A high spatio-temporal resolution land surface temperature (LST) is necessary for various research fields because LST plays a crucial role in the energy exchange between the atmosphere and the ground surface. The moderate-resolution imaging spectroradiometer (MODIS) LST has been widely used, but it is not available under cloudy conditions. This study proposed a novel approach for reconstructing all-sky 1 km MODIS LST in South Korea during the summer seasons using various data sources, considering the cloud effects on LST. In South Korea, a Local Data Assimilation and Prediction System (LDAPS) with a relatively high spatial resolution of 1.5 km has been operated since 2013. The LDAPS model’s analysis data, binary MODIS cloud cover, and auxiliary data were used as input variables, while MODIS LST and cloudy-sky in situ LST were used together as target variables based on the light gradient boosting machine (LightGBM) approach. As a result of spatial five-fold cross-validation using MODIS LST, the proposed model had a coefficient of determination (R²) of 0.89–0.91 with a root mean square error (RMSE) of 1.11–1.39 °C during the daytime, and an R² of 0.96–0.97 with an RMSE of 0.59–0.60 °C at nighttime. In addition, the reconstructed LST under the cloud was evaluated using leave-one-station-out cross-validation (LOSOCV) using 22 weather stations. From the LOSOCV results under cloudy conditions, the proposed LightGBM model had an R² of 0.55–0.63 with an RMSE of 2.41–3.00 °C during the daytime, and an R² of 0.70–0.74 with an RMSE of 1.31–1.36 °C at nighttime. These results indicated that the reconstructed LST has higher accuracy than the LDAPS model. This study also demonstrated that cloud cover information improved the cloudy-sky LST estimation accuracy by adequately reflecting the heterogeneity of the relationship between LST and input variables under clear and cloudy skies. The reconstructed all-sky LST can be used in a variety of research applications including weather monitoring and forecasting.

Yeonsu Lee, Sun Jegal, Siwoo Lee, Bokyung Son, Jungho Im. Korean J. Remote Sens. 2025;41:843-57. https://doi.org/10.7780/kjrs.2025.41.5.11

The transformation of natural land cover into impermeable materials is a common definition of urbanization. However, it should be noted that vertical growth caused by urban renewal has been disregarded, while the horizontal expansion of urban areas has been widely studied. Understanding dynamic land cover change and its impact focusing on the thermal environment is essential for the sustainable management of urban areas. Therefore, the objective of this study is to propose a simple yet effective urban thermal environment investigation strategy that responds to dynamic land cover transformation. The study area is Suwon, Republic of Korea, where the city has recently been experiencing the most extreme urbanization along with explosive population growth. We designed a three-step approach: 1) extraction of temporal local climate zone (LCZ) to monitor the change in land cover due to urban growth over the period 2004 and 2021, 2) retrieval of high spatial-resolution land surface temperature (LST) using artificial intelligence to observe surface energy flux in heterogeneous urban area in detail, and 3) apply the filtering analysis method, which only uses the pixels classified with a higher confidence level, based on deep learning probabilistic approach to reduce the uncertainty of LCZ classification. To construct the temporal LCZ maps with a 30 m spatial resolution and retrieve the downscaled LSTs (DLSTs) with a corresponding spatial resolution, Landsat series satellites, Shuttle Radar Topography Mission (SRTM), and land cover map produced by Ministry of Environment were used. We obtained the following results: First, the overall accuracy (OA) of the LCZ classification was higher than 90% in both 2004 and 2021. Second, the average coefficient of determination (R2) and root mean square error (RMSE) of the DLSTs were greater than 0.9 and less than 1 °C, respectively. Third, through our suggested urban thermal environment investigation strategy, this study could find that the results of LST changes were clearly varied by building height and compactness changes. Especially, change of low-rise building to mid-rise building increased LST significantly. With these findings, we solidly believe the suggested strategy facilitates advanced urban climate studies.

Urban structures are essential for human habitation, yet they profoundly alter surface energy balances, with significant environmental and health implications. Although urban heat island phenomenon is well recognized, the thermal influence of surrounding urban structures and its interaction with climate remain insufficiently understood. This study presents a global analytical framework that quantifies the thermal impact of surrounding structures, analyzes the combined effects of climate and urban morphology, and projects future thermal trajectories across 967 cities worldwide. By integrating climate variables and local climate zones with machine learning, we assess the thermal impacts of surrounding urban structures. Through a comprehensive weighting of the constituent thermal influences, we introduce the city-scale thermal impact of surrounding urban structures (TBE), revealing that climate and morphology jointly contribute to global urban heat. Across climate zones, cities exhibiting high daytime TBE are characterized by low- and mid-rise built form, whereas sparsely built types prevail in cities with low TBE. These patterns persist at night. Future projections show that the spatial patterns of TBE will be distinct from current conditions, with combined climate-structure effects will dominate urban thermal environment in almost half of global cities. These projections also reveal significant regional disparities between the Global South and the Global North. Our findings highlight the role of combined effects in shaping present and future urban heat, informing the need for adaptation strategies tailored to individual cities.

Urban areas alter the surface energy balance, typically resulting in higher temperatures than surrounding rural regions, a phenomenon known as the urban heat island (UHI) effect. The UHI effect exacerbates the risk of heat-related mortality and magnifies the challenges posed by global climate change. While the magnitude of UHIs is influenced by factors such as city size, population, wind, and climate zone, the role of building configurations remains understudied. In this study, we investigate the global impact of building composition on UHIs, focusing on uneven heat effects across climate zones and temporal variations (day/night). We develop a machine learning-based model to estimate UHI intensity, incorporating building composition data from local climate zone (LCZ) classifications, background climate. Our findings reveal significant inequalities in the thermal impact of building composition across climate zones and time. During the daytime, building composition plays a notable role in tropical climates, where compact, low-rise structures exacerbate the local thermal environment through strong heat dissipation. Conversely, at night, the influence shifts to arid climates, with LCZ2 (compact mid-rise) types consistently demonstrating the largest thermal dissipation across all climate zones. The direct implication of these uneven heat impacts is that urbanization and mitigation strategies underscore the critical need for appropriately implemented based on a thorough understanding of the thermal effects of building structures across different climate zones, thereby advancing urban sustainability and resilience.