Deep learning approaches that jointly learn feature extraction have achieved remarkable progress in image matching. However, current methods often treat central and neighboring pixels homogeneously and rely on static feature selection strategies, which fail to account for environmental variations. This results in limited robustness of descriptors and keypoints, thereby affecting matching accuracy. To address these limitations, we propose a robust joint optimization network for feature detection and description in optical and SAR image matching. A Center-Weighted Module (CWM) is designed to enhance local feature representation by emphasizing the hierarchical relationship between central and surrounding features. Furthermore, a Multi-Scale Gated Aggregation (MSGA) module is introduced to suppress redundant responses and improve keypoint discriminability through a gating mechanism. To address the inconsistency of score maps across heterogeneous modalities, we design a position-constrained repeatability loss to guide the network in learning stable and consistent keypoint correspondences. Experimental results across various scenarios demonstrate that the proposed method outperforms state-of-the-art techniques in terms of both matching accuracy and the number of correct matches, highlighting its robustness and effectiveness.

As a key driving factor of land-atmosphere system, land surface temperature (LST) is widely applied in geoscience studies across various fields. Among numerous LST retrieval methods, the Split-Window (SW) algorithm has been widely used because of its advantage of free of atmospheric profile data. However, some satellites provide only one single available thermal infrared (TIR) channel, which limits the direct application of the SW algorithm. To overcome this shortcoming, this study takes Landsat-8 as an example, whose TIR channel-11 is affected by degraded calibration accuracy caused by stray light, and develops a method to construct a virtual channel using MODIS TIR data, enabling the application of the SW algorithm to Landsat-8 data for LST retrieval. During the construction, the angular normalization is adopted to the MODIS TIR data in advance. The validation results derived from the simulated dataset shows that the RMSE of LST retrieval based the virtual channel using the SW method is less than 1.2 K. Further validation with ground-based measurements from the FPK station results in an RMSE of 2.44 K, demonstrating better accuracy than the result from single channel algorithm. Moreover, the angular normalization applied to MODIS data leads to an improvement of 0.36 K in LST retrieval accuracy. The results demonstrate the advantages of LST retrieval from Landsat-8 data with virtual channel and extend the applicability of the SW algorithm.

Soil organic carbon density (SOCD) in swamp wetlands is a critical indicator for assessing global carbon stocks. In plateau wetlands, challenges such as dense vegetation cover and fragmented land distribution complicate SOCD research. The availability of high-resolution optical and radar satellite data introduces new possibilities for precise carbon stock predictions. This study proposes a framework that combines multisource remote sensing data with the sparrow search algorithm random forest (SSA-RF) algorithm to predict SOCD in plateau swamp wetlands. It also compares the effectiveness of laboratory spectroscopy and multisource remote sensing in monitoring SOCD. We integrated 24 features from Sentinel-1 (S1), Sentinel-2 (S2), topographic, and climatic data, along with spectral data ranging from 550 to 1400 nm, to construct the SSA-RF model and map the SOCD distribution of swamp wetlands in Dianchi Basin. Additionally, we estimated the total soil organic carbon (SOC) stock in these wetlands. The results indicate that the multisource remote sensing SSA-RF model (S1+ S2+ topographic + climatic SSA-RF) achieved an$R ^{2}$of 0.76, a root-mean-square error (RMSE) of 1.14, a mean absolute error (MAE) of 0.65, and a residual predictive deviation (RPD) of 1.98. Compared to the spectral model, this model improved the$R ^{2}$by 0.24 and the RPD by 0.5. Relative to the S1+ S2 SSA-RF model, the$R ^{2}$is increased by 0.15, and the RMSE is decreased by 0.46. The total SOC stock of the swamp wetlands in the Dianchi Basin was estimated to be$1.32\times 10^{5}~t$. This study provides a new framework for predicting carbon stocks in plateau wetlands, offering a reference for global wetland carbon sink assessments.

The land surface temperature (LST) downscaling is a valuable technique for obtaining high-spatiotemporal resolution LST data. By incorporating object-level or pixel-level residuals, the spectral information of the downscaled LST can be better restored, leading to improve the prediction accuracy. However, pixel-level residuals may introduce uncertainty in the residual results, whereas object-level residuals may overlook detailed land cover change information within objects, potentially resulting in discontinuities and abrupt changes in the predicted results. To address these challenges, this study proposed an integrated object- and pixel-based residual compensation framework (IOPRCF) to effectively restore the land cover change information within fine-resolution LST, thereby generating more accurate high-temporal and spatial resolution LST data. The IOPRCF was rigorously tested in 12 geographical diverse regions in China using three methods: thermal sharpening algorithm (TsHARP), random forest (RF), and simple and effective downscaling (SED). These tests encompass multiple time intervals and scaling scales to comprehensively validate the effectiveness and applicability of IOPRCF in the field of LST downscaling. The results demonstrate that, when compared with the original benchmark algorithms, TsHARP, RF, and SED with IOPRCF achieved average increases in$R^{2}$of 0.035, 0.035, and 0.028, respectively, average decreases in root mean square error (RMSE) of 0.110, 0.120, and 0.133 K, respectively, and average increases in SSIM of 0.017, 0.014, and 0.016, respectively. Moreover, the downscaled LST data produced by these algorithms with IOPRCF exhibited more similar spatial structures and captured more detailed information. Furthermore, the proposed IOPRCF demonstrated strong robustness and transferability across different LST downscaling algorithms and segmentation models. Finally, IOPRCF is expected to enhance the capabilities of LST downscaling algorithms for recovering land cover information, effectively supporting the generation of high-quality, high-spatiotemporal resolution LST globally.

Accurately estimating sensible heat (H) and latent heat (LE) in mountainous areas is a significant challenge due to the influence of complex topography factors. Currently, most models have been developed to estimate surface heat flux for flat surfaces without considering the effect of complex geometric terrain structures. In this study, the Surface Energy Balance Algorithm for Land coupled with a mountainous net surface radiation (Rn) calculation method (MSEBAL) was proposed to accurately estimate H and LE in the upstream catchment regions of the Heihe River Basin (HRB). The Rnwas estimated by correcting the solar incoming radiation components using topographic factors, including slope, aspect, sky view factor (SVF), and terrain configuration factor (TCF). The SEBAL and MSEBAL models were applied to satellite remote sensing data from Landsat 8 images and ground-observed datasets. In situ measurements from the eddy covariance (EC) system of the A’rou superstation were used to validate the estimation accuracy of Rn, LE, and H by MSEBAL. The results show that Rn, LE, and H estimated by the MSEBAL exhibit good consistency with the validation of in situ measurements. The Rnestimated by MSEBAL showed a decrease in RMSE from 164.32 to 51.31 W/m2and a reduction in absolute bias from 154.50 to 10.50 W/m2compared to SEBAL. The H and LE estimated by MSEBAL exhibit low RMSE and bias, with values of 31.96 and -17.93 W/m2for H, and 35.67 and -6.18 W/m2 for LE, respectively, compared to SEBAL. The spatial pattern of surface heat fluxes exhibited variations with complex terrain changes. H and LE were found to be higher at mountain peaks, while lower values were observed in valleys. Additionally, H and LE were greater on east and south-facing slopes that receive more solar radiation compared to west and north-facing slopes. This study provides an effective tool for estimating surface heat fluxes over mountainous regions.

Accurate spatial quantification of cropland soil organic carbon (SOC) in plateau lake basins is crucial for assessing the carbon sequestration potential in ecologically fragile regions. This study developed a machine learning (ML) framework that integrates multi-source satellite-derived environmental covariates (topography, climate, vegetation, soil properties, and parent materials) to estimate SOC distribution in the Erhai Lake basin. Using 432 topsoil samples (0–20 cm), we systematically compared 15 models, including conventional ML approaches (e.g., random forest, support vector machine, and light gradient boosting machine) and deep learning (DL) models (e.g., long short-term memory, recurrent neural network, and multilayer perceptron). The results showed that DL models achieved higher predictive accuracy than conventional ML models, reducing RMSE by 0.1680 g kg⁻¹ and increasing R², RPIQ, and CCC by averages of 0.0225, 0.1143, and 0.0253, respectively, although conventional ML models exhibited greater robustness. Elevation and temperature were identified as dominant factors controlling SOC spatial patterns, with higher concentrations clustered in the western and northern subbasins. Greater prediction uncertainty in the northwestern and eastern margins was associated with complex terrain heterogeneity. Spatially explicit SOC mapping derived from the integration of multi-source satellite data and ML models offers innovative approaches for carbon management in ecologically fragile lacustrine agroecosystems.

All-day surface downwelling longwave radiation (SDLR) is essential for studying surface radiation balance and the greenhouse effect. However, estimating SDLR under cloudy-sky conditions poses significant challenges due to the reliance on accurate cloud parameters acquisition. Most existing satellite-based algorithms for cloudy-sky SDLR retrieval are limited to daytime operations, as passive satellites are unable to provide critical data, such as cloud optical thickness, during night-time data that are crucial for estimating cloud base parameters. This study introduces an all-day SDLR retrieval algorithm based on cloud top temperature (CTT) and constructs a cloud impact factor based on cloud top height to regulate the radiative forcing contribution of clouds. The newly established algorithm was validated using ground-based measurements from Atmospheric Radiation Measurement (ARM) sites. Compared to algorithms that directly utilize CTT, the new approach incorporating the cloud impact factor demonstrated higher accuracy, with a bias of ﹣1.32 W/m2and an RMSE of 17.75 W/m2. Subsequently, datasets from the FengYun-4A (FY-4A) geostationary satellite were employed to retrieve all-day cloudy-sky SDLR with hourly resolution. Verification against Baseline Surface Radiation Network (BSRN) site measurements revealed a bias of 9.04 W/m2and an RMSE of 28.30 W/m2. Compared to previous retrieval schemes that rely on cloud base parameters, the newly proposed algorithm can estimate all-day cloudy-sky SDLR using only cloud top parameters. This reduces the reliance on cloud base parameters, facilitating a broader application of the algorithm.

Cloud base height (CBH) is crucial for determining cloud radiation effects, and uncertainty in CBH retrieval can lead to significant errors in the estimation of surface downwelling longwave radiation (SDLR) under cloudy-sky conditions. This study proposes a new CBH retrieval model utilizing top of atmosphere (TOA) reflectance and brightness temperature observations from the Fengyun-4A (FY-4A) satellite, employing the random forest algorithm. The accurately retrieved CBH is then used to estimate cloudy-sky SDLR, accounting for the radiative effects of the entire cloud layer. The CBH retrieval algorithm relies solely on TOA observations and cloud top parameter information, enabling CBH retrieval even in the absence of cloud optical thickness. The verification results indicate that the newly proposed CBH retrieval algorithm achieves good accuracy, with a bias of ﹣0.19 km and an RMSE of 1.52 km. Compared to existing studies, the new algorithm performs well in both water cloud and ice cloud phases. Based on the accurately retrieved CBH, this study employs an estimation scheme that uses cloud effective temperature to characterize the radiation contribution of the entire cloud layer, enabling accurate estimation of cloudy-sky SDLR. Ground-based measurements from the Baseline Surface Radiation Network (BSRN) and the National Tibetan Plateau Data Center (TPDC) sites were used to validate the SDLR estimation scheme. The results showed that the cloudy-sky SDLR estimated using the retrieved CBH achieved reasonable accuracy, with a bias of 7.15 W/m2and an RMSE of 27.42 W/m2.

Land surface temperature (LST) is a key factor in monitoring and improving thermal environments. However, conventional LST retrieval algorithms have not sufficiently accounted for the cavity and adjacency effects caused by the three-dimensional (3D) structures in urban settings. In this study, we propose an urban multiple scattering radiative transfer model (UMS-RTM). Based on this model, we develop an urban nonlinear split-window (UNSW) algorithm to retrieve urban land surface temperature (ULST) from Landsat-9 satellite data. The UMS-RTM optimizes the thermal radiation transfer process by correcting the cavity and adjacency effects. Analysis shows that land surface emissivity (LSE) and sky view factor (SVF) are the primary factors influencing these effects. The cavity effect increases the effective LSE by 0.01 to 0.08, while the adjacency effect raises the ground-leaving brightness temperature (BT) by 0.82 K to 3.51 K. The UNSW algorithm’s coefficients were calibrated across various LST, water vapor content (WVC), and SVF groupings to eliminate atmospheric effects and correct for cavity and adjacency effects. Sensitivity analyses of instrument noise, WVC, effective LSE, and SVF uncertainties demonstrated the reliability of the UNSW algorithm. Validation using simulated data showed that the ULST retrieved by the UNSW algorithm had a root-mean-square error (RMSE) of 0.35 K. When applied to Landsat-9 satellite data, the UNSW algorithm revealed that conventional algorithms and LST products overestimate ULST by 0 K to 2 K, with overestimations exceeding 1 K in areas with low SVF. The UNSW algorithm provides more accurate ULST retrieval and finer spatial distribution details.

In radiometric measurements, the emissivity of natural objects exhibits a dependence on the viewing angle. Ignoring the angular effect of surface emissivity can increase the uncertainty of land surface temperature (LST) retrievals. To mitigate this issue, we evaluated the simulation performance of 11 parametric kernel-driven models (KDMs) and developed directional emissivity models using MYD21 and MYD03 products. Afterward, the directional and classification-based emissivities were input into the refined GSW algorithm to retrieve LSTs with and without considering angular effects (LST_GSW_DE and LST_GSW_CE, respectively). Coupled with the MYD21 LST product (LST_TES), three LSTs were evaluated via SURFRAD in situ data and ERA5-Land products. The main findings were as follows: (1) The RMSEs of directional emissivity simulated by different KDMs ranged from ˜0.0003 to ˜0.001, and their performance differences were generally slight, indicating that parameterized KDMs demonstrate reliable simulation performance in satellite-based directional emissivity modeling. (2) The directional emissivity simulation performances of different KDMs were ranked as follows: dual-kernel model (with both hotspot and base shape kernels) ≥ multikernel model > single-kernel model. The USEA and GUTA-sparse models exhibited advantages over the other KDMs when simulating impervious surfaces during the daytime. (3) We evaluated the three types of retrieved LSTs via SURFRAD in situ data. The rankings of the RMSE and MBE values were consistent: LST_TES was optimal, followed by LST_GSW_DE and LST_GSW_CE, with average RMSEs of 2.47 K, 2.62 K, and 2.80 K, respectively. Furthermore, we evaluated the three types of retrieved LSTs against the ERA5-Land data, and the rankings of the RMSE and MBE values were also consistent: LST_TES was comparable to (slightly better than) LST_GSW_DE in some seasons and consistently better than LST_GSW_CE. The average RMSEs were 2.45 K, 2.52 K, and 2.60 K. In addition, the RMSE and MBE values at different VZAs for the three LSTs increased with increasing VZA, especially when the VZA was greater than 40°. The results demonstrated that it is feasible to use KDMs to simulate directional emissivity from satellite data, offering theoretical interpretability and addressing the issues of discrete and missing emissivity data. Future studies could be devoted to establishing new KDMs or kernels that conform to different land surface and solar illumination conditions to improve the LST retrieval accuracy.

A constellation of high-resolution thermal infrared (TIR) missions is expected to be launched in the upcoming years. Land surface temperature (LST), as a key parameter retrieved from TIR observations, constrains the variations in energy and water exchanges in the surface-atmosphere continuum. The widely used temperature and emissivity separation (TES) algorithm stands as a promising candidate for LST retrieval from these future missions due to the availability of$\ge 3$TIR bands. To explore the possibilities of further refinements of TES, a revisit of the TES algorithm in terms of the error propagation from different sources is necessary. Until now, the respective uncertainties introduced by the three modules in TES (i.e., the normalized emissivity method (NEM), the ratio algorithm (RATIO), and the maximum minimum difference (MMD) module) remain unclear. In addition, the controversy over the performances of TES on gray and nongray bodies is still unresolved. To address these research gaps, a comprehensive simulation analysis was conducted for the ECOsystem and Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS) sensor to quantify the independent impact of each error source in TES on LST retrieval accuracy, including: 1) sensor measurement noise; 2) atmospheric correction errors; 3) the NEM and RATIO modules; and 4) the MMD module. The respective responses of gray and nongray bodies to these factors were also compared based on the simulation dataset. Furthermore, the influence of the calibration scale of the minimum emissivity ($\varepsilon _{\min }$)–MMD relationship (i.e., cavity effect within vegetation canopies) was evaluated using the ECOSTRESS observations at 11 vegetated ground sites. The simulation analyses revealed that the error in atmospheric correction is the dominant impact factor significantly affecting the performances of all the other modules in TES, followed by the deviation from the regressed relationship in the MMD module. The measurement noise has minor impacts when it is well-controlled (e.g., NEdT$\le 0.1$K), and uncertainties caused by the NEM and RATIO modules are negligible. The performance discrepancy of TES over gray and nongray bodies is insignificant under low measurement noise, a condition anticipated to be met by the majority of current and future sensors. Regarding the calibration scale, the benefit of cavity effect correction is not evident according to the evaluation using the ground measurements. Based on the analyses, it is recommended that more efforts should be put into refining the atmospheric correction module and improving the fitting of emissivity samples to the$\varepsilon _{\min }$–MMD curve. In contrast, the expected benefits of refining the NEM and RATIO modules appear minimal.

Hyperspectral imagery (HSI) delivers detailed spectral information critical for remote sensing applications such as high-accuracy land cover classification, quantitative parameter retrieval, and environmental monitoring. However, satellite-borne HSI often suffers from limited spatial resolution owing to inherent sensor constraints, whereas airborne HSI is constrained by restricted spatial coverage. Reconstructing high-resolution HSI from multispectral imagery emerges as a promising strategy to address these challenges. In this study, we propose the Frequency-Enhanced Spatial-Spectral Network (FESSN), a computationally efficient architecture that innovatively integrates multi-domain fusion across spatial, spectral, and frequency domains to achieve superior reconstruction performance. A key innovation is the neural network-driven frequency enhanced modulation (FEM), which adaptively refines spectral amplitudes and phases via fast Fourier transform, providing interpretable, parameter-efficient enhancements to bridge spatial-spectral modeling gaps. Meanwhile, a Mamba-based Multi-Scale Spatial Fusion module (MMSAF) that seamlessly integrates local features with long-range dependencies, and a U-shaped Spectral Module (USEM) that integrates Mamba and attention mechanisms to model inter-group and intra-group relationships, while adhering to spectral sparsity priors. The experimental results demonstrate FESSN outperforming six state-of-the-art methods in metrics like RMSE (up to 5.31% improvement), PSNR, SAM, ERGAS, and SSIM. Downstream tasks in land cover classification further validate its utility, positioning FESSN as a breakthrough in efficient, high-fidelity HSI reconstruction. To facilitate reproducibility and further research, the code will be publicly available at https://github.com/KustAIRS/TGRS-FESSN.

The wide range acquisition of all-weather land surface temperatures (LSTs) from passive microwave (PMW) remotely sensed data contributes to understanding the land-atmosphere interactions, surface energy balance, and the global water cycle. Although significant progress has been made in PMW-based LST retrieval using statistical models, physical models, and machine learning methods, there remains a need to propose a model with high accuracy alongside strong physical interpretability and good generalization ability. This article aims to develop a physics-constrained deep learning (DL) hybrid model to obtain accurate LSTs under nonprecipitation clouds and then compare it with the pure physical and DL models. The hybrid model is developed by incorporating the physical loss function into the convolutional neural network, inheriting the advantages of the physical and DL models. Results show that the constructed model achieved good performance with a root mean square error (RMSE) of 1.60 K and a mean absolute error (MAE) of 1.26 K in the simulated data. Sensitivity analysis revealed that the hybrid model is less sensitive to input parameters than the pure physical and pure DL models and exhibits robustness across varying land surface and atmospheric conditions. Furthermore, during the evaluation using U.S. Surface Radiation Budget (SURFRAD) site data, the hybrid model yielded RMSEs of 4.37 and 3.48 K for day and night, respectively, with Advanced Microwave Scanning Radiometer 2 (AMSR2) and ERA5 data from 2012 to 2024, while outperforming the other two models at each SURFRAD site. The spatiotemporal applicability of the hybrid model further highlighted its superior generalization ability in mapping LSTs. We believe that the evident strength of the developed model over the traditional pure physical and DL models is attributed to its good accuracy, robustness to uncertainties in input parameters, and physical interpretability, which will benefit the other parameter estimates.

This study presents a framework for predicting the total carbon storage of wetlands based on machine learning models that integrate Sentinel-1 (S1), Sentinel-2 (S2), the Digital Elevation Model (DEM), and climatic data. The results indicated that the Random Forest (RF) model outperformed the Support Vector Machine (SVM) and Extreme Gradient Boosting (XGB) models in prediction accuracy. With an R-squared value of 0.67 for aboveground biomass carbon density, 0.75 for belowground biomass carbon density, and 0.65 for soil organic carbon density. The predictive performance utilizing multi-source data is significantly superior to that of single indicators, and the inclusion of climate data enhances the model’s predictive capabilities. The total carbon storage of wetlands in the Yunnan-Kweichow Plateau is estimated at 55.5 MtC, comprising 13.8 MtC in aboveground biomass carbon, 7.6 MtC in belowground biomass carbon, and 34.1 MtC in soil organic carbon.

Local incidence angle (LIA) normalization is an important method to improve the accuracy of active microwave remote sensing-based soil moisture retrieval in mountainous areas. In this study, the differences between two commonly used synthetic aperture radar (SAR) backscatter LIA normalization methods, cosine-based and liner-based, were compared using simulated and Sentinel-1 SAR data. The influence of two backscatter normalization methods on soil moisture retrieval in the dual-temporal dual-channel (DTDC) algorithm is analyzed. The results show that the difference between the normalized backscatter by the two methods is less than 0.3 dB in most cases. Despite the simplicity of the methods, both angle normalization techniques can rectify variations in backscattering induced by the LIA effect and improve the accuracy of soil moisture retrieval.

Mountainous land surface temperature (MLST) is a key parameter for the research of mountainous climate change. In this study, an improved method was developed to retrieve the MLST by considering the multiple scattering between pixels over rugged mountainous surfaces based on sky view factors (SVF). The method was applied to the Landsat-9 thermal infrared (TIR) remote sensing data by using the single channel (SC) algorithm. Due to the lack of measured data, the discrete anisotropic radiative transfer (DART) model is used to validate the accuracy of MLST retrieved by the proposed method. The results show that the MLST retrieved has high precision. The findings demonstrated the need for multiple scattering between pixels was taken into account in the retrieval of high-precision MLST under certain conditions.

Yunnan Province is one of the global biodiversity conservation hotspots. Due to the fragmented terrain and spatial heterogeneity of the plateau, land cover (LC) mapping in this region has always been challenging. In this study, we integrated 1593 scenes of Sentinel-1 and 8320 scenes of Sentinel-2 (S2) imagery, proposed a sample data generation strategy based on domain information rules, and developed a two-step feature optimization method. We also constructed a tile-based classification model to mitigate the impact of spatial heterogeneity in plateau regions. By comparing the performance of 45 different combinations of predictive variables, we analyzed the effectiveness of high-dimensional predictive variable optimization. We evaluated the impact of predictive variable sets, sensor types, classifier algorithms, and sample thresholds on LC mapping. The results indicate that: 1) the two-step feature optimization method can improve the classification accuracy of machine learning (ML). With optimal features, the model’s overall accuracy (OA) reached 92.21%, an increase of 0.64%–8.16% compared to using raw features alone; 2) the near-infrared band is a highly effective spectral predictor and slope significantly impacts the spatial distribution of LC in Yunnan Province; and 3) changes in sample thresholds affect the stability of prediction models and fivefold cross validation shows that under the optimal sample threshold scenario, prediction accuracy improves by 0.95%–2.76%; and 4) the tile-based classification model reduces the errors compared to traditional global models, mitigating heterogeneity effects. This study provides a new approach for classification in heterogeneous plateau regions and offers valuable data support for biodiversity conservation efforts.

This article proposes a combination method of nonlinear split-window (NSW) algorithm and temperature and emissivity separation (TES) algorithm to estimate land surface temperature (LST) from the remotely sensed data observed by the Advanced Geosynchronous Radiation Imager (AGRI) onboard Fengyun-4B (FY-4B), China’s second-generation meteorological geostationary satellite. The atmospheric radiation transfer model MODTRAN5.2 is used to simulate the AGRI thermal infrared (TIR) channel satellite observations in ten different viewing zenith angles (VZAs) from 0° to 70°. The optimal thermal channel combination and coefficients of the NSW algorithm are determined using a statistical regression method according to the grouping of the mean emissivity, the atmospheric water vapor content (WVC), and the LST. ERA5 reanalysis data provide atmospheric profiles for atmospheric correction, and then, the land surface emissivity (LSE) could be estimated according to the TES algorithm. The combination of Channel-12 (centered at$8.55~\mu \text{m}$) and Channel-14 (centered at$12.00~\mu \text{m}$) or the combination of Channel-13 (centered at$10.80~\mu \text{m}$) and Channel-14 (centered at$12.00~\mu \text{m}$) depends on different groups and VZAs. The statistical regression analysis showed that the root-mean-square error (RMSE) between the simulated and estimated LST is less than 0.7 and 1.8 K with the determined emissivity under VZA = 0° and VZA = 60°, respectively. Compared with the MODIS LST products (MYD11A1), the retrieved LST image has a similar spatial distribution, with the RMSE of 1.71 K.

Estimating the orientations of remote sensing images is a very important step in remote sensing image matching and is now gradually receiving widespread attention. However, due to the inability to explicitly define the standard orientations of feature points, the current methods still produce feature point orientation estimation errors, resulting in reduced matching accuracy. In this letter, we propose a multiscale unsupervised orientation estimation method with transformers, in which we use a multiscale feature extraction module to aggregate rich semantic features and a transformer-based attention mechanism module to address robust feature extraction in weakly textured regions while predicting the orientations of feature points through a carefully designed loss function. We set up image matching experiments on remote sensing images in different scenes for comparison purposes, and the experimental results show that our proposed method achieves substantially improved orientation estimation accuracy and improved image matching performance.

Mountainous land surface temperature (MLST) is one of the key surface feature parameters in studying mountainous climate change. However, for the high-spatial-resolution thermal infrared remote sensing images, the current land surface temperature (LST) retrieval algorithms were developed without enough accounting for terrain geometry and adjacent effect, which is not suitable to retrieve LST over rugged mountainous surfaces. To overcome this problem, a new method was developed to estimate the small-scale self-heating parameter (SSP) of mountainous pixels to quantify the proportion of internal thermal radiation intercepted, and the mountainous canopy effective land surface emissivity (MLSE) was defined and modeled based on SSP. A novel mountainous canopy multiple scattering thermal infrared radiative transfer (MMS-TIR-RT) model based on SSP and sky-view factor (SVF) was developed to eliminate thermal radiance contribution from inside/adjacent pixels and the atmosphere, and to restore the thermal radiation characteristics of pixels themselves. Based on this model, a new framework of mountainous single-channel algorithm (MSC) was developed for MLST retrieval from thermal infrared (TIR) data of Landsat-9 TIRS-2 sensor. In accordance with simulated data analysis, SSP, SVF, atmospheric water vapor content (WVC), land surface emissivity (LSE) of target pixel, and mean LST and LSE of the proximity pixels are main influence factors on the magnitude of the topographic effect and adjacent effect (T-A effect). Among them, SSP plays a decisive role in the mountainous canopy effective emissivity when the emissivity of the original material is low. The retrieval LST differences (δLST) between the MSC algorithm and conventional single-channel algorithm (SC) (without considering the T-A effect) from Landsat-9 TIR images are related to SSP and SVF. The results showed that the inversion of LST can be overestimated by up to 4.5 K without considering the T-A effect correction in the rugged mountainous areas. Comparing brightness temperature (BT) at the top of atmosphere (TOA) simulated by the Discrete Anisotropic Radiative Transfer (DART) model and TOA BT from radiance of TIRS-2 band 10, there are good consistencies between the spatial distributions at the three sub-regions, with the root-mean-squared error (RMSE) less than 0.62 K. The findings demonstrated the need for T-A effect to be taken into account in the retrieval of high precision MLST.

Optical and synthetic aperture radar (SAR) image keypoint detection is an important foundation for multimodal remote sensing image matching. The influence of nonlinear radiometric differences and geometric deformation between optical and SAR images leads to low repeatability of existing keypoint detection methods. To address the problem that existing keypoint detection methods cannot provide the required homonymous points for heterogenous image matching, we propose a keypoint detection method (SKD-Net) for optical and SAR images, and improve it in terms of both network structure and network optimization. First, we propose a multilevel attention Siamese network, which is composed of multiple convolutional modules and transformer modules with shared weights to extract common features at different levels for keypoint detection. We introduce a transformer module in the keypoint detection pipeline and fuse shallow and deep features to obtain more spatial and rich semantic information to facilitate heterogeneous image keypoint detection. Then, to ensure that the detected keypoints have more homonymous points and localization accuracy, we propose a position consistent loss. Unlike previous loss functions, our designed position-consistent loss function takes the differences between heterogeneous image score maps into account, and it autonomously selects the optimized correct point pairs to enable the network to perform correct learning. Finally, extensive experiments show that our detection method outperforms the current state-of-the-art keypoint detection methods in terms of repeatability, localization accuracy, and matching performance. Our source code is available at https://github.com/zhangschen/ SKD-Net.

Surface upward longwave radiation (SULR) is a key parameter that regulates surface radiation budget balance and matter-energy exchange. However, the state-of-the-art SULR retrieval methods based on remotely sensed data are only effective under clear skies, which mean that existing methods are unable to generate spatiotemporal continuous SULR product at regional or global scale. Herein, taking the advantage of long-pending abundant ground-based radiation observations, satellite products and meteorological reanalysis data, a data-driven random forest (RF) method is proposed to retrieve the instantaneous SULR under all-sky conditions. Based on spectral samples of different surface types and simulation results from the moderate resolution atmospheric transmission (MODTRAN), spectral transformation is carried out to transform SULR of various measured domains into the defined 4~100 μm domain at first. SULR and surface downward shortwave radiation (SDSR) observations from seven stations of the Surface Radiation Budget Network (SURFRAD) and nine stations of the Baseline Surface Radiation Network (BSRN) are used in model’s training and testing procedures, and the RF model achieves a high accuracy with the root-mean-square error (RMSE) of 10.45 W/m2on test set. In model evaluation, ground measurements from 14 stations of FLUXNET have been used, and the overall RMSE is 18.40 W/m2. In the actual application process, SDSR is estimated by remotely sensed data of Meteosat Second Generation (MSG). The accuracy of RF model has been validated with the observations from five stations of BSRN in 2021, and RMSEs are 17.00, 10.94, 12.17, 27.89 and 12.54 W/m2, respectively. Validation result shows that the data-driven method is capable of estimating SULR under all-sky conditions with a high accuracy. Finally, sensitivity analysis has been carried out, and the established RF model keeps robust even though there are great uncertainties among input parameters.

Net surface shortwave radiation (NSSR) is a key parameter that drives the surface material exchange and energy balance. Herein, we propose an improved artificial neuron network (ANN) with parameterized (ANN-P) method to first calculate the albedo at the top of atmosphere (TOA) by considering the surface non-Lambertian effect. Subsequently, the NSSR is estimated based on the relationship between TOA broadband albedo and the Earth's surface-absorbed shortwave radiation using a parameterized method under clear sky. The modeling process is implemented with Chinese Gaofen-5 (GF-5) visible/near-infrared channels data simulated via MODTRAN. For comparison, a previously reported lookup table (LUT) with parameterized (LUT-P) method and an ANN method are also employed. The performances of all these methods are evaluated. In terms of model simulation part, the root-mean-square errors (RMSEs) are 15.01 (17.07), 10.04 (13.67), and 20.39 (29.99) W/m2for land, water, and snow/ice surfaces, respectively, for the ANN-P (versus LUT-P) method. Their mean bias errors (MBEs) are within 0.9 W/m2. With respect to the direct ANN method, it shows the highest accuracy yet relatively large deviation for water surface. Additionally, the sensitivity analysis of water vapor content (WVC) confirms that the ANN-P method is more stable than the LUT-P and ANN methods and is, thereby, recommended for clear-sky NSSR estimation. Finally, the ground validations indicate that the mean RMSEs (MBEs) for the LUT-P, ANN-P, and ANN methods are 49.33 (-3.01), 47.55 (1.75), and 104.24 (-75.72) W/m2, respectively.

Different from bidirectional reflectance, temperature variation takes some time with the change of illumination. However, previous thermal radiation directionality (TRD) models have less considered the influence of this temperature inertia (TI) effect. By using the concept of conversion component, this article proposed an improved geometric optical (GO) model, called MGP_TI model. This model considers the TI effect by further dividing the background component into the continuously sunlit, continuously shaded, converted from sunlit to shaded, and converted from shaded to sunlit backgrounds. Upon combining with in situ measurements and a comprehensive simulated data set of component temperatures and prescribing three levels of TI and six observation times, the TI influence on TRD modeling was comprehensively analyzed. Results indicated that: 1) the overall absolute and relative greatest influence were 0.34 °C and 6.9%, respectively, suggesting that the TI influence on the value of TRD was less significant compared with the land surface temperature (LST) retrieval accuracy and the TRD extent; 2) the TI would weaken TRD on the direction of sun motion, whereas it enhanced the TRD on the opposition direction, and the primary influence was enhancing first and then weakening during the period from 10:30 to 15:30, which were determined by the differences in conversion component fractions; and 3) the TI effect could also result in the delay of the hotspot, and the occurrence and degree of the delay were influenced by the TI strength, local solar time and temperature differences of sunlit/shaded components.

The Fraunhofer line discrimination (FLD) principle is widely used for retrieving solar-induced chlorophyll fluorescence (SIF), which assumes that the spectral reflectance is smooth and can be modeled using simply mathematical function. However, the changes in the sun and observation geometry and atmospheric properties result in the `hump' or `dip' of the reflectance spectrum in the oxygen A-band. This leads to overestimations or underestimations in the SIF retrieval. The principal component analysis (PCA) algorithm is one of the main approaches used for satellite-based SIF retrieval, which can acquire reflectance characteristic information due to directional effect with large datasets. This paper attempts to test whether the errors caused by FLD method can be eliminated using the PCA algorithm. The results show that the PCA algorithm performs well in all conditions, with root mean square error less than 0.005, indicating that the bias caused by the changes in sun and observation geometry could be eliminated with PCA algorithm.

The angular effect is a primary obstacle for wide applications of land surface temperature (LST) products. Current directional thermal radiation models do not fully consider the difference between visible/near infrared and thermal radiative, i.e. thermal inertial effect, and are not practical enough. Therefore, this study proposed a practical method for angular normalization of LST products based on the component temperature and fractional vegetation cover (FVC). Analyzing with simulated data indicated that the proposed method could improve the LST retrieval accuracy caused by angular effect from 1.2 K to 0.8 K. In addition, the retrieval accuracy of component temperature would affect the performance of the proposed method whereas the retrieval accuracy of component emissivity had almost no effect on the performance.

Net surface shortwave radiation (NSSR) is a key parameter for the estimation of surface energy budget. This paper proposes a method to directly estimate the NSSR from simulated Chinese Gaofen-5 (GF-5) data without using any ancillary information. Firstly, the narrowband reflectances of visible/near infrared channels at the top of the atmosphere (TOA) were converted to the TOA broadband albedo. Secondly, by categorizing the land surface into three types, the NSSR was estimated under clear and cloudy skies separately based on the relationship between TOA broadband albedo and the Earth's surface absorbed shortwave radiation. The estimation error of the absorption coefficient for each land type is lower than 0.05. Finally, by employing a look-up-table acquired in the process of narrowband-to-broadband conversion, and the parameters in the NSSR estimation model for each land type, the performance of the proposed method was evaluated, where the root mean square errors (RMSEs) were 25.85 (13.97) W/m2, 20.39 (7.97) W/m2, and 40.54 (11.26) W/m2for land, ocean and snow/ice surfaces for clear (cloudy) skies, respectively.

Up to now, there are four widely used retrieval algorithms for retrieving sun-induced chlorophyll fluorescence (SIF) from plant photosynthesis: the standard FLD method (FLD), the modified FLD (3FLD), the improved FLD (iFLD) and the spectral fitting method (SFM). This paper attempts to compare the four different sun-induced chlorophyll fluorescence retrieval algorithms using simulated and field-measured data. The results show that the SFM and the iFLD methods provide more accurate SIF estimations with root mean square error (RMSE) less than 0.1, using the simulated data. However, when the field-measured data are used, the SFM method is better and the iFLD method becomes unstable, which suggests that the SFM method is appropriate to retrieve SIF from field data measured using a spectrometer instrument with spectral resolution lower than Inm.

Long-term global land surface temperature (LST) is a very important data source for climate change study. By adding a quadratic term of two adjacent channels' brightness temperature difference, this paper proposed to use a refined generalized split-window (GSW) algorithm to retrieve LST from series NOAA-AVHRR data. Results of the simulation analysis and the sensitive analysis indicated that the refined GSW method had a high retrieval accuracy and a robust performance. The overall root mean square errors (RMSEs) varied from 0.55 K to 0.59 K for NOAA 7-AVHRR to NOAA 19-AVHRR data. In terms of the wet atmosphere, the refined algorithm had a better ability than the GSW algorithm, and the proportion of sub-ranges with RMSE below 0.5 K was 61.7%. Most RMSE errors were within 0.2 K and 0.7 K for sensor noise$(\mathrm{NE}\Delta \mathrm{T})=0.1\ \mathrm{K}$and$\mathrm{NE}\Delta \mathrm{T}=0.2\ \mathrm{K}$, respectively, compared with the cases of no$\mathrm{NE}\Delta \mathrm{T}$. Given the uncertainties of emissivity around 1%, the errors were mainly within [0.9K, 1.2K] for dry atmosphere and [0.3K, 0.7K] for wet atmosphere.

This paper proposed a workflow to estimate land surface temperature (LST) from unmanned aerial vehicle (UVA) loaded thermal infrared imager FLIR data. The radiance received at the FLIR's sensor (ZENMUSE XT) was assumed to be the sum of the radiance from the land surface itself and the reflected downward atmospheric radiation, ignoring the upward atmospheric radiation due to UVA's low-altitude flying. The total radiation from the land and the downward atmospheric radiation were calculated from the brightness temperature extracted from the thermal image shot on the land surface and into the sky, respectively, over a farmland field at Shunyi District, Beijing, on October 27, 2017. The emissivity of the winter wheat was measured by the Portable Fourier transform thermal infrared spectrometer (102F). Finally, LST in a target scene was estimated. The range of LST in this area is around 27.8~36.5 °C. A well textual feature was depicted owing to the relatively high resolution of the UAV data.

This work addressed the estimation of Land Surface Temperature (LST) from Chinese Gaofen-5 (GF-5) satellite Thermal Infrared (TIR) data, using a Generalized Split-Window (GSW) algorithm. The numerical values of the GSW coefficients were obtained using a statistical regression method from synthetic data simulated with an accurate atmospheric radiative transfer model MODTRAN 5 over a wide range of atmospheric and surface conditions. The LST, mean emissivity, and atmospheric Water Vapor Content (WVC) were divided into several tractable sub-ranges to improve the fitting accuracy. The experimental results showed that the combination of two adjacent channels CH8.20(centered at 8.20 μm) and CH8.63(centered at 8.63 μm) was comparable with the combination of two adjacent channels CHlO.SO (centered at 10.80 μm) and CH11.95(centered at 11.92 μm) for estimating LST using the GSW algorithm, with Root Mean Square Errors (RMSEs) below 0.8 K, provided that the Land Surface Emissivities (LSEs) are known. Particularly, for the high emissivity surfaces under wet and hot atmospheric conditions , two not adjacent channels combination of CH8.63and CH11.95could also be used to estimate LST with RMSEs within 0.5 K.

Continuous monitoring of high spatial resolution evapotranspiration (ET) is critical for water resources management at both regional and local scales. This research employs a multi-sensor satellite data fusion approach (ESTARFM: Enhanced Spatial and Temporal Adaptive Reflectance Fusion Model) combined with a Two-Source N95 model and a constant evaporative fraction method to compute daily ET at 30 m spatial resolution. Two schemes are followed: the first scheme is to apply ESTARFM on the LST data to estimate daily ET at 30 m spatial resolution. The second scheme is to apply ESTARFM on the ET derived from MODIS and Landsat 8 images. The results show that the ET fused by both schemes is in good agreement with the reference ET data from the Landsat 8, while the first scheme (applying the ESTARFM on LST) is observed with more variations.

This paper attempted to retrieve leaf area index (LAI) from Gaofen (GF)-5 satellite simulation data using 6 common used vegetation indices. The canopy reflectances from 0.4-2.5μm were simulated from the combination of vegetation leaf model PROSPECT and four-stream scattering by arbitrarily inclined leaves (4SAIL) model. GF-5 satellite spectral response functions (SRF) were used to calculate the band reflectances in visible and near infrared regions. Polynomial regression was used to establish the relationships between the vegetation indices and LAI, and coefficient of determination (R2) and root mean square error (RMSE) were used to evaluate the relationships. The results showed that DVI among those indices is the best to retrieve LAI from GF-5 data with R2of 0.964. The results also showed that the retrieval accuracy can be as high as 0.338.

This paper proposes a different thermal channel combination split-window (DTCC-SW) method to estimate the land surface temperature (LST) and sea ST (SST) from the Chinese Gaofen-5 (GF-5) satellite thermal infrared (TIR) data. A nonlinear combination of two adjacent channels CH8.20(centered at 8.20 μm) and CH8.63(centered at 8.63 μm) was proposed to estimate LST for low-emissivity surfaces. A nonlinear combination of two adjacent channels, CH10.80(centered at 10.80 μm) and CH11.95(centered at 11.92 μm), was developed to estimate LST and SST for high-emissivity surfaces under dry atmospheric conditions, and a nonlinear combination of two channels, CH8.63and CH11.95, was used to estimate LST and SST for high-emissivity surfaces under wet atmospheric conditions. The numerical values of the DTCC-SW coefficients were obtained using a statistical regression method from synthetic data simulated with an accurate atmospheric radiative transfer model moderate spectral resolution atmospheric transmittance mode 5 over a wide range of atmospheric and surface conditions. The LST (SST), mean emissivity, and atmospheric water vapor content were divided into several tractable subranges to improve the fitting accuracy. The experimental results and the preliminary evaluation results showed that the root-mean-square error between the actual and estimated LSTs (SSTs) is less than 0.7 K (0.3 K), provided that the land surface emissivities are known, which indicates that the proposed DTCC-SW method can accurately estimate the LST and SST from the GF-5 TIR data.

Directional anisotropy limits the application of land surface temperature (LST) and a simplified parametric model to effectively estimate directional brightness temperature (DBT) in the thermal infrared is critical. This study used a widely validated four-stream scattering by arbitrarily inclined leaves (4SAIL) model as a benchmark to evaluate the performance of the kernel bidirectional reflectance distribution function (BRDF) model and the three-kernel-model. Results showed that the two kernel-driven models can fit the DBT well and the maximum root mean square error (RMSE) is 0.13°C. The kernel BRDF model has a wider application scope including canopies of uniform, spherical, plagiophile and planophile LIDF with low LAI and hotspot. When LIDF is planophile and plagiophile, two models can reach the best fitting effect and the worst effect is the canopy with erectrophile LIDF. Under a specified LIDF, the relationship between fitting accuracy and LAI is negative while hotspot parameter is positive.

Evapotranspiration (ET) is one of the most significant components in the water and heat transfer between land and atmosphere. This paper develops an improved constant evaporative fraction (EF) method through a theoretical derivation to improve the upscaling of remote sensing instantaneous latent heat flux (LE) to daily scale. Preliminary results show that our improved constant EF upscaling method can significantly reduce the underestimation of the daily LE upscaled using the conventional constant EF upscaling method. More validation work will be conducted to test the robustness of our improved EF method for the upscaling of remote sensing instantaneous LE estimates to daily scale.

Quantification of land surface evapotranspiration (ET) at daily or longer time scales is of great significance in agricultural ecosystem and hydrologic cycle. Temporal upscaling of instantaneous remote sensing-based ET to daily or longer time scales is generally only based on a single instantaneous estimate. A test is made to use two instantaneous ET estimates for the daily upscaling. The results show that the temporal upscaling using two instantaneous ET estimates is superior to that using only single instantaneous ET estimate for the constant extraterrestrial solar radiation ratio (Rp) method, the constant global solar radiation ratio (Rg) method, and the constant evaporative fraction (EF) method. The largest improvement of daily ET estimation occurs when instantaneous ET in the morning is combined with that in the afternoon for the Rpand Rgmethods, while the for EF method the optimal combination comprises of two moments in the afternoon.

Thin cirrus clouds can reduce land surface long-wave transmission and re-emit energy at a colder temperature and thus making it difficult to estimate downwelling surface longwave radiation (DSLR) from satellite data. In this study, a simulation database is established in terms of radiances observed at the top of the atmosphere (TOA), cloud optical thickness (COT), atmosphere water vapor content (WVC) and height of the cirrus bottom (HCB) and DSLR. And the back propagation (BP) artificial neural network (ANN) was used to estimate DSLR from remotely sensed data for cirrus cloudy skies. Results show that the BP model with TOA thermal radiance, COT, WVC and HCB as inputs provides a practical and efficient tool for remote sensing applications to estimate DSLR under thin cirrus clouds with root mean square error (RMSE) of 11.66 W/m2.

This paper aims to extend the generalized split-window (GSW) algorithm in land surface temperature (LST) retrieval to atmospheres with air temperature inversion (ATI) near the Earth surface boundary. Simulation analysis shows that the influence of ATI on the LST retrieval of the GSW algorithm becomes larger when the ATI intensity increases. To further analyze the influence, all ATI atmospheric profiles are extracted from the Thermodynamic Initial Guess Retrieval (TIGR) cloud-free database. Combining the ATI atmospheric profiles and the GSW coefficients, we find that the LST retrieval error caused by ATI is larger than 0.3 K. To reduce the LST retrieval error associated with the ATI in the GSW algorithm, a quadratic equation as a function of ATI intensity is proposed. To validate the proposed method, some in situ measurements observed at the Hailar site are used. The results show that the proposed method could improve the LST retrieval accuracy by 0.47 K for atmospheres under ATI conditions.

This paper attempts to retrieve leaf water content (LWC) by developing new vegetation indices from the combination of the near-infrared (NIR) and middle-infrared (MIR) spectral reflectances. The expanded vegetation leaf model PROSPECT-VISIR and the widely validated four-stream scattering by arbitrarily inclined leaves (4SAIL) model are employed to simulate canopy reflectance in 0.4–5.7 μm region with various leaf water content scenarios. Change of standard deviation of the canopy reflectance with respect to wavelength is used to analyze the sensitive of the spectral reflectance to the LWC. The results show that the spectral reflectances at 1.405μm, 1.875μm, 2.015μm, and 4.375μm are most sensitive to the change of LWC, and the difference vegetation index (DVI) combined by spectral reflectances in 1.405μm and 4.375μm is the best index to retrieve LWC with root mean square error (RMSE) of 0.0008 g/cm2.

Land surface temperature is an important driving force in the exchange of water, heat, and even CO2at the surface-atmosphere interface in the desert regions. The rapid and continuous measurements of land surface temperature are meaningful to the ecological and environmental researches. A physically based single-frequency and double-polarization algorithm for retrieving land surface temperature is developed in this study. The 18.7 GHz vertically polarized emissivities are firstly estimated from the Polarization Ratio (PR, defined as the ratio of the horizontal to vertical brightness temperature at the same frequency) at 18.7 GHz. And then the estimated emissivities can be directly used to retrieve land surface temperature without considering the atmospheric effect. A preliminary validation is done in the Taklimakan desert. The retrieved land surface temperatures are compared to the infrared land surface temperature products for all the year of 2007 with a Root Mean Square Error (RMSE) of 3.05 K.

Ambient irradiance is pivotal to be considered for field measurements of soil emissivity with Portable Fourier Transform Infrared Spectro-radiometer (102F). Usually, a diffusely reflecting gold plate which has a near-Lambertian behavior was used to acquire the ambient irradiance. Because of the measurements of soil and ambient irradiance are not synchronized, It can generate errors on determination of soil emissivity, especially for the erratic cloud and instantaneous wind which can make the ambient irradiance a sharp change. In this study, four conditions about the ambient radiances were 30% underestimated, 50% underestimated, 30% overestimated and 50% overestimated to assess the impacts of ambient irradiance on determination of soil emissivity. Preliminary research shows that ambient irradiance has more impacts on determination of soil emissivity in 8-10um than it in 10-13um. In 8-10um, the relative difference of soil emissivity can be more than 0.005 when the ambient irradiance was 30% overestimated. And it can reach up to 0.01 when the ambient irradiance was 50% overestimated. The error magnitudes are related to soil types. By contrast, the impacts of ambient irradiance are not obviously in 10-13um. Similar results can be seen in the ambient irradiance were underestimated conditions.

This paper analyzes the influence of the anomalous temperature occurred at the near surface boundary layer of the atmosphere on the land surface temperature (LST) retrieval with the generalized split-window algorithm (GSW). The coefficients in the GSW algorithm corresponding to a series of overlapping ranging of the mean emissivity, the atmospheric water vapor content, and the LST are derived using a statistical regression method from the numerical values simulated with an accurate atmospheric radiative transfer model MODTRAN 4 over a wide range of atmospheric and surface conditions. The simulation analysis shows that the LST can be estimated by the GSW algorithm with the root mean square error (RMSE) increasing by larger than 0.2 K when atmospheric anomalous profiles are involved. Taking into account the angular dependence of the top of the atmosphere radiance, six different viewing zenith angles (VZAs) are used in the simulations. Results show that the RMSEs become larger when the VZAs change form 0°to 60°.

An algorithm has been developed for retrieving instantaneous microwave land surface emissivity using brightness temperature and precipitable water vapor data. Unlike previous algorithms, the new technique does not need infrared land surface temperature as the input data, and overcomes the limitation of previous algorithms under cloudy conditions. Compared with the values from physical retrieval algorithm, the result demonstrates that this new algorithm has a Root Mean Square Error of 0.038 and a bias of 0.012. Although the accuracy is worse than 1%, this new algorithm presents the potential to obtain the instantaneous microwave land surface emissivity under both cloud-free and cloudy conditions, which can be applied in some weather prediction models.

The midinfrared (MIR) spectral region (3-5 μm), which penetrates most haze layers in the atmosphere and is less sensitive to variations in atmospheric water vapor, seems to be appropriate for retrieving land surface temperature (LST). However, there are currently few studies of LST retrieval with MIR data because it is difficult to eliminate solar irradiance from the total energy measured in the MIR during the daytime. This paper proposes a physics-based method to retrieve LST from MODIS daytime MIR data. The bidirectional reflectivity describing the reflected solar direct irradiance is determined using the method by Tang and Li. The directional emissivity, representing the surface emitted radiance, is determined by a kernel-driven bidirectional reflectance distribution function model, i.e., RossThick-LiSparse-R. Intercomparisons using the MODIS-derived LST product MYD11_L2, for the Baotou experimental site in Urad Qianqi, Inner Mongolia, China, have a maximum root-mean-square error (RMSE) of 1.69 K and a minimum RMSE of 1.31 K, for four scenes of MODIS images. Furthermore, in situ LSTs measured at the Hailar field site in northeastern Inner Mongolia, China, were also used to validate the proposed method. Comparisons of the LSTs retrieved from MODIS daytime MIR data and those calculated using in situ measurements have a bias and RMSE of -0.17 K and 1.42 K, respectively, which indicates that the proposed method can accurately retrieve LST from MODIS daytime MIR data.

Because of the complex influences of cirrus clouds on the estimation of Land Surface Temperature (LST), the traditional LST retrieval algorithms can only be used for clear-sky conditions and there is no LST when the pixel is identified as clouds by cloud mask algorithm. To retrieve LST under cirrus clouds, a three-channel algorithm what is dependent on cirrus optical depth (COD) and effective radius was proposed. The simulated data showed that the daytime LST could be retrieved using the three-channel algorithm with a root mean square error of less than 3.0 K when COD (at 12 μm) was less than 0.7 and viewing zenith angle was less than 60°. Compared with the results of the traditional clear-sky two-channel LST retrieval algorithm, where the maximum RMSE was 17.8 K, the algorithm proposed in this study could significantly improve the accuracy of the daytime LST retrieved using satellite thermal-infrared data.

Evapotranspiration (ET) is one of the most significant components in the water and energy transfer between land surface and atmosphere at regional and global scales. this study aims to explore the underlying mechanism in the surface temperature versus fractional vegetation cover (Ts-Fr) space for regional ET and evaporative fraction (EF) estimation through a physically-based soil-vegetation-atmosphere transfer (SVAT) simulation. It also investigates the effect of vegetation type and physiology on the relationship between EF and Tsunder deep-layer water-saturated and water-stressed conditions. The preliminary results show that in the Ts-Frspace surface EF varies linearly with surface temperature when root zone layer is not water-stressed. However, the linear relationship may be different between one vegetation type and another. When root zone layer is water-stressed, the variation of root zone layer soil water content has a negligible effect on the canopy temperature but the EF can be significantly influenced.

This work estimated firstly net surface shortwave radiation (NSSR) from MODIS/Aqua data with six visible and near infrared channels by re-parameterizing the methodology proposed by Tang et al. (2006). Comparison of the estimated NSSR with those simulated actual one showed that the root mean square error (RMSE) is 34.1 W/m2. To validate the proposed parameterization scheme, some field measurements made at seven sites of the Surface Radiation Budget Network (SURFRAD) in October, 2008 were used. The result showed that the RMSE is 53.33 W/m2. To accurately capture the diurnal variation of NSSR for cloudy skies, a simple and practical linear regression model by combing the instantaneous NSSRs estimated from MODIS/Terra at local solar time 10:30 AM and MODIS/Aqua at 13:30 PM has been proposed to estimate the daily average net surface shortwave radiation (DANSSR). The results showed that the RMSE between the estimated DANSSR and those calculated from the seven SURFRAD measurements for cloudy days in 2008 is 42.59 W/m2.

This paper retrieves the Land surface temperature (LST) from MODIS mid-infrared data. Considering that the daytime mid-infrared satellite data contains both reflected radiance due to sun irradiance and emitted radiance from the surface and the atmosphere, this paper estimates the bidirectional reflectivity in mid-infrared channels firstly, and then derives the directional emissivity with the linear kernel-driven BRDF model. Finally based on the radiative transfer equations in mid-infrared channels, the LST is retrieved. The retrieved LSTs are preliminarily validated with the MODIS LST product MYD11B1. The results show that the root mean square error (RMSE) between the two estimated LST is below 1.9 K and the Bias is below 1.10 K. In addition, some in situ measurements are also used to validate the retrieved LST. The results show that the RMSE is 2.06 K and Bias is 0.73 K.

Based on analyzing the influence of atmospheric water vapor content (WVC) on coefficients determination in the generalized split-window (GSW) algorithm, it is found that the coefficients are relatively monotonic variable with the increasing of WVC, which were proposed to determine the coefficients as implicit linear functions. To improve the land surface temperature (LST) retrieval accuracy in the GSW algorithm, the WVC is proposed to determine the coefficients as an explicit parameter in this work. The results show that the proposed method can acquire relatively high accurate LST if WVC is known. The root mean square errors (RMSEs) between the actual LST and those estimated with the proposed method are lower than those retrieved with the coefficients in the GSW algorithm.

To compare and evaluate the performance of iterative spectrally smooth temperature and emissivity separation method (ISSTES) and linear emissivity constraint temperature and emissivity separation method (LECTES) on land surface temperature (LST) and land surface emissivity (LSE) estimation, the simulation data for hyperspectral infrared spectroradiometer are used. The results reveal that the LST can be retrieved within the accuracy of 1 K at various conditions for both methods. However, the 0.01 accuracy of LSE depends on the method selected and the noise level. The ISSTES method should be taken full consideration when it used to retrieve LSE for the warm and wet atmosphere. It is advised that the ISSTES method is used for cold and dry atmosphere and the LECTES method for warm and wet atmosphere. The noises in the ground measurements may be have more effects on the accuracies of LST and LSE than those in the atmospheric downwelling measurements.

Soil moisture is a key variable that influences the redistribution of the radiant energy and the runoff generation and percolation of water in soil. Knowledge of soil moisture temporal-spatial variations is important in a wide range of studies. This study aims to investigate the temporal-spatial variations of soil moisture using microwave polarization difference index (MPDI). The AMSR-E/Aqua Daily Global Quarter-Degree Gridded Brightness Temperature at 10.65 GHz channel was used to calculate the MPDI. In addition, the AMSR-E/Aqua Daily L3 Surface Soil Moisture was used in this study. The temporal and spatial patterns between the MPDI and soil moisture were analyzed. The results indicate that the temporal and spatial patterns of the MPDI are consistent with those of soil moisture. The MPDI reflects the temporal and spatial variations of soil moisture.

Land surface temperature (LST) is a critical parameter for numerical weather forecasting, drought monitoring, water resources management and global climate change studies. Because of the supercooled temperature, the cirrus cloud can significantly reduce the LST retrieved from thermal infrared data. This paper focused on analyzing and reducing the influence of thin cirrus cloud on the accuracy of LST retrieved using the generalized split-window (GSW) algorithm. A correction method was proposed with the LST retrieval error expressed as linear functions of cirrus optical depth (COD). The slopes of the linear functions were further written as the combination of the difference and mean of two used channels emissivities and cirrus cloud top height (CTH). The results showed that the LST retrieval accuracy could be significantly improved with root mean square error (RMSE) of LST changing from 14.4 K before LST error correction to 1.8 K after LST error correction for COD equivalent to 0.3.

The total atmospheric precipitable water vapor amount (TWV) is a key variable for the study of the Earth's climate. This paper develops an algorithm to estimate the TWV over clear skies from the Medium Resolution Spectral Imager (MERSI) data in the near-IR channels. The MODTRAN 4 code is used to simulate the top of the atmospheric radiances for the MERSI channels. The results show that the proposed algorithm is suitable to estimate TWV form the absorbing channel centered at 0.940 μm and the atmospheric window channels centered at 0.865 μm and centered at 1.030 μm by the radiances over the clear pixels, with relative differences in the range of 10%-15%.

To evaluate the radiometric characteristics of the thermal infrared channels of Visible and InfraRed Radiometer (VIRR) aboard Chinese second generation polar-orbiting meteorological satellite FengYun-3B (FY-3B), the inter-calibration of those thermal infrared channels with high spectral resolution data acquired by the Atmospheric InfraRed Sounder (AIRS) aboard Aqua is carried out in this paper. Four steps, i.e. subsetting, collocating, transforming and regressing, were used to calculate the inter-calibration coefficients. The collocation data were picked out with a series of thresholds: the absolute viewing zenith angle differences less than 10°, the absolute viewing azimuth angle differences less than 20°, and absolute time differences less than 40 minute. The results on June 1st, 2012 reveal that the VIRR/FY-3B measurements are highly linearly related to the convolved AIRS/Aqua measurements. However, calibration discrepancies exist between VIRR and AIRS channels. When brightness temperatures in VIRR channels change from 270 K to 300 K under a normal condition, the AIRS-VIRR temperature adjustment linearly varies from -0.79 K to -2.32K for VIRR channel 4, from 0.14 K to -1.42 K for VIRR channel 5, respectively.

Soil moisture saturation index (SMSI) is an important indicator that demonstrates the status of the soil water content for drought monitoring. However, at present, most of the methods to calculate the SMSI from the in situ measurement data are inadequate or inaccurate. This paper proposed a simple method to determine the SMSI from the remotely sensed data. Combining the theory of thermal inertia and triangle method, the apparent thermal inertia and fractional vegetation cover can construct a triangular space. In this space, SMSI can be determined easily. Validation was performed with in situ measurements for 19 meteorological stations in the study area. Results indicated that the method can obtain the accurate soil water status that reflects the variation in soil moisture to some extent and is suitable for monitoring the regional surface soil moisture.

Land surface temperature and emissivity separation is a critical process for land surface temperature (LST) retrieval from hyperspectral thermal infrared data. This paper compared the iterative spectrally smooth temperature/emissivity separation (ISSTES) and the correlation based temperature/emissivity separation (CBTES) methods for land surface temperature and emissivities retrievals from simulated data under typical atmospheres and different land surface covers. The paper also compared both methods for retrieving low emissivities with simulated data. For typical land cover types, neglecting the instrumental noise, ISSTES is more accurate than CBTES with root mean square error (RMSE) of LSTs less than 0.0005K for the ISSTES and 0.1K for the CBTES. For low emissivity material, considering instrumental noise, both methods have large errors, but the CBTES performs much better.

Land surface microwave emission is mainly a function of soil moisture and surface roughness. However, the relationship between vertical and horizontal polarization land surface emissivities is not fully understood. This study attempts to develop a parameterized relationship to relate the emissivities at different polarizations for bare surfaces. A microwave emission database is simulated for bare surfaces with a wide range of surface roughness and dielectric properties using the Dobson model and the Advanced Integral Equation Model (AIEM) at 10.65 GHz under the configuration of the Advanced Microwave Scanning Radiometer-Earth Observing System (AMSR-E). By analyzing the factors that influence microwave emission, parameterized relationships between vertical and horizontal polarization emissivities are established. With the proposed relationships, the effects of soil moisture and surface roughness on the soil microwave emission signal can be separated. Simulated results using the proposed relationships are compared with those of the AIEM. These results show that the proposed relationships are accurate, with absolute root mean square errors (RMSEs) of 0.0025, and they can be used as a reliable boundary condition to retrieve other surface geophysical parameters. Combining this relationship with the calculated soil moisture, the RMSE of the estimated soil moisture is 0.44% using simulated data. As an example, observations of AMSR-E are used to estimate the variation in soil moisture in Saharan Africa in 2004. By comparing with independent soil moisture data, the result shows that the proposed relationship is promising for retrieving surface geophysical parameters from microwave observations.

This paper presents an improved algorithm for simultaneously retrieving both land surface emissivity (LSE) and land surface temperature (LST) using data from the Spinning Enhanced Visible and Infrared Imager (SEVIRI) on board the MSG-2 satellite. First, the temperature-independent spectral index-based method for LSE retrieval is reviewed and improved in terms of three aspects: atmospheric correction, fitting of the bidirectional reflectivity model, and retrieval of the LSE in SEVIRI channel 10. Then, the generalized split-window method with seven unknown coefficients is used to derive the LST. Finally, this improved algorithm is applied to several MSG-2/SEVIRI data sets over a study area with geospatial coverage of latitude 30 ° N-45 ° N and longitude 15 ° W-15 ° E, and using detailed cases, the modifications to the original LSE/LST retrieval methods are shown to be effective and reasonable. In addition, the SEVIRI-derived LSTs are cross-validated primarily using the Moderate Resolution Imaging Spectroradiometer-derived validated LST data extracted from the MOD11B1 product on two clear-sky days (August 22, 2009 and July 3, 2008). The validation results indicate that more than 70% of the differences are within 2.5 K and that the LST differences tend to be lower at night than in the day, which may result from the homogeneous thermal conditions at night.

Land surface temperature (LST) is crucial for a wide range of applications such as meteorology, climatology, and hydrology. In this study, we develop a method to normalize the Terra-MODIS LST to the same local solar time. An empirical relationship is established to estimate the slope of LST versus local solar time from the MSG-SEVIRI brightness temperature at the top of the atmosphere during the period 10:00-12:00 and 21:00-23:00 local solar time. This relationship is then used to normalize the Terra-MODIS LST to the same local solar time. The results indicate that the spatial variations of the MODIS LST caused by different local solar time are removed after the temporal normalization. The temporal normalized LST may become more suitable for global climate studies.

The crowning objective of this research are to analyze precipitation character of Cb for different dynamical characters in Huai river basin(HRB) with China's first operational geostationary meteorological satellite FengYun-2C (FY-2C) data. Firstly, 5 cloud patch dynamic parameters with respect to life stage and moving parameters are derived based on the Cb tracking method the author has proposed by combing artificial neural network (ANN) cloud classification[1], and cross-correlation-based approach to track Cb patch motion. Secondly, Cb precipitation over different life cycles and motion characters are analyzed. The result shows that: 1) Rain probability has a similar variation to rain rate, and rain rate is generally not more than 6 mm/hour, and probability is randomly higher than 50%. 2) Both rain rate and probability of single Cb is lower than that of complicated Cb which involves cell-merger and cell-split of some minor Cb patches. 3) Motion features such as horizontal moving speed of cloud patch (HMSP), horizontal moving direction of cloud patch (HMDP), and vertical moving character of cloud patch (VMCP) have no obvious impact on rain.

To resolve uncertainties in evapotranspiration (ET) estimates caused by the retrieval error of remotely sensed data, this study develops an evaporative fraction (EF) parameterization based on surface energy balance and the assumption of generally invariant EF during the daytime. EF is deduced as a function of temporal change of surface temperatures, temporal change of air temperature, temporal change of net radiation, and fractional vegetation cover. The EF parameterization is evaluated by the simulated data from a soil-vegetation-atmosphere transfer model with a coefficient of determination (R2) of 0.786 and a root mean square error (RMSE) of 0.117. When the EF parameterization is used to estimate the daily ET of the Yucheng station in North China by in situ measurements, the estimated results are acceptable with an RMSE of 0.7 mm (relative RMSE of 25%) and an R2of 0.837.

This paper proposed two methods to estimate the instantaneous downwelling surface longwave radiation (RL, D) using the MODIS measurements observed at the top of the atmosphere (TOA) over the Tibetan plateau region for clear-sky conditions. One is the method proposed by [2] and refined in this work, and the other is an artificial neural network (ANN) method. The upwelling surface longwave radiation (RL, U) was estimated using the Stefan-Boltzmann law with MODIS surface temperature/emissivity products (MOD11_L2). The two methods are all based on the atmospheric transfer simulation. The net surface longwave radiation (Rn, l) can then be obtained by differing the RL, Dand the RL, U. The results showed that the RMSEs of the estimated RL, Dand measured RL, Dwith the first method are smaller than those of with the ANN method for the sites over the Tibetan plateau region.

This letter presents a method to calculate the width ω over the half-period of the cosine term in a diurnal temperature cycle (DTC) model. ω deduced from the thermal diffusion equation (TDE) is compared with ω obtained from solar geometry. The results demonstrate that ω deduced from the TDE describes the shape of the DTC model more adequately around sunrise and the time of maximum temperature than ω obtained from solar geometry. Additionally, taking into account the physical continuity of land surface temperature (LST) variation, a day-to-day temporal progression (DDTP) model of LST is developed to model several days of DTCs. The results indicate that the DDTP model fits in situ [or Spinning Enhanced Visible and Infrared Imager (SEVIRI)] LST well with a root-mean-square error (RMSE) less than 1 K. Compared with the DTC model, the DDTP model slightly increases the quality of LST fits around sunrise. Assuming that only six LST measurements corresponding to the NOAA/AVHRR and MODIS overpass times for each day are available, several days of DTCs can be predicted by the DDTP model with an RMSE less than 1.5 K.

Soil moisture (SM) is a major concern in the earth science. In previous studies, in the process of retrieval of SM from remote sensing data, surface roughness effects on the retrieval of SM is obvious and a priori knowledge of surface roughness is necessary. In this paper, a simple method to retrieve SM from passive microwave data is proposed. Using the proposed method, surface roughness effects on soil moisture retrieval can be reduced. Result of sensitivity analysis shows it can be a promising method to retrieve SM. Both simulated data and actual data have been used to retrieve SM with the proposed method in this work. The result shows that the SM can be obtained with a RMSE of 1.14% from the simulated data and 1.7% from actual data.

This study firstly analyses the spatial representation of LAS (Large Aperture Scintillometer)-observed heat fluxes for a complex surface; and then evaluates the performance of SEBS model applied to a complex surface in comparison with in situ measurements. The results showed that LAS observation is indeed more stable than EC measurements even for complex surfaces, and the sensible heat flux from LAS is less than that from EC observations because of some land types with more evapotranspiration included into LAS footprint. SEBS overestimated latent heat flux at QYZ station in southern China because of the underestimation of H, but SEBS-estimated turbulent fluxes are more consistent with the LAS measurement.

An operational statistical method suitable for nearly real-time estimate of land surface and atmospheric parameters was developed and applied to the Infrared Atmospheric Sounding Interferometer (IASI) observations. The proposed method utilized three steps to solve the ill-posed problems and to stabilize the solution in a fast speed regression manner: 1) the atmospheric profiles and land surface emissivity spectra were expressed by their eigenvectors to reduce the number of unknowns; 2) a ridge regression procedure was introduced to improve the conditioning of the problem and to lessen the influence of noises; 3) a set of optimal channels was selected to decrease the effect of forward model errors or uncertainties of trace gases, and to increase computational efficiency. The retrieval results using the independent simulated data indicate the proposed method is promising. The root mean squared error (RMSE) of land surface temperature is 3.5 K, the RMSE of land surface emissivity at the selected channels is 0.01, and the RMSE of atmospheric temperature and moisture profile are about 2.0K and 0.001g/g, respectively.

MODIS land surface temperature (LST) products have been widely used in numerous applications. Each pixel within the MODIS LST products is acquired at different local solar time even though they are in the same granule. A temporal consistency and spatial comprehensiveness data set will benefit us in the utilization of the LST products in related applications and researches. In this study, a diurnal temperature cycle (DTC) model was employed to normalize the MODIS LSTs to the same local solar time. The MODIS LSTs were derived from the Terra/MODIS and Aqua/MODIS LST products (MOD11_L2 and MYD11_L2, respectively). The results at daytime only are presented because the larger LSTs heterogeneity makes the comparison of LSTs before and after the temporal normalization much clearer. The preliminary results indicate that the spatial variations of the MODIS LSTs caused by different local solar time are removed after the temporal normalization. The temporal normalized LSTs may become more suitable for the analysis of land surface processes.

Total precipitable water (TPW) is an important atmospheric parameter in many applications. A method was proposed to estimate TPW from thermal infrared hyperspectral data. First, 21 channel groups were selected to retrieve TPW. Then, two indices, namely, the differenceand the ratio-depth in each channel group, were used as the measurement of the water vapor absorption. By multivariate regression, the relationship between the TPW and the indices was established. Finally, this relationship was applied to the simulated thermal infrared hyperspectral data. Results showed that the root mean square error (RMSE) of the model is 0.102 g·cm-2, and the relative error is 8.1%. The proposed method needs to be further refined in the future work, including the complete elimination of the Earth's emission in the retrieval.

This work aimed to refine the Moderate Resolution Imaging Spectroradiometer (MODIS) based snow cover algorithm for the Hindu Kush-Himalayan (HKH) region. Taking into account the effect of the atmosphere and terrain on the satellite observations at the top of the atmosphere (TOA), particularly in heavily rugged Tibet plateau region, the surface reflectances were retrieved from the TOA reflectances after atmospheric and topographic corrections. To reduce the effects of the snow/cloud confusion, a normalized difference cloud index (NDCI) model was proposed to discriminate snow/cloud pixels, apart from use of the MODIS cloud mask product MOD35. Furthermore, MODIS land surface temperature (LST) product MOD11_L2 have been used to ensure better accuracy of the snow cover pixels. Comparisons of the resultant MODIS snow cover with those obtained respectively from high resolution Landsat ETM+ data and the MODIS snow cover product MOD10_L2 for the Mount Everest region at different seasons, showed overestimation of the MOD10_L2 snow cover with the differences of 50%, whereas the improved algorithm can estimate the snow cover for HKH region more precisely with absolute accuracy of 90%.

This paper makes an attempt to establish a generalized neural network for simultaneously retrieving atmospheric profiles and surface temperature from hyperspectral thermal infrared data. To generate the simulated data covering the whole actual situations, the distributions of surface material, temperature and atmospheric profiles are elaborated carefully. The simulated at-sensor radiances are divided into two sub-ranges, one in atmospheric window and another in water absorption band. The simulated data are transformed in the eigen-domain in both sub-ranges and used as the network inputs. The atmospheric profiles, surface temperature and emissivity are used as the outputs after the eigen-domain transformation. The validation of the trained network indicates that a RMSE of surface temperature around 1.6K, a RMSE of temperature profiles around 2K in troposphere and a RMSE of total water content around 0.3g/cm2can be obtained. The results from the net can be used as initial guess of the physical retrieval model.

This paper makes an attempt to address the scaling problem of Leaf Area Index (LAI) and to analyze the propagation of scaling effect of LAI. On the basis of the Taylor series expansion and following the general scaling procedure, it is demonstrated that the magnitude of the scaling effect is the product of the degree of the non-linearity of the retrieval model and the spatial heterogeneity of input variables involved in this model. Finally, a scaling correction model is proposed to correct for the scaling effect of LAI. The validation using the simulated data indicates that the proposed scaling correction model of LAI gives promising accuracy if the spatial heterogeneity is well characterized by its wavelet variance. The RMSE and relative error of retrieved LAI induced by the scale effect can be greatly reduced after scaling correction. The scaling propagation analysis of LAI reveals that the scaling effects caused by several non-linear components may compensate for each other, which would enhance our confidence in using LAI product over heterogeneity areas.

Land surface temperature (LST) and emissivity are required for many applications. Several methods have been proposed to retrieve these two parameters from hyperspectral data, some of which are based on the spectral smoothness of emissivity. To analyze the sensitivity of those methods to various measurement errors, hyperspectral TIR data are first simulated using radiative transfer model 4A/OP (Operational Release for Automatized Atmospheric Absorption Atlas) with different atmospheric profiles and surface parameters, and then the sensitivity of the Downwelling Radiance Residual Index method to different sources of error is analyzed. In terms of resulting errors in LST, results show that: 1) the method is not very sensitive to the uncertainties of atmosphere. An error of 1.47 g/cm2on water vapor content for a sub-arctic summer atmosphere (2.1 g/cm2) only leads to an error of 1.8 K for rock2 (the worst case). 2) Satisfactory results are obtained by this method over heterogeneous land surface. LST retrieval error is less than 0.3 K for all atmospheres.

Land surface temperature, land surface emissivity and atmospheric profiles are all of great importance in many applications. As the at-sensor radiances are dependent on both the land surface parameters (temperature and emissivity) and atmospheric conditions, it is difficult to simultaneously retrieve these parameters with a high accuracy from multi-spectral radiances measured at satellite level. However, some studies have recently shown that hyperspectral thermal infrared data could be used to derive these parameters simultaneously from space. This paper tries to explore the possibilities to recover with an acceptable accuracy both the geophysical properties and the atmospheric parameters from the hyperspectral thermal infrared data using the neural network technique. The results show that the land surface temperature can be obtained with a RMSE=0.24 K and the atmospheric profiles can also be retrieved with relatively high accuracy. However, further work has to be performed to improve the retrieval accuracy in the near future.

The central problem of temperature and emissivity separation (TES) is, as Realmuto had pointed out, that we obtain N spectral measurements of radiance and need to find N+1 unknowns (N emissivities and one temperature), if the atmospheric perturbations are well corrected for. Thus, one constraint must be found in the retrieval to obtain the realistic solution for the temperature/emissivity separation. A new index called `Downwelling Radiance Residual Index' (DRRI) is proposed to provide this type of constraint. Tests with the simulated hyperspectral thermal infrared (TIR) data sets demonstrate that this new index can provide an accurate and fast Temperature/Emissivity separation.

A radiative transfer model to compute brightness temperatures in the microwave region for the soil-vegetation- atmosphere system has been developed in this study. Considering the various atmospheric conditions, a microwave brightness temperature database is generated for a wide range of surface dielectric constant and roughness properties under AMSR-E sensor configurations by using the soil-vegetation-atmosphere radiative transfer model. From the land surface emissivities in the simulated database, the linear relationships of surface emissivities between different channels for the advanced microwave scanning radiometer-earth observing system (AMSR-E) are established in this study. The analysis of the stability and the root-mean-square error (RMSE) for these linear relations indicates that the relationship using three channels is the best of all. The cloud-free AMSR-E actual satellite microwave data combining with MODIS land surface temperature product are used to validate these relationships. The linear relationship derived from the simulated database is in agreement with that from AMSR-E data. This emissivity model as a constrain condition of the radiative transfer equation is applied to retrieval LST and emissivity, and the RMSE of the results is lower than 1 K for LST, 0.0025 for emissivity. So the relationships of emissivity between different channels are very useful to derive land surface parameters directly from AMSR-E data, which will promote the application of AMSR-E data in many fields, such as climate, hydrology, and ecology.

This paper proposed to study a Vegetation Index (VI) with surface bi-directional reflectivities in MODIS near-IR and mid-IR channels. Considering the fact that the observations in mid-IR at satellite altitude during daytime consist of a combination of both reflected radiance due to sun irradiance and emitted radiance from both the surface and the atmosphere, a brief description of estimating the land surface bi-directional reflectivity in mid-IR channels from MODIS data was given. Sensitivity analysis of the vegetation index with regard to the variations of horizontal visibility was performed. The results showed that the proposed vegetation index is much less sensitive to haze in the atmosphere than the Normalized Difference Vegetation Index (NDVI). In addition, in order to compare the traditional NDVI with the new VI, NDVI and VI were calculated using MODIS data acquired over a region of Western Europe (Spain), for cloud-free days, covering a period of one month, July, 2006. The result of this comparison indicated that the proposed new vegetation index is feasible to descript the properties of vegetation and to determine the classification of vegetation, especially in the areas covered by dense smoke or industrial pollution.

Surface soil moisture is a key variable in computing several important variables of the land energy and water budget (albedo,hydraulic conductivity etc). At the same time,surface soil moisture affects the diurnal change of surface temperature. Meteorological satellite data have great potential for providing estimation of surface soil moisture with high temporal resolution on a daily basis. This paper compares some relationship between the parameters derived by fitting land surface temperature (LST) with its diurnal cycle model and surface soil moisture,The results showed that lag time (the difference of the time corresponding to maximum surface temperature and that to maximum solar net short-wave radiation) is most correlated to surface soil moisture.