Accurate snow depth (SD) data are essential for understanding and simulating hydrological processes, particularly in regions with uneven snow distribution and complex terrain, such as the Third Pole (TP). However, existing SD products often suffer from low spatial resolution, introducing uncertainties in regional snowpack and runoff modeling. In this study, we developed an integrated downscaling framework that combines deep learning (DL) with physical constraints to improve the resolution and accuracy of SD data for the TP region. First, a residual network incorporating multifactor spatial-terrain relationships is used to generate preliminary 500-m DL-downscaled SD data. Subsequently, by incorporating both snow cover fraction (SCF) data and snow depletion curve (SDC) as dual physical constraints, the new algorithm significantly improves the estimation accuracy in shallow snow areas. The proposed algorithm significantly enhances SD estimation accuracy, reducing the root-mean-square error (RMSE) from 1.99 to 0.75 cm, which represents a 62% improvement over the purely DL-based downscaling algorithm (RMSE = 1.02 cm). Notably, the improved algorithm demonstrates enhanced capability in terrain feature representation while simultaneously minimizing misclassification in snow-free areas and reducing overestimation in shallow snow regions. The results demonstrate progressive improvement in SD estimation accuracy across diverse geographical environments through stepwise validation of each downscaling step, confirming that integrating DL-based downscaling with physical constraints yields complementary advantages. Although the new algorithm significantly improves the accuracy of SD data, it is still underestimated in complex terrain areas and areas with SD exceeding 7 cm. Future research needs to deeply integrate physical constraint methods, comprehensively consider physical processes such as snow density, water content changes, and energy balance, and establish a more complete downscaling model.

Passive microwave remote sensing has witnessed unprecedented progress in soil moisture (SM) estimation over the decades. However, it is challenging to estimate SM accurately due to the insufficient understanding of microwave emission mechanisms. A ground-based radiometry experiment is undertaken over an agricultural field, toward the reexamination and improvement of the microwave emission models and retrieval algorithms of SM and vegetation water content (VWC). This article reports the preliminary analysis of the experimental data and the calibration of the$\tau $–$\omega $model against the collected measurements. First, the collected multifrequency dual-polarized brightness temperature (TB) reflects the temporal variation of surface SM and a significantly negative correlation between them is observed, with the coefficient of determination ($R^{2}$) and slope (S) of a linear fitting line ranging from 0.036 to 0.367 and from −16.7 to −81.5, respectively. Surface roughness and VWC impact the relationship between TB and SM, with variable$R^{2}$and S observed. Second, the calibrated parameters have improved the model performance, with$R^{2}$greater than 0.65 and root mean standard error (RMSE) less than 4.7 K at all frequencies and polarizations. The parameter values are frequency- and polarization-dependent, and the best performance of the model simulation is observed at V-polarization of L- and Ku-bands, with$R^{2} =0.80$and RMSE =4.69 K at the L-band and$R^{2} =0.74$and RMSE =2.9 K at the Ku-band. Overall, the experiment has provided valuable datasets for calibrating forward models and the calibrated model will facilitate the improvement of surface parameters (e.g., SM and VWC) retrieval.

Constrained by the limitations of remote sensing data in terms of temporal resolution, spatial resolution, and time series availability, there is currently a lack of effective long-term, high-spatiotemporal-resolution snow cover extent (SCE) products for studying snow cover changes. In this article, an SCE downscaling algorithm named SCPOT, which integrates a similar conditional probability (SCP) and Otsu’s method, is proposed. The algorithm is tested based on Moderate Resolution Imaging Spectroradiometer (MODIS) SCE and Advanced Very High Resolution Radiometer (AVHRR) SCE over the Tibetan Plateau. The SCP is defined as the probability that two pixels at corresponding positions across different scales correspond to the same snow conditions, and Otsu’s method is used to classify images by finding the optimal threshold via maximizing the interclass variance. During SCPOT testing, the SCP between 500-m MODIS and 5-km AVHRR SCEs from 2017 to 2018 was calculated, and the optimal segmentation thresholds for the SCP were determined via Otsu’s method. Then, based on the SCP and Otsu’s thresholds, the AVHRR SCE from 2015 to 2016 was downscaled to obtain 500-m resolution SCE, and the missing pixels were filled with space-time cubes and multivariate data. Evaluated with contemporaneous MODIS SCE and Landsat-8 SCE as reference data, the proposed downscaling algorithm has higher accuracy than nearest-neighbor resampling does, demonstrating feasibility in producing long-term SCE products with high spatiotemporal resolution via the algorithm.

Snow phenology is a crucial indicator that captures the dynamic changes in snow cover, which play a significant role in shaping hydrological processes and influencing ecosystem functioning. Recent climate change has affected the temporal dynamics of snow accumulation and ablation processes, particularly on the Tibetan Plateau. However, accurately quantifying how meteorological conditions influence snow phenology, especially the nonlinear and time-lagged effects, remains challenging. To address this challenge, we present a novel survival analysis model, a state-of-the-art approach used in medical research, to examine the complex effects of meteorological conditions on snow onset date (SOD) and snow end date (SED). Rigorous validation demonstrates that incorporating nonlinear and time-lagged relationships enhances both the accuracy and interpretability of the model, providing deeper insights into snow cover dynamics on the Tibetan Plateau. Specifically, our findings indicate that meteorological factors show an average delay of 11−12 days for SOD and SED across the Tibetan Plateau. A 1°C increase in temperature or a 1 W/m² increase in shortwave radiation reduces the probability of SOD by 10.7% and 1.7%, while increasing the probability of SED by 8.2% and 0.6%. Conversely, a 1 mm increase in precipitation or a 1 m/s decrease in wind speed increases the probability of SOD by 11.2% and 25.0%, and decreases the probability of SED by 12.5% and 17.6%, respectively. In addition to enhancing the quantitative understanding of how various meteorological factors influence snow phenology, the proposed model presents a promising approach for forecasting snow cover dynamics under future climate change scenarios.

The snow cover occurrence index (SCOI), defined as the ratio of the number of times that a pixel is classified as snow to the number of times that the pixel is observed in optical remote sensing data over a given year, can effectively mitigate the influence of clouds and holds great potential for extracting the annual snow duration and glacier extent in mountainous regions. The SCOI of the Qinghai-Tibet plateau (QTP) is calculated and analyzed on the basis of Landsat images from 1985 to 2021. The results indicate the following: 1) the evaluation based on station snow depth reveals that the SCOI is stable when the number of combined years reaches 5; 2) the SCOI has a strong correlation with snow cover days (SCD) determined from Moderate Resolution Imaging Spectroradiometer (MODIS) snow cover products; and 3) the SCOI has good potential for glacier extraction and exhibits a high level of consistency with glacier boundary survey data. Overall, owing to the higher spatial resolution and longer duration of the Landsat-based SCOI, it can accurately describe the distribution characteristics and changes in snow cover and glaciers in complex mountainous areas.

Volume scattering (VS) estimation plays a critical role in microwave emission modeling of the snowpack. However, it is challenging to obtain VS accurately for different frequencies by using the microwave emission model of layered snowpacks (MEMLS), which is one of the representative microwave emission models. This article develops a new VS method to consider frequency and exponential correlation length based on a snowfield campaign from November 2015 to April 2016 in Altay, China. Compared with the commonly used empirical and improved Born approximation (IBA) algorithms, the proposed VS algorithm exhibits better performances at both 18 and 36 GHz with a wide range of snow grain sizes. The bias of brightness temperatures at vertical polarization from the proposed algorithm against the observed brightness temperatures are 1.1 K and −0.4 K at 18 and 36 GHz, respectively; the root mean square errors (RMSEs) are 1.8 K and 2.6 K, respectively. The RMSEs decreased by 16.2 K at 18 GHz and 6.5 K at 36 GHz compared with those from the empirical methods and by 2.1 K and 22.2 K compared with those from the IBA. This work demonstrates that the VS difference between 18 and 36 GHz is larger and the dependence of VS on grain size is weaker than those represented by existing methods.

Improving the accuracy of remotely sensed soil moisture (SM) is a challenging and popular topic. Quantifying the uncertainty in the SM inversion process and enhancing the confidence of SM retrieval are promising ways to address this challenge but have received little attention. We present a Bayesian probabilistic inversion algorithm that can simultaneously retrieve SM, surface roughness, and vegetation optical depth and quantify the uncertainty in the inversion. The proposed algorithm is evaluated using airborne polarimetric L-band multibeam radiometer (PLMR) observations. We use three combinations, 3-angular observations at V-polarization (3CV), 3-angular observations at H-polarization (3CH), and 6-channel observations (6CA), to identify the optimal configuration for SM retrieval by taking advantage of PLMR’s dual-polarization and multiple angles. Uncertainties are quantified by introducing multiple uncertainty quantification metrics into Bayesian posterior distributions of SM retrievals. The estimates are validated against multiscale ground-based measurements, including manual measurements and wireless sensor network (WSN) measurements, and the spatial representativeness of the ground-based reference regarding the validation of pixel-scale SM retrievals is discussed. The 6CA attempt yields the best SM estimates (correlation coefficient (R)$\ge0.864$, root mean square error (RMSE)$\le0.04~\text{m}^{3}/\text{m}^{3}$, and unbiased RMSE (ubRMSE)$\le0.035~\text{m}^{3}/\text{m}^{3}$), while the 3CH attempt yields the lowest uncertainty. In addition, dense manual measurements are more representative than sparsely distributed WSN measurements. Overall, combining dual-polarized observations yields the best SM estimates but introduces additional uncertainty. This study highlights uncertainties quantification in SM inversion and thus provides confidence in SM inversion, facilitating improved SM retrieval algorithms.

The snow water equivalent (SWE, a measurement of the amount of water contained in snow packs) is an important variable in earth systems. Microwave remote sensing provides a possible solution for estimating the SWE globally. To support radar SWE retrieval, the snow backscattering theory needs to be studied; the forward simulation model needs to be validated against natural snow observations. In this study, a one-winter experiment to observe the time series backscattering coefficient of snow-covered bare soil is reported. This is the first long time series snow-covered soil backscattering experiment that was measured by an imaging radar. The backscattering coefficient was observed at three frequencies covering the X-band and dual-Ku bands, which are of great interest to the snow remote sensing community and are used for SWE estimation in mountains. The calibration of the synthetic aperture radar (SAR) system was conducted manually and carefully to ensure high-quality radar observation data. The observations from our experiment show that in general, the time series backscattering signature of snow-covered terrain is mainly driven by soil freezing, snow grain size growth, and snow accumulation processes. The time series observations for dry snow are modeled by backscattering models with model inputs directly calculated from field measurements. Our simulation results indicate that the time series radar backscattering at three frequencies and four polarizations can be simulated with high accuracy, including the cross-polarization channels. This study provides some key understanding of the time series signature of radar backscattering from snow and provides some key implications for SWE retrieval from radar observations.

Validation of existing satellite soil moisture (SM) products are critical for retrieval algorithms refinement and high-quality SM datasets achievement. Based on distributed ground observations, accuracy performance of two L-band (1.41GHz) passive microwave remotely sensed SM products of SMAP enhanced level 3 (L3) (SMAP-L3) and the newly published SMOS-INRA-CESBIO (SMOS-IC) are evaluated in Heihe River Basin (HRB) over various land surface types in Northwest China. Results reveals that: both two L-band SM products underestimate over grassland and cropland land cover types in HRB, and the underestimation obviously increases with the increasing vegetation effect. Totally, the SMAP L3 product behaves better in capturing the SM variation with the averaged ubRMSE values of 0.03m3/m3and high R values above 0.6 in ascending orbits for different land cover types in HRB, while the SMOS-IC retrievals are strongly affected by the severe RFI contamination and algorithm uncertainties which need to be handled before application, especially for the descending product.

In this study, a model investigation of a ground-based active and passive microwave experiment for snow and frozen soil is presented. The experiment is carried out from October 2017 to March 2018 in Xinjiang, China. Ground based SAR and microwave radiometers are used to measure the multiple frequency and multiple polarization backscattering coefficient and brightness temperature of snow covered soil. Microwave scattering and emission model of snow and soil are used to study the measurement results, and the microwave signature of snow and frozen soil are studied by model simulations, and this is the fundamental of snow parameter retrieval from active and passive microwave observations.

Well design and coordinated implementation of validation activity is necessary to evaluate the accuracy of remote sensing product. However, validation is not a straightforward task and remain many challenges. A generally recognized difficult issue is the inconsistence between sparse observations and remote sensing pixels, strong spatial and temporal variations of surface variables, and the intrinsic heterogeneity of land surfaces. Thus, to develop, design and conduct reasonable validation schemes and activities to acquire ground truth at pixel scale over heterogeneous land surfaces is urgently needed. This contains, from the perspective of measurement, integrating various ground observations collected at multi-scale, in order to validate different types of RSPs from site to network, especially for those land surface variables with strong spatial-temporal variations. To this end, a dedicated validation initiative has been launched in China since 2011. The main scientific objectives and research contents are to develop mathematical approaches for spatial sampling optimization to acquire the ground truth at pixel scale over heterogeneous land surfaces, to form a series of recognized and practicable technical specifications to guide validation of various RSPs, and to establish a prototype of national validation network for long term operation. Specific validation activities, such as HiWATER, were conducted from site to network, through multi-scale observations collected from multi-platform and multi-source sensors, to experimentally examine those proposed methodologies and guidelines. Following the experience of these validation exercises, we are coordinating a Chinese validation network to use standardized and recognized technical specifications in implementing future validation attempts, aiming to extend validation exercises from point scale to regional scale and to national scale across different zones.

A global sensitivity analysis utilizing the extended Fourier amplitude sensitivity test is used to determine the parameter sensitivity of the L-band microwave emission of the biosphere (L-MEB) model. The results are analyzed from two perspectives of calibration and inversion. First, the parameters of surface soil moisture, soil roughness factor, vegetation optical depth at nadir, and effective land surface temperature are the four most sensitive parameters in the L-MEB model, demonstrating their possibility to be retrieved in the multiparameter retrieval approaches. Then, the high total sensitivity index (TSI) values of surface soil temperature in the analyses emphasize the importance of high-precision land surface temperature data in the surface soil moisture retrievals, especially for rougher or more vegetated surface conditions. Finally, our analysis indicates that TSI values are high for the soil surface roughness and vegetation optical depth model parameters but low for the vegetation structure, single scattering albedo, and soil roughness coefficient model parameters at incidence angles near nadir. This suggests that calibration experiments performed at small incidence angles may be appropriate for some but not all of the model parameters, which characterize the effect of soil surface roughness and vegetation on the terrestrial brightness temperature. Consequently, new calibration procedures that account for the different relative sensitivities of these model parameters at larger incidence angles may need to be developed in the future.

The Soil Moisture Ocean Salinity (SMOS) mission has been providing L-band multi-angular brightness temperature observations at a global scale since its launch in November 2009 and has performed well in the retrieval of soil moisture. The multiple incidence angle observations are not obtained at fixed values and the resolution and accuracy change with the grid locations over SMOS snapshot images. Radio frequency interference issues and aliasing at lower look angles increases the uncertainty of observations and thereby affects the soil moisture retrieval that utilizes observations at specific angles. In this study, we propose a processing chain that uses a mixed objective function based on SMOS L1c data products to refine the characteristics of multi-angular observations. The approach was validated using simulations from a radiative transfer model and analyzed over three external targets: Amazon rainforest, Sahara desert, and Antarctic ice. These results could provide insights for selecting and utilizing external targets as part of the upcoming Soil Moisture Active Passive (SMAP) mission.

Historic freeze-up and break-up dates of lake can record the regional climate variability. Observations of lake ice from different regional or country's stations are inconsistent and their number was declined in past two decades, which makes climate change analysis difficult. This paper proposes a dynamic threshold method to derive the freeze- up and break-up dates of big lakes using passive microwave brightness temperature data. Lake ice information from 35 big lakes in Northern Hemisphere is retrieved in past three decades. These freeze-up and break-up dates were validated from 18 stations' observations. The linear correlation coefficients between observations and retrievals are 0.926 for freeze-up dates and 0.936, respectively.

A simultaneous airborne, satellite and ground based remote sensing experiment which is aiming to improve the observability, understanding, and predictability of hydrological and related ecological processes at catchmental scale is implemented in a typical inland river basin of northwest China. The experiment is composed of the cold region, forest, and arid region hydrological experiments as well as a hydro/meteorological elements and Doppler radar precipitation observation experiment. Airborne microwave radiometers at L, K and Ka bands, hyperspectral imager, thermal imager, and lidar are used. Various satellite data are collected. Based on these observations, the remote sensing retrieval models and algorithms of water cycle variables can be developed or improved, and a catchment-scale land/hydrological data assimilation system is going to be developed.

For the estimation of snow water resources in China, the passive microwave remote sensing data (SSM/I) have been used to retrieve the snow depth information. Due to the overestimation of snow depth in West China (especially the Tibet Plateau regions), the global retrieval algorithm of snow depth from SSM/I data is modified by ground observations in West China region. Both of the global and modified algorithms adopt the negative gradient of brightness temperature at 19 and 37GHz frequencies at horizontal polarization that were generated the snow layer's scattering feature. However, other surface features such as the precipitation, cold desert, frozen ground, may have similar scattering signature to snow cover. Grody's decision tree, containing various filters, is used to separate the scattering signature of snow cover from other scattering signature. For the snow cover pixels after the filters technique, we retrieved the snow depth by global and modified algorithms in East and West China, respectively. The retrieval results have been assessed by the MODIS snow products (MODISC1 snow products) with two methods of accuracy assessment, i.e. the overall accuracy and Kappa analysis based on the error matrix. The two snow-products agree very well based on the overall accuracy and Kappa coefficients. Finally, the operational scheme has been performed to obtain the snow depth datasets in China in a complete hydrological year.

A field campaign was executed in a semi-arid grassland of northwest China from July 11th-July 15th, 2002. According to the VALERI (Validation of Land European Remote Sensing Instruments) sampling procedures, the leaf area index (LAI) were intensively measured within a homogenous 3/spl times/3 km/sup 2/ square by using LAI-2000 and TRAC instrument. A quarter scene of Landsat7 ETM+ with acquisition times close to the field campaign time was processed by proper geo-registration and atmospheric correction. Three kinds of spectral vegetation index including NDVI, SR and MSAVI in the sampling area were derived from the corrected ETM+ image. The two sets of LAI data measured with LAI-2000 and TRAC instrument at the same site were inter-compared. The relationships between the measured LAI and vegetation indices were investigated as well. The results elicit that the statistical relationships between measured LAI and the different vegetation indices are consistent. Among them, NDVI seems the most promising estimator for the extraction of LAI. In addition, the LAI-2000 seems to perform better for LAI measurement in the semi-arid grassland than the TRAC instrument.