Satellite Land Surface Temperature Products Research Articles

Reconstruction of Hourly FY-4A AGRI Land Surface Temperature under Cloud-Covered Conditions Using a Hybrid Method Combining Spatial and Temporal Information

Land Surface Temperature (LST) products obtained by thermal infrared (TIR) remote sensing contain considerable blank areas due to the frequent occurrence of cloud coverage. The studies on the all-time reconstruction of the cloud-covered LST of geostationary meteorological satellite LST products are relatively few. To accurately fill the blank area, a hybrid method for reconstructing hourly FY-4A AGRI LST under cloud-covered conditions was proposed using a random forest (RF) regression algorithm and Savitzky-Golay (S-G) filtering. The ERA5-Land surface cumulative net radiation flux (SNR) reanalysis data was first introduced to represent the change in surface energy arising from cloud coverage. The RF regression method was used to estimate the LST correlation model based on clear-sky LST and the corresponding predictor variables, including the normalized difference vegetation index (NDVI), the normalized difference water index (NDWI), surface elevation and slope. The fitted model was then applied to reconstruct the cloud-covered LST. The S–G filtering method was used to smooth the outliers of reconstructed LST in the temporal dimension. The accuracy evaluation was performed using the measured LST of the representative meteorological stations after scale correction. The coefficients of determination derived with the reference LST were all above 0.73 on the three examined days, with a bias of −1.13–0.39 K, mean absolute errors (MAE) of 1.46–2.4 K, and root mean square errors (RMSE) of 1.77–3.2 K. These results indicate that the proposed method has strong potential for accurately restoring the spatial and temporal continuity of LST and can provide a solution for the production and research of gap-free LST products with high temporal resolution.

Global hourly, 5 km, all-sky land surface temperature data from 2011 to 2021 based on integrating geostationary and polar-orbiting satellite data

Abstract. Land surface temperature (LST) plays a dominant role in the surface energy budget (SEB) and hydrological cycling. Thermal infrared (TIR) remote sensing is the primary method of estimating LST globally. However, cloud cover leaves numerous data gaps in satellite LST products, which seriously restricts their applications. Efforts have been made to produce gap-free LST products from polar-orbiting satellites (e.g., Terra and Aqua); however, satellite data from limited overpasses are not suitable for characterizing the diurnal temperature cycle (DTC), which is directly related to heat waves, plant water stress, and soil moisture. Considering the high temporal variability in LST and the importance of the DTC, we refined the SEB-based cloudy-sky LST recovery method by improving its feasibility and efficiency and produced a global hourly, 5 km, all-sky land surface temperature (GHA-LST) dataset from 2011 to 2021. The GHA-LST product was generated using TIR LST products from geostationary and polar-orbiting satellite data from the Copernicus Global Land Service (CGLS) and the Moderate Resolution Imaging Spectroradiometer (MODIS). Based on ground measurements at the 201 global sites from the Surface Radiation Budget (SURFRAD), Baseline Surface Radiation Network (BSRN), Fluxnet, AmeriFlux, Heihe River basin (HRB), and Tibetan Plateau (TP) networks, the overall root-mean-square error (RMSE) of the hourly GHA-LST product was 3.31 K, with a bias of −0.57 K and R2 of 0.95. Thus, this product was more accurate than the clear-sky CGLS and MODIS MYD21C1 LST samples. The RMSE value of the daily mean LST was 1.76 K. Validation results at individual sites indicate that the GHA-LST dataset has relatively larger RMSEs for high-elevation regions, which can be attributed to high surface heterogeneity and input data uncertainty. Temporal and spatial analyses suggested that GHA-LST has satisfactory spatiotemporal continuity and reasonable variation and matches the reference data well at hourly and daily scales. Furthermore, the regional comparison of GHA-LST with other gap-free hourly datasets (ERA5 and Global Land Data Assimilation System, GLDAS) demonstrated that GHA-LST can provide more spatial texture information. The monthly anomaly analysis suggests that GHA-LST couples well with global surface air temperature datasets and other LST datasets at daily mean and minimum temperature scales, whereas the maximum temperature and diurnal temperature range of LST and air temperature (AT) have different anomalous magnitudes. The GHA-LST dataset is the first global gap-free LST dataset at an hourly, 5 km scale with high accuracy, and it can be used to estimate global evapotranspiration, monitor extreme weather, and advance meteorological forecasting models. GHA-LST is freely available at https://doi.org/10.5281/zenodo.7487284 (Jia et al., 2022b) and http://glass.umd.edu/allsky_LST/GHA-LST (last access: 10 February 2023; Jia et al., 2022c).

A new East African satellite data validation station: Performance of the LSA-SAF all-weather land surface temperature product over a savannah biome

We describe a new satellite data validation facility located in a savannah biome at the International Livestock Research Institute (ILRI) Kapiti Research Station (Kenya). The facility is focused on satellite land surface temperature (LST) and is equipped with multiple ground-viewing infrared radiometers across four sites. The in-situ LST observations are upscaled to match satellite LST products using a geometric illumination model. The in-situ sensor network represents a step-forward in LST validation in East Africa and savannah biomes. To our knowledge this is the first time that such an extensive network of LST radiometers and supporting measurements has been installed in sub-Saharan Africa, or a savannah. With this network we capture surface heterogeneity in a manner that has not previously been possible. The LST ground data from this station collected between October 2018 and March 2019 is used to evaluate the new Land Surface Analysis Satellite Application Facility (LSA-SAF) all-sky LST product (MLST-AS) that blends clear-sky infrared-retrieved LSTs with LSTs derived from a land surface energy balance model to fill gaps due to cloudy conditions. Comparison against the in-situ LSTs indicates overall accuracy, precision, and root-mean-square error (RMSE) of MLST-AS to be 2.02 K, 1.38 K and 3.64 K respectively. The infrared-retrieved LST component of MLST-AS under clear skies has an accuracy, precision and RMSE of 1.16 K, 0.8 K and 3.16 K respectively. The energy balance model-based component of MLST-AS has performance statistics of 3.02 K, 1.38 K and 4.16 K. The MLST-AS energy balance model component is observed to perform worse when surface moisture is present, underestimating night-time and daily maximum temperatures by between 2 and 4 K in the 24 h following surface water deposition as precipitation or dew.

A Machine Learning Based Downscaling Approach to Produce High Spatio-Temporal Resolution Land Surface Temperature of the Antarctic Dry Valleys from MODIS Data

To monitor environmental and biological processes, Land Surface Temperature (LST) is a central variable, which is highly variable in space and time. This particularly applies to the Antarctic Dry Valleys, which host an ecosystem highly adapted to the extreme conditions in this cold desert. To predict possible climate induced changes on the Dry Valley ecosystem, high spatial and temporal resolution environmental variables are needed. Thus we enhanced the spatial resolution of the MODIS satellite LST product that is sensed sub-daily at a 1 km spatial resolution to a 30 m spatial resolution. We employed machine learning models that are trained using Landsat 8 thermal infrared data from 2013 to 2019 as a reference to predict LST at 30 m resolution. For the downscaling procedure, terrain derived variables and information on the soil type as well as the solar insolation were used as potential predictors in addition to MODIS LST. The trained model can be applied to all available MODIS scenes from 1999 onward to develop a 30 m resolution LST product of the Antarctic Dry Valleys. A spatio-temporal validation revealed an R2 of 0.78 and a RMSE of 3.32 ∘C. The downscaled LST will provide a valuable surface climate data set for various research applications, such as species distribution modeling, climate model evaluation, and the basis for the development of further relevant environmental information such as the surface moisture distribution.

Evaluation of Four New Land Surface Temperature (LST) Products in the U.S. Corn Belt: ECOSTRESS, GOES-R, Landsat, and Sentinel-3

Asa critical indicator of crop water stress, land surface temperature (LST) has been widely used in agricultural applications for monitoring and quantifying crop stress and crop water use. Given the availability of several newly operational satellite LST products in recent years and an increasing demand for reliable LST for agricultural applications, a systematic and thorough study of evaluation of different satellite LST products for croplands is highly needed. This article, thus, evaluated both the new satellite LST products including ECOsystem Spaceborne Thermal Radiometer on the International Space Station (ECOSTRESS), GOES-R, Landsat Provisional and Sentinel-3, and several mainstream LST products including MODIS Land Surface Temperature/Emissivity product (MOD11A1 and MYD11A1), MODIS/Aqua Land Surface Temperature/3-Band Emissivity product (MYD21A1), and VIIRS/NPP Land Surface Temperature and Emissivity LST (VNP21A1), for agricultural landscapes in the U.S. Corn Belt. The evaluation was benchmarked on in situ measurements from 11 SURFRAD or eddy covariance sites in the growing seasons of 2018 and 2019. Results showed that the nighttime and daytime biases of all LST products on different sites were generally within ±2 °C and ±3 °C, respectively. Regarding the daytime LST, the highest agreement with ground observations was achieved by ECOSTRESS with an overall absolute bias MYD11A1 a MOD11A1 (0.37/day) > Sentinel-3 (0.26/day) > Landsat Provisional (0.12/day) > ECOSTRESS (0.07/day). Holistically considering factors of bias, RMSE, spatial resolution and cloud-free data availability of different LST products, MOD11A1, and MYD11A1 are relatively appropriate for agriculture-related applications.

Intercomparison of In Situ Sensors for Ground-Based Land Surface Temperature Measurements.

Land surface temperature (LST) is a key variable in the determination of land surface energy exchange processes from local to global scales. Accurate ground measurements of LST are necessary for a number of applications including validation of satellite LST products or improvement of both climate and numerical weather prediction models. With the objective of assessing the quality of in situ measurements of LST and to evaluate the quantitative uncertainties in the ground-based LST measurements, intensive field experiments were conducted at NOAA’s Air Resources Laboratory (ARL)’s Atmospheric Turbulence and Diffusion Division (ATDD) in Oak Ridge, Tennessee, USA, from October 2015 to January 2016. The results of the comparison of LSTs retrieved by three narrow angle broadband infrared temperature sensors (IRT), hemispherical longwave radiation (LWR) measurements by pyrgeometers, forward looking infrared camera with direct LSTs by multiple thermocouples (TC), and near surface air temperature (AT) are presented here. The brightness temperature (BT) measurements by the IRTs agreed well with a bias of <0.23 °C, and root mean square error (RMSE) of <0.36 °C. The daytime LST(TC) and LST(IRT) showed better agreement (bias = 0.26 °C and RMSE = 0.67 °C) than with LST(LWR) (bias > 1.1 and RMSE > 1.46 °C). In contrast, the difference between nighttime LSTs by IRTs, TCs, and LWR were <0.47 °C, whereas nighttime AT explained >81% of the variance in LST(IRT) with a bias of 2.64 °C and RMSE of 3.6 °C. To evaluate the annual and seasonal differences in LST(IRT), LST(LWR) and AT, the analysis was extended to four grassland sites in the USA. For the annual dataset of LST, the bias between LST (IRT) and LST (LWR) was <0.7 °C, except at the semiarid grassland (1.5 °C), whereas the absolute bias between AT and LST at the four sites were <2 °C. The monthly difference between LST (IRT) and LST (LWR) (or AT) reached up to 2 °C (5 °C), whereas half-hourly differences between LSTs and AT were several degrees in magnitude depending on the site characteristics, time of the day and the season.

Component radiative temperatures over sparsely vegetated surfaces and their potential for upscaling land surface temperature

Ground measured component radiative temperatures are basic inputs for modelling energy and hydrological processes and for simulating land surface temperature (LST) as “viewed” by remote sensors. However, knowledge of factors affecting the component temperatures and about their potential for upscaling LST over sparsely vegetated surfaces with high heterogeneity is still lacking. Here, a MUlti-Scale Observation Experiment on land Surface temperature (MUSOES) was performed under HiWATER over an arid sparsely vegetated surface. Component temperatures were obtained with different instruments on multiple spatial scales; for LST upscaling, a three-dimensional scene model was employed for two forest stations (MFS and PFS) and a two-dimensional model for a shrub station (SUP). Results show that intrinsic characteristics contribute to the temperature variability between different components and even within a single component. Using a thermal infrared (TIR) imager at MFS, average temperature difference of 24.9 K between sunlit bare soil and tree canopy was found; different components exhibit different internal temperature differences at direction-level and pixel-level. Furthermore, illumination conditions, viewing directions, and instrument types significantly affected the measured component temperatures. The measurements of the TIR radiometer and the imager can deviate considerably (e.g. 14.9 K for sunlit bare soil at MFS). When the longwave radiometers were selected as target sensors, the component temperatures measured by the imager exhibit good potential for LST upscaling: the upscaled LST has MBD/RMSD values of 2.0 K/2.3 K at MFS and 2.0 K/2.5 K at PFS. The TIR radiometer’s measurements introduce large uncertainties into LST upscaling at MFS and PFS, but result in good accuracy at SUP, mainly due to its simpler land cover and surface structure. Findings from this study can benefit our understanding of factors affecting observations of component temperatures and the LST upscaling process and are, therefore, relevant for further studying the evaluation of satellite LST products.

Laboratory calibration and field measurement of land surface temperature and emissivity using thermal infrared multiband radiometers

Accurate ground measurements of land surface temperature (LST) are necessary for validating satellite LST products. In order to provide reliable data, ground radiometers must be calibrated with reference to an international standard, and radiometric temperatures must be corrected for land surface emissivity. As opposed to water, land surface emissivity is not usually known for many ground covers, so an emissivity value has to be assumed, assigned from spectral emissivity libraries or measured for each land cover and spectral band considered. The aim of this study is to show the laboratory calibration and the methodology for simultaneous field measurements of LST and emissivity employed in the comparison experiment held in the project Fiducial Reference Measurements for validation of Surface Temperature from Satellites (FRM4STS) funded by the European Space Agency. We used multiband CE-312 radiometers (five narrow bands plus one broad band in the 8–13 μm window) to simultaneously retrieve LST and band emissivities by means of the temperature-emissivity separation (TES) method for different ground covers (clover, tarmac, soil, gravel and sand). The TES method requires near-simultaneous measurements of ground-leaving radiances and sky downwelling radiances; the latter being measured in the field using a gold reflectance panel. For each surface cover, TES provided band emissivities in the CE-312 bands and LST from continuous radiance measurements performed over time. As a result of the experiment, we present the laboratory calibration of the CE-312 radiometers carried out against a traceable, reference blackbody, and the LST series and band emissivity values for the ground covers considered, together with a detailed LST uncertainty analysis including the uncertainties associated to the calibration of ground radiometers, the emissivity estimation by means of the TES method, and the sky radiance measurements, among others. According to these results, the total LST uncertainty was estimated as 0.4-0.6 K for the ground covers measured during the experiment.

The GLASS Land Surface Temperature Product

Land surface temperature (LST) is one of the most important geophysical parameters at the earth's surface. Satellite LST products are beneficial for scientific communities with a variety of applications. The Global Land Surface Satellite (GLASS) LST product has been generated for 1983, 1993, 2003, and 2013 from historical NOAA-7 and NOAA-14 AVHRR data and recent Terra/Aqua MODIS data. The spatial resolution of the instantaneous GLASS LST product is 0.05° for 1983 and 1993 and is 1 km for 2003 and 2013. The algorithm for generating the GLASS LST product is based on a multialgorithm ensemble approach, which combines nine split-window algorithms with good performance in training, testing, and sensitivity analysis. The validation results based on in situ measurements demonstrate that the GLASS-AVHRR LST product has an accuracy of 2.89 K at Barrow site; the GLASS-MODIS LST product has an accuracy of 1.82-2.15 K at six grassland/cropland SURFRAD sites at nighttime, which is a similar accuracy to the official MODIS LST product. Systematic underestimation of LST at Desert Rock, NV (arid shrubland) has been found and has been attributed to uncertainties associated with land surface emissivities and other input datasets. Further intercomparison indicates that the GLASS-MODIS LST product agrees well with the official MODIS LST, with mean bias deviations/root mean square deviations of -1.19 K/1.87 K and 1.89 K/2.27 K for the two granules under examination. Nevertheless, the validation of the GLASS LST product is currently in a preliminary stage, and more in-depth examinations are still needed.

Spatiotemporal Reconstruction of Land Surface Temperature Derived From FengYun Geostationary Satellite Data

The FengYun-2F (FY-2F) geostationary satellite land surface temperature (LST) and its diurnal variation are important when evaluating climate change, the land-atmosphere energy budget, and the hydrological cycle. However, the presence of clouds generates numerous meaningless pixels that constrain the potential application of the available satellite LST products. These pixels covered by cloud are assigned –2, and otherwise are the LST values, based on the result of a double-channel threshold cloud detection algorithm. This paper proposes a combined temporal and spatial information reconstruction method for the missing FY-2F LST data reconstruction with a good spatial continuity, where cloud detection has already been undertaken. Compared with the methods used in the past, the main characteristics of the proposed method are: 1) the consideration of a free parameter $\delta T$ when modeling the diurnal temperature cycle curve; 2) the introduction of the genetic algorithm for solving the parameters; 3) the adoption of the spectral multimanifold clustering algorithm for clustering the multitemporal geostationary satellite LST data; and 4) the accurate and efficient combined temporal and spatial reconstruction method. The proposed combined temporal and spatial reconstruction method was tested and quantitatively assessed with both simulated and real data experiments, using the FY-2F LST products. The results indicate that the combined reconstruction method is accurate to within about 2 °C, which can significantly improve the practical value of FY-2F LST datasets.

Robust reconstruction of missing data in Feng Yun geostationary satellite land surface temperature products

Robust reconstruction of missing data in Feng Yun geostationary satellite land surface temperature products

Intercomparison of Operational Land Surface Temperature Products Derived From MSG-SEVIRI and Terra/Aqua-MODIS Data

Accuracy assessment of land surface temperature (LST) products is critical to facilitate their use in various studies. As an alternative method for assessing the accuracy of LST products, a new satellite LST product is compared with a heritage LST product to validate and determine the uncertainties in the satellite-derived LST approach. In this study, we propose a method for the intercomparison of the Meteosat second generation-spinning enhanced visible and infrared imager (MSG-SEVIRI) and Terra/Aqua-moderate resolution imaging spectroradiometer (MODIS) LST products. The intercomparison was performed by verifying the collocation in space, temporal concurrence, viewing geometry alignment, and spatial homogeneity between the two LST products. The discrepancies between the SEVIRI and MODIS LST products were investigated over different seasons, times of day, and surface types. SEVIRI LST values are generally higher than MODIS LST values, with positive biases during daytime (approximately 2–4 K) and nighttime (approximately 1–2 K). Significant variability of the daytime LST discrepancies with season, time of day, and surface type is observed. Compared with the daytime LST discrepancies, the nighttime LST discrepancies are less dependent on season, time of day, and surface type.

Integrated fusion of multi-scale polar-orbiting and geostationary satellite observations for the mapping of high spatial and temporal resolution land surface temperature

Land surface temperature (LST) and its diurnal variation are important when evaluating climate change, the land–atmosphere energy budget, and the global hydrological cycle. However, the available satellite LST products have either a coarse spatial resolution or a low temporal resolution, which constrains their potential applications. This paper proposes a spatio-temporal integrated temperature fusion model (STITFM) for the retrieval of high temporal and spatial resolution LST from multi-scale polar-orbiting and geostationary satellite observations. Compared with the traditional fusion methods for LST with two different sensors, the proposed method is able to fuse the LST from arbitrary sensors in a unified framework. The model was tested using LST with fine, moderate, and coarse-resolutions. Data from the Landsat Thematic Mapper (TM) and Enhanced Thematic Mapper Plus (ETM+) sensors, the Moderate Resolution Imaging Spectroradiometer (MODIS), the Geostationary Operational Environmental Satellite (GOES) Imager, and the Meteosat Second Generation (MSG) Spinning Enhanced Visible and Infrared Imager (SEVIRI) were used. The fused LST values were evaluated with in situ LST obtained from the Surface Radiation Budget Network (SURFRAD) and the Land Surface Analysis Satellite Application Facility (LSA SAF) project. The final validation results indicate that the STITFM is accurate to within about 2.5K.

Land-surface temperature retrieval at high spatial and temporal resolutions based on multi-sensor fusion

Land-surface temperature (LST) is of great significance for the estimation of radiation and energy budgets associated with land-surface processes. However, the available satellite LST products have either low spatial resolution or low temporal resolution, which constrains their potential applications. This paper proposes a spatiotemporal fusion method for retrieving LST at high spatial and temporal resolutions. One important characteristic of the proposed method is the consideration of the sensor observation differences between different land-cover types. The other main contribution is that the spatial correlations between different pixels are effectively considered by the use of a variation-based model. The method was tested and assessed quantitatively using the different sensors of Landsat TM/ETM+, moderate resolution imaging spectroradiometer and the geostationary operational environmental satellite imager. The validation results indicate that the proposed multisensor fusion method is accurate to about 2.5 K.

Validation of GOES-R Satellite Land Surface Temperature Algorithm Using SURFRAD Ground Measurements and Statistical Estimates of Error Properties

Validation of satellite land surface temperature (LST) is a challenge because of spectral, spatial, and temporal variabilities of land surface emissivity. Highly accurate in situ LST measurements are required for validating satellite LST products but are very hard to obtain, except at discrete points or for very short time periods (e.g., during field campaigns). To compare these field-measured point data with moderate-resolution (1 km) satellite products requires a scaling process that can introduce errors that ultimately exceed those in the satellite-derived LST products whose validation is sought. This paper presents a new method of validating the Geostationary Operational Environmental Satellite (GOES) R-Series (GOES-R) Advanced Baseline Imager (ABI) LST algorithm. It considers the error structures of both ground and satellite data sets. The method applies a linear fitting model to the satellite data and coregistered “match-up” ground data for estimating the precisions of both data sets. In this paper, GOES-8 Imager data were used as a proxy of the GOES-R ABI data for the satellite LST derivation. The in situ data set was obtained from the National Oceanic and Atmospheric Administration's SURFace RADiation (SURFRAD) budget network using a stringent match-up process. The data cover one year of GOES-8 Imager observations over six SURFRAD sites. For each site, more than 1000 cloud-free match-up data pairs were obtained for day and night to ensure statistical significance. The average precision over all six sites was found to be 1.58 K, as compared to the GOES-R LST required precision of 2.3 K. The least precise comparison at an individual SURFRAD site was 1.8 K. The conclusion is that, for these ground truth sites, the GOES-R LST algorithm meets the specifications and that an upper boundary on the precision of the satellite LSTs can be determined.