Downscaling Land Surface Temperature Research Articles

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.

Analysis of spectral indices-based downscaled land surface temperature in a humid subtropical city

ABSTRACT The present study analyses the seasonal influence of error estimated in downscaled land surface temperatures (LSTs) in a humid subtropical city using Landsat 8 data of summer and winter seasons in 2021. Thermal sharpening (TsHARP) algorithm is one of the most frequently used downscaling techniques which is originally based on normalised difference vegetation index (NDVI). This study assesses the capability of the TsHARP technique with a separate combination of four selected spectral indices (modified normalised difference water index, normalised difference bareness index, normalised difference built-up index [NDBI], and NDVI), and by determining the root mean square error (RMSE) and mean error produced by the sharpened LST. Besides, sharpened LST has also been estimated by combining the four spectral indices. It is observed that NDBI provides the most effective output (RMSE is 1.11 [30 m], 1.05 [120 m], 1.02 [240 m], and 0.99 [480 m] in summer, whereas RMSE is 0.61 [30 m], 0.59 [120 m], 0.57 [240 m], and 0.56 [480 m] in winter). NDBI-based sharpened LST generates the best relationship (R = 0.565 in summer and R = 0.537 in winter) with surface features. Fallow land generates the best relationship (R = 0.512 in summer and R = 0.530 in winter) with sharpened LST. The summer season (R = 0.438) generates a better relationship between surface features and sharpened LST than the winter season (R = 0.409).

A physical method for downscaling land surface temperatures using surface energy balance theory

Fine-resolution land surface temperature (LST) derived from thermal infrared remote sensing images is a good indicator of surface water status and plays an essential role in the exchange of energy and water between land and atmosphere. A physical surface energy balance (SEB)-based LST downscaling method (DTsEB) is developed to downscale coarse remotely sensed thermal infrared LST products with fine-resolution visible and near-infrared data. The DTsEB method is advantageous for its ability to mechanically interrelate surface variables contributing to the spatial variation of LST, to quantitatively weigh the contributions of each related variable within a physical framework, and to efficaciously avoid the subjective selection of scaling factors and the establishment of statistical regression relationships. The applicability of the DTsEB method was tested by downscaling 12 scenes of 990 m Moderate Resolution Imaging Spectroradiometer (MODIS) and aggregated Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) LST products to 90 m resolution at six overpass times between 2005 and 2015 over three 9.9 km by 9.9 km cropland (mixed by grass, tree, and built-up land) study areas. Three typical LST downscaling methods, namely the widely applied TsHARP, the later developed least median square regression downscaling (LMS) and the geographically weighted regression (GWR), were introduced for intercomparison. The results showed that the DTsEB method could more effectively reconstruct the subpixel spatial variations in LST within the coarse-resolution pixels and achieve a better downscaling accuracy than the TsHARP, LMS and GWR methods. The DTsEB method yielded, on average, root mean square errors (RMSEs) of 2.01 K and 1.42 K when applied to the MODIS datasets and aggregated ASTER datasets, respectively, which were lower than those obtained with the TsHARP method, with average RMSEs of 2.41 K and 1.71 K, the LMS method, with average RMSEs of 2.35 K and 1.63 K, and the GWR method, with average RMSEs of 2.38 K and 1.64 K, respectively. The contributions of the related surface variables to the subpixel spatial variation in the LST varied both spatially and temporally and were different from each other. In summary, the DTsEB method was demonstrated to outperform the TsHARP, LMS, and GWR methods and could be used as a good alternative for downscaling LST products from coarse to fine resolution with high robustness and accuracy.

Downscaling land surface temperature with high-resolution image: a difference mean distribution (DMD) post-processing method

ABSTRACT Land surface temperature (LST) inversion using remote sensing is a crucial tool to study the land surface thermal environment. To improve the resolution of the inverted LST, a downscaling method is used but it often reduces the accuracy. Here, a difference mean distribution (DMD) method was developed to post-process the downscaled LST. Landsat 8 and Sentinel 2 images were used to test and Field measured LSTs in Tianjin, China were used to verify the method. Results show DMD method reduces MAE and RMSE by 1.36°C and 1.50°C, respectively indicating that it can improve the accuracy of the downscaled LST.

Assessment of the TsHARP method for spatial downscaling of land surface temperature over urban regions

Nowadays, acquiring high spatial and temporal land surface temperature (LST) is one of the significant factors in urban climate studies. This study aims to assess the sharpening thermal imagery (TsHARP) method, which is developed for determining the LST in urban regions. Whereas the original TsHARP method correlates the Normalized Difference Vegetation Index (NDVI) with the LST, the current study seeks to evaluate the performance of the TsHARP method by correlating between the LST and NDVI over urban areas. Additionally, it is intended to investigate the relationship between the impervious surfaces (IS) and LST. At this point, the Advanced Spaceborne Thermal Emission and Reflection Radiometer ASTER image of the year 2003 was used to retrieve the IS as well as the LST of the Kuala Lumpur city, Malaysia. To retrieve the main LST, IS and NDVI, ASTER Level1B data were used. On the other hand, ASTER land surface temperature products (AST08) was utilized to evaluate the downscaled LST (DLST) image. With selection of the ~1/4 of the pixels with lowest coefficient of variation (CV), the LST has more correlation with the IS in comparison with the NDVI, while the correlation between the IS and LST was increased in this study. The IS index illustrates the most optimal correlation for the urban LST (R2 = 0.59). Furthermore, the TsHARPIS-based method depicted a good correlation between the DLST and the ASTER LST product (RMSE = 1.5 °C, R2 = 0.83) compared to the NDVI (RMSE = 4 °C, R2 = 0.73). Consequently, it is found that the application of TsHARPIS-based method to ASTER remote sensing data is a robust/low-cost approach for spatial downscaling of the LST over urban regions.

Downscaling MODIS nighttime land surface temperatures in urban areas using ASTER thermal data through local linear forest

• A new nighttime LST downscaling method using local linear forest (LLF) was proposed. • Feature-selected ASTER nighttime LSTs were identified as important input kernels. • The LLF-based model predicted LST with high linearity, especially for both extremes. • DLST revealed more distinct characteristics of urban climate than MODIS LST. Spatial downscaling effectively produces high spatiotemporal resolution land surface temperature (LST) in urban areas. Although nighttime LST is an essential indicator in urban thermal research, few LST downscaling studies have focused on nighttime in fine resolution. This study proposed a novel approach using local linear forest (LLF) to downscale 1 km Moderate Resolution Imaging Spectroradiometer (MODIS) nighttime LSTs to 250 m spatial resolution in three cities: Rome, Madrid, and Seoul. First, we used Least Absolute Shrinkage and Selection Operator (LASSO) to select a set of past clear-sky ASTER LSTs (ALST) which showed a high spatial correlation with the target MODIS LST. Downscaling models were then developed using input kernels of the selected ALSTs and eight auxiliary variables: normalized difference vegetation index (NDVI), elevation, slope, built-up area percentage, road density, population density, wind speed, and distance from the built-up weighted center of the study area. Three schemes were evaluated: scheme 1 (S1) using only auxiliary variables as input kernels with a random forest (RF) model; scheme 2 (S2) using selected ALSTs and auxiliary variables as input kernels with an RF model; and scheme 3 (S3) using input kernels as in S2 but with the LLF model. Validation was performed using bias-corrected ALSTs for seven reference dates in the three cities. LLF-based S3 showed the highest accuracy with an average correlation coefficient (R) ∼ 0.94 and Root Mean Square Error (RMSE) ∼ 0.64 K while maintaining the dynamic range of the original LST at the finer resolution. The downscaled LST (DLST) based on S3 effectively depicted the nocturnal thermal spatial pattern in greater detail than the other two schemes did. The S3-based DLST also showed a relatively high spatial correlation with the in-situ nighttime air temperature within the cities. When compared to the original 1 km LST, S3-based DLST showed larger surface urban heat island intensity for the urban-type surfaces and a higher temporal correlation with nighttime air temperature.

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 < 0.2 for the bare soil, 0.2 ≤ NDVI ≤ 0.5 for the partial vegetation, and NDVI > 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.

Quantification of landscape metrics effects on downscaled urban land surface temperature accuracy of satellite imagery

Improving the spatial resolution of Land Surface Temperature (LST) obtained from satellite images is of great importance in various applications, especially in urban areas. Aiming to bridge the gap in the research background, this study attempted to quantify the effect of landscape metrics on the accuracy of urban LST downscaling. The research data included a collection of satellite images from Landsat and Terra satellites and auxiliary data of nine European cities, namely Lisbon, Madrid, Bucharest, Vienna, Prague, Paris, Copenhagen, Stockholm, and Helsinki. In the first stage, surface biophysical, topographic, and thematic property maps affecting the spatial distribution of LST were prepared. Secondly, the effect of the endogenous (influence of intra-pixel properties) and exogenous (influence of landscape metrics) variables of each surface property on LST was evaluated in warm and cold seasonal conditions. Finally, the effect of landscape metrics on the accuracy of LST downscaling from 960 m to 30 m was quantified. The results showed that the mean absolute correlation coefficient (r) values between the endogenous (exogenous) effects of surface biophysical, topographic, and thematic properties and LST in the cold season were 0.25 (0.12), 0.19 (0.06), and 0.09 (0.04), respectively. For the warm season, these values were 0.55 (0.24), 0.11 (0.03), and 0.23 (0.09), respectively. The mean r (root-mean-square error (RMSE)) between the actual LST and downscaled LST (DLST) obtained by considering only the endogenous effects in the cold and warm seasons were 0.79 (0.66) and 0.87 (1.38 °C), respectively. Considering the simultaneous effect of endogenous and exogenous variables in LST downscaling, these values reached 0.83 (0.47) and 0.91 (0.99 °C), respectively. The greatest effect of landscape metrics on the accuracy of LST downscaling was associated with built-up lands. Accounting for landscape metrics in the downscaling process significantly increases the accuracy of downscaled LST.

Downscaling land surface temperature: A framework based on geographically and temporally neural network weighted autoregressive model with spatio-temporal fused scaling factors

Downscaling land surface temperatures (LST) from satellite imagery is essential for many fine-scale applications. However, the accuracy of the downscaling is often limited by different environmental and geographical conditions. In this work, a novel LST downscaling framework is proposed to improve the accuracy, especially for heterogeneous areas with varying land covers and complex terrains. The framework focuses on downscaling the MODIS LST from 1 km to 100 m, using the proposed geographically and temporally neural network weighted autoregression (GTNNWAR) model with spatio-temporal fused scaling factors derived from Landsat 8 imagery and digital surface models (DSM). To tackle the issues of the non-stationarity of the scaling factors in heterogenous areas, a region-adaptive parameterization approach is first applied. Then, the GTNNWAR invokes a two-stage deep neural network to estimate the regression coefficients, resulting in the adaption of varying weights for the scaling factors to raise the prediction performance. Moreover, the GTNNWAR is incorporated with a spatial autoregressive model which intakes the neighbor effects so that the overall accuracy can be further improved. Prior to the actual downscaling with the GTNNWAR, a filter-based fusion method is applied to ensure the spatio-temporal consistency of scaling factors is high enough for the neural networks to converge. The results suggest that the proposed framework exhibits high accuracy at the boundaries of different land covers and complex terrains. Compared with several other downscaling algorithms in three case study areas (Beijing and Zhangye in China, Netherlands–Germany in Europe), the proposed framework outperforms with a 28% improved R-squared (R2) and a root mean square error (RMSE) of 1.02 K. In addition, the downscaled LST has R2 over 0.63 for the UAV observations (Guangdong). It is concluded that the proposed framework has high reliability and robustness to provide LST datasets with high spatio-temporal resolutions in a wide range of land types.

A new nonlinear method for downscaling land surface temperature by integrating guided and Gaussian filtering

Land surface temperature (LST), retrieved from thermal infrared (TIR) bands of remote sensing satellites, is an important parameter for various climate and environmental models. TIR bands detect a range of low-energy wavelengths, resulting in a coarser spatial resolution than other multispectral bands, and limiting applicability in heterogeneous urban regions. In this study, a new nonlinear method for LST downscaling, called Three Layers Composition (TLC), was proposed. The TLC integrates large-scale temperature variations, re-constructed detailed characteristics of LSTs, and strong boundary information. The performance of TLC is compared with disaggregation of radiometric surface temperature (DisTrad), thermal imagery sharpening (TsHARP), and random forest (RF) for a complex landscape in Beijing city, which has agriculture, forest, and massive impervious surfaces. The scale effects on the downscaled LSTs (DLST) were analyzed from the aspects of spatial resolution and spatial contexts. The experimental results indicate that the nonlinear algorithms (TLC and RF) perform better than linear methods (DisTrad and TsHARP). Indicated by coefficient of determination (R2), centered root-mean-square error (CRMSE), and correlation coefficient (CC), TLC (R2 = 0.901, CRMSE = 0.319, CC = 0.951) was the most effective and workable technique for predicting LSTs, followed by RF (0.768, 0.502, 0.874), TsHARP (0.544, 0.652, 0.734), and DisTrad (0.518, 0.751, 0.719). Larger experimental regions and larger ratios between initial and target resolution weaken the accuracy of DLST. TLC indicated a stronger ability to resist the influence of such scale effects. Traditional downscaling methods (DisTrad, TsHARP, and RF) are trained with global LST-predictor relationships and predict the DLST point by point, which can result in significantly biased estimates for very high or very low temperatures. Addressing this issue, TLC advantageously preserves the texture similarity between LST and its predictors, and yields more precise DLST, which showed higher consistency with the reference LST. Considering high accuracy and low computation time, TLC may be a safe technique for LST downscaling in other regions and different remote sensing sensors.

Correcting Scaling Effect in Downscaling Surface Temperature at High Resolutions With a Multiple Regional Correction Approach

There exists a scaling error in currently downscaling land surface temperature (DLST) processes and it is necessary to develop substantial methods specially for reducing it. In this letter, a multiple regional correction (MRC) approach with multiple correction terms (mCTs) was proposed to correct scaling errors in DLST processes. The test results indicate that (1) the proposed approach can effectively correct the scaling effects in DLST processes at high resolutions by reducing root mean square error by ~ 55%; (2) all initial resolutions considered for optical data (0.5 m – 4 m) were effective for developing the MRC approach with mCTs in correcting the scaling errors. Overall, this novel approach can significantly improve the accuracy of DLST maps at very high resolutions.

Reducing Scaling Effect on Downscaled Land Surface Temperature Maps in Heterogenous Urban Environments

The literature review indicates that a scaling effect does exist in downscaling land surface temperature (DLST) processes, and no substantial methods were specially developed for addressing it. In this research, the main aim is to develop a new method to reduce the scaling effect on DLST maps at high resolutions. A thermal component-based thermal spectral unmixing (TSU) model was modified and a multiple regression (REG) model was adopted to create DLST maps at high resolutions. A combined variance of red and NIR bands at a very high resolution with a difference image between upscaled LST and DLST was used to develop a new method. With two case data sets, LSTs at coarse resolutions were downscaled by using the modified TSU model and the REG model to create DLST results. The new method with a correction term expression (a linear model created by using a semi-empirical approach) was used to improve the DLST maps in the two case study areas. The experimental results indicate that the new method could reduce the root mean square error and the mean absolute error >30% and >33%, respectively, and thus demonstrate that the proposed method was effective and significant, especially reducing the scaling effect on DLST results at very high resolutions. The novel significance for the new method is directly reducing the scaling effect on DLST maps at high resolutions.

Downscaling Land Surface Temperature Based on Non-Linear Geographically Weighted Regressive Model over Urban Areas

Land surface temperature (LST) is a vital physical parameter in geoscience research and plays a prominent role in surface and atmosphere interaction. Due to technical restrictions, the spatiotemporal resolution of satellite remote sensing LST data is relatively low, which limits the potential applications of these data. An LST downscaling algorithm can effectively alleviate this problem and endow the LST data with more spatial details. Considering the spatial nonstationarity, downscaling algorithms have been gradually developed from least square models to geographical models. The current geographical LST downscaling models only consider the linear relationship between LST and auxiliary parameters, whereas non-linear relationships are neglected. Our study addressed this issue by proposing an LST downscaling algorithm based on a non-linear geographically weighted regressive (NL-GWR) model and selected the optimal combination of parameters to downscale the spatial resolution of a moderate resolution imaging spectroradiometer (MODIS) LST from 1000 m to 100 m. We selected Jinan city in north China and Wuhan city in south China from different seasons as study areas and used Landsat 8 images as reference data to verify the downscaling LST. The results indicated that the NL-GWR model performed well in all the study areas with lower root mean square error (RMSE) and mean absolute error (MAE), rather than the linear model.

Assessment of urban cooling effect based on downscaled land surface temperature: A case study for Fukuoka, Japan

The phenomenon of urban areas becoming warmer than surrounding rural areas, known as urban heat island, is one of the biggest urban issues due to human activities. The study adopted random forests (RF) based downscaling land surface temperature (LST) to assess the surface urban heat island (SUHI) of Fukuoka, Japan. The LST images were first downscaled to 250 m spatial resolution, and then the seasonal and diurnal SUHI was identified. The results implied that the SUHI in summer daytime is the most significant while in winter nighttime is the least significant in the area. Through the bivariate Moran's I test and zonal statistics, the cooling effect of the green-blue spaces were evaluated. The forest has the most substantial urban cooling effect with the average LST change occurs at a distance of about 1 km (maximum 6.32 K). The cooling effect of the sea is significant during daytime, whilst that of the lake/river and the cropland/grassland is not obvious. We concluded the urban population and land use and land cover (LULC) are significantly spatial correlated to UHI, and are important factors of considering the urban heat island (UHI) mitigation strategies.

Landsat-8 and Gaofen-1 image-based inversion method for the downscaled land surface temperature of rare earth mining areas

Land surface temperature is an important indicator for characterizing the habitat conditions and ecological disturbances of ion-adsorbed rare earth mining areas. Because the ions in an ion-adsorbed type rare earth mining area are scattered and the area of a single mining site is small, it is important to obtain practical surface temperature data for monitoring the ecological environment of the area. In this study, Landsat-8 and Gaofen-1(GF-1) satellite images were used as data sources. In view of the characteristics of rare earth mining areas, and the thermal infrared data were used to retrieve the land surface temperature (LST), a land surface temperature downscaling model TDFSU based on remote sensing data fusion and mixed pixel decomposition is proposed. ultimately verifying the quantitative and qualitative accuracy of the downscaled results. The results show that the model can effectively improve the spatial resolution of surface temperature and reflect the spatial differences of surface temperature. This model has high feasibility and applicability in ionic rare earth mining areas.

Development of a 250-m Downscaled Land Surface Temperature Data Set and its Application to Improving Remotely Sensed Evapotranspiration Over Large Landscapes in Northern China

Satellite-derived land surface temperature (LST) is critical for retrieving terrestrial evapotranspiration (ET); however, its availability is limited by low spatial resolution and inclement weather conditions. This study develops a spatiotemporal regression strategy that can downscale 1-km Moderate Resolution Imaging Spectroradiometer (MODIS) LST product to 250-m resolution and simultaneously gap-fill the missing values. The proposed methodology synergistically uses random forest (RF) model and geographically weighted regression, which are, respectively, available for demonstrating the nonlinear correlation between LST and explanatory variables and for calibrating the RF-derived residuals. The study is conducted across a region of ~1.49 million square kilometers in northern China. The coupled model creates a 250-m spatial resolution LST product with the root-mean-square error (RMSE) of 2.32 and 1.87 K when compared with field observations and reference Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) LST, respectively. Meanwhile, it minimizes the constraint of LST availability due to inclement weather conditions with RMSE of 2.69 and 2.31 K relative to field observations and reference images, respectively. The results further reveal that remote-sensing-derived ET using the 250-m downscaled LST data is fairly accurate with the relative errors of 6%–9% as evaluated with flux measurements. The 250-m modeled ET retrievals exhibit a more intense hydrological response to the water use conditions compared with the 1-km remotely sensed ETs and Noah land surface model ETs. This study may benefit land surface hydrology research and water resource management.

Assessing scaling effect in downscaling land surface temperature in a heterogenous urban environment

It is confirmed that there exists a scaling effect in downscaling land surface temperature (DLST) processes. However, a literature review indicates that quantifying and fully understanding scaling effect in DLST processing remains unclear. In this study, a main goal is to quantify, assess and understand scaling effect in downscaling LST product processes at different higher spatial resolutions and spatial extents. A machine leaning model and a traditional multivariate regression model were adopted with corresponding scaling factors extracted from ASTER 15–30 m optical multispectral data and Airborne Imaging Spectrometer for Different Applications (AISA) 2 m hyperspectral visible-near infrared data. MODIS 990 m LST and ASTER 90 m LST products were downscaled to high and very high resolution LST maps. In addition, ETM+ 60 m retrieved LST and Thermal Airborne Broadband Imager (TABI) 2 m retrieved LST and its upscaled LSTs were used to verify higher resolution DLST maps. The experimental results demonstrate that scaling effect in downscaling LST processes is significant, especially downscaling LST to high and very high resolution LST maps. One innovation point derived from findings by assessing the scaling effect in DLST processing is that when DLST processes are at spatial resolutions beyond a range (20–30 m in this study) measured from semivariograms, the processes are safe and their results are reasonable and reliable, and thus their scaling effect may be ignored, but when spatial resolutions and spatial extent lag distance within the range, the DLST processes are not safe, and their results are not reliable and thus the scaling effect has to be considered. Therefore, it is recommended that before conducting a DLST processing project, a range needs to be calculated by plotting semivariograms with high or very high resolution images (better to include three visible and one NIR bands), then DLST processes may be conducted at spatial resolutions lower than and equal to the range.

Downscaling NLDAS-2 daily maximum air temperatures using MODIS land surface temperatures.

We have developed and applied a relatively simple disaggregation scheme that uses spatial patterns of Land Surface Temperature (LST) from MODIS warm-season composites to improve the spatial characterization of daily maximum and minimum air temperatures. This down-scaling model produces qualitatively reasonable 1 km daily maximum and minimum air temperature estimates that reflect urban and coastal features. In a 5-city validation, the model was shown to provide improved daily maximum air temperature estimates in the three coastal cities, compared to 12 km NLDAS-2 (North American Land Data Assimilation System). Down-scaled maximum temperature estimates for the other two (non-coastal) cities were marginally worse than the original NLDAS-2 temperatures. For daily minimum temperatures, the scheme produces spatial fields that qualitatively capture geographic features, but quantitative validation shows the down-scaling model performance to be very similar to the original NLDAS-2 minimum temperatures. Thus, we limit the discussion in this paper to daily maximum temperatures. Overall, errors in the down-scaled maximum air temperatures are comparable to errors in down-scaled LST obtained in previous studies. The advantage of this approach is that it produces estimates of daily maximum air temperatures, which is more relevant than LST in applications such as public health. The resulting 1 km daily maximum air temperatures have great potential utility for applications such as public health, energy demand, and surface energy balance analyses. The method may not perform as well in conditions of strong temperature advection. Application of the model also may be problematic in areas having extreme changes in elevation.

Generation of spatially complete and daily continuous surface soil moisture of high spatial resolution

Surface soil moisture (SSM), as a vital variable for water and heat exchanges between the land surface and the atmosphere, is essential for agricultural production and drought monitoring, and serves as an important boundary condition for atmospheric models. The spatial resolution of soil moisture products from microwave remote sensing is relatively coarse (e.g., ~40 km × 40 km), whereas SSM of higher spatiotemporal resolutions (e.g., 1 km × 1 km and daily continuous) is more useful in water resources management. In this study, first, to improve the spatiotemporal completeness of SSM estimates, we downscaled land surface temperature (LST) output from the China Meteorological Administration Land Data Assimilation System (CLDAS, 0.0625° × 0.0625°) using a data fusion approach and MODIS LST acquired on clear-sky days to generate spatially complete and temporally continuous LST maps across the North China Plain. Second, spatially complete and daily continuous 1 km × 1 km SSM was generated based on random forest models combined with quality LST maps, normalized difference vegetation index (NDVI), surface albedo, precipitation, soil texture, SSM background fields from the European Space Agency Soil Moisture Climate Change Initiative (CCI, 0.25° × 0.25°) and CLDAS land surface model (LSM) SSM output (0.0625° × 0.0625°) to be downscaled, and in situ SSM measurements. Third, the importance of different input variables to the downscaled SSM was quantified. Compared with the original CCI and CLDAS SSM, both the accuracy and spatial resolution of the downscaled SSM were largely improved, in terms of a bias (root mean square error) of −0.001 cm3/cm3 (0.041 cm3/cm3) and a correlation coefficient of 0.72. These results are generally comparable and even better than those in published studies, with our SSM maps featuring spatiotemporal completeness and relatively high spatial resolution. The downscaled SSM maps are valuable for monitoring agricultural drought and optimizing irrigation scheduling, bridging the gaps between microwave-based and LSM-based SSM estimates of coarse spatial resolution and thermal infrared-based LST at a 1 km × 1 km resolution.

Downscaling Land Surface Temperature from MODIS Dataset with Random Forest Approach over Alpine Vegetated Areas

In this study, we evaluated three different downscaling approaches to enhance spatial resolution of thermal imagery over Alpine vegetated areas. Due to the topographical and land-cover complexity and to the sparse distribution of meteorological stations in the region, the remotely-sensed land surface temperature (LST) at regional scale is of major area of interest for environmental applications. Even though the Moderate Resolution Imaging Spectroradiometer (MODIS) LST fills the gap regarding high temporal resolution and length of the time-series, its spatial resolution is not adequate for mountainous areas. Given this limitation, random forest algorithm for downscaling LST to 250 m spatial resolution was evaluated. This study exploits daily MODIS LST with a spatial resolution of 1 km to obtain sub-pixel information at 250 m spatial resolution. The nonlinear relationship between coarse resolution MODIS LST (CR) and fine resolution (FR) explanatory variables was performed by building three different models including: (i) all pixels (BM), (ii) only pixels with more than 90% of vegetation content (EM1) and (iii) only pixels with 75% threshold of homogeneity for vegetated land-cover classes (EM2). We considered normalized difference vegetation index (NDVI) and digital elevation model (DEM) as predictors. The performances of the thermal downscaling methods were evaluated by the Root Mean Square Error (RMSE) and the Mean Absolute Error (MAE) between the downscaled dataset and Landsat LST. Validation indicated that the error values for vegetation fraction (EM1, EM2) were smaller than for basic modelling (BM). BM model determined averaged RMSE of 2.3 K and MAE of 1.8 K. Enhanced methods (EM1 and EM2) gave slightly better results yielding 2.2 K and 1.7 K for RMSE and MAE, respectively. In contrast to the EMs, BM showed a reduction of 22% and 18% of RMSE and MAE respectively with regard to Landsat and the original MODIS LST. Despite some limitations, mainly due to cloud contamination effect and coarse resolution pixel heterogeneity, random forest downscaling exhibits a large potential for producing improved LST maps.