Accurate Land Surface Temperature Research Articles

Revealing the urban heat Island: Investigating spatiotemporal surface temperature dynamics, modeling, and interactions with controllable and non-controllable factors

The unchecked growth of urban populations has spurred rapid and unsustainable urban expansion, exacerbating environmental degradation both locally and globally. This research offers a comprehensive analysis of the Urban Heat Island (UHI) phenomenon in Bangalore and Hyderabad, India, focusing on its spatial and temporal distribution and its correlation with air pollutants. Conducted over the course of summer and winter seasons from 2001 to 2021, this study advances our understanding of UHI dynamics in rapidly urbanizing regions, shedding light on the complex interplay between urbanization, climate, and air quality. The primary objective of this study is to assess and compare the predictive performance of baseline Random Forest (RF) and Extreme Gradient Boosting (XGBoost) models for forecasting Land Surface Temperature (LST) in Hyderabad, India, while examining the impact of integrating spatial information into these models. Incorporating an extensive array of input variables, ranging from spectral indices to terrain features, land use/land cover (LULC) data, atmospheric parameters, and spatial data, the analysis reveals significant relationships between LST and various factors. Notably, Aerosol Optical Depth (AOD) and Normalized Difference Built-up Index (NDBI) exhibit strong positive correlations with LST, while Enhanced Vegetation Index (EVI) demonstrates a pronounced negative correlation. Similarly, Modified Normalized Difference Water Index (MNDWI) displays the strongest positive correlation with LST (r = 0.49), whereas EVI exhibits the most notable negative correlation (r = −0.67). Additionally, Modified Bare Soil Index (MBI) demonstrates a noteworthy negative correlation (r = −0.48) with LST for Bangalore. Atmospheric components such as CO, NO2, and O3 also show significant positive correlations with LST, with NO2 displaying the highest correlation during winter in urban areas. XGBoost emerges as the superior model, achieving high R2 values of 0.84 for Bangalore and 0.88 for Hyderabad. Integration of spatial information consistently enhances the performance of both RF and XGBoost models, resulting in reduced Mean Squared Error (MSE) and Root Mean Squared Error (RMSE) values. However, predictions over water bodies tend to be underestimated. The study underscores the importance of incorporating spatial data for accurate LST prediction, with implications for environmental monitoring and management. Despite the advancements made, further research is needed to elucidate the specific mechanisms through which spatial information improves model performance, suggesting avenues for future investigation and refinement of modeling methodologies. At the intersection of rapid urban growth and environmental conservation, these findings are poised to inform policymakers, urban planners, and researchers, guiding them towards innovative and sustainable urban development strategies aimed at mitigating the adverse impacts of rapid urban expansion while promoting ecosystem well-being. The study contributes to spatial modeling strategies and refines LST prediction methodologies, offering valuable insights for addressing the challenges posed by urbanization and climate change in rapidly urbanizing regions.

Solar zenith angle-based calibration of Himawari-8 land surface temperature for correcting diurnal retrieval error characteristics

The geostationary Himawari-8 satellite offers a unique opportunity to monitor sub-daily thermal dynamics over Asia and Oceania, and several operational land surface temperature (LST) retrieval algorithms have been developed for this purpose. However, studies have reported inconsistency between LST data obtained from geostationary and polar-orbiting platforms, particularly for daytime LST, which can arise from variations in viewing geometries and inherent differences in sensor types and LST algorithms. Despite this, previous research has primarily focused on analysing the directionality of LST without thoroughly exploring systematic differences between platforms. Hence, we presented a Solar Zenith Angle-based Calibration (SZAC) method to harmonise the daytime component of a split-window retrieved Himawari-8 LST (referred to here as the baseline) with the MODerate-resolution Imaging Spectroradiometer (MODIS) LST. SZAC describes the spatial heterogeneity and magnitude of diurnal LST discrepancies from different platforms, which is anticipated to complement typical directionality analyses. We evaluated the harmonised LST data, referred to as the Australian National University LST with SZAC (ANUSZAC), against MODIS LST and the Visible Infrared Imaging Radiometer Suite (VIIRS) LST, as well as in-situ LST from the OzFlux network. Two peer Himawari-8 LST products from Chiba University and the Copernicus Global Land Service were also collected for comparison. The median daytime bias of ANUSZAC LST against Terra-MODIS LST, Aqua-MODIS LST and VIIRS LST was 1.52 K, 0.98 K and −0.63 K, respectively, which demonstrated improved performance compared to baseline (5.37 K, 4.85 K and 3.02 K, respectively) and Chiba LST (3.71 K, 2.90 K and 1.07 K, respectively). All four Himawari-8 LST products showed comparable performance of unbiased root mean squared error (ubRMSE), ranging from 2.47 to 3.07 K, compared to LST from polar-orbiting platforms. In the evaluation against in-situ LST, the mean values of bias (ubRMSE) of baseline, Chiba, Copernicus and ANUSZAC LST during daytime were 4.23 K (3.74 K), 2.16 K (3.62 K), 1.73 K (3.31 K) and 1.41 K (3.24 K), respectively, based on 171,289 hourly samples from 20 OzFlux sites across Australia between 01/Jan/2016 and 31/Dec/2020. In summary, the SZAC method offers a promising approach to enhance the reliability of geostationary LST retrievals by incorporating the spatiotemporal characteristics observed by accurate polar-orbiting LST data. SZAC can also reduce LST angular effects due to its capability of quantifying spatial inhomogeneity of surface heat dynamics. Furthermore, it is possible to implement SZAC using LST data acquired by geostationary satellites in other regions, e.g., Europe, Africa and Americas. This could improve our understanding of the error characteristics of coincident imageries from geostationary and polar-orbiting platforms, allowing for targeted refinements and global harmonisations to further enhance applicability.

Validation of Remotely Sensed Land Surface Temperature at Lake Baikal’s Surroundings Using In Situ Observations

The accuracy of Land Surface Temperature (LST) products retrieved from satellite data in mountainous-coastal areas is not well understood. This study presents an analysis of the spatial and temporal variability of the differences between the LST and in situ observed air and surface temperatures (ISTs) for the southeastern slope of Lake Baikal’s surroundings. The IST was measured at 12 ground observation sites located on the southeastern macro-slope of the Primorskiy Ridge (Baikal, Russia) within an elevation range of 460–1656 m above sea level from 2009 to 2021. LST was estimated using 617 Landsat (7 and 8) images from 2009–2021, taking into account brightness temperature, surface emissivity and vegetation cover fraction. The comparison of the LST from satellite data and the IST from ground observation showed relatively high differences, which varied depending on the season and site type. A neural network was suggested and calibrated to improve the LST data. The corrected remote-sensed temperature was found to reproduce the IST very well, with mean differences of about 0.03 °C and linear correlation coefficients of 0.98 and 0.95 for the air and surface IST.

Correcting land surface temperature from thermal imager by considering heterogeneous emissivity

It is fundamental to obtain accurate land surface temperature (LST) to study surface energy process. Infrared thermal imagers are commonly used for deriving LST on the basis of radiance measurements. However, when deriving LST from brightness temperature of a blackbody in thermal imagers, thermal imagers only allow setting a fixed land surface emissivity (LSE). This causes uncertainty in retrieving thermal infrared (TIR) temperature from heterogeneous surfaces with varied LSE, such as those covered in vegetation. Corrections can be made using the Normalized Vegetation Index (NDVI). However, commercial thermal imagers provide only red (R), green (G), and blue (B) bands without a near infrared band so NDVI cannot be calculated on the same instrument and is commonly not available. We propose an alternative method to estimate LSE using RGB-based vegetation index. Thereafter the estimated LSE was used to correct the TIR temperature derived from the fixed LSE. An experiment was conducted to validate the proposed correcting method. The results show that 1) the corrected LST values were closer to the ground truth, with a mean absolute error (MAE) of 0.41 ± 0.34 °C, whereas the MAE was 0.75 ± 0.56 °C for the uncorrected LST; 2) the more heterogeneous the surface, the greater the difference between the corrected and uncorrected LST values, indicating the necessity of LST correction when applying thermal imagers over heterogeneous surface.

Accuracy assessment of land surface temperature retrievals from remote sensing imagery: pixel-based, single and multi-channel methods

Land Surface Temperature (LST) measurements using remote sensing data are conducted using various methods, including the Single Channel Method (SCM) and the Multi-Channel Method (MCM). Emissivity, which significantly affects LST, is determined using methods such as the Correction-Based Emissivity Method (CBEM) and the Non-Linear Based Emissivity Method (NBEM). Given these factors, validating LST measurements is crucial for assessing the accuracy of data processing results. This study aims to compare LST data retrieved using SCM and MCM with various emissivity calculation methods and assess their accuracy compared to single measurements with an infrared thermometer, measurements using a grid plot, and measurements using a thermal camera with the assistance of an Unmanned Aerial Vehicle (UAV). The results show that SCMJM&S and SCMAC did not produce optimal accuracy (>2.5K) when using the emissivity estimates from CBEM and NBEM. In contrast, MCM that adopted MCMSko, MCMQin, and MCMMao generally produced high-accuracy LST with an average difference of <1 K from actual temperatures, specifically in shrub pixels. MCM with the NDVI threshold method (NDVITHM) showed excellent results in distinguishing between vegetation and non-vegetation objects. In conclusion, MCM provides more accurate LST than SCM. Nevertheless, selecting the appropriate method should consider heterogeneity, homogeneity, and regional physiography.

Improving HJ-1B/IRS LST Retrieval of the Generalized Single-Channel Algorithm with Refined ERA5 Atmospheric Profile Database

Land surface temperature (LST) is a fundamental variable of environmental monitoring and surface equilibrium. Although the HJ-1B infrared scanner (IRS) has accumulated many observations, further application of HJ-1B/IRS is limited by the lack of LST products. This study refined the ERA5 atmospheric profile database, instead of the widely used traditional TIGR atmospheric profile database, and simulated the coefficients of the generalized single-channel (GSCs) algorithms to improve LST retrieval. GSCs can be divided into the GSCw and GSCwT algorithms, depending on whether the input is atmospheric water vapor content (WVC) or in situ near-surface air temperature and WVC. Land surface emissivity (LSE) was obtained from the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) Global Emissivity Dataset (GED) and vegetation/snow cover products. Then, the retrieved LSTs were evaluated using the LSTs from the RTE algorithm, TIGRw/TIGRwT profiles, and in situ near-surface air temperature from the HiWATER experiment in China from 2012 to 2014. The bias (root mean square error (RMSE)) values are displayed as ERA5wT &lt; RTE &lt; ERA5w &lt; TIGRwT &lt; TIGRw. The accuracy of ERA5wT, with a bias (RMSE) of 0.02 K (2.30 K), is higher than that of RTE, with a bias (RMSE) of 0.74 K (2.47 K). The accuracy of RTE is preferable to that of ERA5w, with a bias (RMSE) of 0.89 K (2.48 K), followed by TIGRwT, with a bias (RMSE) of −1.18 K (2.50 K), and then, TIGRw, with a bias (RMSE) of 1.60 K (2.77 K). In summary, the accuracy of LST obtained by GSC from the refined ERA5 atmospheric profiles is higher than that obtained from the TIGR profiles. The accuracy of LST obtained by GSCwT is greater than that obtained by GSCw. The accuracy of LST obtained using in situ near-surface air temperature is higher than that obtained using ERA5 air temperature. The accuracy of LSEASTER is slightly better than that of LSEMOD21. The aforementioned conclusions can provide scientific support to generate HJ-1B/IRS LST products.

Assessment of Land Surface Temperature from the Indian Cities of Ranchi and Dhanbad during COVID-19 Lockdown: Implications on the Urban Climatology

An apparent increase in average global temperature is evident globally, and India is no exception. With the recent decade (2011–2020) arguably being the warmest, significant challenges due to rapid climate change have gained attention. However, notable spatial-temporal changes, especially with regard to land surface temperature (LST), were observed during the COVID-19 pandemic lockdown period, when a comparatively cooler climate was experienced in many urban centers. Assessment of LST, crucial in many heat-balance, land use, and climate change models research studies, depicts the near-surface hotness of the Earth’s temperature at a given location. Thus, this study utilizes satellite remote-sensing data to investigate the spatial-temporal variations of LST pre and post-lockdown imposed during the COVID-19 outbreak. Unlike many existing research studies on the metropolitans of India, the study considers developing Indian cities, Ranchi and Dhanbad, as its study area. Accurate LST computation was performed using existing LANDSAT-8 OLI/TIRS images and judged using other parameters (NDVI, LSE) obtained directly from the thermal infrared bands. The LST assessment successfully estimated temperature variations in Ranchi and Dhanbad, depicting a significant drop in temperature coinciding with the lockdown period and subsequent increase in urban temperature post-pandemic, indicating a meaningful relationship between human activities and urban surface temperature.

Retrieval and validation of the dawn-dusk land surface temperature from FY-3E MERSI-LL

ABSTRACT Land surface temperature (LST) is an important physical parameter that reflects changes in surface processes. Remote sensing makes it possible to obtain global or regional LSTs with high spatial and temporal resolution. Feng-Yun 3E (FY-3E) is the world’s first civilian dawn-dusk orbiting meteorological satellite on board the Medium Resolution Spectral Imager – Low-Light (MERSI-LL) sensor capable of acquiring LST at 250 m spatial resolution at dawn and dusk. By complementing with existing polar-orbiting satellites, FY-3E can fill the satellite observation gap within the 6-hour assimilation window and further enhance the observation capability of LST. To retrieve accurate LST from MERSI-LL, the generalized split-window (GSW) method was developed. The GSW coefficients were derived from the simulation dataset produced by Moderate Resolution Transmittance Code 5.2 with an atmospheric profile database. As a key input to the GSW algorithm, the land surface emissivity was dynamically estimated by the Advanced Spaceborne Thermal Emission and Reflection Radiometer Global Emissivity Dataset in combination with the fractional vegetation cover. The water vapour content is calculated in real time by the split-window covariance/variance ratio method. The comparison between retrieved LSTs and in situ LSTs shows that the GSW algorithm can effectively retrieve the LSTs of vegetation cover surfaces. The root mean square error ranged from 0.79 to 1.79 K at dawn and from 1.32 to 1.92 K at dusk.

Land Surface Temperature Estimation from Landsat-9 Thermal Infrared Data Using Ensemble Learning Method Considering the Physical Radiance Transfer Process

Accurately estimating land surface temperature (LST) is a critical concern in thermal infrared (TIR) remote sensing. According to the thermal radiance transfer equation, the observed data in each channel are coupled with both emissivity and atmospheric parameters in addition to the LST. To solve this ill-posed problem, classical algorithms often require the input of external parameters such as land surface emissivity and atmospheric profiles, which are often difficult to obtain accurately and timely, and this may introduce additional errors and limit the applicability of the LST retrieval algorithms. To reduce the dependence on external parameters, this paper proposes a new algorithm to directly estimate the LST from the top-of-atmosphere brightness temperature in Landsat-9 two-channel TIR data (channels 10 and 11) without external parameters. The proposed algorithm takes full advantage of the adeptness of the ensemble learning method to solve nonlinear problems. It considers the physical radiance transfer process and adds the leaving-ground bright temperature and atmospheric water vapor index to the input feature set. The experimental results show that the new algorithm achieves accurate LST estimation results compared with the ground-measured LST and is consistent with the Landsat-9 LST product. In subsequent work, further studies will be undertaken on developing end-to-end deep learning models, mining more in-depth features between TIR channels, and reducing the effect of spatial heterogeneity on accuracy validation.

A General Paradigm for Retrieving Soil Moisture and Surface Temperature from Passive Microwave Remote Sensing Data Based on Artificial Intelligence

Soil moisture (SM) and land surface temperature (LST) are entangled, and the retrieval of one of them requires a priori specification of the other one. Due to insufficient observational information, retrieval of LST and SM from passive microwave remote sensing data is often ill-posed, and the retrieval accuracy needs to be improved. In this study, a novel fully-coupled paradigm is developed to robustly retrieve SM and LST from passive microwave data, which integrates deep learning, physical methods, and statistical methods. The key condition of the general paradigm proposed by us is that the output parameters of deep learning can be uniquely determined by the input parameters theoretically through a certain mathematical equation. Firstly, the physical method is deduced based on the energy radiation balance equation. The nine unknowns require the brightness temperatures of nine channels to construct nine equations, and the solutions of the physical method equations are obtained by model simulation. Based on the derivation of the physical method, the solution of the statistical method is constructed using multi-source data. Secondly, the solutions of physical and statistical methods constitute the training and test data of deep learning, which is used to obtain the solution curve of physical and statistical methods. The retrieval accuracy of LST and SM is greatly improved by smartly utilizing the mutual prior knowledge of SM and LST and cross iterative optimization calculations. Finally, validation indicates that the mean absolute error of the retrieved SM and LST data are 0.027 m3/m3 and 1.38 K, respectively, at an incidence angle of 0–65°. A model-data-knowledge-driven and deep learning method can overcome the shortcomings of traditional methods and provide a paradigm for retrieval of other geophysical variables. The proposed paradigm not only has physical meaning, but also makes deep learning physically interpretable, which is a milestone in the retrieval of geophysical remote sensing parameters based on artificial intelligence technology.

An extended SW-TES algorithm for land surface temperature and emissivity retrieval from ECOSTRESS thermal infrared data over urban areas

Land surface temperature (LST) over urban areas is an essential parameter in the monitoring of urban thermal environment. Urbanization has resulted in a more complex geometric structure of urban surface than natural surface. However, in the conventional LST retrieval algorithms, the impact of urban geometric structure is usually neglected. In this study, an extended algorithm that combines the split-window algorithm (SW) and the temperature and emissivity separation algorithm (TES) was proposed to retrieve LST and land surface emissivity (LSE) over urban areas simultaneously from the ECOsystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS) thermal infrared (TIR) data. In the extended SW-TES algorithm with considering urban geometry effect, the brightness temperature at ground level was obtained using the SW algorithm, and the LST and LSE were retrieved simultaneously by using the TES algorithm. The results indicate that LST (LSE) retrieved by the extended SW-TES algorithm with urban geometry effect correction has generally lower (higher) values than that without urban geometry effect correction. The LST differences mostly vary from 0.3 K to 1.2 K, whereas the LSE differences are mainly concentrated between −0.01 and − 0.06. Furthermore, the root mean squared error (RMSE) between LST/LSE retrieved by the extended SW-TES algorithm with urban geometry effect correction and LST/LSE obtained from the ECOSTRESS level-2 LST/LSE product decrease with the increase of sky view factor (SVF). The largest RMSE is approximately 1.6 K for LST and 0.03 for LSE in the SVF of 0.1, which represents an urban pixel surrounded by high buildings within its adjacent pixels. There are no evident differences between LST/LSE retrieved by the extended SW-TES algorithm with urban geometry effect correction using five and three ECOSTRESS TIR bands. The LST differences are mainly concentrated between −0.2 K and 0.2 K, whereas the LSE differences mostly range from 0.001 to 0.004. The findings of this study indicate that it is necessary to consider urban geometry effect for accurate LST and LSE retrieval over urban areas.

Land Surface Temperature Retrieval From Landsat 9 TIRS-2 Data Using Radiance-Based Split-Window Algorithm

The thermal infrared sensor-2 (TIRS-2) carried on Landsat 9 is the newest thermal infrared (TIR) sensor for the Landsat project and provides two adjacent TIR bands, which greatly benefits the land surface temperature (LST) retrieval at high spatial resolution. In this article, a radiance based split window (RBSW) algorithm for retrieving LST from Landsat 9 TIRS-2 data was proposed. In addition, the split-window covariance-variance ratio (SWCVR) algorithm was improved and applied to Landsat 9 TIRS-2 data for estimating atmospheric water vapor (AWV) that is required for accurate LST retrieval. The performance of the proposed method was assessed using the simulation data and satellite observations. Results reveal that the retrieved LST using the RBSW algorithm has a bias of 0.06 K and root-mean-square error (RMSE) of 0.51 K based on validation with the simulation data. The sensitivity analysis exhibited a LST error of <1.75 K using the RBSW algorithm when the uncertainties in input parameters (i.e., AWV, emissivity, and at-sensor radiance) were considered. There is a marginal discrepancy between LST retrievals using the estimated AWV and moderate resolution imaging spectroradiometer AWV, with a difference in bias of −0.14 K and RMSE of 0.22 K, which indicates that the improved SWCVR method can provide an optional means to obtain AWV inputted in LST retrieval. With regard to the validation using the in situ measurements, the retrieved LST from Landsat 9 TIRS-2 data exhibits a bias of 0.44 K and RMSE of 1.98 K, respectively, showing a higher accuracy than the USGS Landsat LST product with bias of 1.21 and RMSE of 2.56 K. In conclusion, the proposed algorithm combining RBSW algorithm and improved SWCVR algorithm for LST retrieval from Landsat 9 has a good accuracy without dependence on external atmospheric data, and it is expected to be a reliable method for LST generation from Landsat 9 TIRS-2 data.

Examining the Relationship between Land Use/Land Cover (LULC) and Land Surface Temperature (LST) Using Explainable Artificial Intelligence (XAI) Models: A Case Study of Seoul, South Korea

Understanding the relationship between land use/land cover (LULC) and land surface temperature (LST) has long been an area of interest in urban and environmental study fields. To examine this, existing studies have utilized both white-box and black-box approaches, including regression, decision tree, and artificial intelligence models. To overcome the limitations of previous models, this study adopted the explainable artificial intelligence (XAI) approach in examining the relationships between LULC and LST. By integrating the XGBoost and SHAP model, we developed the LST prediction model in Seoul and estimated the LST reduction effects after specific LULC changes. Results showed that the prediction accuracy of LST was maximized when landscape, topographic, and LULC features within a 150 m buffer radius were adopted as independent variables. Specifically, the existence of surrounding built-up and vegetation areas were found to be the most influencing factors in explaining LST. In this study, after the LULC changes from expressway to green areas, approximately 1.5 °C of decreasing LST was predicted. The findings of our study can be utilized for assessing and monitoring the thermal environmental impact of urban planning and projects. Also, this study can contribute to determining the priorities of different policy measures for improving the thermal environment.

Identifying and characterizing frequency and maximum durations of surface urban heat and cool island across global cities

Surface urban heat island (SUHI) has been widely reported from a local to global scale. However, variations and controls of temporal indicators for SUHI and SUCI (surface urban cool island) remain unclear. This paper firstly reconstructed the seamless daily LST (land surface temperature) based on ATC-SKT (annual temperature cycle-skin temperature) and comprehensively validated for SUHI applications across 1112 global cities. Based on the seamless daily LST, this paper further characterized the spatiotemporal variations of the frequency (SUHIF and SUCIF) and maximum duration (SUHID and SUCID) and investigated the impacts from related factors, inconsideration of the different characteristics of SUHI and SUCI. There are five major findings. (1) The seamless daily LST reconstructed based on ATC-SKT is validated through pixel-based temperature and city-based SUHII accuracy assessments. (2) The selection of the frequency threshold is based on robustness for LST accuracy, approximation to SUHII global average, and mitigation of frequency saturation. (3) The average daytime SUHIF is 214 days/year, with 44 % of cities exhibiting SUHI occurrences for almost every day in summer. The nighttime SUHIF is 175 days/year, with increasing latitudinal variations from equatorial to polar regions. The daytime SUCIF is 41 days/year, with the greatest average frequency of 172 days/year exhibited in arid regions. (4) The average SUHID is 147 days at daytime and 58 days at nighttime, with relatively opposite geographical distributions between day and night. (5) Greater vegetation difference at daytime and greater albedo difference at nighttime result in more occurrences and longer consecutive durations of SUHI, with opposite effects on the temporal indicators of SUCI. Furthermore, the improvements in daily SUHII and impacts from the maximum duration were discussed. This paper aims to identify and highlight the period with significant SUHI and SUCI effects across global cities for further mitigation.

An Analyze of Urban Temperature Using Energy Balance Algorithm for Land (SEBAL) in Yogyakarta City

This study examines the Land Surface Temperature (LST) using the Surface Energy Balance Algorithm for Land (SEBAL) model in Yogyakarta. SEBAL is relied upon for its accurate LST predictions because it takes into account the influence of vegetation and soil. This study identified LST in various land cover/land use (LULC) types extracted from Landsat 8 remote sensing images recorded in April 2019 (wet day) and June 2019 (dry day). The LULC classification results in the study area show that built-up land is the dominant land use, with 93.30% of the total area, and the rest is non-developed land (vegetation, open land, and water body). The average LST value on a wet day is 26.79 °C, while on a dry day, it is 30°C. The highest temperature occurs on the dry day, 35.17 °C, and the lowest on the wet day, which is 13.63°C. The correlation between LST and LULC shows the same pattern on the two different days, in which the value of vegetation temperature is lower than that of open and developed land. This research proves that vegetation influences a decrease in land surface temperature. Judging from the dominant land use being the built-up area in Yogyakarta, urban planners need to consider increasing green open spaces.

Estimation of Urban-Rural Land Surface Temperature Difference at Different Elevations in the Qinling-Daba Mountains Using MODIS and the Random Forest Model.

Land surface temperature (LST) variations are very complex in mountainous areas owing to highly heterogeneous terrain and varied environment, which complicates the surface urban heat island (SUHI) in mountain cities. Previous studies on the urban heat island (UHI) effect mostly focus on the flat terrain areas; there are few studies on the UHI effect in mountainous areas, especially on the influence of elevation on the SUHI effect. To determine the SUHI in the Qinling–Daba mountains (China), MODIS LST data were first preprocessed and converted to the same elevations (1500 m, 2000 m, 2500 m, 3000 m, and 3500 m) using a digital elevation model and the random forest method. Then, the average LSTs in urban land, rural land, and cultivated land were calculated separately based on the ranges of the invariable urban, rural, and cultivated areas during 2010–2018, and the urban, rural, and cultivated land LST difference were estimated for the same elevations. Results showed that the accuracy of LST estimated using the random forest method is very high (R2 ≥ 0.9) at elevations of 1500 m, 2000 m, 2500 m, 3000 m and 3500 m. The difference in urban, rural, and cultivated lands’ LST has a trend of decrease with increasing elevation, meaning that the SUHI weakens at higher elevations. The average LST of urban areas is 0.52–0.59 °C (0.42–0.57 °C) higher than that of rural and cultivated areas at an elevation of 1500 m (2000 m). The average LST of urban areas is 0.10–1.25 °C lower than that of rural and cultivated areas at elevations of 2500 m, 3000 m, and 3500 m, indicating absence of the SUHI at those elevations.

Prediction of MODIS land surface temperature using new hybrid models based on spatial interpolation techniques and deep learning models.

Land surface temperature (LST) prediction is of great importance for climate change, ecology, environmental and industrial studies. These studies require accurate LST map predictions considering both spatial and temporal dynamics. In this study, multilayer perceptron (MLP), long short-term memory (LSTM) and an integrated machine learning model, namely Convolutional LSTM (ConvLSTM), were utilized for one step ahead LST prediction. Data were gathered from 1-day (MYD11A1) and 8-day composite (MYD11A2) Moderate Resolution Imaging Spectroradiometer (MODIS) sensors, which have 1-km × 1-km spatial resolution. Considering the inability of MODIS sensors to provide LST data under cloudy conditions, Inverse DISTANCE WEIGHTING (IDW), natural neighbor (NN), and cubic spline (C) methods were used to overcome the missing pixel problem. The proposed methods were tested over the Northern part of Adana province, Turkey, and the performances of the models were quantitatively evaluated through performance measures, namely, Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). The selected datasets range from 01 January 2017 to 01 November 2020 and from 01 January 2015 to 01 November 2020 for daily LST and 8-day composite LST, respectively. While 60% of the datasets were used as training set, the remaining 40% were used as validation (20%) and test (20%) sets. RMSE maps were generated to evaluate the pixelwise performance of the proposed method. On the other hand, the best average RMSE and MAE for the daily test set were obtained from the combination of ConvLSTM and NN (NN-ConvLSTM) as 3.62°C and 2.85°C, respectively, while they were acquired 3.57°C and 2.69oC from the combination of MLP and NN (NN-MLP) for the 8-day composite LST test set. The results revealed that the proposed hybrid models could be used for one step ahead spatiotemporal prediction of LST data.

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.

Assessment of Land Surface Temperature Estimates from Landsat 8-TIRS in A High-Contrast Semiarid Agroecosystem. Algorithms Intercomparison

Monitoring Land Surface Temperature (LST) from Landsat satellites has been shown to be effective in the estimation of crop water needs and modeling water use efficiency. Accurate LST estimation becomes critical in semiarid areas under water scarcity scenarios. This work shows the assessment of some well-known Single-Channel (SC) and Split-Window (SW) algorithms, adapted to Landsat 8/TIRS, under the conditions of a high-contrast semiarid agroecosystem. The recently released Landsat 8 Level-2 LST product (L8_ST) has also been included in the performance analysis. Ground measurements of surface temperature were taken for the evaluation during the summers of 2018–2019 in the cropland area of the Barrax test site, Spain. A dataset of 44 ground samples and 11 different L8/TIRS dates/scenes was gathered, covering a variety of crop fields and surface conditions. In addition, a simplified Single Band Atmospheric Correction (L-SBAC) was introduced based on a linearization of the atmospheric correction parameters with the water vapor content (w) and a redefinition of the emissivity threshold for the emissivity correction in the study site. The best results show differences within ±4.0 K for temperatures ranging 300–325 K. Statistics for the L-SBAC result in a RMSE of ±1.8 K with negligible systematic deviation. Similar results were obtained for the other SC and SW algorithms tested, whereas an overestimation of 1.0 K was observed for the L8_ST product because of inappropriate assignment of emissivity values. These results show the potential of the proposed linearization approach and set the uncertainty for LST estimates in high-contrast semiarid agroecosystems.

Detecting Geothermal Resources in a Plateau Area: Constraints From Land Surface Temperature Characteristics Using Landsat 8 Data

Geothermal energy is a kind of clean energy, which attracts more attention. The detection of geothermal resources is inseparable from regional geothermal prospects. Land surface temperature (LST) is an indispensable parameter for geothermal exploration, but the retrieval accuracy of LST for complex and remote areas is currently a major challenge. In this article, based on Landsat 8 remote sensing data, the characteristics of surface temperature retrieval methods are systematically reviewed, and the differences among these three algorithms are researched by using them to detect the surface temperature in the study area, which is Kangding County, Sichuan Province, China. Then the experimental results of the three algorithms are verified by using long-time (more than 1 year) measured data from the two monitoring sites, and the monitoring sites are situated in Zhonggu and Lao Yulin area of Kangding County. The results show that the radiative transfer equation (RTE) has the highest accuracy, and the mean error is 0.372°C; mono-window algorithm (MW) has a mean error of −0.606°C; and the split-window (SW) algorithm has the lowest accuracy, with a mean error of −2.07°C. The experimental results were used to select an algorithm with relatively high accuracy and low sensitivity. At the same time, a time series was used to perform temperature retrieval for this study area from November 2016 to December 2017 to evaluate the applicability of the method. The result shows that the RTE has the highest accuracy in mid-winter and a relatively low accuracy in summer in Kangding County. The purpose of this article is to establish a suitable method for high-precision surface temperature retrieval in plateau areas and to provide technical support for exploring geothermal resources or evaluating geothermal potential in these areas.