Urban areas are typically warmer than their non-urban surroundings, a phenomenon known as urban heat island (UHI); yet, intra-urban temperature contrasts remain under-characterized, especially within and between slums. Conventional meteorological stations, often located at city peripheries (e.g., airports), poorly capture neighbourhood-scale microclimates. Earth Observation (EO) provides land surface temperature (LST), which is widely used to characterize urban heat islands. However, LST is not a direct proxy for near-surface air temperature (Tair), as the LST-Tair relationship is modulated by several interacting factors such as heat storage, vegetation, ventilation, and urban canopy effects among others. We investigate whether intra-slum Tair differences can be modelled using EO covariates alone and test potential transferability of such models across African cities. Quality-controlled, citizen-generated Tair observations collected between January and March 2024 are aggregated to 50 × 50 m grid cells and used as the target variable in a Random Forest (RF) model. EO-derived features (albedo and radiative properties, urban morphology indicators and vegetation index), computed on the same grid, serve as covariates for modelling intra-urban Tair to generate high-resolution heat maps within and between slums in Nairobi (Kenya) and Lagos (Nigeria). Trained in Nairobi, the model attains a low out-of-sample mean absolute error, however a direct transfer of the Nairobi-trained model to Lagos underperforms, indicating that zero-shot transferability is limited. Instead, robust performance requires a domain-adaptation strategy. Incrementally adding a modest fraction of Lagos samples to the training set improves cross-city generalisation, yielding comparable accuracy in both cities. Beyond demonstrating the value of low-cost sensors and citizen science for actionable climate information, we release an open, reproducible pipeline to guide urban-cooling interventions and benchmark resilience policies where evidence is most limited.

The new generation of high-resolution nightlight sensors (e.g., SDGSat-1, Jilin-1) with a spatial resolution of 10 m and below can capture the spatial patterns and light colours of Artificial Light at Night (ALAN). Our research leverages these new generations of high spatial and spectral resolution nightlight images and in situ data collection by citizen science and through nightlight mobile applications to accurately assess electricity access and reliability in Sub-Saharan African cities. More than 50% of Sub-Saharan Africa’s population (around 600 million people) has no access to electricity, and a large part (estimated at 80%) does not have access to stable electricity- data coming from the global dataset "Global Electrification Project" (World Bank). Existing global electricity access datasets use low-resolution satellite data (such as VIIRS) and, therefore, remain uncertain and misrepresent that many of these cities have universal access to electricity. In reality, many residents lack formal grid connections and face unreliable electricity supply. We conducted local fieldwork in informal settlements to capture access gaps and relate the results to high-resolution ALAN data. Surveyed areas experiencing unstable access to electric power, leading to frequent outages, such as Nigeria's average of over 32 monthly outages. Results show that field observations in combination with high-resolution night light images can provide a more accurate and nuanced understanding of electricity distribution and reliability to understand the gaps in SDG-7 across sub-Saharan Africa. Our study provides insights into how global monitoring of the multiple dimensions of urban poverty can include access to electricity as an essential indicator. Furthermore, emphasises incorporating nighttime light observations into global urban poverty monitoring to include electricity access as an essential indicator. It also outlines the need for advanced satellite-based sensors to support comprehensive urban poverty mapping, in line with the European Space Agency-funded “NightWatch” project.

Living-condition deficits remain widespread in rapidly growing cities, yet their spatial distribution is poorly measured beyond metropolitan areas, limiting progress toward Sustainable Development Goal (SDG) 11.1 on adequate, safe and affordable housing. Here we conceptualize morphological deprivation as an area-based form of disadvantage reflected in built form, residential density, road connectivity and infrastructure conditions, and generate globally comparable, neighborhood-scale estimates across 5,132 cities in Africa, Asia, and Latin America and the Caribbean. Using harmonized geospatial datasets and supervised learning trained on eight benchmark cities, we estimate that 395 million people (20.2% of the covered urban population) live in morphologically deprived neighborhoods. While relative prevalence is highest in Africa (43.6%), substantial populations are affected across all regions. Notably, over one-third (136 million) reside in small and medium-sized cities, revealing a large but underrecognized burden and underscoring the need to prioritize these cities in data, finance and policy efforts to achieve SDG 11.1. This study provides neighborhood-scale mapping of populations living in infrastructurally and environmentally constrained urban areas, identified from the physical form of the built environment, across 5,132 cities in Africa, Asia, and Latin America and the Caribbean. It reveals that over one-third of this burden falls on small and medium-sized cities, highlighting a substantial yet overlooked obstacle to achieving SDG 11.1.

Urban livability is shaped by dominant values, often economic or aesthetic, and power dynamics that often overlook the lived experiences of deprived urban area (DUA) residents. As a result, conventional livability indicators risk reinforcing existing inequalities unless these are grounded in inclusive and participatory approaches. To address this issue, we developed lightweight deep learning models - 'AI-voters' - trained on livability preferences from both DUA residents and city planners, using open-source satellite imagery. Applied in Ghana's Greater Accra Metropolitan Area, our approach reduced data requirements to map urban livability by 90% through a two-step urban form sampling strategy that enabled scalable participatory mapping. Training separate 'AI-voters' for planners and DUA residents revealed systematic differences: planners not only disagree among themselves but also consistently assign higher livability scores and overlook the preferences of DUA residents, such as avoiding coastal area exposure. The AI-voters mirrored human-voter behavior based on physical urban features such as greenery and building density, especially when trained on the preferences of DUA residents, demonstrating their potential as scalable proxies for local insights. These results highlight the importance of integrating community perspectives into AI models trained to map urban livability to expose hidden spatial inequities and promote more inclusive urban development.

Rapid urbanization across many regions worldwide has significantly contributed to the growth of deprived urban areas (DUAs), often called slums or informal settlements. The lack of reliable geospatial information on their location and extent in many cities continues to hinder efforts aimed at improving living conditions. This study addresses this critical information gap by exploring a User-and Data-centric Artificial Intelligence (AI) approach for accurately mapping these areas to support Sustainable Development Goal (SDG) Indicator 11.1.1. In col laboration with local communities, governments, and international stakeholders, we co-designed an AI-driven strategy leveraging open Earth Observation (EO) and geospatial data acrosseight cities worldwide. Instead of re lying solely on algorithmic precision, our method prioritizes local knowledge, iterative validation, and adaptive data collection. To achieve this, we developed a tailored multi-branch encoder-decoder convolutional neural network capable of integrating multi-modal data sources. Our approach incorporates an agile and iterative model refinement process, ensuring continuous feedback loops between AI design, data collection, and validation. Recognizing the importance of stakeholder engagement, we developed the IDEAtlas collaborative data collection platform-https://portal.ideatlas.eu/- to enhance data quality and inclusivity. The resulting dataset (IDEABench) is publicly available at https://doi.org/10.17026/PT/X4NJII to facilitate continued research and development. Findings indicate that fusing multi-spectral EO data with urban morphometric features, particularly Sentinel-2 imagery and built-up density, provides the highest accuracy for identifying DUAs. Furthermore, improvements in reference data quality through the IDEAtlas platform led to increased mapping precision. However, the sig nificant variability in accuracy across cities underscores the complexity of the task and suggests the need for supplementary geospatial data to complement EO-driven analysis. The code used in this study is available at https://github.com/IDEAtlas/ai-dua-mapping.

Machine learning models using Earth observation data are vital for mapping deprived urban areas, enabling scalable application and transfer across cities and time where ground data is scarce. Yet, their practical impact is often hindered by reduced user trust in the underlying data quality. A scoping review of deprived urban area mapping studies reveals systemic shortcomings in spatial data quality practices, even if studies report high performance accuracy. Shortcomings include a lack of operationalizing semantic definitions to reference labels, difficulty in reproducibility due to restricted data access, and sampling strategies that ignore spatial autocorrelation. Together, these issues can inflate reported performance estimates and overstate real-world generalization. To address these challenges, this study introduces a Spatial Data-Centric Quality Assessment framework. The framework extends established geographic information standards and existing practices in relevant domains to provide suggested reporting items for data provenance, primary use case of the dataset, semantic consistency, spatiotemporal consistency, representativeness, annotation quality, and critical ethical considerations to ensure the protection of vulnerable populations during mapping. The framework bridges the gap between technical research and practical application by mandating transparent documentation and rigorous evaluation of spatial data quality. It ensures that datasets used for machine learning are reliable, transparent, and trustworthy to inform urban policy and planning.

Urban heat exposure is intensifying due to climate change and urbanisation, with disproportionate impacts on vulnerable populations. Unfortunately, many urban areas, particularly informal settlements, lack sufficient data for detailed analysis to understand these impacts. Traditional air temperature measurement methods—such as meteorological stations—are sparsely distributed in African cities, typically located on city outskirts (e.g., airports), and fail to capture localized temperature variations. This study explores the use of low-cost sensors and citizen science initiatives to measure air temperature with higher spatial resolution in informal and surrounding formal settlements. A two-stage process is employed to evaluate data quality: first, statistically assessing biases in low-cost sensor (LCS) measurements, and second, employing Monte Carlo simulations to quantify uncertainties. The resulting data reveals significant temperature differences between informal settlements and surrounding formal areas, with informal settlements consistently exhibiting higher temperatures. This approach not only highlights the value of low-cost sensors and citizen science in generating high-resolution temperature data but also provides insights into thermal inequalities between different urban environments.

In growing cities, deprived neighborhoods house large numbers of residents, yet their extent and distribution remain poorly quantified, complicating implementation of SDG 11.1.1. We present the first global, neighborhood-scale spatial estimates of morphological deprivation, covering 5,132 cities in 103 countries across Africa, Asia, and Latin America & the Caribbean (LAC) home to 3.2 billion people. Neighborhood units and built-environment indicators from the City Segments v1 dataset were combined with segment-level labels from the eight-city IDEABench benchmark to train a supervised model, which was then applied to classify each segment as morphologically deprived or non-deprived. The mapped cities contained 1.96 billion residents, of whom 349 million (17.8%) lived in deprived segments, with the highest regional shares in Africa and substantial burdens in Asia and LAC. Morphologically deprived populations spanned the urban hierarchy, with about one-third living in small and medium cities, revealing important gaps in current deprivation monitoring.

Environmental hazards are key determinants of urban liveability, shaping the safety, health, and resilience of residents. This study investigates the intersection of urban livability and flood exposure by integrating remote sensing, citizen science, and AI-driven analysis across three African countries: Ghana, Kenya, and Mozambique. Using Sentinel-1 satellite imagery, open geospatial datasets, and advanced deep learning techniques, a citizen-derived perceived livability index was created which was then combined with rapid flood exposure modelling through FastFlood. The results reveal that areas with the lowest livability scores -characterized by poor housing conditions, limited service access, and minimal green spaces- are also consistently the most exposed to frequent and severe flooding. In Nairobi, for instance, approximately 35% of built-up areas are flood-prone, with informal settlements like Kibera and Mathare facing disproportionate risks. Citizen science efforts validated the flood models, underscoring the critical role of local knowledge in capturing fine-scale flood dynamics invisible to remote sensing alone. The project demonstrates that liveability and environmental risk are deeply interrelated, and contribute to worsening urban vulnerability. By combining community mapping with scalable Earth Observation methods, this work delivers actionable methods for urban planners, humanitarian organizations, and local policymakers. Our results stress the importance of planning strategies that prioritize investments in flood mitigation, nature-based solutions, and resilient infrastructure for the most at-risk communities. Such communities are often omitted in official data. The needs and views of such vulnerable communities need to be included in supporting sustainable and inclusive urban development under increasing climate pressures.

Urban deprivation mapping is critical for addressing inequalities and achieving Sustainable Development Goal (SDG) 11.1.1, which focuses on ensuring access to adequate housing and services in urban areas. This study introduces a geospatial framework to operationalize previously conceptualized urban Domains of Deprivation related to unplanned urbanization, limited infrastructure, and limited services within city segments at the city-scale. Leveraging open, global datasets, including Google’s V3 building footprints and 2.5D building heights, the model assigns deprivation scores (ranging from 0 to 6) based on binary thresholds derived from median values. Validation against reference slum boundaries provided by the IDEAMAPS network achieved an F1- score of 0.45 for high-deprivation areas. The results highlight the spatial distribution of deprivation across Nairobi and demonstrate the reliability of dense building indicators for identifying informal settlements. The framework demonstrates computational efficiency, enabling citywide analysis using accessible resources, and highlights its potential to inform urban planning and targeted interventions through scalable geospatial methodologies aligned with SDG 11.1.1.

The growth of deprived urban areas (DUA), often associated with slums or informal settlements, is one of the consequences of rapid urbanization. Earth Observation (EO) data provides valuable information for mapping and monitoring such urban areas to assess the Sustainable Development Goal (SDG) indicator 11.1.1. Previous studies show that building density is one of the most informative morphometric variables to map DUA. However, building density when available are often mono-temporal and lack information about the exact date it was assessed. To address this gap, we present a deep learning-based approach that integrates building density regression from EO data to guide the learning process of a pixel-wise classification network. Our methodology optimizes the combined loss function of a dual-output semantic segmentation model, balancing classification and regression tasks, using Sentinel-1 and Sentinel-2 as input. This balance improves the accuracy of building density predictions, which, in turn, enhances the detection of DUAs and model interpretability. We evaluated our approach in Salvador (Brazil) and Nairobi (Kenya), achieving improvements of 9.75% and 0.60%, respectively, compared to previous studies.

Slums and informal settlements are among the most common forms of urban development in Africa. These areas often lack essential infrastructure and services, face socioeconomic disparities, and are increasingly vulnerable to climate risks such as floods. The Space4All project aims to promote sustainable urban development by integrating advanced geospatial technologies with locally grounded approaches to analyze the intersection of livability and flood exposure in pilot cities in Ghana, Kenya, and Mozambique. To develop a scalable approach, we use open data such as Sentinel-2 satellite imagery. The methodology combines state-of-the-art AI models, Earth Observation data with Citizen-generated data collected via a custom app. The results highlight deprivation hotspots, revealing the intersection of spatial inequalities and flood exposure. Flood exposure is assessed through a comprehensive approach that integrates local knowledge gathered from workshops in informal settlements with flood models and historical rainfall data. This research provides actionable insights for urban planners, policymakers, and NGOs to prioritize targeted interventions and investments, fostering resilience through community-driven approaches.

Over one billion people globally live in slums, informal settlements, and other deprived areas. However, maps of deprived areas are often unavailable or over-simplistic, distinguishing only between slums and formal areas. Recent research advocates for a multidimensional approach to better account for the complexity of deprivation. Previous studies have mapped the unplanned urbanization domain of deprivation using morphometrics derived from building footprint data. This study explores leveraging Earth observation data for scalability and regular updates by using high-resolution satellite imagery and deep learning to map indicators for morphological informality in Nairobi, Kenya. The proposed model combines a ResNet backbone with three classification heads to map the two indicators: irregular settlement layout (ISL) and small, dense structures (SDS), alongside building presence. The model was trained on automatically generated reference data using building footprint morphometrics and clustering, and its outputs were validated through community-sourced annotations obtained via participatory action research. The results demonstrate the potential of high-resolution satellite imagery for mapping ISL (F1 80.90) and SDS (F1 78.73). Nonetheless, further research is required on the geographic transferability of the proposed method.

This study addresses the challenge of accurately mapping informal settlements, which are home to over a billion people globally. Current maps often simplify these areas into binary categories, ignoring the nuanced dimensions of deprivation. The research focuses on ”unplanned urbanization,” a key domain in informal settlement mapping, and proposes a method to classify morphological informality into three deprivation levels (low, medium, and high) based on two subdomains: small, dense structures (SDS) and irregular settlement layouts (ISL). The methodology involves analyzing building footprints and road network data using urban morphometrics, clustering these metrics into subdomains with k-means, and validating results with community-sourced reference data. Tested in Nairobi, Kenya, and Lagos, Nigeria, the model achieves good performance (F1 > 65 for indicator maps) but faces challenges in the medium informality class, particularly in Nairobi, where community feedback diverges significantly. Despite an overall accuracy of 48 % for Nairobi and 60 % for Lagos, the model offers a framework for continuous improvement. This work highlights the value of integrating local perspectives into mapping efforts and provides a scalable, transferable approach for identifying levels of morphological informality.

Addressing the challenge of mapping deprived urban areas (DUAs) globally requires both technical innovation and user engagement. The IDEAtlas project developed a novel approach to monitor DUAs by combining advanced Earth Observation (EO) technologies with a user-centered design. Central to this is our IDEAtlas User Portal, a data platform that provides scalable, accessible, and participatory mapping solutions in support of Sustainable Development Goal (SDG) 11.1.1. The portal provides outputs from a custom Multi-Branch Convolutional Neural Network (MB-CNN) model trained on freely available Sentinel-1 and Sentinel-2 data, enriched with ancillary open datasets such as building footprints. Presently, the portal offers binary settlement maps, with future plans to include deprivation severity indices, and multi-temporal urban growth maps for the period 2019 to 2023. A unique feature of the IDEAtlas platform is its two-tier access model: an open section offering 100 by 100 m gridded outputs for public use, and a protected section for sensitive, city-level data. In the protected section, users can validate outputs. The user validation is an essential element of our Living Labs and is done in the form of continuous stakeholder collaboration. Through Living Labs and iterative user feedback across eight global cities, the portal has demonstrated how involving local governments, NGOs, and community organizations can enhance data quality, relevance, foster ownership, and empower decision-making. Early user engagements, such as the update of Argentina's RENABAP informal settlement registry, highlight the portal's operational value. By coupling technological scalability with participatory validation. Thus, the IDEAtlas User Portal represents a significant step toward inclusive, evidence-based urban planning and policymaking.

Africa, as a major climate change hotspot, faces severe impacts, including extreme temperatures. Notably, urban areas are unequally affected by these impacts. The urban poor are particularly vulnerable to extreme temperatures, because of the environmental and physical characteristics of their neighbourhoods, and their limited resources to develop coping strategies. Limited knowledge exists of the spatial patterns of thermal inequalities within neighbourhoods. Our overall scientific objective is to explore the potential of Earth Observation (EO) to study how and why urban dwellers in the Global South (focusing on Africa) with different levels of deprivation are divergently exposed to varying temperatures and extreme heat, and to quantify the urban population exposed to such conditions. We make use of several state-of-the-art EO/AI models, and employ innovative in situ data collection methods together with local stakeholders through Citizen Science. We rely as far as possible on open or low-cost satellite imagery (e.g., Sentinel-1/2, Landsat, ECOSTRESS) for scalability and transferability, and we implement Machine Learning (ML) methods, including Deep Learning (DL). Results highlight significant local differences in thermal exposure, emphasizing the need to understand and communicate these spatial patterns to support the development of cost-effective adaptation strategies.

Many studies are pointing to the fact that cities are experiencing higher temperatures than non-built-up areas. Yet limited can be found on thermal inequalities in the context of vulnerable groups, specifically linked to people living in deprivation. Here, we study heat patterns across vulnerable groups living in deprivation as an important effort that should be paralleled to the other urban climate studies and aim at answering two primary questions: (1) how temperature varies within and across deprived areas, and (2) what the key driving factors are for such variation. We conduct intensive in-situ measurements by involving local residents in air temperature traverse across deprived neighbourhoods and modelling the pattern of air temperature with spatial covariates. We also compare different modelling techniques while securing the interpretability of the air temperature pattern by using understandable spatial covariates, which is especially informative for mitigation and adaptation, and linking scientific exploration and practical solutions.

AbstractIn response to the “Leave No One Behind” principle (the central promise of the 2030 Agenda for Sustainable Development), reliable estimate of the total number of citizens living in slums is urgently needed but not available for some of the most vulnerable communities. Not having a reliable estimate of the number of poor urban dwellers limits evidence-based decision-making for proper resource allocation in the fight against urban inequalities. From a geographical perspective, urban population distribution maps in many low- and middle-income cities are most often derived from outdated or unreliable census data disaggregated by coarse administrative units. Moreover, slum populations are presented as aggregated within bigger administrative areas, leading to a large diffuse in the estimates. Existing global and open population databases provide homogeneously disaggregated information (i.e. in a spatial grid), but they mostly rely on census data to generate their estimates, so they do not provide additional information on the slum population. While a few studies have focused on bottom-up geospatial models for slum population mapping using survey data, geospatial covariates, and earth observation imagery, there is still a significant gap in methodological approaches for producing precise estimates within slums. To address this issue, we designed a pilot experiment to explore new avenues. We conducted this study in the slums of Nairobi, where we collected in situ data together with slum dwellers using a novel data collection protocol. Our results show that the combination of satellite imagery with in situ data collected by citizen science paves the way for generalisable, gridded estimates of slum populations. Furthermore, we find that the urban physiognomy of slums and population distribution patterns are related, which allows for highlighting the diversity of such patterns using earth observation within and between slums of the same city.

Earth Observation (EO) data provides valuable information to localize and monitor deprived areas for the assessment of Sustainable Development Goals (SDGs). We propose a semantic segmentation model that uses a regression output to an important engineered feature as a guide to the weights learning process of the model.

Light pollution has increased globally, with 80% of the total population now living under light-polluted skies. In this Review, we elucidate the scope and importance of light pollution and discuss techniques to monitor it. In urban areas, light emissions from sources such as street lights lead to a zenith radiance 40 times larger than that of an unpolluted night sky. Non-urban areas account for over 50% of the total night-time light observed by satellites, with contributions from sources such as transportation networks and resource extraction. Artificial light can disturb the migratory and reproductive behaviours of animals even at the low illuminances from diffuse skyglow. Additionally, lighting (indoor and outdoor) accounts for 20% of global electricity consumption and 6% of CO2 emissions, leading to indirect environmental impacts and a financial cost. However, existing monitoring techniques can only perform a limited number of measurements throughout the night and lack spectral and spatial resolution. Therefore, satellites with improved spectral and spatial resolution are needed to enable time series analysis of light pollution trends throughout the night. Increasing light emissions threaten human and ecological health. This Review outlines existing measurements and projections of light pollution trends and impacts, as well as developments in ground-based and remote sensing techniques that are needed to improve them.