ABSTRACT Accurately identifying extreme land surface temperature (LST) zones—heat cores (HCs, areas with persistent high temperatures) and cold cores (CCs, areas with persistent low temperatures)—is crucial for urban climate resilience. Traditional approaches fail to capture subtle LST variations and hierarchical structures. This study introduces the Local Contour Tree Analysis (LCTA), a graph theory–based method that captures topological hierarchies and dynamic interactions, surpassing conventional fixed-threshold techniques. Applied to western Shenzhen, China, it revealed HCs' northward interannual migration from southern clustering through dispersion to northern clustering, while CCs remained stable as primary cooling sources. Seasonally, HCs exhibited stronger clustering in summer and greater dispersion in winter; CCs maintained structural layering year-round with reduced heterogeneity in winter. Structurally, HCs were simple and compact, while CCs were complex, multi-layered, and interconnected. Urban thermal deterioration concentrated in stable HCs due to impervious surface expansion and vegetation loss, while improvements occurred around CCs through vegetation recovery. These shifts reflected urbanization trends and revealed scale-specific characteristics tied to development stages. Findings elucidate HC and CC mechanisms, informing green-blue infrastructure planning via targeted vegetation corridors to enhance cooling connectivity and counteract HC expansion, offering actionable heat mitigation strategies for rapidly developing urban regions.

This study investigates the sensitivity of satellite-based brightness temperature measurements at 3.74, 10.85, and 12.00 μ m with respect to the identification of water, ice, and mixed-phase clouds (MPC). Radiative transfer simulations computed by SCIATRAN reveal that the directional brightness temperature difference at 3.74 μ m ( Δ BT 3 . 74 ), which is dependent on scattering, enables water clouds and MPC separation from ice clouds. For water clouds and MPC, Δ BT 3 . 74 typically exceeds 2 K, whereas for ice clouds it remains below 2 K. To separate MPC from water clouds, we introduce the Liquid Cloud Probability Index (LCPI) based on cloud top temperature and absorption differences between water and ice at 10.85 and 12.00 μ m. LCPI values generally exceed 0.4 for water clouds but fall below 0.4 for many MPC cases. The Δ BT 3 . 74 and LCPI approach is validated using the Sea and Land Surface Temperature Radiometer (SLSTR) dual-view data collocated with the 2B-CLDCLASS-LIDAR cloud phase product, showing over 90% accuracy in water and ice phase classification, and approximately 60% for MPC. This dual-view, multi-channel method enhances the detection of cloud phases, offering improved results for brightness temperature measurements. • Directional brightness temperature difference at 3 . 74 μ m can be used to separate water and ice clouds. When used with brightness temperature differences at 10 . 85 μ m and 12 . 00 μ m , mixed-phase clouds can be identified. • Mixed-phase clouds occurring at lower cloud top heights are challenging to separate from water clouds.

Urban heat exposure patterns and domain-specific executive function in adolescents.

PGDM: Physically guided diffusion model for land surface temperature downscaling

Purpose As climate warming and urban heat exposure continue to intensify, accurately identifying the cooling effect of vegetation is essential for developing effective urban climate adaptation strategies. However, the relationship between the Normalized Difference Vegetation Index (NDVI) and Land Surface Temperature (LST) is highly sensitive to spatial scale. Inappropriate scale selection may lead to misjudgments of urban cooling potential due to the Modifiable Areal Unit Problem (MAUP). This study aims to reveal the scale dependency of the NDVI–LST relationship and assess its implications for urban green space planning and climate adaptation decision-making. Design/methodology/approach Taking Shanghai as a case study, this research constructs multi-scale grids while maintaining consistent spatial resolution of remote sensing data. Land use change information from 2000 to 2024 is integrated into the analysis. By combining the Random Forest model with the SHAP interpretation method, this study examines the statistical relationship, nonlinear response structure and variations in cooling turning points of NDVI–LST across different scales and land use contexts. This approach enables a systematic assessment of how spatial aggregation methods influence the identification of vegetation cooling effects. Findings The results show that as the analysis scale shifts from coarse to fine, local heterogeneity and contextual effects are significantly enhanced. The nonlinear response structure of NDVI–LST and its cooling thresholds undergo systematic changes with scale. At coarse scales, local fluctuations are smoothed out and the overall response becomes more stable. At fine scales, nonlinear characteristics become more pronounced and spatial differences more prominent. From the MAUP perspective, scale selection not only affects the assessment of statistical correlations but may also introduce biases in evaluating the cooling efficiency of green spaces within climate adaptation planning. Originality/value This study proposes a reusable multi-scale assessment framework from the MAUP perspective. It reveals the scale sensitivity of the NDVI–LST relationship and its contextual dependency across different land use backgrounds. The findings highlight the critical value of multi-scale information in urban green infrastructure planning and heat risk management. This provides methodological innovation and practical references for developing evidence-based urban climate adaptation strategies.

Methods such as multi-factor geographically weighted machine learning (MFGWML), geographically neural network weighted regression (GNNWR), and its enhanced version—geographically convolutional neural network weighted regression (GCNNWR)—have improved spatial non-stationarity modeling. However, they are over-reliant on spatial proximity weighting and underutilize spatial neighborhood information. Therefore, this study proposes a novel Geographically Conditioned Multiscale Convolutional Neural Network (GCMCNN) that downscales MODIS land surface temperature (LST) data to 100 m resolution by integrating two- and three-dimensional surface features. GCMCNN enhances the capture of spatial heterogeneity through a geographically aware (geo-aware) mechanism conditioned by location and attributes, while improving the characterization of spatial correlation by developing a multiscale convolutional neural network (CNN). GCMCNN model outperformed four benchmark models (random forest, MFGWML, GNNWR, and GCNNWR) in LST downscaling, reducing RMSE by 15.4–26.16% and MAE by 15.05–23.59%, and increasing R2 by 39.81–148.09%. GCMCNN improved the model performance of CNN through its geo-aware mechanism and multi-level spatial feature extraction and fusion, reducing RMSE and MAE by 6.6% and 8%, respectively, and increasing R2 by 7.5%. Furthermore, spatial block cross-validation better reflects the model’s true generalization ability than random cross-validation. This study highlights the importance of accurately modeling spatial heterogeneity and correlation in LST downscaling.

In arid and semi-arid countries, accurate agricultural drought characterization is crucial for climate-resilient planning and efficient early warning systems. Therefore, by integrating remote sensing data, CMIP6 projections, and machine learning (ML) models, the study assesses the spatio-temporal dynamics, driving factors, and future evolution of agricultural drought in the Borena Zone, southern Ethiopia. The Vegetation Health Index (VHI) was primarily targeted using several drought-related indicators, such as the Normalized Difference Vegetation Index (NDVI), Modified Normalized Difference Water Index (MNDWI), Land surface temperature (LST), and Temperature Condition Index (TCI) to detect the drought. Additionally, Shapley Additive Explanations (SHAP) and ML models were utilized to differentiate drought dynamics and performance for two seasons: March-May (MAM) and August-November (ASON). The findings show that the observed distribution of rainfall during both seasons spans from less than 20mm month⁻¹ in the southern lowlands throughout a large portion of the southern basin, and that LST variation increases to 35.1°C during MAM and rises to 38.1°C during ASON, making it greatly susceptible to evapotranspiration. Consequently, the southern zone's root-zone soil moisture decreases from 0.30 to 0.35m³ m⁻³ in MAM to 0.25-0.30m³ m⁻³ in ASON, along with notable decreases in NDVI and MNDWI. The ML models show strong drought prediction performance when targeting VHI; the Receiver Operating Characteristic Area Under the Curve (ROC-AUC) and Cohen's kappa of XGBoost achieve 0.997, and 0.72 during MAM, and Random Forest achieves approximately 0.988 and 0.78 during ASON, respectively. TCI and NDVI are the most important predictors of drought severity according to SHAP-based feature attribution. Integrating future drought forecasts from a bias-corrected CMIP6 model (CNRM-CM6-1) under SSP2-4.5 and SSP5-8 scenarios indicates a significant expansion and intensification of drought. Mild drought is projected to affect up to 36.2% of the zone under SSP2-4.5, whereas under SSP5-8.5, moderate, severe, and extreme drought are projected to increase by 18.87%, 5.03%, and 3%, respectively, by the late 21st century. These robust results highlight escalating risks to rain-fed agriculture, rangeland productivity, and water resources, underscoring the urgency of machine-learning-based drought early-warning systems and targeted adaptation strategies in the Borena Zone.

A Bayesian regression modeling framework for identifying land surface temperature and evapotranspiration treatment effects in UAV and airborne data

ABSTRACT Tropical forests are crucial ecological resources but are face severe threats from human activities. Assessing forest cover is essential for monitoring and for the development of effective sustainable forest management approaches. This study analyses the changes in forest canopy density and land surface temperature (LST) in Kolasib district of Mizoram, India, over a period of three decades (1995–2025), utilizing Landsat satellite imagery via Google Earth Engine (GEE). The Forest Canopy Density (FCD) model was computed by integrating the Normalized Difference Vegetation Index (NDVI), Advanced Vegetation Index (AVI), Barren Index (BI), and Shadow Index (SI). A Mann-Kendall trend test and Pearson correlation were also applied to the long-term data. The results revealed a 1.83% increase in non-forest areas and a 0.21% decrease in dense forest, primarily driven by infrastructure and shifting cultivation. Conversely, moderate forest increased by 2.41% due to vegetation recovery and land‑use policies. The Mann-Kendall trend analysis shows that LST is increasing by 0.042°C annually, accumulating in a 1.26°C increase over the three decades. Although NDVI has improved over recent years, the thermal footprint remains elevated, proving that forest conversion permanently alters the regional microclimate. These findings emphasize the necessity of sustainable land-use planning for continued forest management and conservation.

ABSTRACT Accurate evapotranspiration (ET) estimation in an energy balance model, such as SEBAL, depends on the selection of hot and cold anchor pixels. This selection is highly subjective and varies among modellers, leading to substantial uncertainty and frequent inconsistencies in ET outputs. Manual identification is time-consuming and often problematic, with incorrect anchor-pixel choices producing unrealistic energy partitioning and even negative ET estimates. The sensitivity analysis in this study revealed that hot anchor pixels with identical land surface temperature (LST) but at different spatial locations can yield significantly different, and sometimes negative, ET estimates, underscoring the need for an optimal, physically representative anchor-pixel pair. Therefore, this study introduces an automated optimization framework using a genetic algorithm that selects an anchor pixel pair from the candidate anchor pixel pool, defined based on the NDVI-LST feature space, to maximize the evaporative fraction over agricultural areas while maintaining energy balance closure. The approach eliminates subjectivity in anchor pixel selection and ensures consistent energy-balance computation. The model performance was evaluated using 10-day cumulative ET estimates and the water balance model, ORYZA2000, which was parameterized with detailed field-level crop, water, soil and meteorological information. Optimized SEBAL estimates showed strong agreement with independent ET estimates from field-parameterized water balance model, with R2 values between 0.89 and 0.94 and RMSE values of 5.21–8.56 mm. In contrast, ET estimates using manual anchor pixel selection exhibited lower performance with R2 values between 0.77 and 0.83 and RMSE values between 7.69 and 10.52 mm. Moreover, the negative ET values obtained during the manual approach were automatically excluded while executing optimized SEBAL. The proposed method is automated and ensures consistent ET estimates, thereby greatly improving the operational reliability and applicability of the SEBAL model across a large spatial domain.

This study investigates vegetation dynamics in a semiarid mountain ecosystem in southwestern Saudi Arabia from 1990 to 2024 by integrating multisource remote sensing data, bioclimatic variables, and machine learning models. Trends in vegetation greenness (NDVI), water content (NDWI), and land surface temperature (LST) were quantified via Kendall's τ and Sen's slope estimators. To capture fine-scale climatic heterogeneity in complex terrains, CHELSA bioclimatic variables were downscaled from ~ 1km to 30m resolution via random forest regression. These predictors were used to model NDVI trend patterns through a comparative framework including a baseline artificial neural network (ANN), metaheuristic-optimized ANN variants (ANN-PSO and ANN-GWO), and ensemble models (random forest and XGBoost). Model performance was assessed via independent validation, error metrics (RMSE, MAE, R²), bootstrap uncertainty analysis, and spatial residual diagnostics. All the models exhibited strong predictive ability, with XGBoost achieving the highest accuracy (RMSE ≈ 0.051; R² ≈ 0.92) and the lowest residual spatial autocorrelation. The spatial results indicate dominant greening across ~ 73-74% of the area, whereas ~ 25% of the area exhibits degradation concentrated in thermal and moisture-stressed zones. These findings highlight the value of integrated spectral-climatic modeling for monitoring vegetation changes in semiarid mountainous environments.

Abstract Urban heat islands (UHIs) pose significant environmental challenges in rapidly urbanising regions, highlighting the need for high-quality spatial datasets to better characterise urban thermal patterns. The Local Climate Zones (LCZ) approach provides a standardised land-use framework for UHI research and is known to strongly correlate with Land Surface Temperature (LST). In this study, 27 LCZ maps for five Southeast Asian cities were generated using cloud-free Landsat 8/9 scenes. Corresponding LST maps were retrieved to construct a paired dataset. Spearman’s rank correlation analysis confirmed a similar relationship between LCZ classes and LST existed in all pairs, shown by Spearman’s r ranging from -0.675 to -0.874, indicating a strong negative direction. When evaluated using the same testing polygons, our LCZ maps consistently outperformed the existing global LCZ dataset. A fully convolutional network was trained using the dataset to perform pixel-wise LST regression using LCZ maps, achieving an RMSE of 1.371℃ and an R 2 of 0.891 under random-split setting and RMSE values ranging from 1.212 °C to 1.623 °C across individual city validations. The results highlight the substantial predictive performance of LCZ information for spatial LST estimation and demonstrate the suitability of the dataset for UHI machine-learning applications. The dataset is publicly available at https://doi.org/10.5281/zenodo.18223368 . Graphical Abstract Based on the graphical abstract snapshot, this study utilises Landsat 8 and 9 to develop a paired Local Climate Zones (LCZ) and Land Surface Temperature (LST) for five major Southeast Asian Cities. For each city, all Landsat 8/9 scenes from 2013 to 2025 were retrieved and clipped to the city extent and checked for cloud cover, if the cloud cover is less than 10%, the scene is selected for LCZ mapping through a Convolutional Neural Network (CNN) using our carefully-selected LCZ polygons, if the validation overall accuracy of the mapped LCZ scene is greater than or equal to 70%, the corresponding LST scene is calculated and clipped to form an LCZ and LST pair. Spearman’s rank correlation analysis resulted in an average Spearman’s rank of -0.783 across all generated pairs. The formed SEA-LCZ-LST dataset was tested for pixel-wise LST regression from LCZ using a Fully Convolutional Network (FCN) and achieved a RMSE of 1.371℃ and R2 of 0.891 under random-split setting. These findings demonstrate the dataset’s suitability for machine-learning-based UHI applications in the Southeast Asian region.

Deyang City, situated in a typical mountain-plain transition zone of Southwest China, faces high ecological fragility due to its steep topographic gradient, frequent seismic disturbances (e.g., the 2008 Wenchuan earthquake), and intensive anthropogenic pressures from rapid urbanization and agricultural production. Taking this representative area as a case study, this study utilized multi-source remote sensing data from 2000, 2005, 2010, 2015, and 2020. A comprehensive Land Degradation Index (LDI) was constructed by integrating NDVI, Surface Albedo, Salinity Index, and Land Surface Temperature. Using transition matrix, Theil-Sen median trend analysis with Mann-Kendall tests, Geodetector models, and Coefficient of Variation (CV), we systematically diagnosed the spatio-temporal evolution, trends, stability, and driving factors of land degradation. The results show that: (1) From 2000 to 2020, the LDI exhibited a spatial pattern of “higher in the northwest, high and dispersed in the center, and low in the southeast.” The combined proportion of “Low” and “Slight” degradation rose from 43.33% to 62.86%, though a distinct rebound occurred in 2010 due to the Wenchuan earthquake. (2) Transition matrix analysis revealed a positive evolution, with the L1 grade increasing by 18.01%, reflecting long-term ecological recovery. (3) Trend analysis indicated a regional LDI decline (β = −0.0173/5years), with improvement areas (31.16%) exceeding degradation areas (12.13%). (4) Geodetector results identified temperature and precipitation as foundational drivers, while per capita GDP and population density gained explanatory power during rapid economic expansion. Universal bivariate and nonlinear enhancements suggest land degradation results from complex multi-factor coupling. (5) CV analysis showed high stability (90.75% of area with CV < 0.1), but the northwestern mountain-plain transition zone exhibited elevated ecological vulnerability due to seismic disturbance and human activities. This study provides an analytical framework for diagnosing land degradation in mountain-plain composite regions, offering a scientific basis for ecological restoration and sustainable land management policies.

Urban heat intensification threatens environmental quality and human health, particularly in arid urban areas where cooling resources are scarce. While impervious surface area is a well-established driver of urban land surface temperature (LST), the influence of built-up spatial patterns beyond their total area remains poorly understood. This study aimed to disentangle the effect of built-up pattern after accounting for built-up area on LST using a dual-scale analysis in an arid urban-barren mosaic. The spatial pattern of Sentinel-2 10m-derived built-up areas, measured as Area, Mean Patch Size (MPS), Shape Index (SHI), Nearest Neighbor Distance (ENN), and Number of Patches (NP), was extracted within each 100m Landsat thermal pixel, creating a dataset of 8,130 observations. Landsat-derived summer LST served as the dependent variable in a set of flexible regression models designed to evaluate both the total extent of built-up land and the spatial structure of urban patches. Results showed that adding configuration metrics to Area-only models improves LST prediction accuracy (R² = 0.589 to 0.693) and reduces RMSE (2.792°C to 2.409°C). Additional models examined how different spatial metrics interact, allowing the influence of built-up patterns to be evaluated independently of total built-up area. Unsupervised clustering (average silhouette width = 0.295) revealed that, when controlling for area, Compact configurations, low NP with high MPS, SHI, and ENN, were on average 2.7°C warmer than Fragmented or Balanced types. These findings demonstrate that, after adjusting for built-up extent, built-up pattern exerts a measurable and spatially distinct influence on LST, underscoring its importance for climate-responsive urban design in arid regions.

Geospatial and machine learning approaches for malaria risk mapping in flood-prone districts: Implications for public health decision-making.

Glacial lake outburst floods (GLOFs) represent one of the most severe climate-induced hazards in high-mountain regions. The Shisper Glacier in Hassanabad Village, Hunza, Pakistan, has experienced repeated surge events and rapid ice-dammed lake expansion, posing escalating risks to downstream communities and critical infrastructure. This study presents a multi-temporal, multi-index remote sensing and GIS-based assessment of Shisper Glacier dynamics and associated GLOF hazards for the period 2020-2024. Landsat-8 OLI imagery was used to compute the normalized difference snow index (NDSI), normalized difference water index (NDWI), modified normalized difference water index (MNDWI), normalized difference vegetation index (NDVI), land surface temperature (LST), land use and land cover (LULC), and green-red-NIR-shortwave infrared (GRZI) composite indices. Elevation change analysis was performed using the Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM), while NASA POWER data provided climatological context for temperature, precipitation, and humidity patterns. The ice-dammed lake area at the glacier's snout was delineated using MNDWI-based water body extraction from Landsat-8 OLI imagery, and lake volume was estimated using a depth-area empirical scaling relationship, consistent with established methodologies for ice-dammed lakes in the Karakoram. Results indicate that the debris-covered glacier expanded from 161 km2 in 2015 to 197 km2 in 2018, while the ice-dammed lake exceeded 0.1 km2 with an estimated volume between 1 × 106 m3 and 10 × 106 m3, representing a substantial and growing GLOF hazard. Rising land surface temperatures and changing precipitation patterns are identified as key drivers of accelerated glacial melt. Approximately 80% of Hassanabad Village's infrastructure, including the Karakoram Highway Bridge and a nearby hydropower facility, lies within the projected GLOF inundation zone. The novelty of this study lies in the first integrated application of seven spectral indices combined with DEM-based surge mapping and infrastructure exposure assessment for Shisper Glacier within a unified hazard framework. These findings underscore the urgent need for continuous satellite-based monitoring, early warning systems, and risk-informed land-use planning in the Hunza region.

Opisthorchis viverrini and soil-transmitted helminths (STH) remain major public health challenges in Thailand due to their widespread distribution. This study aimed to map and predict the prevalence of O. viverrini and STH infections across Thailand and to identify key environmental, climatic, and socio-economic factors influencing their geographical distribution. Bayesian geostatistical logistic regression models were fitted to national survey data to estimate infection risk and generate high-resolution spatial predictions. Demographic (sex, age), environmental (land surface temperature, vegetation indices, altitude), climatic (minimum and maximum temperature), and socio-economic (nighttime light intensity) variables were included as covariates in the models, and their associations with infection risk were quantified using posterior odds ratios and 95% Bayesian credible intervals (95% CrI). Opisthorchis viverrini infection was significantly associated with sex and age. Males had higher odds of infection than females (OR: 1.57; 95% CrI: 1.24-1.97). Individuals aged 25-59 years (OR: 4.75; 95% CrI: 3.22-7.21) and ≥ 60 years (OR: 4.69; 95% CrI: 3.10-7.29) had similarly elevated risks compared with those < 25 years. Minimum temperature (TMIN) was significantly negatively associated with infection risk. For each 1°C increase in TMIN, the probability of O. viverrini infection decreased by 33% (OR = 0.67, 95% CrI: 0.49-0.88). Hookworm infection was more common among males (OR: 1.65; 95% CrI: 1.42-1.93) and individuals aged 25-59 years (OR: 1.86; 95% CrI: 1.52-2.28). For Ascaris lumbricoides, infection risk was lower in individuals aged 25-59 years (OR: 0.42; 95% CrI: 0.18-0.98) and negatively associated with land surface temperature (OR: 0.66; 95% CrI: 0.39-0.98) and nighttime light intensity (OR: 0.97; 95% CrI: 0.93-0.99). No statistically significant associations were detected for Trichuris trichiura. Spatial predictions showed that O. viverrini was concentrated in northeastern Thailand, whereas STH infections were most prevalent in the south. Distinct spatial heterogeneity was observed in the distribution of O. viverrini and STH infections in Thailand. Demographic factors were consistently associated with O. viverrini and hookworm infection, whereas selected environmental and socio-economic correlates were associated with A. lumbricoides. These findings support geographically targeted surveillance and control strategies focused on high-burden areas and vulnerable populations.

Unlocking the cooling effect of urban green spaces is an effective strategy to mitigate urban heat islands and extreme heat exposure. However, the cooling potential of optimized green configurations in riparian buffer zones remains largely unknown. In this study, we integrated land surface temperature (LST) data derived from remote sensing, machine learning models, and scenario analysis method to quantify the potential effects of green space configurations in different riparian buffer zones (30–200 m) on urban cooling. We selected three cities, Chongqing, Changsha, and Wuhan as study areas, and found that forests and cropland have cooling effects of −4.08 °C ± 0.71 °C and −2.09 °C ± 0.27 °C, respectively, compared with temperature on impervious surfaces. In addition, our simulated results show that different cities respond differently to the cooling effects of green space configurations. In Chongqing, the riparian afforestation scenario has the potential to reduce citywide temperature by −0.10 to −0.51 °C. In contrast, Changsha and Wuhan exhibit more pronounced cooling effects in rooftop greening scenarios, which have the potential to reduce temperatures by −0.25 °C–1.51 °C. Attribution analysis further indicates that land cover changes, such as afforestation, are the primary drivers of urban cooling, while the cooling mechanisms within riparian buffer zones across the three cities exhibit significant nonlinear interaction characteristics. Overall, our results demonstrate that optimizing green space configurations in riparian buffer zones is one of the key strategies to achieve urban cooling.

The Tibetan Plateau (TP), known as the "Third Pole", is highly sensitive to climate change. Fine-scale daily mean land surface temperature (DMLST) data are needed to study regional climate processes, permafrost dynamics, glacier melt, and land surface energy balance. However, existing DMLST products over the TP generally have spatial resolutions of 1 km to 0.05°, which are too coarse to resolve local processes in glacier forefields, permafrost slopes, and other heterogeneous landscapes. Here, we developed a partition-based spatiotemporal fusion framework that integrates ERA5-Land reanalysis data with MODIS and Landsat LST products to generate a gapless 100 m-resolution DMLST dataset over the TP for 2019. Validation against ground-based observations shows that the dataset achieves a root mean square error (RMSE) of 3.13 K, a mean absolute error (MAE) of 2.42 K, a mean bias error (BIAS) of -0.81 K, and a coefficient of determination (R²) of 0.93 at the daily scale. Compared with an existing 1 km product, the proposed dataset improves error metrics by approximately 0.50-0.60 K. The 100 m resolution improves the representation of slope aspect, microtopography, land-cover contrasts, local temperature gradients, and transition zones in glacierized and other typical regions. This dataset provides a high-resolution and gapless DMLST resource for quantitative studies of land surface energy balance, permafrost degradation, glacier melt, and related environmental processes over the TP.

Exposure to heat in urban settings is a significant and growing public health concern. Yet, few studies have longitudinally measured everyday heat exposure using GPS-derived activity space data, paired with fine resolution satellite-derived heat measures, and no studies have measured effects on salivary cortisol in a community-dwelling sample. We address these gaps using data collected 2019–2023 across 11 neighborhoods in Detroit. Participants wore a GPS device and accelerometer over a one-week period. They completed a survey on demographics and mental health (anxiety and depression symptoms) and provided salivary cortisol samples (used to calculate diurnal slope) each year. We used a novel land surface temperature fusion product from ConstellR at daily cadence and 30-m spatial resolution to calculate heat exposure. We linked the seasonal time series of 2-band Enhanced Vegetation Index values calculated from the Harmonized Landsat-Sentinel-2 product with accumulated growing degree-days calculated from the ConstellR fusion product to model the land surface phenology at each pixel and used these modeled EVI2 values to estimate the exposure to greenness. We then tested whether higher heat exposure during routine daily outdoor activities was associated with poorer mental health or salivary cortisol data and whether these effects were attenuated by exposure to greenness. Among participants with ≥ 60 minutes of outdoor non-vehicle time, we observed a positive association between heat and anxiety (coef = 0.007, p-value = 0.062) and depression (coef = 0.007, p = 0.036) symptoms, after full confounder adjustment. Greenness was not independently, significantly associated with mental health measures and did not alter the effect size of heat exposure. We also observed a significant, positive association between heat exposure and cortisol slope (coef < 0.001, p-value = 0.029), indicative of higher stress. The consistency in our findings strengthens the evidence of the mental health impacts of everyday heat exposure, beyond extreme heat events. These findings have urban planning implications for increasing tree canopy, lowering impervious surfaces, providing heat shelters, and other urban cooling strategies. City-level or neighborhood-level heat action plans may help reduce the broader health impacts of rising temperatures in a warming climate.