Surface soil moisture (SSM) plays a crucial role in hydrological processes, ecosystem dynamics, and agricultural management. Currently, high spatial resolution SSM estimation primarily relies on machine learning methods. However, in heterogeneous environments, the challenges associated with hyperparameter optimization, computational efficiency and uncertainty control compromise the robustness of these methods. To address this issue, this study introduces and evaluates an integrated strategy that combines Gaussian Process Bayesian Optimization (GP-BO) with machine learning for high-resolution SSM retrieval. The results demonstrate that the combined method significantly outperforms conventional optimization methods evaluated by the test sets from Heihe River Basin, Naqu, and Shandian River basins. Notably, the integration of GP-BO with XGBoost turned out to be the optimal combination, improving R² by 0.01–0.19 and reducing ubRMSE by 0.02–1.71 percentage points relative to conventional optimizers, while requiring the least training time for ensemble models. Furthermore, vegetation-specific GP-BO-tuned XGBoost models achieve varying degrees of accuracy improvement and reduced uncertainty across various vegetation, particularly in barren and grassland regions. These findings highlight the effectiveness of GP-BO hyperparameter optimization algorithm in reducing the uncertainties of SSM estimation in heterogeneous environments.

Unmanned aerial vehicle (UAV)-based fine-scale evapotranspiration (ET) estimation is becoming increasingly critical in precision agricultural water management. However, existing UAV-based ET estimation studies often directly transfer satellite-based ET models and parameterization schemes to fine-scale UAV data, which hampers accurate fine-scale ET estimation. Here, we use machine learning (ML)-based alternative estimation schemes to estimate key parameters of aerodynamic roughness length (z0m) and excess resistance (kB-1) in the surface energy balance system (SEBS) ET model. In addition, we use a computational fluid dynamics (CFD) model to provide downscaled meteorological data for the SEBS model. Compared to physical parameterization schemes, ML-based estimates ofz0mandkB-1show improved accuracy, reducing the mean root mean square error (RMSE) forz0mfrom 0.07 m to 0.04 m, and forkB-1from 4.58 to 2.41. Validation against eddy covariance (EC) systems with a source area of hundreds of meters shows that ML-based estimates of latent heat flux (LE) have an RMSE of 39.94 W/m2, which is superior to the RMSE of 77.44 W/m2achieved by physical parameterization schemes. ML-based LE estimates also show comparable accuracy with an RMSE of 41.94 W/m2when using CFD-based meteorological data. A comparison with an optical-microwave scintillometer (OMS) system with a source area spanning kilometers confirmed the importance of CFD-based meteorological data and reduced the mean relative error (MRE) for LE from 26.53% (using site-observed meteorological data) to 22.28%. Our proposed robust framework improves the accuracy of UAV-based ET estimates, thus helping to bridge the scale gap between satellite remote sensing and site-based observations.

Land surface temperature (LST) simulated by land surface model (LSM) can maintain spatial integrity and high temporal resolution. However, the relatively low spatial resolution limits the practical applications of LSM-simulated LST. Traditional downscaling methods often require lots of auxiliary data and suffer from significant loss of spatial details under large scale differences. To this end, we propose a super-resolution (SR) downscaling method based on multi-source reference using a multi-attention multi-residual network (MAMRN). The Moderate Resolution Imaging Spectroradiometer (MODIS) LST was used as the reference to improve the spatial resolution of LSM-simulated China Land Data Assimilation System (CLDAS) LST. Six regions with different land cover types covered the Chinese mainland were selected to test the MAMRN’s performance, and a traditional bilinear interpolation method and three deep learning-based SR methods were used for comparison. Comparative experiments demonstrate that MAMRN achieves improved performance, both visually and quantitatively in all six regions. Specifically, the average Root Mean Square Error (RMSE), Peak Signal-to-Noise Ratio (PSNR) and Learned Perceptual Image Patch Similarity (LPIPS) are 2.06 K, 16.76, and 0.54, respectively. The quantitative evaluation scores surpass those of the comparison methods. The transfer experiments and ablation studies also indicate MAMRN’s superiority and effectiveness. Validation by in-situ LST shows that MAMRN can accurately retrieve LST under both clear-sky and cloudy-sky conditions, with the overall accuracy of clear-sky slightly superior to that of cloudy-sky. In a word, MAMRN can enhance the spatial resolution of CLDAS LST (about 6.25 km) to MODIS (1 km) scale while preserving clear spatial details. This capability is beneficial for generating spatially integrity LST with high spatiotemporal resolution, and contributes to the study of global climate change. Our code can be available at https://github.com/AHU-RS/MAMRN.

The uncertainty of ground truth values at the pixel scale obtained via upscaling directly affects the credibility of remote sensing product validation. This article provides an in-depth analysis of the sources of uncertainty in ground truth evapotranspiration (ET) at the pixel scale. This uncertainty is quantitatively evaluated, and the methods for its control are discussed. The results indicate that the uncertainty from upscaling methods is highest, followed by that from auxiliary data, with that from instrument measurements being the smallest. The relative accuracy of the ground truth ET at the pixel scale for the LAS1–LAS4 (LAS5–LAS7) regions is 89.15%–90.16% (81.56%–82.58%). The accuracy for homogeneous surfaces is relatively high at approximately 90%–93%, whereas for moderately and highly heterogeneous surfaces, it is lower, varying from approximately 81% to 92%. To control uncertainty, precise instrument calibration, strategic positioning, the use of diverse constraints, and robust modeling are recommended to increase measurement accuracy and prediction reliability. This uncertainty study includes the analyses of different sources of uncertainty and quantitatively evaluates the uncertainty of ground truth values over different heterogeneous underlying surfaces. The results can be used to objectively evaluate the accuracy of remote sensing products, thus advancing studies of the uncertainty of ground truth values at the pixel scale and enhancing the scientific, reliable, and systematic validation of remote sensing products. This approach can greatly promote the validation of remote sensing ET products over heterogeneous surfaces.

Quantifying and monitoring land surface evapotranspiration (ET) is an essential task for understanding the earth’s water, energy, and carbon cycles. ET, specifically plant transpiration ($T$), is closely linked to the photosynthesis, which is coupled through stomatal function. However, the mechanistic links between sun-induced chlorophyll fluorescence (SIF) information indicating canopy photosynthetic activity and$T$are complex and difficult to derive empirically. An empirical SIF-$T$relationship at ecosystem scale was developed and coupled to the two-source energy balance model (TSEB-SIF) to estimate the ET and its components,$T$and soil evaporation,$E$. By comparing model predictions with observations from an irrigated cropland site located in a semiarid region, the TSEB-SIF model shows a slightly better performance to the TSEB model in estimating ET, especially under water deficit conditions. Moreover, the TSEB-SIF model more reliably partitioned the$T$from ET, while the TSEB model tended to overestimate the contribution of$T$to ET.

To solve the limitation of coarse spatial resolution of passive microwave surface soil moisture (SSM) product, many spatial downscaling methods has been proposed under the theoretic basis of the land surface temperature (LST)/vegetation index triangle space. However, in most studies, the topographic influences on the downscaling results is rarely analyzed but the impacts should be significant due to the strong effect of topographic changes on LST. To effectively solve this issue, a downscaling approach was developed in this study to disaggregate the Soil Moisture Active and Passive (SMAP) SSM product with the use of the Optical/Thermal infrared (TIR) observations from the Moderate-Resolution Imaging Spectro-radiometer (MODIS) onboard the Terra satellite for a typical mountain region located in the west of U.S. Two steps are included in the approach: (1) normalizing the terrain effect on LST and (2) downscaling passive SSM product based on a machine learning approach. The SNOTEL soil moisture observation network located in the study area was used to validate the downscaled SSM.

Land surface temperature (LST) immensely affects the energy balance and water cycle on the earth's surface. Merging thermal infrared (TIR) and passive microwave (MW) remote sensing provides the possibility to obtain all-weather LST with moderate resolutions. However, due to difficulties in downscaling MW LST, current methods merging TIR LST and MW LST into such an all-weather LST are limited over large areas with very complicated land surfaces (e.g. the Tibetan Plateau). By fully considering the influence of the topography on estimation of merged LSTs, this study revises the recently-developed physical method for generating the 1-km all-weather LST and applies it over the Tibetan Plateau to merge MODIS (1 km) and AMSR2 (10 km) observations. Results show that the merged LST has accuracy of 0.99 K-3.22 K when validated against insitu LSTs from five ground stations with various land cover types. This study would be beneficial for continuously monitoring LST and improving spatio-temporal resolutions for associated land surface process studies requiring high-quality all-weather LST over large scales.

Thermal-infrared remote sensing of land surface temperature (LST) provides valuable information for quantifying root-zone water availability, evapotranspiration (ET) and crop condition. This paper describes the most recent modifications applied to the robust but relatively simple LST-based energy balance model, the Two-Source Energy Balance (TSEB), which solves for the soil/substrate and canopy temperatures that achieves a balance in the radiation and turbulent heat flux exchange with the lower atmosphere for the soil/substrate and vegetation elements. As a result, the TSEB modeling framework is applicable to a wide range in atmospheric and canopy cover conditions. This work illustrates the utility of high resolution LST data providing within-field variability in energy fluxes and evapotranspiration (ET), including modifications made in TSEB to be adapted for structurally complex crops, such as vineyards. Such high resolution spatial information is being used in precision farming applications to assess the impacts of within variability in soil texture, water availability and other stress factors on plant condition and productivity.

Remotely sensed evapotranspiration (RS_ET) products have been applied from regional to global. However, the validation of remote sensing products over heterogeneous land surfaces has been hindered due to the challenges in the theory and methods in recent decades, especially in estimation of “ground-truth” at the satellite pixel scale. In this study, an innovative validation framework including quantification of the spatial heterogeneity, optimization of the ground sampling strategy, multi-scale measurement, upscaling theory, uncertainty analyses, and validation method (direct validation, indirect validation and cross validation), was proposed to validate RS_ET products at different scales. Here, the framework was applied in Haihe and Heihe basin in China. The results showed the proposed validation framework of RS_ET product was reasonable and feasible.

Although many challenges exist, validation of the satellite land surface temperature (LST) product over heterogeneous surface can provide new and in-depth understandings of the product. Lessons learned from the validation are important to improve the satellite LST product. In order to better understand the relationship between LSTs measured through different approaches and instruments and test the possibility to upscale the ground measured LST over heterogeneous surface, a MUlti-Scale Observation Experiment on land Surface temperature (MUSOES) was designed and conducted in an extremely arid region in Northwest China. The experiment was concentrated at two typical sites (i.e. HHL - sparsely forest, and SDQ - open shrubland). It began from July 2014 and have run normally for two years. First results of this experiment have been presented here. MUSOES provides a basis to examine the upscaling of the ground measured LST to the LST at the satellite pixel scale over the heterogeneous surface.

Thermal cameras are helpful instruments for measuring surface temperatures in field experiments. However, previous studies haven't detailed the method of deriving component temperatures of vegetation and soil over heterogeneous surfaces. In addition, the sources contributing to uncertainties in the derived component temperatures require further investigation. We present a study wherein the component temperatures of a dynamically developing maize field were derived from thermal images. The sources influencing the derived component temperatures have been investigated and different parameterization schemes for atmospheric downwelling radiation have been compared. The results demonstrate that the thermal cameras provide a feasible method of deriving the component temperatures. If the thermal camera is mounted at approximately 30 m above the target and then the atmospheric upwelling radiation and transmittance is ignored, a 1.0-2.0 K error for the component temperatures may occur.

To validate satellite-based surface fluxes by ground measurements properly, several numerical simulations were carried out at a homogeneous alpine meadow site and mixed cropland site, considering various atmospheric conditions and different land cover distribution types. By comparing various pixel selection methods, the results showed that footprint was significant in insuring a consistent spatial scale between ground measurements and satellite-based surface fluxes, particularly for heterogeneous surface and high-resolution remote sensing data. Because large aperture scintillometer measurements cover larger areas than eddy covariance (EC) system measurements, the spatial heterogeneity at a subpixel scale in complicated surface should be further considered in validating coarse satellite data. Thus, more accurate validation data and scaling methods must be developed, such as measuring surface fluxes at the satellite pixel scale by a flux measurement matrix or airborne EC measurements.

The determination of the spatial heterogeneity of the regional evapotranspiration over a complex underlying surface in an oasis-desert region is crucial for water resource management in a river basin and aiding in irrigation decisions. The surface energy balance system (SEBS) model has been widely used to estimate surface energy fluxes. However, the parameterization of surface roughness length for momentum transfer (z0m) and heat transfer (z0h) did not perform well for a complex underlying surface. Moreover, it is difficult to estimate surface soil heat flux, i.e., G0, accurately at the regional scale. In this letter, the parameterization schemes of z0m, z0h, and G0were optimized. Measurements from 21 sets of eddy covariance systems were used to validate the model performance. The results show that the revised SEBS model root-mean-square errors (RMSEs) of the satellite-based sensible and latent heat fluxes (H and LE) decreased from 97.2 W · m-2to 56.9 W · m-2and from 102.9 W · m-2to 74.8 W · m-2, respectively, at the footprint scale. At the pixel scale, the RMSEs of the revised model estimates of the H and LE were 40.9 W · m-2and 57.5 W · m-2, respectively. The improved agreements between the estimates and the measurements indicate that the revised SEBS model is appropriate for estimating regional energy fluxes over heterogeneous oasis-desert surfaces. Furthermore, the spatial and temporal patterns of the LE in the middle reaches of the Heihe River were investigated.

The two-source energy balance (TSEB) model was successfully applied to estimate evaporation (E), transpiration (T), and evapotranspiration (ET) for land covered with vegetation, which has significantly important applications for the terrestrial water cycle and water resource management. However, the current composite temperature separation approaches are limited in their effectiveness in arid regions. Moreover, E and T are difficult to measure on the ground. In this letter, the ground-measured soil and canopy component temperatures were used to estimate E, T, and ET, which were better validated with observed ratios of E (E/ET%) and T (T/ET%) using the stable oxygen and hydrogen isotopes, and the ET measurements using an eddy covariance (EC) system. Our results indicated that even under the strongly advective conditions, the TSEB model produced reliable estimates of the E/ET% and T/ET% ratios and of ET. The mean bias and root-mean-square error (RMSE) of E/ET% were 1% and 2%, respectively, and the mean bias and RMSE of T/ET% were -1% and 2%, respectively. In addition, the model exhibited relatively reliable estimates in the latent heat flux, with mean bias and RMSE values of 31 and 61 W · m-2, respectively, compared with the measurements from the EC system. These results demonstrated that a robust soil and vegetation component temperature calculation was crucial for estimating E, T, and ET. Moreover, the separate validation of E/ET% and T/ET% provides a good prospect for TSEB model improvements.