High Resolution Land Surface Temperature Research Articles

Spatial Downscaling of Nighttime Land Surface Temperature Based on Geographically Neural Network Weighted Regression Kriging

Land surface temperature (LST) has a wide application in Earth Science-related fields, and spatial downscaling is an important method to retrieve high-resolution LST data. However, existing LST downscaling methods have difficulties in simultaneously constructing and expressing spatial non-stationarity, spatial autocorrelation, and complex non-linearity during the LST downscaling process, which limits the performance of the models. Moreover, there is a lack of research on high-resolution nighttime land surface temperature (NLST) reconstruction based on spatial downscaling, which does not meet the data needs for urban-scale nighttime urban heat island (UHI) studies. Therefore, this study combined Geographically Neural Network Weighted Regression (GNNWR) with Area-to-Point Kriging interpolation (ATPK) to propose a Geographically Neural Network Weighted Regression Kriging (GNNWRK) model for NLST downscaling. To verify the model’s generality and robustness, this study selected four study areas with different landform and climate type for NLST spatial downscaling experiments. The GNNWRK was compared with four benchmark downscaling methods, including TsHARP, Random Forest, Geographically Weighted Regression, and GNNWR. The results show that compared to these four benchmark methods, the GNNWRK method has higher accuracy in NLST downscaling, with a maximum Pearson’s Correlation Coefficient (Pcc) of 0.930 and a minimum Root Mean Square Error (RMSE) of 0.886 K. Moreover, the validation based on MODIS NLST data and ground-measured NLST data also indicates that the GNNWRK model can obtain more accurate, high-resolution NLST with richer and more detailed texture. This enhances the potential of NLST in studying the effects of urban nighttime heat islands at a finer scale.

Spatiotemporal Land Use/Land Cover Changes and Impact on Urban Thermal Environments: Analyzing Cool Island Intensity Variations

This study pioneers the comprehensive evaluation of the spatiotemporal evolution of land use/land cover (LULC) in Hangzhou city, introducing the novel water body shape index (WBSI) to analyze its seasonal impacts on the urban thermal environment and urban cool island (UCI) effects, uncovering distinct patterns of thermal regulation. It particularly investigates how distance gradients and the water body shape index (WBSI) influence land surface temperature (LST) in the urban core. The region’s climate, featuring hot summers and cold winters, highlights significant seasonal LST variations. Addressing a gap in existing UCI research, the analysis extends beyond the typical large-scale planning focus to include small-scale, high-resolution aspects. Employing remote sensing and geographic information system (GIS) analysis techniques, this study analyzes the seasonal dynamics in Hangzhou’s central urban area. High-resolution LST data, obtained through single-channel inversion and resolution enhancement algorithms, are crucial to this analysis. This study employs the maximum likelihood classification method to analyze land use and land cover changes from 1990 to 2020. This analysis reveals potential drivers of urban thermal environment changes, such as the expansion of residential and commercial areas and the reduction in green spaces. Different regions in LST data are delineated to assess the cool island effect, and the complexity of water body boundaries is quantified using the water body shape index. Spatial and temporal patterns of LST changes are investigated using multivariate regression and time-series analysis models. We identified significant changes in LULC over the past 30 years in Hangzhou, closely correlating with a continuous rise in LST. This observation underscores a clear finding: the strategic importance of blue–green infrastructure in mitigating urban heat, a novel insight that extends the current understanding of urban thermal dynamics. A clear and novel finding of this study is that the intensity of the cool island effect from large water bodies not only diminishes with distance but is intricately influenced by the complexity of their shapes, as quantified by the WBSI, whereas the complexity of their boundaries enhances this effect. Additionally, the regulatory role of the cool island effect is observed to vary seasonally, being most pronounced in summer and less so in autumn and winter, thereby demonstrating a positive impact. In conclusion, our findings innovatively highlight how the specific shapes of water bodies, quantified through the water body shape index (WBSI), emerge as critical, yet previously underappreciated, drivers in modulating the urban thermal environment. This underscores a new avenue for urban planning, advocating for the strategic design of water bodies within urban landscapes. It also finds that spatial factors and seasonal variations significantly affect the intensity of the cool island effect. These findings offer valuable evidence for urban planning and climate change adaptation, emphasizing balancing natural elements with the built environment in urban design.

Summer drought weakens land surface cooling of tundra vegetation

Siberia experienced a prolonged heatwave in the spring of 2020, resulting in extreme summer drought and major wildfires in the North-Eastern Siberian lowland tundra. In the Arctic tundra, plants play a key role in regulating the summer land surface energy budget by contributing to land surface cooling through evapotranspiration. Yet we know little about how drought conditions impact land surface cooling by tundra plant communities, potentially contributing to high air temperatures through a positive plant-mediated feedback. Here we used high-resolution land surface temperature and vegetation maps based on drone imagery to determine the impact of an extreme summer drought on land surface cooling in the lowland tundra of North-Eastern Siberia. We found that land surface cooling differed strongly among plant communities between the drought year 2020 and the reference year 2021. Further, we observed a decrease in the normalized land surface cooling (measured as water deficit index) in the drought year 2020 across all plant communities. This indicates a shift towards an energy budget dominated by sensible heat fluxes, contributing to land surface warming. Overall, our findings suggest significant variation in land surface cooling among common Arctic plant communities in the North-Eastern Siberian lowland tundra and a pronounced effect of drought on all community types. Based on our results, we suggest discriminating between functional tundra plant communities when predicting the drought impacts on energy flux related processes such as land surface cooling, permafrost thaw and wildfires.

Deep Interpolation of Remote Sensing Land Surface Temperature Data with Partial Convolutions.

Land Surface Temperature (LST) is an important resource for a variety of tasks. The data are mostly free of charge and combine high spatial and temporal resolution with reliable data collection over a historical timeframe. When remote sensing is used to provide LST data, such as the MODA11 product using information from the MODIS sensors attached to NASA satellites, data acquisition can be hindered by clouds or cloud shadows, occluding the sensors' view on different areas of the world. This makes it difficult to take full advantage of the high resolution of the data. A common solution to interpolating LST data is statistical interpolation methods, such as fitting polynomials or thin plate spine interpolation. These methods have difficulties in incorporating additional knowledge about the research area and learning local dependencies that can help with the interpolation process. We propose a novel approach to interpolating remote sensing LST data in a fixed research area considering local ground-site air temperature measurements. The two-step approach consists of learning the LST from air temperature measurements, where the ground-site weather stations are located, and interpolating the remaining missing values with partial convolutions within a U-Net deep learning architecture. Our approach improves the interpolation of LST for our research area by 44% in terms of RMSE, when compared to state-of-the-art statistical methods. Due to the use of air temperature, we can provide coverage of 100%, even when no valid LST measurements were available. The resulting gapless coverage of high resolution LST data will help unlock the full potential of remote sensing LST data.

Downscaling Land Surface Temperature Derived from Microwave Observations with the Super-Resolution Reconstruction Method: A Case Study in the CONUS

Optical sensors cannot penetrate clouds and can cause serious missing data problems in optical-based Land Surface Temperature (LST) products. Under cloudy conditions, microwave observations are usually utilized to derive the land surface temperature. However, microwave sensors usually have coarse spatial resolutions. High-Resolution (HR) LST data products are usually desired for many applications. Instead of developing and launching new high-resolution satellite sensors for LST observations, a more economical and practical way is to develop proper methodologies to derive high-resolution LSTs from available Low-Resolution (LR) datasets. This study explores different algorithms to downscale low-resolution LST data to a high resolution. The existing regression-based downscaling methods usually require simultaneous observations and ancillary data. The Super-Resolution Reconstruction (SRR) method developed for traditional image enhancement can be applicable to high-resolution LST generation. For the first time, we adapted the SRR method for LST data. We specifically built a unique database of LSTs for the example-based SRR method. After deriving the LST data from the coarse-resolution passive microwave observations, the AMSR-E at 25 km and/or AMSR-2 at 10 km, we developed an algorithm to downscale them to a 1 km spatial resolution with the SRR method. The SRR downscaling algorithm can be implemented to obtain high-resolution LSTs without auxiliary data or any concurrent observations. The high-resolution LSTs are validated and evaluated with the ground measurements from the Surface Radiation (SURFRAD) Budget Network. The results demonstrate that the downscaled microwave LSTs have a high correlation coefficient of over 0.92, a small bias of less than 0.5 K, but a large Root Mean Square Error (RMSE) of about 4 K, which is similar to the original microwave LST, so the errors in the downscaled LST could have been inherited from the original microwave LSTs. The validation results also indicate that the example-based method shows a better performance than the self-similarity-based algorithm.

Climatic comparison of surface urban heat island using satellite remote sensing in Tehran and suburbs

In this study, we aim to compare the climatic conditions of Surface Urban Heat Island (SUHI) in Tehran and its suburbs using day/night time data from three satellites. A high-resolution Land Surface Temperature (LST) data from MODIS Aqua, Sentinel-3, and Landsat 8 were selected to facilitate this study. The highest values of LST/UHI are observed in downtown Tehran and suburban areas at night. The temperature difference also shows an increase at night in Tehran and the western suburbs, while it decreases during the day. When comparing LST/UHI with altitude in different directions, it is found that urban areas and the south, southeast, southwest, and west suburban areas experience higher temperatures at night. MODIS LST products are more appropriate for checking nighttime SUHI in Tehran's Great area in comparison to other products. Moran's I indicates that the highest positive values occur during seasonal and annual periods at night. The Getis index demonstrates a consistent pattern across all seasons, and this trend persists throughout the year. The seasonal and annual UHI difference between Tehran and its suburbs is 5 °C. The LST diagram reveals that higher temperatures occur during warm months. The temporal NDVI distribution indicates lower NDVI values from June to February and summer to winter. The spatial distribution shows that due to the lack of NDVI index in urban areas, LST/UHI values are higher at night in Tehran compared to the suburbs. UHI is not limited to urban areas but has also spread beyond the city borders. As a result, the highest UHI values are found in downtown Tehran and its southeast, south, southwest, and west suburbs.

Comparison of multi-factor spatial downscaling models for high-resolution LST estimation in mountainous and hilly open-pit mines

The stripping of soil and vegetation by mining activities in mountainous and hilly areas of southern China has caused drastic surface disturbance and serious ecological and environmental problems. The spatial distribution of Land Surface Temperature (LST) in the mine site can better reflect the ecological disturbances caused by mining and is an important parameter for assessing the ecological environment of the mining sites. The classic Distrad thermal pixel decomposition model was improved by using different regression factors and linear and nonlinear fitting methods in this study, and three multi-factor downscaling models were proposed. Finally, four downscaling models were used to estimate the high spatial resolution LST for three three typical open-pit mining areas and explore the contribution of different surface parameters in the downscaling process from qualitative and quantitative perspectives. The results show that: (1) The MTVET (Extremely Randomized Trees regression model with multitype predictor variables) model exhibits the most accurate prediction accuracy in all three mining areas. The spatial distribution of errors shows that the area share of absolute errors between 0 and 1 K exist in 86.48 %, 81.38 %, and 84.51 % of the areas in Ganzhou rare earth mine, Dexing copper mine and Pingxiang coal mine, respectively. Based on the Root Mean Square Error (RMSE) assessment, the three mines are 0.699 K, 0.790 K, and 0.751 K, respectively, which are all controlled within 1 K; (2) The downscaled results of the MTVET model show rich and clear LST spatial texture features when typical mine sites are selected in each mining areas for detail comparison; (3) It was found that the topographic index has the greatest impact on accuracy enhancement, suggesting that topographic factors have an important influence in the study of LST downscaling in mountainous and hilly areas of southern China.

Conversion of satellite passive microwave signals to land surface “skin” temperature for extremely dry deserts

Satellite passive microwave (PMW) observations are important data sources for obtaining land surface temperature (LST) especially for cloudy conditions. Considering the generally poorer spatial representativeness of PMW observations, fusion of PMW-based and thermal infrared-(TIR-) based observations have been considered as an important approach to generate all-weather LST with appropriate resolution. However, the existing penetration difference between PMW-based and TIR-based observations has been an obstacle against this objective. This is especially the case for desert regions, where the penetration depth of PMW observations is larger than under other conditions and the relationship between TIR-based and PMW-based signals are thus more complicated. In this regard, this study develops an easy and high-efficiency model suitable for transforming PMW-based subsurface LST to TIR-like skin surface temperature in four regions across the world characterized by typical arid climate. The developed model employs microwave-derived surface soil moisture (SSM) and vegetation transmissivity (VT) as major inputs to quantify that “resampling depth difference” between TIR-based and PMW-based observations. Results show that the model can significantly reduce that difference (RMSE) from >10 K to around 2–5 K in the typical desert regions. The developed model is beneficial to obtaining high-resolution all-weather LST for the vast desert regions across the world in subsequent studies.

A High-Resolution Land Surface Temperature Downscaling Method Based on Geographically Weighted Neural Network Regression

Spatial downscaling is an important approach to obtain high-resolution land surface temperature (LST) for thermal environment research. However, existing downscaling methods are unable to sufficiently address both spatial heterogeneity and complex nonlinearity, especially in high-resolution scenes (<120 m), and accordingly limit the representation of regional details and accuracy of temperature inversion. In this study, by integrating normalized difference vegetation index (NDVI), normalized difference building index (NDBI), digital elevation model (DEM), and slope data, a high-resolution surface temperature downscaling method based on geographically neural network weighted regression (GNNWR) was developed to effectively handle the problem of surface temperature downscaling. The results show that the proposed GNNWR model achieved superior downscaling accuracy (maximum R2 of 0.974 and minimum RMSE of 0.896 °C) compared to widely used methods in four test areas with large differences in topography, landforms, and seasons. We also achieved the best extracted and most detailed spatial textures. Our findings suggest that GNNWR is a practical method for surface temperature downscaling considering its high accuracy and model performance.

Observed inequality in thermal comfort exposure and its multifaceted associations with greenspace in United States cities

Increasing exposure to heat stress threatens the health and well-being of urban residents. However, existing studies on measuring human thermal comfort exposure remain uncertain without considering fine-scale human-heat interaction and its long-term dynamics. To inform this issue, we proposed a population-weighted exposure assessment framework with the integration of high-resolution land surface temperature and population data to evaluate human exposure to thermal comfort and the associated inequality across 398 major cities over 2000–2020 in the United States, and further explored the multifaceted associations between greenspace and thermal comfort. Results show that 199 United States cities (50.00 %) experience severe heat stress (i.e., thermal comfort exposure <0.44) and 99 of which (24.87 %) are unevenly exposed to heat stress (i.e., Gini index of thermal comfort exposure >0.36). Temporal analysis from 2000 to 2020 reveals that human exposure to thermal comfort decreases by a mean magnitude of −0.00081 yr−1, and the associated inequality level decreases by a mean magnitude of −0.00153 yr−1. By linking urban greenspace and heat exposure, we find that greenspace has multifaceted associations with heat stress, with a highly positive correlation between greenspace and thermal comfort (i.e., comfort regulation by physical cooling effect) and a coincided exposure inequality between greenspace and thermal comfort. This study offers an alternative framework to characterize fine-scale human exposure to thermal comfort across space and time, provides observational evidence of thermal comfort exposure inequality in the United States cities, and highlights the need for prioritizing greening policies and actions to mitigate heat stress and exposure disparity.

A Robust Framework for Resolution Enhancement of Land Surface Temperature by Combining Spatial Downscaling and Spatiotemporal Fusion Methods

Land surface temperature (LST) products with high spatial resolution and short revisiting cycles are crucial for environmental studies. However, due to the tradeoff between spatial and temporal resolutions of satellite observations, such data are not directly available. Spatial downscaling and spatiotemporal fusion methods are existing solutions for this problem, but their robustness is limited under different surface conditions. Here, we propose a Robust Framework for Combining Downscaling and spatiotemporal Fusion methods (RFCDF) to generate the synthesized daily high-resolution LST with high accuracy in different landscapes. RFCDF introduces a novel weighting strategy that determines pixel-level weights using an empirical function under the constraint of the image-level weights of two predictions. We implement the framework using the thermal sharpening algorithm (TsHARP) and Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM) with Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) and Moderate Resolution Imaging Spectroradiometer (MODIS) data in Beijing and Baotou, as well as Landsat 8 and simulated coarse resolution imagery in nine sub-regions with different surface landscapes in Beijing. Our results demonstrate that RFCDF can generate more accurate estimations and preserve more spatial details than either individual or combination methods, improving accuracy by 0.1–0.6 K and 0.4–1.3 K in the two study areas, respectively. Moreover, the proposed framework is robust, reducing the root mean square error of estimations by 8-24% under different surface conditions. RFCDF can also generate dense high-resolution LST time series, which is crucial for studying the surface thermal environment at a finer scale.

A Spatial Downscaling Approach for Land Surface Temperature by Considering Descriptor Weight

Acquiring the satellite land surface temperature (LST) with high spatiotemporal resolutions is pressing in the land surface biophysical process. However, most current LST products hardly satisfy this requirement. LST Downscaling provides an effective way to solve this issue by introducing driving factors, but existing methods usually ignore the weights of descriptors. In this letter, based on the Geographically Weighted Regression (GWR) and Random Forest (RF), a new downscaling method ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i.e</i> ., WGWR) considering the weights of LST descriptors is proposed. To examine the performance of WGWR, the 100-m Landsat-8 TIRS and Terra ASTER LSTs are aggregated to 1000 m as the simulated coarse LSTs, and then the coarse LSTs are downscaled to 100 m using WGWR, RF, and GWR. Meanwhile, the original 100-m LSTs are used as validation references. Results indicate that the proposed WGWR outperforms RF and GWR: for RF (GWR), the RMSEs can be reduced by 0.34 K (0.26 K) in Zhangye and 0.22 K (0.1 K) in Beijing. Compared to RF and GWR, WGWR also yields better image quality: the downscaled LST images have neither obvious smoothing effect nor boundary effect and maintain the details of the image at high spatial resolution. Validation based on <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">in-situ</i> LST indicates that the downscaled LST based on WGWR has better agreement with the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">in-situ</i> LST, and the RMSE is reduced by 0.57 K. The proposed WGWR contributes to obtain high spatio-temporal resolution LSTs and promote hydrological, meteorological, and ecological studies.

Interactions among spatial configuration aspects of urban tree canopy significantly affect its cooling effects

Increasing urban tree canopy (UTC) has been widely recognized as an effective means for urban heat mitigation and adaptation. While numerous studies have shown that both percent cover of UTC and its spatial configuration can significantly affect urban temperature, the pathways governing these relationships are largely unexplored. Here we present a cross-city comparison aiming to fill this gap by explicitly quantifying the pathways on which percent cover of UTC and its spatial configuration affect land surface temperature (LST) using structural equation modeling (SEM), based on UTC mapped from high resolution imagery and LST derived from Landsat thermal bands. We found: 1) Although both the direct and indirect pathways significantly affected LST regardless of scales and cities, the direct pathway played a more important role in affecting LST in Baltimore, Beijing, and Shenzhen. In contrast, an opposite result was found in Sacramento, likely due to the effects of buildings and their interactions with UTC. 2) Similarly, the direct pathway of mean patch size (MPS) and mean shape index (MSI) played a more important role in affecting LST than their indirect effects via altering edge density (ED). Our results highlighted the necessity for discomposing the effects of different spatial configuration variables on LST. Understanding the pathways through which UTC affects LST can provide insights into urban heat mitigation and adaptation.

A high spatiotemporal resolution land surface temperature research over Qinghai-Tibet Plateau for 2000–2020

As ‘the third pole of the world’, the land surface temperature (LST) of the Qinghai-Tibet Plateau (QTP) has a profound impact on the climate of central Asia and even the whole earth. Studying the impact of the LST over QTP depends on long time and high spatiotemporal resolution LST dataset. However, the unavailability of such dataset has hindered LST-related researches: one of the most important reasons is that traditional spatiotemporal fusion methods such as Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM) have heavy computation demands to process big data. To fill this gap, this paper outlines one new cloud spatiotemporal fusion method by combining Google Earth Engine and STARFM to develop for the first time a 3-hourly 30 m LST dataset over the QTP from 2000 to 2020 through fusing Landsat and GLDAS-2.1 derived LST data. The outlined method first fuses the LSTs obtained from Landsat and GLDAS-2.1 data within one year to synthesize the LST of the entire QTP on one base time, and then the base time LST is spatiotemporally fused with GLDAS-2.1 LSTs on the base and prediction times to derive the 3-hourly 30 m QTP's LSTs on prediction times. The outlined method provides a promising technical scheme for batch processing big data in combining traditional spatiotemporal fusion methods with cloud computing platforms. Derived LST dataset is validated by station observations at multiple time and spatial scales to have high accuracy, which provides a guarantee for analyzing water and heat exchange and climate change over the QTP.

Land surface temperature downscaling in the karst mountain urban area considering the topographic characteristics

To obtain high-spatial-resolution land surface temperature (LST) in karst areas, it is necessary to select a downscale regression model with a better simulation effect and the scale factors that can best represent the topographic characteristics of karst mountainous areas. In Guiyang, a typical karst mountain city, two areas are selected as the study area, which is dominated by natural surface and construction land. Based on the data of Landsat-8 Thermal Infrared Sensor (TIRS), Sentinel-2, Advanced Land Observing Satellite Digital Elevation Model (ALOS DEM), and meteorological stations, the scale factors representing bare land: bare soil index and topographic relief: mountain shadow (hillshade), relief degree of land surface (RDLS), solar incident angle, and sky view factor (SVF) are added on the basis of the conventional factors. At the same time, random forest (RF) and extreme gradient boosting (XGBoost) models are used to construct an LST downscaling method that is more suitable for karst mountain cities. After the above steps, the LST product with a spatial resolution of 10 m is finally estimated. The results show that, due to the characteristics of large elevation variation, fragmentation, and high heterogeneity of surface landscape in karst areas, digital elevation model (DEM), RDLS, and SVF factors need to be considered in the downscaling of surface temperature, and the contribution rates of these factors are all more than 6% in the model. In terms of accuracy evaluation of ground temperature, XGBoost model has the highest accuracy with an average absolute error of 1.67K, RF model has an average error of 1.90K, and thermal image sharpening has the worst accuracy with an average error of 2.41K. In terms of accuracy evaluation of ascending scale, the XGBoost model also shows higher accuracy and richer texture details. The research results can provide basic data for the acquisition of high-resolution LST and its intermediate parameters in this area and also provide a method reference for the reduction of high-resolution LST in similar areas.

Improving LST Downscaling Quality on Regional and Field-Scale by Parameterizing the DisTrad Method

Many scientists have been investigating Land Surface Temperature (LST) because of its relevance in water management science due to its direct influence on the hydrological water cycle. This effect stems from being one of the most significant variables influencing evapotranspiration. One of the most important reasons for the evapotranspiration retrieved from MODIS data’s limited suitability for scheduling and planning irrigation schemes is the lack of spatial resolution. As a result, high-resolution LST is required for estimating evapotranspiration. The goal of this study is to improve the resolution of the available LST data, to improve evapotranspiration (ETa) estimation using statistical downscaling with Normalized Difference Vegetation Index (NDVI) as a predictor. The DisTrad (Disaggregation of Radiometric Surface Temperature) method was used for the LST downscaling procedure, which is based on aggregating the NDVI map to the LST map resolution and then calculating the coefficient of variation of the native NDVI map within the aggregated pixel and classifying the aggregated map into three classes: NDVI &lt; 0.2 for the bare soil, 0.2 ≤ NDVI ≤ 0.5 for the partial vegetation, and NDVI &gt; 0.5 for the full vegetation. DisTrad uses 25% of the pixels with the lowest coefficient of variation from each class to calculate the regression coefficients. In this work, adjustments to the DisTrad method were implemented to enhance downscaling LST and to examine the impacts of that alteration on the evapotranspiration estimation. The linear regression model was tested as an alternative to the original second-order polynomial. In using 10% of the pixels instead of the originally proposed 25% with the lowest coefficient of variation values, it is assumed that a group of pixels with a lower coefficient of variation represents a more homogeneous area, thus it gives more accurate values. The downscaled LST map retrieval was validated using Landsat 8 thermal maps (100 m). Applying the modified DisTrad approach to disaggregate Landsat LST to 30 m (NDVI resolution) yielded an R2 of 0.72 for the 10%, 0.74 for the 25% and 0.61 for the second-order polynomial lowest coefficient of variation compared to native LST Landsat, which means that 10% can be used as an alternative. Applying the downscaled LST map to estimate ETa yielded R2 0.84 in both cases, compared to ETa yielded from the native Landsat LST. These results prove that using the robust linear regression provided better results than using polynomial regression. With the downscaled Land Surface Temperature data, it was possible to create detailed ETa maps of the small agricultural fields in the test area.

Localized Downscaling of Urban Land Surface Temperature—A Case Study in Beijing, China

High-resolution land surface temperature (LST) data are essential for fine-scale urban thermal environment studies. Urban LST downscaling studies mostly remain focused on only two-dimensional (2-D) data, and neglect the impact of three-dimensional (3-D) surface structure on LST. In addition, the choice of window size is also important for LST downscaling over heterogeneous surfaces. In this study, we downscaled Landsat-LST using localized and stepwise approaches in a random forest model (RF). In addition, both 2- and 3-D building morphologies were included. Our results show that: (1) The performances of a local moving window and stepwise downscaling are dependent on the extent of surface heterogeneity. For mixed surfaces, a localized window performed better than the global window, and a stepwise approach performed better than a single-step approach. However, for monotonous surfaces (e.g., urban impervious surfaces), the global window performed better than a localized window; (2) That multi-scale geographically weighted regression (MGWR) could provide a possibility for selection of the optimal moving window. 7 × 7 windows derived from MGWR by the minimum bandwidth of predictors, performed better than other windows (3 × 3, 5 × 5, and 11 × 11) in the Beijing area; (3) That the morphology of buildings has a non-negligible impact and scaling effect on urban LST. When building morphologies were included in downscaling, the performance of the RF model improved. Furthermore, the importance of the sky view factor, building height, and building density was greater at a higher resolution than at a lower resolution.

Pansharpened landsat 8 thermal-infrared data for improved Land Surface Temperature characterization in a heterogeneous urban landscape

Reliable urban Land Use Land Cover (LULC) and Land Surface Temperature (LST) characterization are important for land use planning as well as climate and thermal impact analysis in cities. High resolution LULC and LST datasets are required for detailed spatial analysis such as understanding the contribution of small LULC fragments to thermal characteristics of the entire urban landscape. To date, detailed analysis of urban surfaces using satellite imagery has been hindered by low to moderate spatial resolution of freely available sensors of past and existing missions. Hence, this study sought to investigate the potential of pansharpening Landsat 8 data in improving LULC mapping and LST characterization in Bulawayo metropolitan city, Zimbabwe. Six LULC classes namely; bare area, built-up, low density, dense forest, grasslands and water bodies were considered for the study and respective spatially-explicit locational information (GPS coordinates) class samples identified. Supervised image classification using the Support Vector Machines algorithm was adopted for the un-pansharpened (30m) and separately using different pansharpened (i.e. Brovey, Simple Mean, ESR, IHS and Gram-Schmidt) datasets (15m) as input. Results indicated improved discriminability of LULC and LST features as well as increased classification accuracy from all pansharpening methods when compared to the 30m resolution data. Average inter-class separability increased from 1.577 using original data to 1.863 after pansharpening. The simple mean method showed the highest classification accuracy (92.13%), outperforming un-pansharpened data (89.77%) and other pansharpening methods (89.45–91.75%). Pansharpened data showed better accuracy in LST spatial characterization than un-pansharpened thermal data. For each LULC class, mean LST was slightly higher before, than after pansharpening, by between 0.1 and 1 °C. The study further established that pansharpening based on simple mean method did not compromise the spatial correlation between LULC and LST. These findings present a great improvement towards accurately mapping LULC and LST in heterogeneous urban landscapes.

Spatiotemporal Fusion Network for Land Surface Temperature Based on a Conditional Variational Autoencoder

High spatiotemporal resolution land surface temperature (LST) data are essential for dynamic monitoring and prediction in climate change research. Due to the limitations of remote sensing instruments, the current platforms have difficulty achieving a compromise between high spatial and temporal resolutions for LST products. In this study, we propose a spatiotemporal fusion network for land surface temperature based on a conditional variational autoencoder (CVAE-LSTFM). First, an improved network is designed based on the CVAE by reconstructing an encoder and a decoder. To generate fine LST images based on dense time series, a variational inference model is formulated to integrate coarse and fine LST image pairs in variational autoencoded latent space. In addition, a new compound loss function for the proposed training method is designed to reduce the effects of noise and outliers. Then, a pretraining mechanism is adopted to optimize the network training process, and the parameters can be transferred to the training network of the CVAE to accelerate network convergence. Finally, a novel weighting strategy that considers spatiotemporal variations in LST (LST consistency weighting) is employed to solve the spatiotemporal heterogeneity problem caused by the rapid changes in LST. The method is quantitatively tested and evaluated in the Heihe River Basin using FY-4A LST and MODIS LST from September 2019. Compared with two traditional models and two deep learning-based models, CVAE-LSTFM yields lower RMSE (average&#x003C;1.26 K) and LPIPS (learning perceptual image patch similarity, average&#x003C;0.13) values and higher SSIM (structural similarity, average&#x003E;0.96). In practice, CVAE-LSTFM can generate high spatiotemporal resolution LST values (hourly LST with a 1 km spatial resolution) with high accuracy, quality, and robustness.

Spatiotemporal Downscaling Method of Land Surface Temperature Based on Daily Change Model of Temperature

Land surface temperature (LST) is one of the most crucial variables of surface energy processes. However, the trade-off between spatial and temporal resolutions of remote sensing data has greatly limited the availability of concurrently high- spatiotemporal resolution LST data for wide applications. Existing downscaling methods are easily affected by null values of LST data and effective time distribution of high-resolution LST data, resulting in large downscaling errors at sometimes. Within this context, this study proposes a novel spatiotemporal fusion model of LST based on diurnal variation information (BDSTFM) to predict LST data with a high temporal resolution and spatio-temporal continuity based on FY-4A and MODIS. Results indicated that the accuracy of the downscaling results was comparable to that of MODIS LST products. The BDSTFM model exhibited the following characteristics: use low-spatial resolution data to establish a DTC model for scale deduction, and retention of the temporal distribution characteristics of LST data; extend the observation time of high-spatial resolution data to improve the accuracy and stability of the model; add an invalid pixel reconstruction step that considers the LST spatiotemporal continuity, and can obtain a realistic and reliable 1-km seamless LST datasets at hourly intervals under clear skies. Compared to the enhanced spatial and temporal adaptive reflectance fusion model (ESTARFM), 4-parameter DTC model and random forest model, the BDSTFM model attained a higher downscaling accuracy.