paper proposed a downscaling method for the revised QP model with Dual-Channel Algorithm (QDCA) soil moisture product based on microwave polarization difference index (MPDI) from AMSR2 and Land Surface Temperature (LST), Normalized Difference Vegetation Index (NDVI) from Moderate resolution Imaging Spectroradiometer (MODIS) to derive high resolution of soil moisture data (1km). The downscaling method is validated with the in situ soil moisture data over Genhe in China, and the results showed that the R, bias, RMSE between downscaling revised QDCA soil moisture and in situ measurements is 0.176~0.4156, 009~0.050m3m-3and 0.056~0.087m3m-3, respectively. With the different land surface, the accuracy of downscaling soil moisture in grass land cover is higher than the forest land cover.

Ongoing information on snow and its extent is critical for understanding global water and energy cycles. Passive microwave data have been widely used in snow cover mapping for its long-time observation capabilities under all-weather conditions. But assessment of different passive microwave (PM) snow cover area (SCA) mapping algorithms have been rarely reported, especially in China. In this study, the performance of seven well documented successfully applied PM SCA mapping algorithms were tested using in situ snow depth measurements over China. The results shown in this study would contribute to the ongoing effort to improve the performance and applicability of PM SCA algorithms.

Fractional snow cover from Low-Earth-Orbit (LEO) satellites often encounters data gaps mainly caused by cloud obscuration in one or two observations in a single day. Imagers onboard GEOstationary (GEO) satellites have hourly or more frequent observations, making it possible to reduce cloud obscuration significantly. To map daily cloud-free fractional snow cover, we present a way using blended measurements from Terra/Aqua Moderate Resolution Imaging Spectroradiometer (MODIS) and FengYun-2 Visible and Infrared Spin Scan Radiometer (FY-2 VISSR). The fractional snow cover from MODIS is estimated using the Multiple End-member Spectral Mixture Analysis (MESMA), while that from FY-2 VISSR is generated using a simple linear interpolation between snow and snow-free end-members. Quickly-moving clouds can be alleviated mainly by the utilization of FY-2 VISSR multi-temporal data. Then spatio-temporally continuous daily fractional snow cover results from the removal of remaining clouds through the interpolation pixel by pixel in time series of snow fraction.

The main objective of this work is to improve the snow depth (SD) estimation for the Special Sensor Microwave Imager (SSM/I) and Special Sensor Microwave Imager/Sounder (SSMI/S) in China. To avoid systematic bias of different sensors to brightness temperature, it's better to retrieve SD with similar sensors. Meanwhile, long-term dataset is essential, which has significant impact on climate, weather and water resources. Sensors (SSM/I, SSMI/S) that span from 1987 to present are optimal for long time series of SD product generation. However, the accuracy of current SD (SWE) product still can't meet demand of climate and hydrological models. In this study, each grid-cell SD was estimated as the sum of SDs from each land cover algorithm weighted by percentages of land cover types. Through evaluation of this algorithm using measurements from 2005-2006, the root mean square errors (RMSE) are about 3.2, 2.9 and 5.3 cm for farmland, grassland and forest respectively. However, for mixed pixels, the RMSE is 6.6 cm. Finally, reasons about poor performance for mixed pixels were discussed. For further work, incorporating 10 GHz and fractional snow cover (FSC) is expected to improve SD retrievals in China.

For the multi-frequency sensors such as AMSR-E and AMSR-2, the verification results of QP model with dual-channel algorithm (QDCA) soil moisture product are not good in Genhe area. In order to obtain the long time series of soil moisture datasets and improve the accuracy of QDCA in China, this paper improved the vegetation correction method on current QDCA soil moisture algorithm. This paper incorporated the Vegetation water content (VWC) formulation from SMAP (Soil Moisture Active Passive) algorithm to estimate vegetation opacity thickness (VOD), b values and single scattering albedo redefined by different situations. The results showed that R2between revised QDCA retrievals and in situ measurements increased by approximately 18.4% and 10.1% and RMSE decreased by about 5% and 8.5% for the ascending and descending orbits, respectively.

Three MODIS-based fractional snow cover data are evaluated over the Tibetan Plateau from May, 2013 to April, 2015 with Landsat8/OLI data, including MOD10A1 in MODIS snow products collection 6[1, 2], along with MODSCAG[3] and MODAGE[4] fractional snow cover data which were retrieved based on linear spectral mixture analysis algorithm. The significant difference between MODSCAG and MODAGE is the endmember selection approach. This study compared these three products against the `true value' derived from Landsat-8/OLI. Forests, grass and soil region were chosen to carry out the evaluation, as well as the Himalaya Mountain due to its acute topographic heterogeneity. We applied both binary and fractional metrics to evaluate all three of them with the spatial resolution increasing from 500m to 1km, 2km and 5km. We also quantitatively depict the patchiness of snow cover under diverse spatial resolutions in order to analyze the effect of snow patchiness to the accuracy of algorithms under different spatial scales.

The Moderate Resolution Imaging Spectroradiometer (MODIS) fractional snow cover products may have significant errors due to cloud contamination, varying viewing geometry and complex surface properties. To improve snow and cloud discrimination with a particular interest in large sensor viewing angles, we utilize a reinterpretation test accounting for temporal surface variability to discard false positives and recover false negatives. This method is applied to MODIS fractional snow cover products including MOD10A1 and MODSCAG, then evaluated with reference snow cover generated from Landsat-8 Operational Land Imager (OLI) data. Rather than simply implementing evaluation at the normative 500 m spatial resolution, the expansion of pixel size is considered. Preliminary results indicate that this method significantly improves the precision and F-score of these two snow cover products, especially MODSCAG.

QP model with dual-channel algorithm could accurately represent the effect of surface roughness on emission at different polarization under big view angle. The purpose of this paper is to estimate long temporal series soil moisture product based on the QP algorithm, and compared it with JAXA, LPRM in Genhe basin. The results indicate that QP retrieval values are closest to the ground data but it have many missing values and high RMSE (around 0.15m3m-3); JAXA product has good correlation coefficients (around 0.4) but underestimates the ground data. LPRM product overestimates the ground data and it is found to be very noisy and unstable. Finally, In order to improve the model, this paper analysed the tendency of QP retrievals with satellite brightness temperature, and examined the influence of auxiliary data for retrievals.

This paper presented a pixel-based statistical regression method based on a high-resolution snow depth product to improve the accuracy of passive microwave snow depth retrievals. The statistical regression relation was established based on a linear relationship between the snow depth and the brightness temperature (TB) difference. The coefficients of these regression equations were derived using the snow depth product of Snow Data Assimilation System (SNODAS) as a reference. The regression relation was established over the winter period of 2013 and 2014. Passive microwave SD maps can be produced using the regression relation. Then retrieved SD was evaluated by the SNODAS SD product from November to December in 2010. The root mean square error (RMSE) and correlations (R) were computed between the SD retrievals and the SNODAS SD product. The R (mostly greater than 0.55) and RMSE (mostly lower than 16cm) maps showed a good agreement between the retrieved SD and the SNODAS SD product.

The coarse spatial resolution of microwave radiometer measurements hinders its application on land and sea surface parameters estimation. Measurements therefore require better spatial resolution to improve parameters estimation with enhanced resolution and accuracy. In this paper, a linear unmixing method is presented to downscale brightness temperatures (TB) for accurate land surface parameters retrieval. Contributions to brightness temperatures originating from different land surfaces can be identified with high-resolution land-cover images, land surface temperature products, and an antenna gain function. This produces an underdetermined equation set, which can be solved by a constrained linear least-square method with an assumption that the emissivity of each land-cover type over a small localized region is uniform. Finally, downscaled (unmixed) brightness temperatures of each land-cover type are derived from mixed pixels. Simulation results of three numerical experiments validated that the unmixing algorithm is capable of separating the signals of land-cover types from mixed pixels. The unmixing method is then applied to FY3B-MWRI measurements. The resulting downscaled brightness temperature presented enhanced details while keeping the original overall distribution of brightness temperatures. In conclusion, the linear unmixing method is capable of downscaling brightness temperatures.