Passive Microwave Research Articles

A Simulation of Snow on Antarctic Sea Ice Based on Satellite Data and Climate Reanalyses

AbstractAlthough snow plays an important role in the energy and mass balance of sea ice, it is little studied in the Southern Ocean. We present a Lagrangian model of snow on sea ice, CASSIS, that simulates the daily creation and drift of floes. Drifting floes accumulate snow from the atmosphere and the Antarctic ice sheet, and lose snow to the ocean and snow‐ice formation. The depth of snow on Southern Ocean sea ice increases in all sectors between autumn and spring 1981–2021, reaching 40 cm in much of the Weddell Sea, coastal Amundsen Sea and south east Indian Ocean. The root mean square difference between seasonally‐averaged model and ship‐based snow depths is 13.1 cm, and between modeled and airborne snow depths from Operation IceBridge is 13.5 cm. Our model offers an alternative long‐term snow depth record to that from passive microwave (PM) radiometry, which does not capture the seasonal growth of the snow cover. We find that although the average circumpolar snow layer thickness has increased by 16 mm between 1981 and 2021 (P = 0.004), there has been a decrease of 13 mm in the Southern Pacific Ocean (P = 0.133, but significant in spring and autumn), driven by a reduction of summer sea ice extent in this region. Our model paves the way for improved satellite‐based estimates of Antarctic sea ice thickness.

Detecting Tropical Cyclone from the basic overview of life cycle of Extremely Severe Cyclonic Storm, Tauktae

The Extremely Severe Cyclonic Storm (ESCS) Tauktae, which made landfall on the Gujarat coast on May 17, 2021, is discussed in the current study. The analysis is based on INSAT-3D and passive microwave (PMW) images, fo-cusing on the cyclone’s eye characteristics and intensity. The satellite imag-es and products are utilized to determine the cyclone's intensity and the spe-cific characteristics of its eye. The study showcases the variations in intensi-ty and eye features of the cyclone throughout its life span, providing insights into the process of tropical cyclone intensification. The paper is structured to cover the methods for analysis, an overview of ESCS Tauktae's life cycle, the regulation of intensity based on Dvorak's technique, intensity estimation us-ing ADT9.0 and comparison with SATCON method and Indian Meteorologi-cal Department (IMD) provided best track data. Furthermore, the formation of Tauktae's eye, timeframes for eye scenes, comparison with sea surface temperature data, and concluding thoughts are also discussed.

First mapping of polarization-dependent vegetation optical depth and soil moisture from SMAP L-band radiometry

Soil moisture (SM) and vegetation optical depth (VOD) estimates using passive microwave remote sensing at L-band (1.4 GHz) are essential for attaining a better understanding of water exchanges at the land-atmosphere interface. However, current retrieval algorithms often ignore the polarization dependence of vegetation effects. This study proposed a parameter self-calibrating framework for the multi-channel collaborative algorithm (MCCA) and presented a new SM and the first polarization-dependent VOD product based on the dual-polarized L-band observations at a fixed angle (40°) from the NASA Soil Moisture Active Passive (SMAP) mission. The parameter self-calibrating framework utilizes an information theory-based approach to obtain surface roughness and effective scattering albedo globally. Furthermore, the MCCA does not require auxiliary data for vegetation or soil moisture to constrain the retrieval process. Comparison with other SM and VOD products, such as MT-DCA version 5, DCA, SCA-H, SCA-V from SMAP Level-3 products version 8, and SMAP-IB, demonstrate analogous spatial patterns. The MCCA-derived SM exhibits the lowest unbiased root mean square deviation (ubRMSD, about 0.055 m3/m3), followed by SMAP-IB and DCA (0.061 m3/m3), with an overall Pearson's correlation coefficient of 0.744 (SMAP-IB performed best with R = 0.764) when evaluated against in-situ observations from 18 dense soil moisture networks. The MCCA generates VOD values for both vertical and horizontal polarization, demonstrating a slight polarization difference of vegetation effect at the satellite scale. Both VODs exhibit a strong linear relationship with above-ground biomass and canopy height. The polarization difference of VODs is primarily observed in densely vegetated and arid areas.

A climate data record of year-round global sea-ice drift from the EUMETSAT Ocean and Sea Ice Satellite Application Facility (OSI SAF)

Abstract. Sea ice in the polar regions can move several tens of kilometres per day under the actions of winds, ocean currents, and internal stresses. Long-term observations of the rate and patterns of this motion are needed to characterize the full response of the polar environment to climate change. Here, we introduce a new climate data record (CDR) of year-round, global, daily sea-ice drift vectors covering 1991–2020. The motion vectors are computed from series of passive microwave imagery in the winter months and from a parametric free-drift model in the summer months. An evaluation against on-ice buoy trajectories reveals that the RMSEs of the sea-ice drift CDR are small and vary with hemisphere and seasons (2.1 km for Arctic winters, 2.6 km for Arctic summer, 3 to 4 km for the Antarctic sea ice). The CDR is un-biased for Arctic winter conditions. The bias is larger for Antarctic and for summer sea-ice motion. The CDR consists of daily product files holding the dX and dY components of the drift vectors on an Equal-Area Scalable Earth (EASE2) grid with 75 km spacing as well as associated uncertainties and flags. It is prepared in the context of the EUMETSAT Ocean and Sea Ice Satellite Application Facility (OSI SAF) and is readily available at https://doi.org/10.15770/EUM_SAF_OSI_0012 (OSI SAF, 2022).

Helios and Juliette: Two falsely acclaimed medicanes?

The present work analyzes the synoptic, thermodynamic, and microphysics characteristics of two Mediterranean cyclones that occurred in February–March 2023. The analysis is mainly carried out through the use of passive microwave (PMW) satellite measurements, which allow us to follow the cyclones' evolution and state whether Helios and Juliette can be considered as Mediterranean tropical-like cyclones (i.e., medicanes). Both cyclones show a very similar evolution, with a low-stratospheric warm air anomaly during the development phase, followed by the development of a warm anomaly in the low−/mid-troposphere. This feature is often observed in medicanes (e.g., Qendresa, Zorbas), except for few cases (i.e., Medicane Ianos, which shows a warm core (WC) development clearly driven by diabatic processes without a preliminary warming signal in the lower stratosphere and upper troposphere). The analysis carried out highlights that, while Helios maintains this setting through its whole lifetime, Juliette undergoes tropical transition in the final stage of its evolution. As opposed to most medicane cases, the PMW precipitation microphysics diagnostics shows the predominance of shallow clouds, with almost total absence of ice hydrometeors and deep convection in the proximity of the WC center (i.e., within 100 km radius) for Helios and during the initial stage of Juliette. PMW radiometry provides strong indication that diabatic heating plays a role in the WC development when the onset of deep convection features is identified in the proximity of the Juliette cyclone center. Moreover, the PMW cloud-top height product does not show a closed cloud-free eye for Helios, while it is observed for the final stage of Juliette as often happens during medicanes' mature phase. Therefore, we deem that while Helios can be labeled as a warm seclusion, Juliette can be included in the tropical-like cyclone category.

Tracking Mesoscale Convective Systems in IMERG and Regional Variability of Their Properties in the Tropics

AbstractMesoscale convective systems (MCSs) constitute only a fraction of convective systems in the tropics but significantly impact tropical weather and global climate. Early studies used satellite infrared (IR) data to track MCSs and study their properties but only for a short period due to computing limitations. Though valuable, the IR brightness temperature (Tb)‐derived precipitation properties have biases. The recent availability of Integrated Multi‐satEllite Retrieval for Global (IMERG) precipitation mission rainfall data of lesser bias than IR Tb‐based rainfall and access to high‐performance computers motivated us to track MCSs over global tropics for 10 years, using the Forward in Time (FiT) algorithm. Though IMERG is advantageous, it poses challenges to tracking MCSs due to convective systems connected by light rain areas and resolution differences between contributing passive microwave sensors. The precipitation field is smoothed and normalized to overcome these problems; then, the FiT algorithm identifies MCSs and tracks them. Our results show that MCSs contribute ∼70% of annual precipitation, though they are only ∼7% of all tracked systems. MCSs occur more often over the Amazon basin and Maritime Continent than in central Africa, known for high thunderstorm frequency, highlighting the contrasting convective regimes. The large, long‐lived, and intensely precipitating MCSs occur more often over the ocean than land, except for the Amazon basin. Fast‐moving MCSs often occur over West Africa, the Amazon basin, and the western Pacific, whereas slow‐moving MCSs are common over Colombia and the Maritime Continent.

Monitoring of Lake Ice Phenology Changes in Bosten Lake Based on Bayesian Change Detection Algorithm and Passive Microwave Remote Sensing (PMRS) Data.

Lake ice phenology (LIP), hiding information about lake energy and material exchange, serves as an important indicator of climate change. Utilizing an efficient technique to swiftly extract lake ice information is crucial in the field of lake ice research. The Bayesian ensemble change detection (BECD) algorithm stands out as a powerful tool, requiring no threshold compared to other algorithms and, instead, utilizing the probability of abrupt changes to detect positions. This method is predominantly employed by automatically extracting change points from time series data, showcasing its efficiency and accuracy, especially in revealing phenological and seasonal characteristics. This paper focuses on Bosten Lake (BL) and employs PMRS data in conjunction with the Bayesian change detection algorithm. It introduces an automated method for extracting LIP information based on the Bayesian change detection algorithm. In this study, the BECD algorithm was employed to extract lake ice phenology information from passive microwave remote sensing data on Bosten Lake. The reliability of the passive microwave remote sensing data was further investigated through cross-validation with MOD10A1 data. Additionally, the Mann-Kendall non-parametric test was applied to analyze the trends in lake ice phenology changes in Bosten Lake. Spatial variations were examined using MOD09GQ data. The results indicate: (1) The Bayesian change detection algorithm (BCDA), in conjunction with PMRS data, offers a high level of accuracy and reliability in extracting the lake ice freezing and thawing processes. It accurately captures the phenological parameters of BL's ice. (2) The average start date of lake ice freezing is in mid-December, lasting for about three months, and the start date of ice thawing is usually in mid-March. The freezing duration (FD) of lake ice is relatively short, shortening each year, while the thawing speed is faster. The stability of the lake ice complete ice cover duration is poor, averaging 84 days. (3) The dynamic evolution of BL ice is rapid and regionally distinct, with the lake center, southwest, and southeast regions being the earliest areas for ice formation and thawing, while the northwest coastal and Huang Shui Gou areas experience later ice formation. (4) Since 1978, BL's ice has exhibited noticeable trends: the onset of freezing, the commencement of thawing, complete thawing, and full freezing have progressively advanced in regard to dates. The periods of full ice coverage, ice presence, thawing, and freezing have all shown a tendency toward shorter durations. This study introduces an innovative method for LIP extraction, opening up new prospects for the study of lake ecosystem and strategy formulation, which is worthy of further exploration and application in other lakes and regions.

Optimizing Back-Propagation Neural Network to Retrieve Sea Surface Temperature Based on Improved Sparrow Search Algorithm

Sea surface temperature (SST) constitutes a pivotal physical parameter in the investigation of atmospheric, oceanic, and air–sea exchange processes. The retrieval of SST through satellite passive microwave (PMW) technology effectively mitigates the interference posed by cloud cover, addressing a longstanding challenge. Nevertheless, conventional functional representations often fall short in capturing the intricate interplay of factors influencing SST. Leveraging neural networks (NNs), known for their adeptness in tackling nonlinear and intricate problems, holds great promise in SST retrieval. Nonetheless, NNs exhibit a high sensitivity to initial weights and thresholds, rendering them susceptible to local optimization issues. In this study, we present a novel machine learning (ML) approach for SST retrieval using PMW measurements, drawing from the Sparrow Search Algorithm (SSA) and Back-Propagation neural network (BPNN) methodologies. The core premise involves the optimization of the BP neural network’s initial weights and thresholds through an enhanced SSA algorithm employing various optimization strategies. This optimization aims to provide superior parameters for the training of the BP neural network. Employing AMSR2 brightness temperature data, sea surface wind speed data, and buoy SST measurements, we construct the ISSA-BP model for sea surface temperature retrieval. The validation of the ISSA-BP model against the test data is conducted and compared against the multiple linear regression (MLR) model, an unoptimized BP model, and an unimproved SSA-BP model. The results manifest an impressive R-squared (R2) value of 0.9918 and a root-mean-square error (RMSE) of 0.8268 °C for the ISSA-BP model, attesting to its superior accuracy. Furthermore, the ISSA-BP model was applied to retrieve global sea surface temperatures on 15 July 2022, yielding an R2 of 0.9926 and an RMSE of 0.7673 °C for the OISST product on the same day, underscoring its excellent concordance. The results indicate that SST can be efficiently and accurately retrieved using the model proposed in this paper, based on satellite PMW measurements. This finding underscores the potential of employing machine learning algorithms for SST retrieval and offers a valuable reference for future studies focusing on the retrieval of other sea surface parameters.

Impact of microwave observations on the estimation of Arctic sea surface temperatures

The frequent and persistent cloud cover in the Arctic limits the extent to which sea surface temperature (SST) can be retrieved from thermal infrared (IR) satellite sensors. Passive microwave (PMW) observations provide highly complementary information to IR, enabling measurements through non-precipitating clouds, although at a coarser spatial resolution. The differences in coverage, accuracy, footprint size, spatial resolution and error characteristics between IR and PMW SSTs require a systematic assessment of how to best combine IR and PMW SST retrievals. This is provided in this study on the basis of the ESA-CCI PMW SST climate data record (CDR) and an existing IR-based gap-free SST and sea ice surface temperature CDR covering the Arctic (>58°N), where cloud cover is a serious limitation to IR sensors. An important step towards a combined IR and PMW SST CDR is to correct for systematic biases in the PMW and IR SST datasets relative to each other. The PMW SSTs show reduced biases against in situ SSTs compared to the IR SSTs, but for consistency with time periods when no Arctic PMW SSTs were available, the PMW SSTs have been adjusted to the IR SSTs in this study. This is done using a dynamic bias correction to generate a consistent combined IR and PMW Arctic SST CDR for the period 2002–2017. Including PMW SSTs reduces the standard deviations from 0.54 °C, 0.55 °C and 0.47 °C to 0.47 °C, 0.54 °C and 0.41 °C against drifters, moorings and Argo floats, respectively. The improved performance is seen in almost all regions (including those already covered by IR observations), with the largest improvement in IR data sparse regions. The average theoretical uncertainty reduces by 0.08 °C, which is in good agreement with the observed improvement in the standard deviation against drifters. The results are very promising and expected to improve even further in the future with the launch of the Copernicus Imaging Microwave Radiometer (CIMR), which will enable PMW SST retrievals with lower uncertainties and much closer to coasts and sea ice (where the largest uncertainties arise) than what is possible with previous and current PMW radiometers.

Estimation of Lightning Flash Rate in Precipitation Features by Applying Shallow AI Neural Network Models to the GMI Passive Microwave Brightness Temperatures

AbstractThe satellite‐constellation passive‐microwave Brightness Temperature (TB) observations, with global coverage, and more additions from upcoming CubeSats, have been mainly used in surface precipitation retrievals. However, these observations can also be used to indicate the intensity of convective systems. This study attempts to relate Global Precipitation Mission (GPM) Microwave Imager (GMI) TBs to Geostationary Lightning Mapper (GLM) lightning flashes by using 4 years (02/2018–04/2022) of GPM Precipitation Feature (PF) database. GMI TBs are collocated to the GLM lightning counts, and to the ERA5 reanalysis 2‐m air temperature in PFs. Three Artificial Intelligent Neural Network Models (AI‐NN) are trained to classify PFs producing lightning in a 20‐min window respectively for land, ocean, or coast. The flash rate of the determined Lightning producing PFs (LPFs) is then quantified by three other AI‐NNs, each trained for one of the three regions. Though the models clearly capture the global geographical distribution of LPFs with a Probability Of Detection over 90%, high False Alarm Rates are found, ranging from 49.9% over land to 91.5% over the ocean. The importance of TB at each passive microwave channel varies regionally, corresponding to the different microphysical properties in various types of precipitation systems. The global lightning distribution is derived by applying the AI models to global PFs and is well compared to the lightning climatology from Lightning Imaging Sensor and Optical Transient Detector. This suggests that the use of passive microwave TBs can help to fill the gaps in lightning monitoring thanks to their global coverage.

A NEW WEATHER FILTER FOR REDUCING WEATHER EFFECT IN CALCULATING SEA ICE CONCENTRATION FROM AMSR2 DATA

Abstract. Ice concentration is one of the most fundamental parameters of sea ice which can be calculated from brightness temperatures measured by passive microwave radiometers such as AMSR2 onboard satellites. However, the presence of atmospheric water vapor, cloud liquid water, etc. is likely to increase the brightness temperatures over open water. As a result, sometimes, certain ice concentrations are calculated over open water covered with heavy clouds. The authors call this kind of area as “false sea ice” and the phenomenon is called the weather effect. In the AMSR2 ice concentration product, a weather filter is used to reduce the weather effect. However, it is known that false sea ice appears occasionally in some areas even in the summertime in AMSR2 ice concentration products. In this study, the authors have modified the original weather filter and added a new weather filter utilizing the following two equations which are TB(23GHzV)<−0.75×(TB(36GHzV)−TB(36GHzH))+253K and TB(36GHzV)−TB(36GHzH) < 57K. If the brightness temperature of the pixel meets the above two equations, we set the ice concentration of the pixel as 0%. By adding the new weather filter, most of the false sea ice that appeared in the ASMR2 ice concentration products could be rejected. JAXA has decided to add this weather filter for producing AMSR2 standard sea ice concentration products.

Passive microwave Arctic sea ice melt onset dates from the advanced horizontal range algorithm 1979–2022

The onset of the summer melt season is a key stage of the Arctic sea ice seasonal cycle and is an indicator of climate change. Surface melting of the bare or snow-covered sea ice is detected using passive microwave satellite observations. The data set presented here is a 44 year record of Arctic sea ice annual melt onset (MO) dates for 1979–2022 produced using an updated version of the Advanced Horizontal Range Algorithm (AHRA). This data product contains annual maps of the sea ice MO date and a set of descriptive statistics summarizing the data. This paper describes a new update of the AHRA methodology, now AHRA V5, including key changes to the algorithm starting date and sea ice mask methodology to improve estimates of early-season MO dates especially near the sea ice periphery. AHRA V5 data are suitable for monitoring trends in Arctic and regional sea ice MO dates and for process studies of atmosphere-sea ice interactions during the early spring and summer months.

Toward a real validation of passive microwave snowmelt detection algorithms over the Antarctic Ice sheet

Antarctic Ice Sheet (AIS) surface snowmelt can profoundly impact climatological and hydrological processes, as well as energy and mass balances. Passive microwave remote sensing has proven highly effective in detecting snowmelt, and the accuracy of the algorithms is typically assessed using air temperature (Tair). Nonetheless, the occurrence of snowmelt cannot be solely determined by Tair being above the melting point, it can only provide evaluation of the algorithms rather than a real validation. This study is the first to conduct a reliable validation of six commonly used algorithms by using the melt rate data from a surface energy balance model, including the cross-polarized gradient ratio (XPGR) algorithm, the diurnal amplitude variation (DAV) algorithm, the Microwave Emission Model of Layered Snowpacks based (MEMLS-based) algorithm, the constant threshold algorithm (CTA), the Ashcraft and Long Algorithm (ALA), and the M + 30 K (M30 hereafter) algorithm. Results suggest that the M30 algorithm and MEMLS-based algorithm have overall better performances, with high overall accuracies (with a value of 95.81 % and 95.56 %) and high kappa coefficient (with a value of 0.69 and 0.68). Though Tair is not reliable for evaluating the algorithms, a threshold of Tair = − 0.8 ℃ can serve as an alternative criterion in the absence of melt rate data. Additionally, two algorithms with the highest accuracy were selected to investigate interannual snowmelt trends in the Antarctic and the sub-regions. The results show a significant increase in melt extent and density in the Wilkes and Adelie sub-region during the period 1979–2020.

Distant object sensing through tap water with enhanced detection range using a passive microwave resonator and wideband impedance integral

Distant object sensing through tap water with enhanced detection range using a passive microwave resonator and wideband impedance integral

Assimilation of Sentinel-1 Backscatter into a Land Surface Model with River Routing and Its Impact on Streamflow Simulations in Two Belgian Catchments

Abstract Accurate streamflow simulations rely on good estimates of the catchment-scale soil moisture distribution. Here, we evaluated the potential of Sentinel-1 backscatter data assimilation (DA) to improve soil moisture and streamflow estimates. Our DA system consisted of the Noah-MP land surface model coupled to the HyMAP river routing model and the water cloud model as a backscatter observation operator. The DA system was set up at 0.01° resolution for two contrasting catchments in Belgium: (i) the Demer catchment dominated by agriculture and (ii) the Ourthe catchment dominated by mixed forests. We present the results of two experiments with an ensemble Kalman filter updating either soil moisture only or soil moisture and leaf area index (LAI). The DA experiments covered the period from January 2015 through August 2021 and were evaluated with independent rainfall error estimates based on station data, LAI from optical remote sensing, soil moisture retrievals from passive microwave observations, and streamflow measurements. Our results indicate that the assimilation of Sentinel-1 backscatter observations can partly correct errors in surface soil moisture due to rainfall errors and overall improve surface soil moisture estimates. However, updating soil moisture and LAI simultaneously did not bring any benefit over updating soil moisture only. Our results further indicate that streamflow estimates can be improved through Sentinel-1 DA in a catchment with strong soil moisture–runoff coupling, as observed for the Ourthe catchment, suggesting that there is potential for Sentinel-1 DA even for forested catchments. Significance Statement The purpose of this study is to improve streamflow estimation by integrating soil moisture information from satellite observations into a hydrological modeling framework. This is important preparatory work for operational centers that are responsible for producing the most accurate flood forecasts for the society. Our results provide new insights into how and where streamflow forecasting could benefit from high-spatial-resolution Sentinel-1 radar backscatter observations.

Land cover and management effects on ecosystem resistance to drought stress

Abstract. Drought events are projected to become more severe and frequent across many regions in the future, but their impacts will likely differ among ecosystems depending on their ability to maintain functioning during droughts, i.e., ecosystem resistance. Plant species have diverse strategies to cope with drought. As a result, divergent responses of different vegetation types for similar levels of drought severity have been observed. It remains unclear whether such divergence can be explained by different drought duration; co-occurring compounding effects, e.g., heat stress or memory effects; management practices; etc. Here, we provide a global synthesis of vegetation resistance to drought and heat using different proxies for vegetation condition, namely the vegetation optical depth (L-VOD) data from the ESA Soil Moisture and Ocean Salinity (SMOS) passive L-band microwave mission and enhanced vegetation index (EVI) and kernel normalized difference vegetation index (kNDVI) from NASA MODIS. Due to its longer wavelength, L-VOD has the advantage over more commonly used vegetation indices (such as kNDVI, EVI) in that it provides different information on vegetation structure and biomass and suffers from less saturation over dense forests. We apply a linear model accounting for drought and temperature effects to characterize ecosystem resistance by their sensitivity to drought duration and temperature anomalies. We analyze how ecosystem resistance varies with land cover across the globe and investigate the potential effects of forest management and crop irrigation. We compare estimates of ecosystem resistance to drought and heat as retrieved from L-VOD, kNDVI, and EVI products. We find that regions with higher forest fraction show stronger ecosystem resistance to extreme droughts than cropland for all three vegetation proxies. L-VOD indicates that primary forests tend to be more resistant to drought events than secondary forests when controlling for the differences in background climate, but this cannot be detected in EVI and kNDVI. The difference is possibly related to EVI and kNDVI saturation in dense forests. In tropical primary evergreen broadleaf forests, old-growth trees tend to be more resistant to drought than young trees from L-VOD and kNDVI. Irrigation increases the drought resistance of cropland substantially. Forest harvest decreases the drought resistance of forests. Our results suggest that ecosystem resistance can be better monitored using L-VOD in dense forests and highlight the role of forest cover, forest management, and irrigation in determining ecosystem resistance to droughts.

Retrieving Precipitable Water Vapor Over Land From Satellite Passive Microwave Radiometer Measurements Using Automated Machine Learning

AbstractAccurately retrieving precipitable water vapor (PWV) over wide‐area land surface remains challenging. Unlike passive infrared remote sensing, passive microwave (PMW) remote sensing provides almost all‐weather PWV retrievals. This study develops a PMW‐based land PWV retrieval algorithm using automated Machine learning (ML) (AutoML). Data from the Advanced Microwave Scanning Radiometer 2 serve as the main predictor variables and high‐quality Global Positioning System (GPS) PWV data as the target variable. Unprecedentedly large GPS training samples (over 50 million) from more than 12,000 stations worldwide are used to train the AutoML model. New predictors with clear physical mechanisms enable PWV retrieval over almost any land surface type, including snow cover and near open water. Validation shows good agreement between PWV retrievals and ground observations, with a root mean square error of 3.1 mm. This encouraging outcome highlights the potential of the algorithm for application with other PMW radiometers with similar wavelengths.

Applications of Deep Learning-Based Super-Resolution Networks for AMSR2 Arctic Sea Ice Images

Studies have indicated that the decrease in the extent of Arctic sea ice in recent years has had a significant impact on the Arctic ecosystem and global climate. In order to understand the evolution of sea ice, it is becoming increasingly imperative to have continuous observations of Arctic-wide sea ice with high spatial resolution. Passive microwave sensors have the benefit of being less susceptible to weather, wider coverage, and higher temporal resolution. However, it is challenging to retrieve accurate parameters of sea ice due to the low spatial resolution of passive microwave images. Therefore, improving the spatial resolution of passive microwave images is beneficial for reducing the uncertainty of sea ice parameters. In this paper, four competitive multi-image super-resolution (MISR) networks are selected to explore the applicability of the networks on multi-frequency Advanced Microwave Scanning Radiometer 2 (AMSR2) images of Arctic sea ice. The upsampling factor is set to 4 in the experiment. Firstly, the optimal input lengths of the image sequence for the four MISR networks are found, and then the best network on different frequency band images is further identified. Furthermore, some factors, including seasons, sea ice motion, and polarization mode of images, that may affect the super-resolution (SR) results are analyzed. The experimental results indicate that utilizing images from winter yields superior SR results. Conversely, SR results are the worst during summer across all four MISR networks, exhibiting the largest difference in PSNR of 4.48 dB. Additionally, the SR performance is observed to be better for images with smaller magnitudes of sea ice motion compared to those with larger motions, with the maximum PSNR difference of 2.04 dB. Finally, the SR results for vertically polarized images surpass those for horizontally polarized images, showcasing an average advantage of 4.02 dB in PSNR and 0.0061 in SSIM. In summary, valuable suggestions for selecting MISR models for passive microwave images of Arctic sea ice at different frequency bands are offered in this paper. Additionally, the quantification of the various impact factors on SR performance is also discussed in this paper, which provides insights into optimizing MISR algorithms for passive microwave sea ice imagery.

Super Resolution of Satellite-Derived Sea Surface Temperature Using a Transformer-Based Model

Sea surface temperature (SST) is one of the most important factors related to the ocean and the climate. In studying the domains of eddies, fronts, and current systems, high-resolution SST data are required. However, the passive microwave radiometer achieves a higher spatial coverage but lower resolution, while the thermal infrared radiometer has a lower spatial coverage but higher resolution. In this paper, in order to improve the performance of the super-resolution SST images derived from microwave SST data, we propose a transformer-based SST reconstruction model comprising the transformer block and the residual block, rather than purely convolutional approaches. The outputs of the transformer model are then compared with those of the other three deep learning super-resolution models, and the transformer model obtains lower root-mean-squared error (RMSE), mean bias (Bias), and robust standard deviation (RSD) values than the other three models, as well as higher entropy and definition, making it the better performing model of all those compared.

Use of a new Tibetan Plateau network for permafrost to characterize satellite-based products errors: An application to soil moisture and freeze/thaw

Two basic properties of soil moisture, namely water content and its phase status (freeze/thaw, referred to as F/T), strongly influence the pattern and efficiency of water and heat exchanges at the land-atmosphere interface. Therefore, accurate soil moisture and F/T retrievals are crucial to explore the impact of soil moisture and temperature on multi-sphere interactions, especially in the permafrost region of the third pole (Tibetan Plateau, TP), which is highly sensitive to climate changes. However, given the difficulty of installing and maintaining ground stations in harsh environments, dense ground networks are scarce over the TP with an average altitude of 4000 m. In this study, we reported a new soil moisture and temperature wireless sensor network consisting of 10 stations, which were established over the permafrost in the WuDaoLiang region (referred to as SMN-WDL) of the TP in 2019. In situ measurements from the SMN-WDL were used to simultaneously assess satellite-based surface soil moisture (SSM) and F/T products. The reliability of categorical triple collocation (CTC) method was investigated by comparing CTC-derived and ground-based relative performance rankings. The results reveal that the low-frequency (L-band) passive microwave SSM and F/T products outperformed both high-frequency (C/X/Ka-bands) passive and active microwave products, especially SMAP (Soil Moisture Active Passive) SSM and F/T products displayed the best statistical metrics to SMN-WDL ground-based measurements. The existing satellite F/T products had varying degrees of earlier freezing in autumn and later thawing in spring (mainly in ASCAT F/T) within SMN-WDL, which resulted in unsatisfactory thaw accuracy (< 85% in most cases) and overall accuracy (33.16% to 94.92%). In SMN-WDL, using existing F/T products as a masking criterion to filter out frozen soil for SSM validation studies would incorrectly mask out too many data samples and, lead to a decrease of the performance of SSM products. The overestimation of F/T thresholds may be one of the main reasons for the above misclassification of satellite F/T products. Compared to existing F/T products, the overall accuracy of threshold-optimal F/T retrievals can be increased by >10% during nighttime after optimizing thresholds based on ground-based measurements. CTC can provide consistent performance ranking with ground-based results in most cases, but the risk of CTC performing incorrect rankings was higher within the triplet where two products strongly covariate or one of the products has a significant error. This study is the first to use dense ground-based measurements in the permafrost region (SMN-WDL) to characterize the errors of four satellite SSM and F/T products. These results are expected to provide a reference with smaller representativeness errors for the further refinement of satellite retrievals in the permafrost region of the third pole.