Abstract Urban heat is a growing challenge, compounded by climate change and urban development. Yet, we lack a global understanding of heat extremes and where they coincide with population growth. Here, we explore satellite-based land surface temperature and population exposure across 1400 cities worldwide. Over two decades, urban surface temperature rose by 0.03-0.2 °C/year during daytime and 0.01-0.15 °C/year at night. The fastest increases were observed in cold-climate cities and in winter, particularly in Eastern Europe and Western Asia. Extreme temperatures have also become more frequent, particularly hot nights, which now occur 47% more often in arid cities. Compounded by population growth, this translates to a 51% increase in global exposure to extreme daytime heat. Climate warming amplified the exposure increase by 82% relative to population growth alone. Relative importance of population and climate change varies by climate zone, with climate dominating daytime exposure in continental regions (71%), and population growth in arid cities (69%).

GLIDE-SOL is a fully scripted and globally re-deployable Python workflow that operationalizes SOLWEIG for rapid and repeatable thermal-comfort mapping across diverse urban environments. GLIDE-SOL is built on the SOLWEIG radiative balance libraries, but rewrites the surrounding system – including automated input generation, the execution engine, and post-processing – so that the model can be driven by globally available datasets and executed efficiently on GPUs. All inputs (terrain, building morphology, canopy height, land cover, and meteorology) are automatically derived from global products, eliminating local preprocessing while enabling consistent applications from neighborhood-scale analyses to city-wide and multi-city experiments. In addition, GLIDE-SOL introduces lightweight physical diagnostics to improve realism when driven by coarse meteorological forcing, targeting key urban controls on wind and near-surface temperature. The workflow incorporates two physical augmentations: (i) roughness- and obstacle-based directional wind attenuation to approximate near-surface ventilation; and (ii) diagnostic temperature adjustments that combine a simple urban heat island (UHI) cycle with an elevation-based correction using high-resolution DEM information, to better capture nocturnal warming and local lapse-rate effects. To scale to large metropolitan areas, GLIDE-SOL uses explicit domain tiling with cross-tile synchronization to preserve radiative consistency across tile boundaries, enabling meter-scale simulations over tens to hundreds of square kilometers without sacrificing reproducibility. Daily outputs (24 radiative and meteorological fields) are stored as compressed GeoTIFFs to reduce disk usage and accelerate downstream processing. GLIDE-SOL is implemented through three reproducible components: an automated global-input generator; a SOLWEIG execution engine with coordinated tiling; and a post-processing module for systematic sampling, time-series extraction, and visualization. An operational demonstration in Dortmund, using hourly measurements from 25 urban and peri-urban stations and simulations run at 2 m grid spacing between August 2024 and December 2025, shows that incorporating wind attenuation and the diagnostic temperature corrections substantially improves UTCI performance (RMSE reduced from 8.09 to 2.84 °C), alongside improvements in Tmrt, air temperature, and wind speed simulations. By integrating harmonized global inputs with physics-based diagnostics, GPU acceleration, and scalable tiling, GLIDE-SOL supports applications such as operational UTCI nowcasting, retrospective and climatological analyses of heat stress, sensitivity tests of urban morphology and greening strategies, and coordinated multi-city experiments requiring consistent modeling protocols.

Urban heat is a growing concern for public health, energy demand, and urban liveability. High-resolution air temperature (T a ) data are needed for developing effective, localised adaptation strategies, but obtaining such data at the city scale remains a challenge as weather stations are often sparse and unevenly distributed. To address this data gap, we developed a machine learning (ML) framework that creates high-resolution, gridded T a maps using crowdsourced observations and diverse geospatial datasets describing the city, and evaluated its accuracy, transferability, and performance under varying data availability conditions. This framework uses satellite-derived land surface temperature (LST), urban form and fabric datasets, and meteorological variables to train a convolutional neural network (CNN) model. The approach was implemented in Sydney, Australia, producing 30 m gridded T a estimates (at ~10:00 a.m. local time) with high accuracy on an independent held-out test set (R² = 0.97; RMSE = 0.91 °C), surpassing previously reported ML performance. We further assessed the generalisation and spatial transferability of this method, showing that the CNN model maintained strong predictive accuracy for unseen locations across the city (R² = 0.91–0.93; RMSE = 1.27–1.44 °C). Model performance also remained stable when the number of T a stations was significantly reduced by ~80%. Performance was more variable for unseen days (R² = 0.66–0.93; RMSE = 1.52–2.60 °C), indicating that ML-based T a mapping requires training data representative of diverse weather conditions. We find that high-resolution city-descriptive datasets are beneficial but not essential, as comparable accuracy was achieved using only globally available predictors. These results indicate that the proposed framework is transferable to other cities, including those in data-sparse regions. The study provides an effective and scalable approach for city-scale air temperature mapping, which is urgently needed for urban heat assessment, climate adaptation, and public health planning.

With climate change accelerating, global temperatures continue to rise. Urbanization further compounds local and regional changes in climate and causes cities to be warmer than their surroundings. In this work, we quantify the urban warming rates as well as the contribution of urbanization across all densely populated climates, using a multi-level Bayesian model that accounts for the correlation among multi-city data and their uncertainty, and nighttime land surface temperature (LST) data from the Moderate Resolution Imaging Spectroradiometer (MODIS). Our results show substantial latitudinal variation in the 2002-2021 trends, with continental cities warming the fastest (0.75 K decade-1) and tropical the slowest (0.44 K decade-1). In contrast to previous studies, we find no clear urbanization effect in the 2002-2021 MODIS trends. We demonstrate that this discrepancy arises from explicitly accounting for the trend standard errors and show that when these uncertainties are omitted, differences between the urban and rural LST trends appear spuriously significant. This finding remains robust across alternative datasets, methods, and urban boundaries, reinforcing our conclusion that any apparent differences between urban and rural LST trends derived from 19 year of MODIS data cannot be reliably attributed to urbanization at the climate zone level.

Machine learning (ML) and artificial intelligence (AI) have become integral to modern weather and climate science, and their rapid evolution is driving new opportunities for urban climate research. Across the urban climate community, ML methods of varying complexities are being used for tasks ranging from fast point-based predictions to neighborhood- or city‑scale urban climate modelling for one or multiple (bio)meteorological variables. For instance, novel ML and AI approaches support scenario generation for impact studies such as quantifying the effects of urban vegetation, assessing thermal comfort under different warming pathways, or identifying locations for future cool spaces. In addition, ML techniques are also being applied to provide boundary conditions for micro‑scale models. While the field is expanding quickly, it remains highly fragmented. AI/ML in urban climate research spans diverse methods, scales, datasets, and scientific aims, making it difficult to define a common research agenda or benchmarks. The lack of an established interdisciplinary network between AI and urban climate communities further increases the risk of duplicated efforts and missed opportunities for coordinated progress. The newly founded AI4UrbanClimate working group addresses this gap by establishing an international community focused on AI/ML applications in urban climate research. Our goals are to (1) bring people together with similar research interests, (2) develop a common understanding of the scope and landscape of ML-based urban climate research, (3) review existing work across domains, and (4) identify persistent challenges and priorities and initiate and coordinate collective action - such as benchmarking datasets, standardized evaluation metrics, and best‑practice guidelines. As an initial step, we are mapping ongoing activities across the community to enable the development of meaningful benchmarks and identify where joint efforts could accelerate research. We will present the first outcomes from the initiative’s activities, including insights on who is currently involved and thoughts that were shared during the kick‑off meeting and online social event. We will also outline how you can join the AI4UrbanClimate network and contribute to building this emerging community.

The urban heat island (UHI) effect is one of the most studied phenomena in urban climatology. Numerous studies have revealed the heterogeneous nature of air temperature within cities, manifesting as an urban heat "archipelago" with small-scale air-temperature differences and multiple hot and cold spots rather than a single, uniform hot spot in the city core. With the introduction of the local climate zones (LCZs) scheme, close attention has been paid to the definition and description of "urban" and "rural" sites. However, what remains understudied and inconsistent across studies is the "sea level" of the "archipelago," i.e., defining the air-temperature conditions of the surroundings, unaffected by the city. Here, we compare definitions and requirements of that "sea level," and investigate multiple possible data sets for UHI calculation. Most typically, single weather stations, often at airports, are used as a "rural" reference. However, these stations are not ubiquitously available and typically suffer from effects such as a high fraction of impervious surfaces or urban heat advection when located downwind of the city. Crowd weather stations (CWS), which have gained attention in recent years in urban climate studies due to their abundance, are often affected by nearby buildings and are mostly located in urban areas. Besides station-based data, reanalysis products such as ERA5-Land could provide an independent reference, as they are available globally and often do not consider urban areas. In this study, all three data sources were compared against an ideal case of multiple professional weather stations placed around the city. We investigate the two temperate European cities of Paris (France) and Berlin (Germany) during six years (2019-2024), focusing on crowdsourced data from CWS. We find that ERA5-Land air-temperature data is a consistent and ubiquitously available reference for the definition of "rural," providing UHI-calculations most comparable to the ideal case of having multiple professional weather stations. Using it as a universal rural reference allows for comparison between cities and further enables exploiting the potential of CWS, even in regions with few stations and a lack of professionally-operated rural weather stations. Establishing a consistent "sea level" for urban air-temperature and UHI-studies enables comparison between various cities globally and allows for the integration of different data sources, such as local networks of weather stations or CWS, on a larger scale.

Microscale urban climate simulations provide detailed insights into thermal conditions within cities, supporting research and planning for climate adaptation. However, high spatial resolution simulations are computationally demanding, limiting their applicability for large urban domains. City-wide assessments often require coarser grid resolutions, where explicit representation of three-dimensional urban structures becomes impractical due to computational constraints and impacts the quality of the modelling results especially in the case of large-eddy simulations. Another issue arises from the limited availability of high-quality geodata for building resolving simulations.To address these challenges, urban parameterisation schemes can be employed in the PALM model system to capture urban effects while significantly reducing computational costs. This allows for the simulation of larger domains without explicitly resolving three-dimensional structures such as buildings and trees. The required input data for the parametrisation schemes can be generated based on Local Climate Zone (LCZ) maps. Furthermore, the LCZ maps can serve as a basis for idealised three-dimensional setups in cases where high-quality geodata are unavailable.In this study, we systematically compare various urban parameterisation schemes implemented in PALM, as well as an idealised LCZ-based setup, against a baseline scenario with explicitly resolving the three-dimensional urban structure. The simulations are conducted for a summer situation with initial and boundary conditions derived from the ICON-D2 model. Simulations are evaluated using observational data from Dortmund, Germany. Our analysis identifies the suitability and limitations of each approach across different spatial resolutions, providing guidance on when to parameterise and when to resolve urban structures in PALM simulations. Initial results will be presented.

Over the last decade, substantial progress has been made in the development of seamless all-sky Land Surface Temperature (LST) products. However, many agricultural and environmental applications require spatial resolutions finer than the currently available (≥1km). A practical solution to this limitation is to downscale existing all-sky datasets by extending methods originally developed for clear-sky data to also account for cloudy-sky conditions. In this study, we propose such an approach. Our method first generates a synthetic LST image representing average clear-sky conditions at the target spatial resolution, that then adjusts using downscaled residuals derived from the difference between the all-sky and synthetic LST. This approach offers two key advantages: it ensures physically realistic, fine-scale LST patterns by using the synthetic LST image as an initial guess, and amplifies the local weather effects in the all-sky LST, making them easier for the model to learn. To evaluate the method, we use half-hourly all-sky LST from the Spinning Enhanced Visible and Infrared Imager (SEVIRI), which we downscale from ∼5km to ∼1km and validate against independent in-situ and satellite LST. The results demonstrate improved performance relative to the original data across a range of land covers, topographies, and climate conditions, while also highlighting the potential of a single random forest (RF) model to predict the LST under both clear and cloudy conditions.

Background. Climate change causes an increased number of hot days, especially in urban areas. This can cause heat stress in the human body. Despite being a vulnerable group, data on thermoregulation and sweat loss in heat for young children are missing. Therefore, model calculations can serve as an alternative. Objective. This small-scale case study compares climate data from the study site Bochum, eligible as an urban heat island in Central European temperate climate zone, to available climate data of physiological model calculations of young children's thermoregulation to assess transferability. Methods. The climate station in Bochum provides 13 weather variables. For the comparison, data of the summers 2020-'23 were used. Literature was screened for physiological model calculations for children aged 3-6 years in real-life scenarios. Results. The annual temperature in Bochum has risen within the past decades, nevertheless ambient temperature of 30 degrees C has rarely been exceeded in 2020-'23. Only one study using real weather data from Tokyo, Japan was suitable for displaying a model of a young child under heat stress in a real-life scenario. The maximum ambient temperature of the model calculation (36.5 degrees C) was only surpassed once in Bochum during the study period. Conclusion. Thus, the model calculations of Tokyo's climate scenario are not fully transferable to Bochum as a model city in Central European temperate climate zone. Consequently, further research is needed. This analysis can nevertheless serve as an impetus to initiate action on this growing public health problem in young children in Central Europe.

Privately-owned weather stations, Crowd Weather Stations (CWS), offer high spatial and temporal density in many urban regions across the globe, and therefore have been used in a variety of urban climate studies, mostly focusing on single cities. One challenge in crowdsourcing CWS data lies in the fact that the link between measured atmospheric data and (historic-) metadata is often lost due to the limited metadata available from popular CWS networks. This poses challenges in retrieving and analyzing data, as, e.g., past changes in CWS location remain undetected, introducing incorrect data, thus reducing data integrity.We developed an end-to-end workflow for consistently collecting and checking CWS (meta-)data in 257 areas worldwide, covering over 500 urban regions since 2019. The workflow automatically adds newly set-up CWS to the database, as well as consistently handling changes in CWS location. Until now, the database includes over 310,000 CWS with 7 Billion hourly observations of air temperature and relative humidity (mean, maximum, minimum). Over 65,000 changes in CWS location have been detected since 2019. This highlights the importance of continuous metadata updates for this dynamic data source, further enabling the use of the measurements for different applications. Within the database, CWS are linked to additional metadata, including a global digital elevation model, a global Local Climate Zones map, and the Global Human Settlement Layer Urban Center Database.The database was developed using open data and open-source software, combining PostgreSQL, PostGIS, and Timescale, which allows us to manage billions of measurements efficiently. All air-temperature measurements are consistently and continuously quality controlled using the state-of-the-art open R-Package CrowdQC+. The result is a dataset of consistently-processed metadata and measurements with potential for global-scale (intra-)urban climate studies and in-depth city analyses.[MD1] [DF2]

Escalating urban heat, driven by the convergence of global warming and rapid urbanization, is a profound threat to billions of city dwellers. The science directing urban heat adaptation is strongly influenced by studies that use satellite-based land surface temperature (LST), which is readily available globally and address data gaps in cities, particularly in the Global South. LST, however, is a poor surrogate for near-surface air temperature, physiologically relevant human thermal comfort, or direct human heat exposure. This flawed practice leads to issues for several downstream use cases by inflating adaptation benefits, distorting the magnitude and variability of urban heat signals across scales, and thus misguiding urban adaptation policy. We argue that satellite-based LST must be treated as a distinct indicator of surface climate, which, though relevant to the urban surface energy budget, can be frequently decoupled from human-relevant thermal impacts especially during daytime. Only by a disciplined application of this variable, combined with complementary datasets, process-based and data-driven models, as well as interdisciplinary collaboration, can urban adaptation design and policy be effectively advanced.

Urban areas increasingly face heat-related climate risks, necessitating targeted, nature-based interventions such as tree planting to improve resilience, livability, and public health. This study presents a data-driven workflow to identify urban tree planting potential (TPP) in the city of Dortmund, Germany. The approach integrates high-resolution spatial datasets capturing land cover, shading, thermal comfort, population density, and critical infrastructure. All variables were harmonized within a 50 m hexagonal grid, normalized, and combined into a composite TPP score using weighting schemes informed by expert judgment and sensitivity testing. Spatial and non-spatial clustering were applied to group urban areas by shared characteristics, and a connectivity analysis evaluated the spatial coherence of high-potential cells and their relationship to existing green infrastructure. The findings demonstrate the potential to strengthen urban green infrastructure and guide coordinated planting strategies while addressing both ecological and social priorities. The presented workflow offers a flexible, transferable tool to support municipalities in prioritizing effective greening interventions and integrating climate adaptation objectives into urban development planning.

Detailed measurements are indispensable in order to understand small-scale urban climate effects. With professional weather stations (PWS) mostly being available outside of cities with few sites per city, alternative data sources such as crowd-sourced weather data have proven to be valuable. Often the Urban Heat Island (UHI) is studied under ideal calm conditions when its development is strongest. At the same time, it has been shown that wind leads to advection of urban air, impacting regions downwind of urban areas and within the city.We aim to provide insights into the effects of Urban Heat Advection (UHA) in the Urban Canopy Layer (UCL). The metropolitan regions of Paris and Berlin were studied, using four years (2019 - 2022) of quality-controlled crowdsourced air-temperature data from thousands of privately-owned Crowd Weather Stations (CWS). Those data were combined with global ERA5-Land data to overcome gaps in rural CWS coverage and globally-available Local Climate Zone (LCZ) information.It is shown that wind causes increased exposure to urban heat for areas located downwind of the city core, which was derived using a LCZ-weighted centroid detection. For all observed wind directions, classified by dynamically moving wind sectors, differences in spatial patterns were visible with the effect being strongest with regional wind speeds of 3 m·s−1. The results highlight the importance of considering the effects of UHA when studying the UHI to avoid underestimating the exposure to urban heat in downwind areas of the city. The results could be used as a starting point for coupling the conditions in the Atmospheric Boundary Layer with the resulting conditions in the UCL, utilizing a large database with crowdsourced CWS data.

The increasing intensity and frequency of heat waves combined with the urban heat island can create thermal conditions which are hazardous for human health. Numerical urban climate modelling can deliver the necessary information to plan resilient adaptation measures for healthy living conditions in cities under a future climate. However, as a model is always a simplification of the real world, model evaluation with measurement data is important. Traditional measurement networks and campaigns are very often not suitable in active planning processes. Crowdsourcing the required weather data offers the potential to easily evaluate model results at any given time. To identify the potentials and limitations of this approach, the microscale urban climate model PALM is applied to simulate a hot day (Tmax > 30 °C) in a German city. The model results are evaluated with quality controlled crowdsourced air temperature data. The evaluation reveals a good model performance with a high coefficient of determination (R2) of 0.86 to 0.88 and a root mean squared error (RMSE) around 2 K. A temporal pattern in model accuracy is detected with an underestimation of night-time air temperatures. Due to the high number of available stations and the resulting representation of intra-urban temperature variations, the crowdsourced air temperature data proved valuable for model evaluation. Limitations for this approach arise from radiation errors leading to a reduced data quality. Furthermore, measurements from a single station are influenced by microscale and localscale conditions and therefore only the information derived from several stations can be used for evaluation.

Urban heat is characterised by elevated temperatures in cities, resulting not only from global climate change but also from urban development and human activities. Previous research on urban heat has predominantly relied on satellite-derived land surface temperature (LST) data to investigate the changes in near-surface thermal environments. However, the applicability of LST for examining the temporal variation of air temperature is still not well understood. Using crowdsourced air temperature observations and satellite imagery, we explore the temporal variation of air temperature and its relationship with LST in more than 50 populated cities worldwide. Results show that city-average air temperature values are highly correlated with LST. However, the intensity of this correlation differs by season, day/night cycle, and is further influenced by background climate. Using satellite LST data, we expanded our analysis to include over 1500 urban areas and evaluated temperature changes in the past two decades. We observed a general trend of increasing temperatures in cities globally, although the rates of warming vary. The highest rate of temperature change was found in cold climate cities, with a more rapid increase during winter days. These cities are predominantly located in Eastern Europe, extending into parts of Western Asia. These findings provide new insights into the application of satellite-based LST for predicting future air temperature changes and identifying areas most vulnerable to urban overheating.

Urban heat is a growing concern for public health, energy demand, and urban liveability. High-resolution air temperature (T a ) data is needed for developing effective, localised adaptation strategies, but obtaining such data at the city scale remains a challenge as weather stations are often sparse and unevenly distributed within cities. To address this data gap, we developed a machine learning (ML) framework that creates high-resolution, gridded T a maps using crowdsourced observations and diverse geospatial datasets describing the city. This framework uses satellite-derived Land Surface Temperature (LST), urban form and fabric datasets, and meteorological variables to train a Convolutional Neural Network (CNN) algorithm. The approach was implemented in Sydney, Australia, using multi-day observations from 2019 to 2024, producing 30 m gridded T a estimates with high accuracy (R² = 0.97, RMSE = 0.91°C, surpassing previously reported ML performances). We further assessed the generalisation and spatial transferability of this method, which revealed that the CNN model maintained strong predictive accuracy for unseen locations across the city (R² = 0.91–0.93; RMSE = 1.27–1.44°C). Model performance also remained stable when the number of T a stations was significantly reduced by ~ 80%. Performance slightly declined for unseen days (R² = 0.66–0.93; RMSE = 1.52–2.60°C), suggesting the need for incorporating a broader range of weather conditions. We find that high-resolution city-descriptive datasets are beneficial but not essential, as comparable accuracy was achieved using only globally available predictors. These results indicate that the proposed framework is transferable to other cities, including those in data-sparse regions. The study provides an effective and scalable approach for developing city-scale air temperature maps, which are urgently needed for urban heat assessment, climate adaptation, and public health planning.

Global climate change causes rising of the average temperature as well as an increased number of days with extreme heat, even in the temperate climate of Germany. Children are a particularly vulnerable group according to heat stress and consequentially water loss. Even small changes in the water balance can lead to functional impairments. Therefore, quantification of exact water loss as well as the knowledge of hydration status changes in children under heat stress are important and can contribute to renewing water intake recommendations in accordance with climate change. This may promote approaching public health issues. In this study water loss, hydration status data and voluntary water intake of children are collected. Therefore, 3–6 years old children from a German kindergarten are recruited. Weight, urine osmolality and saliva osmolarity are measured twice a day on two different testing days with different temperatures (day 1: hot summer day with heat peak; day 2: fall-like day with neutral temperatures). In-between the tests on each day, the children are exposed to these climate scenarios by playing outside for 2 h just like the usual daily routine. The drinking amount as well as the activity during the observational time is recorded. Additionally, the collected data is compared to previously published data on computed estimation of body core temperature changes and water loss in heat on a theoretically modelled 3-year-old child. The weight difference before and after exposure (in dependence of the water intake), provides information about the water loss (thermoregulation) in the observation period. The comparison of newly collected data with modelled water loss can elaborate variations in climate and physiological responses and provide information about, whether the model calculation can be transferred to German conditions. Furthermore, hydration status changes, as reflected by the urine osmolality, are expected to be different between the two test days depending on the extent to which the higher water loss through sweating is compensated by different water intake through voluntary drinking. By means of the collected data, drinking recommendations for children may be aligned and specified in accordance to different climate scenarios. German Clinical Trials Register (Deutsches Register Klinischer Studien), DRKS00033942,19th August 2024.

Investigating the weekly rhythms of urban heat islands (UHIs) is critical for gaining deep insights into the urban climate changes caused by periodic cycles of human activities. However, the weekly rhythms of both canopy and surface UHIs (termed Ic and Is) remain poorly understood at a large spatial scale. Leveraging daily screen-level air temperature (Ta) and satellite land surface temperature (Ts) observations (2015∼2020), we show that the weekly rhythms of Ic averaged for all selected cities exhibit an evident peak-and-valley pattern across most seasons, with the annual mean Ic reduction during weekends of 0.09 ± 0.01 K (mean ± S.E., p < 0.05) for daytime and 0.10 ± 0.01 K (p < 0.05) for nighttime when compared to weekdays. In contrast, the Is only displays a pronounced weekly rhythm during winter daytime, with the corresponding Is reduction during weekends of 0.07 ± 0.01 K (p < 0.05) relative to weekdays. These findings remain robust against potential observation errors in Ta and Ts. From an intra-city perspective, both Ic and Is reductions during weekends escalate with increasing urban impervious surface percentage. This study illustrates the significance of short-term cycles of human activities in shaping global urban climates.

Urbanization has altered land surface properties driving changes in micro-climates. Urban form influences people’s activities, environmental exposures, and health. Developing detailed and unified longitudinal measures of urban form is essential to quantify these relationships. Local Climate Zones [LCZ] are a culturally-neutral urban form classification scheme. To date, longitudinal LCZ maps at large scales (i.e., national, continental, or global) are not available. We developed an approach to map LCZs for the continental US from 1986 to 2020 at 100 m spatial resolution. We developed lightweight contextual random forest models using a hybrid model development pipeline that leveraged crowdsourced and expert labeling and cloud-enabled modeling – an approach that could be generalized to other countries and continents. Our model achieved good performance: 0.76 overall accuracy (0.55–0.96 class-wise F1 scores). To our knowledge, this is the first high-resolution, longitudinal LCZ map for the continental US. Our work may be useful for a variety of fields including earth system science, urban planning, and public health.