Conference Agenda

While urban populations grow, cities are ultimately confined in space, needing to accommodate diverse social, ecological, and economic functions. Cities worldwide face the challenge of creating integrated urban environments that balance growth ambitions with new standards for green growth, promoting biodiversity, mitigating climate change, and supporting inclusiveness and quality of life.

Urban Nature-Based Solutions (NBS) offer a multifaceted approach to addressing complex urbanization challenges. As cities grapple with limited space amidst burgeoning populations, NBS emerge as indispensable tools for fostering sustainable development. Monitoring and evaluating the impact and potential of NBS activities are inherently challenging due to the complexity of urban environments and the dynamic nature of these solutions. Herein lies the value of EO technology, offering a bird's-eye view of urban landscapes and facilitating continuous monitoring at various scales. EO enables the systematic collection of high-resolution spatial data, providing insights into vegetation dynamics, land use changes, and environmental conditions over time. EO enables near real-time responsiveness to environmental shifts and evaluation of NBS effectiveness, enhancing the resilience of NBS interventions in the face of urban challenges such as climate change and population growth.

Based on the results of a UNEP funded urban NBS activity, we will illustrate how EO enables near real-time responsiveness to environmental shifts and evaluation of NBS effectiveness, hence enhancing the resilience of NBS interventions in the face of urban challenges such as climate change and population growth. We will shed light on the technology and provide practical use cases from around the world for the applied use of EO to underpin urban green management and planning, emphasizing how modern EO technology can be used to create and maintain an accurate and updated urban information.

World from Space is developing UpGreen, a service that accurately assesses, predicts, and proposes urban green infrastructure using advanced Earth Observation and geospatial methods. A comprehensive understanding of the dynamics of urban greenery is ensured by using multi-sensor, multi-resolution, and multi-temporal approaches. The service comprises three subsequently interdependent modules: UpGreen Assessment, UpGreen Prediction, and UpGreen Vision.

UpGreen Assessment utilizes multispectral data and beyond, to provide detailed delineation, segmentation, and allocation of attributes and functions of urban green spaces. The information gathered per green segment shall include for instance: greenery type, vitality, density, height, biomass, cooling effect, connectivity, accessibility, air cooling efficiency, air pollution blocking, shaded area cast and others.

UpGreen Prediction utilizes advanced AI algorithms and vast amounts of EO and other data to forecast future scenarios for urban green infrastructure with confidence.

UpGreen Vision provides actionable insights for optimal urban green planning based on the city's preferred ecosystem services targets. The recommendations include suggesting the most effective greenery placement distribution, types and quantities to maximize environmental and socio-economic benefits.

The service is a technical response to domain requirements gathered in the preceding ESA Feasibility Study. In summary, those are (1) holistic understanding and strategic planning of urban green, (2) trend analysis and forecasting urban green health, (3) data interoperability for better stakeholder engagement. UpGreen demonstration pilot is currently being developed within an ESA project: Development and Verification of Urban Analytics (4000143727/24/I-DT). The business model with go-to-market activities and first partnerships are already set up and fully operational commercial product-as-a-service is scheduled to be completed after the end of the project. A consortium partner ASITIS will be UpGreen's product manager.

UpGreen will assist cities in making informed decisions towards sustainable urban development by enhancing ecosystem services, urban resilience, and citizen well-being through efficient nature-based solutions.

Satellite image processing engineers, biodiversity experts, and artificial light experts in Cerema (Center for Studies and Expertise on Risks, Environment, Mobility, and Urban Planning) worked together to better quantify the pressure put by Artificial Light At Night (ALAN) on biodiversity. The purpose is to identify priority areas to work on public and private lighting. Indeed, most animals are active at night and sensitive to ALAN. Preserving and restoring an ecological network supportive to nocturnal wildlife is imperative.

Satellite imagery shows a substantial potential to map ALAN as it covers large territories several times per year, at different spatial resolution (global, national, regional scale, or even individual lighting sources thanks to very high resolution).

LuoJia 1-01 is a Chinese experimental satellite that has taken night-time images over France in 2018. Its 130 m spatial resolution enables to study ALAN at a neighbourhood level. These freely available radiance data have a low level of processing and must be orthorectified before use. Clouds also need to be detected and masked.

Cerema processed these 217 images over France and combined them to build a nearly national map of ALAN, produced at the departmental level. Some departments are missing because of the lack of acquisitions, or the dense cloud cover. Orthorectification was achieved with Ground Control Points (GCP) from the Copernicus High Resolution Layer (HRL) Impervious Built-Up (IBU), as ALAN and imperviousness were shown to be strongly correlated.

The nearly national map of France ALAN produced will be freely available for all public entities. Such a map can be produced over other European countries, depending on the available LuoJia 1-01 data.

This map is interesting for biodiversity and lighting experts. Facing these data with other sources (local taxes database or biodiversity maps) can provide even more valuable information.

Green Urban Sat is a two-year project (2022-2024) labelled and co-financed by the Space for Climate Observatory (SCO) and conduced by Cerema (Center for Studies and Expertise on Risks, Environment, Mobility, and Urban Planning) in collaboration with LIVE/A2S laboratory, TerraNIS company, and the city of Nancy. In this project, methods and tools were developed in order to generate a geospatial database describing precisely the urban vegetation cover, which will help to better evaluate some ecosystem services. The urban environment is highly disparate and used to be set aside due to spatial resolution issues. Urban vegetation maps are usually very basic, and only made of two classes : high and low vegetation. With the arrival of Very High Resolution (VHR) satellites such as Pléiades, it is now possible to infer urban issues with stronger accuracy. Green Urban Sat framework discriminates three vertical layers of vegetation, then detects several vegetation formations (isolated tree, wood, narrow band of trees, etc.).

The project is based on stereoscopic Pléiades imagery, acquired at different periods of the year, enabling to derive several indicators describing the vegetation (height, surface, orientation, vegetation type, landscape type, NDVI, etc.). This set of quantitative attributes and indicators will allow to feed a group of decision aid tools for municipalities. A demonstrator is being produced over the city of Nancy (France), and the method is duplicable over other cities.

Urban growth prediction is essential for sustainable urban planning, requiring accurate and reliable models. Satellite-based Earth observation (EO) time series data offer valuable insights into past and allow to conclude on future trends of urban development. However, existing models often struggle to incorporate detailed local information, leading to inaccuracies in growth predictions, or require plenty additional datasets. To address this challenge, we utilize the Spatio-Temporal Matrix (STM) approach, which leverages EO data to generate spatial and temporal characteristics of neighborhoods.

In this study, we applied the STM-based model coupled with a multi-layer perceptron (MLP) for settlement growth prediction in Huê (Vietnam), Surat (India), Ho-Chi-Minh City (Vietnam), and Abidjan (Ivory Coast). Our research aims to assess the model's ability to predict settlement growth while avoiding restricted or intra-urban areas without incorporating additional datasets besides multitemporal settlement layers (World Settlement Footprint – WSF).

Using the WSF-evolution dataset, we achieved promising results, indicating the STM-based model's effectiveness in settlement growth prediction. Compared to baseline results of the SLEUTH model, our approach produced more realistic growth patterns, minimizing predictions in restricted areas without the need for additional layers.

This study highlights the potential of the STM-based model as a reliable tool for urban growth prediction based on EO information products, offering enhanced accuracy and independence from external datasets. By providing insights into future urban development while respecting local constraints, our approach contributes to more sustainable urban planning practices.

Urban areas necessitate effective management strategies to mitigate natural risks and protect communities. By harnessing Sentinel-1 radar images, the European Ground Motion Service (EGMS), an integral component of the Copernicus program, provides comprehensive insights into ongoing ground deformation processes with high spatial resolution and precision of the measurements.

The EGMS is produced over the Copernicus participating countries and is updated annually. The most recent data release covers the period January 2018 – December 2022. The product is made available to users under an open data policy through a dedicated data visualization, interaction and download platform: the EGMS Explorer. Every year, the EGMS produces a massive amount of ground motion data, equal to 13 billion measurement points, each containing a value of the ground motion velocity, a time series describing its evolution over time, and a series of quality parameters.

The EGMS is a deferred-time, multi-purpose mapping, and – to a certain extent – monitoring tool for active ground motion in urban areas. The density of measurement points is the highest in the urban environment, where the so-called permanent scatterers are abundant (buildings, bridges, and man-made structures in general). The EGMS shall be intended as a baseline product, which shall be complemented by in situ or other remotely sensed measurements to achieve the high level of precision of measurements that is sometimes required to assess a single building or linear infrastructure. Nonetheless, the EGMS is certainly a powerful resource to spot areas of ongoing deformation over an entire city, region, or country.

The presentation will showcase some EGMS use cases to convince existing and new users about the added value of the product in enabling the identification of areas prone to instability, the improvement of decision-making processes, the prioritization and cost-saving of, and the implementation of targeted land-use regulations.

Nowadays, urban environments encounter a plethora of diverse challenges and threats due to the effects of Climate Change (CC). The European Cities, are constituted as significantly sensible environments with high susceptibility to threats such as increased temperatures (Urban Heat Island Effect), air quality degradation, extreme weather phenomena, urban flash floodings and an overall increased risk of public health issues. To this end, Earth Observation (EO) products and applications serve a pivotal role in the timely monitoring of urban environments by providing an analytical perspective of several urban-related variables. The combination of EO with novel digital tools and Decision Support Systems (DSS) could play a determinant role in assisting decision-makers by means of providing access to relevant data and insights. The Horizon 2020 research and innovation project entitled “Development of a Support System for Improved Resilience and Sustainable Urban areas to cope with Climate Change and Extreme Events based on GEOSS and Advanced Modelling Tools - HARMONIA GA 101003517” introduces a series of innovative digital tools along with a novel urban planning DSS that exploits the Harmonia multiparametric risk assessment methodology for a spectrum of different urban perils to eventually offer comprehensive and tangible urban recommendations for mitigating future hazard-driven adverse impacts. The proposed solution can exploit dynamically updated EO services for a series of urban perils and is offered as a web-based application with a user-friendly interface, capable to handle and visualize multidimensional (4D) geospatial information. The overall methodology and the capabilities of the system are demonstrated in four different and diverse European urban environments, i.e., the cities of Milan, Piraeus, Sofia, and Ixelles by prioritizing tangible recommendations for the most effective mitigation strategies.

Local Climate Zone (LCZ) classification and satellite image processing for mapping Urban Heat Islands (UHI) offer promising prospects for improving the cities resilience to climate change by identifying sensitive areas and proposing recommendations for urban greening strategies and promoting a more efficient and sustainable approach to urban planning.

Our prototype of LCZ generation is based on a method combining the vector approach developed by Olivier Montauban at city scale in Strasbourg in 2019 and the LCZ classification algorithm of Cerema (Cerema Sat’ 2021), after having evaluated all the needed data at the level of Aix-en-Provence and Florence. To ensure reproducibility, accessibility, and availability of data in both study areas, the retrieving of morphological indicators and Land Use and Land Cover relies mainly on raster and vector data in Open Source except for Pléiades data.

Using satellite images from Landsat 8, Aster, and MODIS, we analyzed the land surface temperature (LST) and UHI during summer (June to August) 2022 and 2023 in the conurbation areas. The objective was to evaluate the impact of these variables on the variability of urban hot spots (UHS) and on the level of thermal comfort, using the Urban Thermal Field Variance Index (UTFVI) in each type of Local Climate Zone.

This study demonstrates the complex links between LCZ, LST, UHI and UHS. The spatio-temporal evolution of LST provides information on areas that are particularly likely to become UHI in the future.

The Urban Atlas product of the Copernicus Land Monitoring Service help monitor and understand urban areas and support effective urban planning and policy making.

The service is enhancing the Urban Atlas suite of products for the reference years 2021-2024.

The upgraded suite will include the traditional land cover and land use (LC/LU) status and change layers 2018-2021 and 2021-2024 and Building Block Heights (BBH) for 2021 and 2024.

A new integration within the Green Urban Areas class will be applied on the LC/LU product for 2021 and 2024, and it will allow users to distinguish between public and private green spaces using an innovative approach based on available in-situ and ancillary data. The Street Tree Layer product will continue to be part of the product suite but for reasons of consistency it will be extracted from the Small Landscape Features component of CLMS.

The methodology applied for the creation of the new products involves integrating Sentinel-2 time series for extracting information on change and basic land cover classification and comparing the resulting product with Very High Resolution data for detecting small changes and precise outline. As fully automated, the methodology allows for a reduction of the update cycle from 6 to 3 years.

The distinction of the public vs private character of Green Urban Areas will allow the benefits of inner-city green areas for urban ecology and quality of life to be properly assessed, overcoming a limitation in the previous Urban Atlas concerning the determination of whether a green urban space is public or private. This new layer is also a direct response to requirements expressed by users in the preparation for the 2021 update.

The upgraded Urban Atlas products are designed to address the critical need for up-to-date, detailed information on urban expansion, land use changes, and environmental shifts within urban areas.

In addition to the technological advancements, CLMS is committed to ensuring these products are accessible and beneficial to a wide range of users. To ensure this, the project includes a robust user engagement program, featuring training webinars, presentations at external events, and a dedicated Helpdesk for support.

The World Bank is interested in conducting least-cost electrification studies in developing countries with a view toward universal electricity access. Accurate and up-to-date knowledge of existing electrical transmission grid infrastructure is required for this purpose. To improve the quality of this data NEO has developed a novel

smart-tracing algorithm to detect and trace electrical towers in Very High Resolution (VHR) satellite imagery. This smart-tracing approach uses existing open datasets alongside a deep learning model for object detection. The method is scalable and adaptable to arbitrary regions with satellite image coverage.

The method makes full use of existing open datasets such as nightlight satellite imagery and land use map etc. as inputs to derive probability map of transmission grid presence. This probability map guides the smart tracing algorithm to search and determine the power tower tracing direction. The power tower detection is done by a well-trained deep learning model. The method is designed to be adaptive to input imagery and be flexible to handle regional difference in terms of landscape variation and dataset availability, which makes it highly feasible to replicate to other regions in the world provided satellite imagery coverage.

In an era marked by rapid urbanization and escalating environmental challenges, the imperative for sustainable urban development is critical. Central to this challenge is the strategic development of urban green spaces, serving as city lungs and enhancing urban resilience and community well-being.

"Let’s Park", a Greek NGO, leads this movement, advocating for the creation and expansion of urban green areas. It distinguishes itself by engaging a diverse array of stakeholders, from individual citizens and community groups to businesses and local government bodies. Utilizing a bottom-up approach, "Let’s Park" leverages participatory design and co-creation, alongside modern technologies like satellite remote sensing, artificial intelligence, and advanced web technologies, to drive its initiatives.

The organization's efforts unfold through two main avenues. Firstly, it has launched a comprehensive crowdsourcing web platform connecting citizens and initiatives with municipal authorities and company ESG departments. This platform fosters a seamless exchange of ideas, resources, and support, ensuring projects are community-led and aligned with environmental and social objectives. Secondly, "Let’s Park" provides Earth Observation (EO) services, such as Land Use/Land Cover and urban sprawl mapping, crucial for urban planning and decision-making, enabling effective monitoring and impact assessment of urban parks.

Integrating crowdsourcing with EO technologies, "Let’s Park" enhances urban environments and redefines urban planning and environmental management paradigms. This collaborative approach advances the urban greening agenda and aligns with global sustainability goals. "Let’s Park" serves as a scalable, replicable model for developing resilient, healthy, and livable cities in the climate change era, setting a benchmark for future urban development worldwide.

Persistent Scatterer Interferometry (PSI) is a well-established technique in the field of multitemporal A-DInSAR (Advanced Differential Synthetic Aperture Radar Interferometry). The strength of PSI lies in its ability to detect and monitor ground displacements over large areas with sub-centimeter accuracy. However, the effectiveness of PSI in identifying deformation phenomena depends on the available PS density, which is related to the sensor resolution and site-specific characteristics.

To overcome the inherent limitation of the PSI technique, our study proposes an integrated analysis that combines displacement information extracted by multi-band SAR satellites for evaluating landslides in the Central Apennines, Italy. This area, affected by the 2016-2017 seismic sequence, requires a comprehensive understanding of landslide-related deformations and their interaction with urban areas to provide valuable insights for risk assessment and mitigation strategies. Based on the strain tensor, the developed data fusion method combines data with different orbital geometries to obtain synthetic datasets characterized by the integrated deformation velocities along the horizontal and vertical components. Compared to the single-sensor PS approach, the method identifies deformation patterns that might not be visible from an individual data source, grouping data with similar movement patterns over time to increase spatial coverage and reinforce information content. The multi-sensor analysis provides insights into the underlying causes of the processes by identifying areas experiencing similar deformation.

Landslide hazard data were combined with buildings’ vulnerability and real-estate market value to achieve a comprehensive risk assessment. Structural resistance estimates retrieved physical vulnerability, while the economic value was calculated through official government estimates of asset market values. By combining multi-sensor hazard data with asset market value and vulnerability estimates, our study aims to provide an integrated approach to landslide risk analysis, enhance disaster resilience, and inform urban planning practices of the selected test sites.

Due to the Sustainable Development Goals (SDGs) being designed at both national and globally applicable level, it is challenging to adequately reflect the local context and characteristics in different urban regions without fully utilising big earth data. To effectively address this issue, this work localized 73 indicators for the 13 SDGs, and utilizing big earth data, conducted a comprehensive assessment and prediction of the urban sustainable development status in 18 cities in Hainan province from 2010 to 2030. Our analysis specifically focused on indicator score, goal score, SDG index, SDG spatial spillover effect, and trade-offs and synergies. The results indicated an overall upward trend in sustainable development in Hainan province, predicting achievement of the SDGs by 2030. The SDG index score and spatial spillover effect showed a pattern of 'high in the north and south, low in the middle'. Although the SDG synergies are generally stronger than trade-offs, the trade-off effects develop at a faster pace. More specifically, the average SDG index of each city increased from 29.85 in 2010 to 60.09 in 2018, with a projected score of 89.76 by 2030. During 2010-2018, the synergy-to-trade-off ratio declined from 3.91 to 1.84, driven by a trade-off growth rate 2.03 times higher than the synergy. Based on the SDGs imbalances, governmental policies should foster essential transformations across sectors for sustainable development. Our work provides a valuable localized case method, and data support for monitoring sustainable development at the global urban level.

Characterizing Built-up Areas (BA) is crucial in making cities and human settlements safe, resilient, and sustainable, supporting the Sustainable Development Goal (SDG #11). Synthetic Aperture Radar (SAR) data is a potent resource for BA mapping due to strong coherent backscatter from diverse human-made targets, distinct texture patterns, and sensitivity to its geometric characteristics. With the advent of the Sentinel-1 C-band SAR mission, dual polarimetric (dual-pol) SAR data has been widely exploited for several land cover applications. These data sets provide wide-swath SAR images at an impressive spatial and temporal resolution with a distinct cross-pol response of BA rotated to the radar line of sight (LOS).

This study exploits these advantages of dual-pol SAR data by introducing a set of descriptors that are helpful in the enhanced characterization of BA. (1) a built-up index from Single Look Complex (SLC) and Ground Range Detected (GRD) dual-pol SAR data: DpRBIS and DpRBIG, (2) a target characteristic parameter from dual-pol SLC and GRD SAR data: α(S) and α(G), and (3) scattering power components from dual-pol SLC and GRD SAR data: Pd−l, Pu, and Ps−l.

The descriptors obtained from SLC SAR data are capable of characterizing different types of built-up areas in diverse scenarios, overcoming the significant challenge of inaccurate identification of BA or buildings oriented to radar LOS and mixed BA. However, descriptors derived from GRD SAR data may pose certain challenges in identifying the oriented and mixed BA. The DpRBIS and DpRBIG range between 0 and 1, with BA having contrasting high values than other land cover targets. Similar significant variations between the values of α(S) and α(G) is observed for built-up (close to 90◦) and non-built-up areas. Likewise, we observe that the “dihedral-like” (Pd−l) power component predominates over built-up targets, facilitating its discrimination from other land cover targets.

The ERC Synergy (ERC-SyG) Project urbisphere aims to forecast feedbacks between weather/climate and cities, by exploiting new synergies between spatial planning, remote sensing, modelling and ground-based observations, and incorporating city dynamics and human behaviour into weather and climate forecasts/projections. The urbisphere field campaign in Berlin, Germany, provides new information on the impact of cities on the urban- and regional-scale boundary layer using data measured across a wide range of scales during the course of a full year (Autumn 2021 to Autumn 2022).

During an intensive thermal infrared (TIR) observation campaign in August 2022, sensors included five TIR cameras (Optris 640 Pi and Optris400 Pi) mounted on the ground and a building roof, SatelliteVu MIR (Mid-Infrared) sensor mounted on an aircraft, and Anafi Parrot Thermalsensor mounted on an UAV (Unmanned Aerial Vehicle). These measurements were complemented by satellite observations from Sentinel-3 SLSTR, MODIS, ASTER, ECOSTRESS and Landsat. Thus, data from the intensive observation period (IOP) offer a wide range of spatial resolutions (<1 m to 1 km), many collected over the same location and many at the same time. The sensors differ in their respective fields of view, their wavelengths, and their accuracies. In this contribution, we provide an overview of the TIR and MIR observations, their spatial and temporal coverages, and initial results for evaluating the spatial and temporal variability of surface temperature during the IOP.

Acknowledgement

This work is part of the urbisphere project (www.urbisphere.eu), a synergy project funded by the European Research Council (ERC-SyG) within the European Union’s Horizon 2020 research and innovation program under grant agreement no. 855005. Special thanks to the Chair of Climatology at Technische Universität Berlin for providing equipment, ensuring access to observation sites and to all those who contributed to the field work: Fred Meier, Kai König, Josefine Brückmann.

In the context of aging infrastructure, limited funding for comprehensive inspections, and the escalating risks associated with extreme weather events, evaluating the structural integrity of numerous bridge assets is a challenging task for transport managers. Recent progress in Synthetic Aperture Radar (SAR) satellites and Interferometric SAR (InSAR) techniques has led to cost-effective and high-quality measurements of infrastructure deformations. Specifically, Multi Temporal (MT) InSAR can detect displacements at the millimetre level, providing monitoring data comparable in accuracy with conventional terrestrial methods. Moreover, MT-InSAR offers broader spatial coverage, frequent updates, operates in all weather conditions, provides day-and-night acquisitions, and allows for retrospective monitoring. As a result, MT-InSAR holds the promise of complementing traditional methods for assessing bridge conditions, particularly in regional-scale evaluations. However, integrating MT-InSAR observations with an understanding of structural behaviour remains challenging due to SAR's inherent geometric differences from traditional monitoring systems. To address this, we propose a method that considers anticipated asset motion and its alignment with the satellite flight path to assess MT-InSAR sensitivity towards expected displacement directions and establish specific damage indicators. The proposed method is used for bridge analysis in the Netherlands, utilising displacement data derived from a temporal series of 3-m resolution TerraSAR-X imagery. Findings can enhance our understanding of structural behaviour and aid in proactive maintenance, ultimately contributing to more resilient infrastructure systems.

The European Commission (EC) Knowledge Centre on Earth Observation (KCEO) undertakes deep dive (DD) assessments on selected EU policy areas. DD methodology consists of the following steps: 1. definition of key actors involved, 2. collection and assessment of needs of policy Departments across the entire policy cycle, 3. Earth observation (EO) value chain analysis, 4. translation of policy needs into technical requirements, 5. fitness for purpose analysis of existing products and services and 6. gap analysis and final report production including recommendations and relevant information to contribute to filling the knowledge gaps. One of the DDs that KCEO is currently undertaking is on urban climate change adaptation. As a result of co-design activities carried out with policy Departments, four use cases were identified:

1. the role of urban greening for managing flood and drought risks and mitigating the urban heat island (UHI) phenomenon, in relation to the Biodiversity Strategy for 2030, proposed by DG ENV

1. monitoring temperature and its impacts (floods, drought and UHI) at a temporal and spatial resolution adapted to Cohesion Policy 2021-2027 reporting, proposed by DG REGIO

1. coastal flooding, coastal eutrophication, coastal erosion and geomorphological change and coastal marine heat waves, in relation to the Marine Spatial Planning Directive, proposed by DG MARE

1. compact cities and function of cities in strategic corridors in Africa and measuring and mapping urban vulnerability to make cities resilient in EU partner countries, to support the Urban Development Technical Facility in the context of Global Gateway, proposed by DG INTPA

In addition to the use cases, this DD touches upon the relevant EU Missions and defines a set of indicators for urban climate change adaptation, defining the role of EO, the related Key Type Measures and risk components and considering both the climate change impacts and adaptation action outcomes.

Nature-based solutions in urban environments can provide cooling effects, decrease air pollution, and improve mental health, amongst others important ecosystem services and health-related benefits. Ambitious plans, such as the pledge to plant 3 billion trees in the EU, the European Green Deal, or the Green City Accord support this direction. Their implementation, however, requires transformative changes on the ground to overcome business as usual approaches. The Urban ReLeaf project delivers change by bringing public authorities and citizen groups together to shape urban green infrastructure actions. Six pilot cities co-create citizen-centric innovations for the democratisation of urban greenspace monitoring and the wider policy making process in pursuit of urban climate resilience. This presentation showcases the stories of the six cities and their approaches to designing citizen-powered and multi-sensor data ecosystems to support decision making. Athens is undergoing a greening transformation with a new tree registry the combines Earth Observation (EO) and crowdsourcing methods to provide critical data for better management of greenspaces. Cascais engages citizens in sharing perceptions and thermal comfort levels while using greenspaces to validate the effectiveness of its parks. Meanwhile in Dundee, a city facing increasing grey infrastructure in deprived areas, actions to enhance the accessibility of greenspaces are co-developed with citizens and stakeholders. Mannheim has a heat action plan to safeguard its most vulnerable residents but has identified critical data gaps. Citizen observations of trees and thermal comfort, when integrated with EO and official data streams, will aid the delivery of climate adaptation measures. Riga engages diverse audiences to address concerns about air quality and greenspace usage, through the use of low-cost sensor networks. Finally, in Utrecht, data on temperature, humidity and heat stress, collected by and for citizens, will help them reduce the urban heat island effect and shape effective mitigation strategies.

In urban environments, nature-based solutions and the integration of natural capital into planning are crucial to creating climate resilient cities. These strategies act as carbon sinks, improve biodiversity and overall well-being, providing cleaner air, reducing heat and mitigating flood risk. Incorporating these elements into urban planning ensures sustainable development and climate resilience, highlighting the importance of enhancing and protecting our urban ecosystems and forests.

Lobelia Earth responds to these environmental challenges with an innovative platform that leverages web technology, satellite data from various sources (optical, radar and LiDAR) and climate science. Through big data analysis methods and the application of artificial intelligence and landscape ecology analysis methods, we measure key indicators such as soil moisture, aboveground biomass (AGB), phenology, fragmentation and connectivity. This multifaceted approach permits an accurate assessment of urban biodiversity and natural capital, facilitating informed decision making for sustainable management and climate change mitigation in urban and peri-urban forests.

These efforts are essential to address environmental degradation and promote sustainable growth that respects the limits of our natural resources. Effective measurement and management of human impact on ecosystems is vital to mitigate deforestation, ground degradation and biodiversity loss. Integrating natural capital into corporate and economic decisions permits sustainable practices that benefit both the medium and economic development, contributing to the global fight against climate change and ensuring a sustainable future and responsible environmental management.

HeatScape Resolve offers a comprehensive solution for recognizing and predicting urban heat islands (UHI) using Earth observation (EO) data, benefiting real estate developers and municipalities. It operates through three key stages: assessing the current state of UHI, simulating UHI changes post-urban development, and validating UHI post-development. This process yields localized predictions of UHI intensity tailored to specific local climate zones (LCZs), aiding in reducing building cooling energy loads and facilitating sustainable urban public space planning through detailed microclimate mapping. The study elucidates how EO-derived urban scape attributes inform inputs for predictive simulation models and outline user requirements for the service. Additionally, it will demonstrate the integration of EO aided predictive UHI services for city-scale sustainability assessments.

The commercial development activity is performed under a programme of, and funded by, the

European Space Agency and is carried out under the ARTES BASS programme.

Launched in October 2023 and funded by ESA's FutureEO programme, the Synthetic Data for Earth Observation (SD4EO) project wants to pioneer the use of synthetic satellite imagery to address a critical challenge in Earth Observation applications: the scarcity of comprehensive, accurately labelled reference datasets. The project, led by GMV in partnership with the University of Valencia, leverages advanced Computer Graphics-based simulation techniques and generative Artificial Intelligence (AI) to generate high-quality synthetic satellite imagery that mirrors real-world data with precise annotations. This innovative approach aims to substantially minimise the need for time-consuming manual labelling, while enhancing the precision and availability of reference data for training AI models in EO applications.

The SD4EO project addresses three use cases, two of which are related to sustainable urban development challenges. The first regards the human settlements categorisation for energy consumption monitoring and management. By producing synthetic images that accurately represent various urban structures, SD4EO wants to contribute to a more effective global energy demand mapping using Sentinel-2 imagery. The second use case focuses on the monitoring of photovoltaic panels. As solar energy becomes increasingly integral to urban landscapes, there is a growing need for up-to-date and accurate data on residential solar installations. The project generates synthetic imagery of photovoltaic panels under varying conditions to assist in detecting and evaluating their status. This initiative is beneficial for optimising solar energy use, informing energy policy, and managing power grid needs.

In addition, the SD4EO project aims to encourage cooperation between the AI, Computer Graphics, and EO scientific communities. The synthetic datasets and the associated code will be made available under open-friendly licences to foster innovation. This pragmatic strategy seeks to streamline EO analytics, setting a precedent in using synthetic data to overcome the limitations of conventional EO data annotation and analysis techniques.

Responding to global warming and adapting to climate change effects such as heat waves and drought is a key priority of European and national-level Climate Change Adaptation strategies. Regional and city administrations aim to reduce climate change-related health risks and increase human well-being through adequate planning measures such as establishing green and blue infrastructure. Changes in land use (LU) and land cover (LC) play an important role in determining local climate characteristics. Urban Climate, for instance, differs from the surrounding natural areas, showing higher air and surface temperatures, known as the Urban Heat Island Effect, mainly related to changes in the surface radiative properties. By leveraging LULC data, Sentinel 2 data, meteorological data, climate models and other auxiliary datasets and integrating Land surface temperature (LST) stemming from ECOSTRESS, we developed a multi-sensor/data multi-resolution downscaling algorithm grounded in the physical representation of LST [1, 2].

Our methodology leverages a super-pixel Convolutional Neural Network (CNN) architecture in two pivotal steps: firstly, modelling LST at its sensor-specific resolution to create a dense time series (e.g., 70m in the case of ECOSTRESS), secondly, the further refinement and downscaling to 10m resolution. Employing the GHSS2Net-derived model, optimized for a 5x5 super-pixel input, using 2x2 convolutional kernels without pooling layers to preserve contextual information, our downscaling approach achieves significant advancements in spatial prediction accuracy (R² ≳ 0.9) without sacrificing temporal consistency [3]. This efficiency is attributed to the model's design, which prioritizes contextual information retention through its unique convolutional structure and employs dropout regularization and batch normalization to enhance performance.

The integration of downscaling techniques into spatial planning activities while considering various climate scenarios provides a novel avenue for assessing and visualizing the impacts of urban development and LULC changes on local climates. By enhancing the resolution and accuracy of LST and climate data, our methodology supports the development of targeted, evidence-based urban adaptation and resilience strategies and facilitates policy development and communication with the broader public.

Our contribution would be in poster format.

[1] Matzarakis, A., Rutz, F. & Mayer, H. Modelling radiation fluxes in simple and complex environments: basics of the RayMan model. Int. J. Biometeorol. 54, 131–139 (2010).

[2] Rigo, G., Parlow, E. Modelling the ground heat flux of an urban area using remote sensing data. Theor. Appl. Climatol. 90, 185–199 (2007).

[3] Corbane, C., Syrris, V., Sabo, F. et al., Convolutional neural networks for global human settlements mapping from Sentinel-2 satellite imagery. Neural Comput & Applic (2020).

Cities face many challenges in terms of health, well-being and liveability. In order to make the right decisions municipalities, citizens and other stakeholders require good quality information on our habitat. Remote sensing data are very suitable for deriving both complete and up to date information. NEO has developed standardized services to monitor the urban habitat. These services provide object-based information on trees, pavement and buildings, on a national level and are an addition to the available toolboxes such as the Copernicus Land Monitoring Services. The services provide essential elements for the effective application of a digital twin approach.

Using laser altimetry data, aerial imagery, satellite data and AI techniques a national tree register has been developed that contains all trees in the Netherlands. The added value is that the registry is more complete and up to date than other registries. This completeness is important for addressing climate and heat stress issues. The registries are kept up to date using multiple acquisitions per year of very high resolution optical satellite imagery (a.o. Pleiades Neo).

With this portfolio of services actionable insights for climate adaptation and liveability are derived. Insights into buidling density, proximity of trees and nearest green area (including distance, per building) and coverage of tree crowns per neighbourhood are indicators that serve policymakers in assessing liveability. The monitoring capabilities of the services enable monitoring indicators and policy outcomes in the future (or the past, using historic data).

The services are also applied by other stakeholders involved in managing urban affairs, such as utility companies, contractors and engineering companies. Good quality data helps improving infrastructure and public space whilst preserving the trees as important elements in a healthy habitat.

The standardized services have international potential and are in the process of being applied internationally, e.g. France.

Urban green infrastructure plays a pivotal role in climate regulation by offering various ecosystem services. One crucial metric in understanding human thermal exposure is the mean radiant temperature (Tmrt), which encompasses the spatial and temporal variations of radiation exposure. In the context of urban microclimate models like SOLWEIG, the accurate characterization of trees is essential, whether incorporating existing trees or assessing the cooling effects of new greenery. Currently, urban tree structures are usually generalised inside of the model due to the lack of detailed measurements and scientific knowledge about urban tree growth.

Various vegetation types exhibit distinct effects on the attenuation of direct shortwave radiation through shading. Leaf Area Index (LAI), tree height, and trunk height significantly determine shade patterns and solar attenuation. This abstract proposes the utilisation of state-of-the-art Terrestrial LiDAR Scanning (TLS) techniques to parameterize these structural properties for the precise implementation of existing trees within urban microclimate models.

This enhanced structural understanding of urban trees will facilitate the creation of more realistic tree models, allowing for a comprehensive assessment of their impact on human thermal comfort.

SOLWEIG operates as a 2.5-dimensional model, where x and y coordinates and associated attributes (e.g. height, emissivity or reflectivity) are utilised for the calculation of Tmrt. TLS allows for the highest degree of parameterisation of urban trees within the given raster environment. By conducting a sensitivity analysis on the modelled Tmrt, we will explore the impact of tree and trunk height, canopy area and volume, and radiation transmissivity of vegetation.

This research will provide valuable guidance on the TLS data collection of tree parameters essential for evaluating current cooling effects. Which in turn leads to the identification of tree species with significant cooling potential, and determining the size at which a tree substantially contributes to human thermal comfort.

The world is constantly changing and becoming more complex in all aspects. The changes have unpredictable impacts and implications at various scales. The increasing complexities in urbanized systems pose challenges in their comprehension and management. The intricate structures and large spatial scales make visualization and analysis arduous. Conventional methods may fall short in accurately representing the situation, thereby impeding detailed analysis and decision-making processes. Consequently, communities seek new synergies to access timely, updated, standardized, reliable, user-friendly, and actionable information to make well-informed decisions. There is a pressing need for innovative monitoring systems to streamline and simplify urbanization processes. It is imperative to democratize information, integrate it into society, and ensure its accessibility to all. Earth Observation (EO)-based space techniques have become more comprehensible, accessible, and dependable for analyzing Earth resources and monitoring urban environments. The outcomes of numerous application-oriented studies have been promising thus far. This research primarily addresses the requirements of public and research organizations, with a particular focus on utilizing Imaging Radar Systems for spatial feature extraction and remote mapping of sensitive areas in Turkey. The major concern is the expanding urbanization. The study's multidisciplinary approach incorporates EO technologies, particularly integrating SAR and optical satellite imagery data to produce detailed maps of land surface features. The study primarily aims to monitor and delineate spatial features of urban developments to enhance understanding of environment, land and sea changes including land use, and analyze their connections to reference data. The basic methodology involves detecting, delineating, recognizing, identifying, and interpreting urban features. Leveraging 3D visualization, color manipulation, and the three-dimensional sensing capabilities of Synthetic Aperture Radar (SAR) data sets facilitates a better understanding of the surface, aiding in discriminating, locating, and mapping meaningful spatial information in the study area. The integration of spatially enhanced SAR and optical imagery data yields significant combined analysis results, providing highly acceptable outcomes for operational use, even in cases where extensive ground studies have been conducted. The imagery data taken from different wavelength bands are used in models that explain the processes controlling the development and configuration of new land surfaces in the region. The SAR and optical data results demonstrated the links between surface developments and remote sensing in the visible, infrared, and microwave spectra. By combining SAR and optical imagery data, land and sea surface features, objects, structures, patches, and changes are effectively mapped. Color composite analysis of SAR-enhanced images enabled optimal extraction of spatial information in the study region. Multi-seasonal and multi-year color composite SAR images highlight changes by displaying them in different color tones. EO-based systems facilitate timely identification, enabling proactive intervention, preventive measures, and documentation of urban development. This approach enhanced the precision and effectiveness of change detection. Multi source imaging systems serves as valuable data for Geographic Information Systems (GIS) analysis and can be a crucial data source for local and state governments, real estate companies, financial businesses, and individuals to make informed decisions. The integration of Space based EO multi source data offered novel information and innovative aspects for land and sea surface mapping studies. As a conclusion, EO systems play a pivotal role in encouraging research and user community activities in the vital domain of urban development mapping and change detection.

The nexus between growing urban population, increased water consumption and exacerbation of land subsidence is well understood. Interferometric Synthetic Aperture Radar (InSAR) data are effective in providing a reliable quantification of the resulting deformation at surface. However, less explored is how such satellite-based information can combine with authoritative statistics on groundwater usage and contribute to improve management practices.

In Mexico, several hotspots of land subsidence are well-known, but a country-wide mapping exercise was not available yet. Furthermore, the aquifer-system management reports issued by the National Waters Commission (CONAGUA) are a valuable resource for understanding the impact of groundwater exploitation. Using ~1700 Sentinel-1 IW SAR images acquired in 2019–2020, we perform the largest ever-made InSAR survey of land subsidence over Central Mexico. Our 700,000 km² study area encompasses the whole Trans-Mexican Volcanic Belt and several major states, hosting >85.2 million inhabitants (i.e., ∼68% of the Mexican population).

Using the parallelized Small BAseline Subset (SBAS) multi-temporal InSAR approach in ESA’s Geohazards Exploitation Platform (GEP), we estimate present-day subsidence rates for ~35.7 million coherent targets and identify yet unmapped and well-known hotspots, e.g.: −45 cm/year in Mexico City. We also compute compaction volume rates at 321 aquifer-systems. These generally correlate well with CONAGUA’s modelled and/or measured groundwater deficits, extractions and storage changes.

We derive semi-theoretical relationships between groundwater balance parameters and land subsidence for the whole Central Mexico and its main hydrological-administrative regions, which enable the assessment of ground compaction rates and volumes resulting from groundwater exploitation, and thus can inform groundwater management strategies towards climate change adaptation and future needs of a growing population. We finally discuss how InSAR-derived subsidence risk maps produced at city level prove valuable to help regional authorities in quantifying properties and population at risk.

Full paper: Cigna & Tapete 2022, Geophysical Research Letters, 49(15), https://doi.org/10.1029/2022GL098923

The Urban Heat Island (UHI) refers to the warmer temperatures experienced by a city compared to its rural surroundings. The Local Climate Zones (LCZ) concept is a suitable description of local scale landscape types used to study the UHI. This description was first published by Stewart & Oke in 2012. This approach allows UHI studies to be more comparable, regardless of prevailing local urban planning, building materials, city size and geographical location. Based on this concept, Cerema has developed, through research projects, the SatLCZ methodology, which is now operational, to determine the vulnerability of urban environments during summer heat waves. SatLCZ uses very high resolution (VHR) satellite images from the Pleiades and SPOT missions. This method, which can be replicated in any city, divides areas into homogeneous elementary typo-morphological units based on their climatic behaviour. The SatLCZ maps will enable local actors to better understand how their territory reacts to the UHI in order to implement solutions in their urban fabric: de-pollution, renovation of the built heritage, planting of vegetation, adaptation of mobility, etc. Moreover, to better support public planning policies in their fight against the urban heat island phenomenon, LCZ mapping can also provide basic indicators such as imperviousness and vegetation cover, but also, if sufficiently detailed data are available, a socio-economic vulnerability index. Finally, this method has allowed the production of a national map covering French cities with more than 50,000 inhabitants. The SatLCZ is now an operational tool that can be used with QGIS3 modellers and is available on the Cerema GitHub

The ONEKANA project addresses the urgent issue of thermal inequalities in African urban areas, exacerbated by climate change. Using Earth observation (EO) technology, including advanced EO/AI models such as Machine Learning and Deep Learning, the project examines the disparate exposure of urban populations in urban Africa, to varying temperatures and extreme heat. By leveraging accessible satellite imagery sources such as Sentinel, Landsat, MODIS and Ecostress, ONEKANA ensures the adaptability and scalability of its methodology in diverse urban contexts. Preliminary results have revealed significant local variations in thermal exposure, delineating clear spatial patterns of heat vulnerability. Simultaneously, the project is pioneering the mapping of slum areas with a systematic approach that fuses open-source EO imagery with unsupervised learning techniques. This method produces refined and up-to-date maps of urban deprivation, circumventing the gaps and unreliability often associated with traditional manual slum delineation. Early analyses highlight the potential of image-derived morphometrics and texture indicators to accurately identify patterns of deprivation. Furthermore, the project advances in the modelling of the distribution of population within the slums. The collection of accurate in-situ population data in Nairobi challenges and revises existing population estimates, significantly improving the accuracy of urban population maps. When combined with thermal exposure assessments, this dataset aims to locate populations most at risk due to thermal disparities. ONEKANA not only enriches the scientific discourse on urban thermal inequalities, but also introduces a replicable framework applicable across diverse urbanities in the Global South. The ultimate goal of the project is to integrate EO-derived insights into urban planning and climate adaptation efforts, thereby strengthening urban resilience to climate-induced thermal extremes. Through scientific rigour and user-centred methodologies, ONEKANA lays the foundation for transformative urban planning that prioritises the well-being of the most vulnerable urban populations.

The assessment of long-term urban transformations is critical for informed decision-making and scientific support during urban reconstruction. While commercial Very High-Resolution imagery is often employed for sustainable urban development planning, Copernicus Sentinel-2 data offer a valuable alternative, providing free images at frequent revisit times. This study investigates the feasibility of using Sentinel-2 data to analyse long-term changes in urban buildings following seismic events. Given Sentinel-2’s spatial resolution limitations for clear identification of alterations, our methodology relies on temporal analysis of a parameter known as Perceived Lightness (PL). PL is derived from Red-Green-Blue reflectance values and captures variations in luminance caused by events like building demolition, reconstruction, or rubble removal. To address PL variation due to vegetation and seasonal changes, our analysis incorporates the Normalized Difference Vegetation Index (NDVI), which assesses green vegetation health. By comparing the PL trend of the building under investigation with a reference PL trend, anomalies are revealed, potentially indicating a change, and defining a possible date of change.

The methodology is applied to Sentinel-2 data from L'Aquila and Amatrice, two central Italian cities struck by earthquakes in the past. The results are promising, and, beyond the post disaster assessment, this methodology can also be employed to track urban development and sprawl over time. Furthermore, it can be potentially adapted to analyse changes in other types of land cover, such as deforestation. The ability to monitor these changes using freely available data can be a valuable tool for environmental monitoring and conservation efforts.

Extreme weather events, sea level rise and coastal erosion are major issues that need to be urgently addressed in European coastal cities. Smart technologies can support Coastal Cities Living Lab (CCLLs) in detecting potential risky conditions in the study area and co-designing adaptation solutions, like Nature Based Solutions (NBS) and Ecosystem-Based Adaptation (EBA). A GIS-based Early-Warning Support (EWS) system and Digital Twin (DT) platform presented here can be deployed in the CCLLs to assist both the planning of long-term climate resilience strategies and decision making in case of risky events. We show how the the effectiveness of these solutions is evaluated by exploiting multi-temporal flood maps obtained from satellite in the coastal area of Massa (Italy). Specifically, we are developing a general analysis tool for testing outputs of hydraulic and hydrological models included in the DT exploiting synthetic aperture radar (SAR) images such ad Sentinel-1 or COSMO-SkyMED, or optical imagery like Sentinel-2 available for major flood events in the study area in the last decade. Information on real case studies are retrieved by municipality report archives, Floods Directive maps, local newspapers and exploration of regional agencies website and Copernicus Services. Flood hazards maps in terms of flood extent and, wherever possible, water depth or flow velocities distributions, simulated by the models implemented in the DT will be thus compared to real flood events mapping and validated. Evaluation of the effectiveness of the procedure and solutions and their possible incorporation in environmental monitoring, early-warning systems, and decision-making processes at different CCLLs will be then assessed through meetings with relevant users. This work is supported by the project SCORE (Smart Control of the Climate Resilience in European Coastal Cities), funded by the European Commission’s Horizon 2020 research and innovation programme under grant agreement No. 101003534.

Climate change will increase the number, duration and intensity of heatwaves. Urban heat islands amplify these climate extremes, further increasing heat stress in cities. Nature-based solutions, like increasing green/blue infrastructure in cities, are proposed to mitigate heat stress by enhancing evapotranspiration. However, it is unknown how urban vegetation and related carbon, energy and water cycles will respond to climate change and how it influences the potential of nature-based solutions to mitigate urban heat stress.

Remote sensing facilitates monitoring of vegetation state and environmental conditions at large scales. Studying urban vegetation, however, requires spatially high-resolution estimates of abiotic and biotic factors. These factors are essential for the parameterization of models to simulate vegetation-mediated processes (e.g. evapotranspiration, gross primary production), or monitoring vegetation health. The required high spatial resolution imposes substantial methodological challenges to cope with the large heterogeneity of urban environments.

Particularly cast shadows and geometric-optical scattering significantly constrain the retrieval of abiotic and biotic factors. If not properly accounted for, shadowed pixels can for example show a decrease of up to 25% in the NDVI compared to fully illuminated pixels. This leads to an underestimation in vegetation health studies or in modelling evapotranspiration products.

Our study aims to demonstrate and quantify the effects caused by shadowing on the retrieval of abiotic and biotic factors from spatially high-resolution data. We particularly focus on leaf area index and Absorbed Photosynthetic Active Radiation, comparing retrievals from in situ measurements, spatial high-resolution airborne data, and operational satellite data to investigate scaling effects. We discuss methodological challenges and needs related to retrieving abiotic and biotic factors in complex environments. We expect that derived insights allow moving towards advanced retrieval techniques for abiotic and biotic factors in complex urban environments and eventually improve our understanding of urban vegetation in the context of climate change.

This work describes preliminary attempts aimed at creating a dataset of daily averages of aerosol optical depth (AOD) on an intraurban scale (300m) using MODIS MAIAC AOD, SEN3 SYN, and AOD from ERA5 reanalysis models. Our preliminary efforts were aimed at understanding the quality of available AOD products by comparing them with daily average measurements provided by the AERONET network for the Italian peninsula during the reference period 2019-2023. MODIS MAIAC AOD proves to be state-of-the-art in satellite AOD reconstruction, while SEN3 SYN correlates less and shows a significant bias when compared with AERONET.

Our efforts are oriented in two directions: a) evaluating whether SEN3 SYN is mature enough to deliver unbiased AOD products on an intraurban scale on a daily, weekly, and monthly basis, possibly using other sources of information such as DEM, latlon, and LST; b) performing data imputation of missing observations using AOD from reanalysis models such as ERA5.

Regarding AOD correction on an urban/intraurban scale, we are evaluating pixel-based approaches such as linear/nonlinear/GAM regression algorithms fueled by the combined use of SEN3 SYN, MODIS MAIAC, ERA5 AOD, and auxiliary data such as MODIS land surface temperature and climatic data.

Our findings demonstrate that there is room for further improvement of AOD products by imputing missing AOD values and by further calibrating AOD using regression models fed with available AOD estimates and auxiliary data.

This work is part of a collaborative project funded by ASI and called APEMAIA (Assessment of PM Exposure at the intra-urban scale in preparation for the MAIA mission). The project is designed to investigate the potential of MAIA by developing a multi-modular system for extracting PM concentrations at the intra-urban scale using Artificial Intelligence techniques.

Building anomaly and displacement detection are critical for ensuring the safety and longevity of structures.

Based on the progress of the RepreSent project [1][2], the unsupervised building anomaly detection methods based on GNN autoencoders and LSTM autoencoders using PS-InSAR have been successfully developed and demonstrated their effectiveness in detecting three types of building anomalies caused by step, noise, and trend displacements for Rome (Italy).

The purpose of the current study is to enhance the ability to detect building anomalies. Given the varied and changing nature of urban environments, we aim to expand the area of study from Rome (Italy) to Bucharest (Romania). This expansion allows us to better understand the patterns of anomalies across different urban landscapes. By using the recently released European-wide Building Footprint Datasets in our models, we expect to deepen our knowledge of the relationship between various building attributes (e.g., construction year, height, seismic risk level) and the anomalies detected. We also plan to refine our anomaly detection by applying signal decomposition techniques to minimize prediction errors, particularly those associated with noise. Furthermore, our goal is to advance our detection methodology by not only identifying the occurrence of anomalies but also predicting their timing and duration.

The dataset focuses on Bucharest, the capital of Romania, which faces a significant challenge due to numerous buildings from the late 19th century that have structurally deteriorated over time and do not comply with current seismic standards [3]. According to the latest statistics released on March 29th, 2024, by the Bucharest Municipal Administration, over 2700 buildings are at risk of collapse in the event of an earthquake.

This work is supported by the European Space Agency with contract as part of the RepreSent project under the Grant 4000137253/22/I-DT.

References:

[1] ESA RepreSent project website, https://eo4society.esa.int/projects/represent/.

[2] Kuzu, Rıdvan Salih, et al. "Automatic Detection of Building Displacements Through Unsupervised Learning From InSAR Data." IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (2023).

[3] Ianoş, Ioan, et al. "Mapping accessibility in the historic urban center of Bucharest for earthquake hazard response." Natural Hazards and Earth System Sciences Discussions 2017 (2017): 1-24.

The purpose of this contribution is to leverage the integration of medical and Earth observation (EO) data with artificial intelligence (AI) to estimate the potential impact of future transmissible diseases, with a specific focus on cancer patients. By combining medical treatment progress and monitoring the living environment from space, this research aims to provide valuable insights into the estimation of healthcare requirements and resource allocation from various perspectives. The future aim of this research is to support local authorities in organizing effective medical schemes while enabling central authorities to develop resilient plans for future pandemics.

Both remote sensing and medical fields benefited a lot from the machine learning methods, originally developed for computer vision and multimedia. We investigate the applicability of the same data mining-based machine learning technique for exploring the structure of both Earth observation (EO) and medical image data.

EO images were acquired for a city with high diversity in term of builds, transportation infrastructure, green areas, etc. The EO dataset consists of images from multispectral and radar sensors. In addition, medical images were acquired by computed tomography (CT). The methodology has been tested by several experts, and the results were checked by either comparing them with reference data or through the feedback given by these experts in the field.

The results of such a study could help in the identification of possible causes for which in a city there is a higher number of patients with a particular disease (e.g., patients with cancer) towards another city where the number is smaller, considering the structure of the city (e.g., the surface of the green areas, number of hospitals, the surface of the industrial area, type of industry). Such a perspective could be ground-breaking in medical epidemiology.

Douala, a city situated on the coast of Cameroon in the Gulf of Guinea, is characterised by its low elevation above sea level and sedimentary geology, making it particularly susceptible to erosion, subsidence, and sea level rise. Currently, Douala City and its surrounding mangrove forests experience alarming rates of coastal erosion, frequent flooding, complete land loss, and evidence of subsidence from regional and continental research. This raises critical questions and reveals numerous research gaps, such as the need to better understand current coastal city dynamics; approaches for monitoring and predicting Douala's low coastland changes; the need to understand the rates, causes, and patterns of subsidence; and lastly, the understanding of the combined impacts of multiple factors on coastal city vulnerability. Therefore, this study aims to fill these knowledge gaps by investigating, understanding, and estimating the causes, consequences, and coastal vulnerability of land subsidence. In this study, remote sensing data, InSAR analysis, spatial analysis, and statistical analysis were used to assess the actual land subsidence rate, determine the factors influencing land subsidence, estimate the influence of land use change on subsidence processes, and establish an integrated vulnerability assessment for the coastal areas of Douala. The findings of this study indicate an average rate of subsidence amounting to 2.9 mm/year, which is indicative of subsidence occurring in all areas of the city. Furthermore, the effects of land use were observed to be dependent on the period and rate of change. These results will be of great importance in gaining a more comprehensive understanding of the dynamics of Cameroon's mangrove landscape and the susceptibility of coastal infrastructure to subsidence, coastal retreat, and potential flooding events. These findings can be utilised to develop sustainable management strategies for the coastal zone of Douala.

Natural hazards and severe weather events represent an increasing threat to both human lives and property. With climate change, extreme weather events are projected to occur more often, increasing the risks of damage to buildings and infrastructure across many regions of the world.

Earth observation satellites can play a crucial role in disaster response and management, offering unprecedented access to large-scale views of affected areas. In particular, deep learning techniques have great potential for automated building damage detection from satellite imagery. Consequently, several recent studies proposed new network architecture and demonstrated their effectiveness on the popular xView2 Building Damamage Assessment (xBD) dataset featuring bi-temporal very-high-resolution image pairs of multiple disasters. Although achieving strong performance, many of these methods are highly engineered, including the winning solution of the xView2 competition. From a practical perspective, however, there is a high demand for simple and robust methods with good generalization ability.

Therefore, our work focuses on simplifying the xView2 winning solution, keeping only the vital components while retaining accurate building damage detection performance. Thereafter, the xBD dataset splits from the xView2 competition are rearranged to eliminate spatial overlap between training and test locations. We evaluate several recent building damage detection methods on the proposed split and shed light on the limited generalization ability of existing methods under more realistic scenarios. Furthermore, we hypothesize that minor and major building damages have distinctive characteristics across different disaster types, hampering the generalization to new areas. To that end, we propose a novel method incorporating readily available disaster-type information into the building damage prediction pipeline. We empirically demonstrate that our strong baseline conditioned on disaster-type information outperforms state-of-the-art methods on the proposed realistic split of the xBD dataset.

Extreme heat endangers human health and well-being and impairs the use of public spaces. Dortmund’s Integrated Climate Adaption Master Plan prioritizes actions and measures to improve heat resilience. This project supports the city of Dortmund (Germany) in attaining this goal, by deploying a state-of-the-art biometeorological sensor network and developing a nowcasting service for monitoring thermal comfort across the city. The project aims to pioneer the integration of thermal comfort data in smart-city ecosystems and provide actionable insights for the development of Dortmund’s Heat Action Plan. Modeled, remotely sensed, and in-situ data will be used to provide near-real-time information regarding the outdoor thermal conditions. City-officials of Dortmund are involved in the design of the dashboard, and the weather station network, ensuring they meet their needs. The collected data will be used in a series of on-ground actions, supporting the evaluation of existing climate adaptation measures, and the design of new ones. These actions include the mapping of areas with high potential for planting trees, the investigation of changes in human behavior during hot days, and the assessment of backyard greening strategies. To engage with the local stakeholders, promote the role of citizen scientists, and disseminate the project, a series of workshops and on-site events are planned, such as climate comfort labs, mobile measurement campaigns, or climate walks with citizens. The overall goal of the project is for the city of Dortmund to adopt and integrate the developed network and nowcasting service into its smart-city ecosystem.

With global warming and increasing urbanization, especially tropical regions will experience seasons of extreme heat, the impacts of which require better mitigation measures from urban planners. This was especially evidenced in 2023 with the occurrence of el Nino, and related extreme temperatures for prolonged periods of time, causing amongst other problems, health challenges to the urban communities.

The European Space Agency’s Global Development Assistance (GDA) Urban project aims to foster the mainstreaming of EO into International Financing Institute (IFIs) programmes in themes such as urban resilience. In this context, a case study focused on mapping and analyzing UHIs within Sargodha city, Pakistan for an Asian Development Bank (ADB) project was undertaken. Sargodha is a densely populated city experiencing temperatures above 40^oC in the summer months. In addition, rapid urban sprawl/urbanization has resulted in city expansion, and reduced green areas.

EO data was used to derive Surface Urban Heat Island Intensity (SUHII). SUHII is an indicator for the increased local temperature due to the urbanization/imperviousness effect and allows decoupling the absolute temperature into two components. This helps in modelling urban heat. The method entails estimating the linear correlation via land surface temperature, as measured via Landsat 8, and impervious surfaces, using the world settlement footprint. Additional analyses included a time series exhibition throughout the year of 2021, creating a simple model for predicting the increased SUHII beyond the recorded temperatures and a basic greening simulation - if some areas were to be transformed into parks.

For the Sargodha use case, the datasets (vectors/city blocks) were favorably received by ADB and used in planning park locations. Future work aims to increase outreach further by building a geospatial dashboard for showcasing urban heat and SUHII, with accompanying supporting geodata (population density) and analyses in another 1-2 cities.

Global monitoring platforms are key tools to evaluate changes in urban areas and to facilitate sustainable urban development. In this context, the Urban Thematic Exploitation Platform (UTEP) is a leading web-based platform that provides thematic products and indicators to policy makers, urban planners, and other stakeholders. UTEP takes advantage of distributed high-level cloud computing infrastructures and provides multiple functionalities to enable users to easily access, visualise, and process EO data.

In order to keep serving the needs of the urban user community, the platform is currently evolving by following the “algorithm-as-a-service” paradigm following the Best Practice for EO Application Package as defined by the Open Geospatial Consortium and the EO Exploitation Platform Common Architecture (EOEPCA). This Best Practice supports developers that want to adapt and package their existing algorithms written in a specific language to be reproducible, deployed and executable in different platforms. The Application Package, encoded as a Common Workflow Language (CWL) document, comprehensively describes the full data processing application. Developers build container images that encapsulate their application and command line tool(s),which are then published on container registries for easy access and deployment.

UTEP currently supports the ingestion, metadata extraction and calibration of free and commercial EO data from more than 40 missions (Radar and Optical). EO data is exposed using the SpatioTemporal Asset Catalog, where single band assets are classified under Common Band Names.

In this work we use an algorithm for multi-mission and multitemporal analysis in urban areas to demonstrate how an EO processor can be easily integrated and deployed in UTEP. This exercise also showcases how the platform can handle massive processing using scalable cloud resources. Thanks to this evolution of the platform, UTEP offers to users a cost-effective production environment ready to host other thematic processors from researchers and service providers.

Urban areas show significant differences with natural environments regarding their effects on weather and climate through land surface processes attributed to the characteristics of urban form, function and presence of pollutants from anthropogenic activities among others. These effects include elevated surface and air temperatures, increased boundary layer height and increased CO₂ concentrations. In turn, urban elements such as trees which have biophysical mechanisms driven by meteorological conditions show altered behavior compared to their natural environment-based counterparts. These differences are conceptually well-established, however observing these and accounting for their implications is not straight-forward. In this direction, this study investigates the urban effects on tree phenology for Berlin. In particular, we compute phenology-based parameters (Start of Season, Peak of Season, End of Season) over a NWP (Numerical Weather Prediction) model grid configuration, moving outwards from the city center. A cloud based approach is implemented to achieve this, leveraging the capabilities of GEE (Google Earth Engine), using the Sentinel-2 time-series for 2023 and ESA’s World Cover product. The total area covered expands well beyond the boundaries of Berlin, in order to capture potential urban-rural differences, and explore similarities as well as diverging outcomes compared to other studies assessing relevant vegetation traits in an urban climate context.

As urban areas globally face unprecedented challenges, the importance of leveraging advanced technologies for sustainable urban planning has never been more critical. In this framework, the CITYNEXUS project is specifically designed to confront and mitigate existing urban challenges in the City of Copenhagen, such as traffic congestion, air quality, and urban livability by enabling the near-real time simulation of the effects of infrastructural, land-use and policy changes on air quality, mobility and health. Specifically, CITYNEXUS is one of the first 5 Use Cases of the Destination Earth (DestinE) Core Service Platform (DESP).

Central to the activity is the integration of diverse datasets, including crowd-sourced human mobility information gathered from smartphones, governmental and para-governmental data, Google Environmental Insights Explorer air quality data, and EO(-based) layers from the DestinE Data Portfolio. These datasets play a pivotal role in the development of state-of-the-art models capable of conducting in-depth analyses and forecasts on the potential impacts of proposed urban interventions, facilitated through robust 'what-if' scenario simulations.

While targeting the City Copenhagen, CITYNEXUS aspires to broader applicability and scalability in various urban contexts within Denmark and potentially worldwide, striving to contribute to the adoption of digital twin technology and space-based data in data-driven urban planning and environmental management. By empowering urban planners and policymakers with the ability to simulate and critically evaluate the ramifications of varied urban interventions, the project not only addresses specific urban planning challenges but also showcases a scalable solution that could inspire cities globally in the pursuit of sustainable development.

Studying changes in land cover and land use (LCLU) within urban environments provides critical insights into the dynamics of our cities. Particularly, monitoring the evolution of LCLU, such as deforestation, mining, agriculture, or other anthropogenic activities, can significantly alter the landscape and soil composition. These alterations may lead to soil erosion or instability, increasing the risk of landslides or soil subsidence, which, in turn, can damage roads and railways built in or near affected areas.

Remote sensing data, especially multispectral and multitemporal optical imagery, is instrumental in accurately monitoring LCLU changes globally. Automated approaches, particularly within artificial intelligence and machine learning, have shown impressive capabilities in identifying LCLU classes using such imagery, facilitating the study of global changes.

Planetek Italia is actively developing an operational infrastructure monitoring service to support predictive maintenance of roads, railways, and bridges. This initiative involves analysing multiclass LCLU changes (i.e. changes from urban class to other and vice versa) with time-series multispectral Sentinel-2 data and employing supervised machine learning approaches. Supported by the Italian Space Agency (ASI) under the I4DP Market project (Innovation for Downstream Preparation Market), this initiative aims to provide a comprehensive solution aligned with guidelines for risk classification, safety assessment, and monitoring of existing bridges outlined by the Italian Ministry of Infrastructure and Transport.

The proposed operational service builds upon the existing Rheticus^® Safeway, developed by Planetek within the Horizon 2020 Safeway project concluded in February 2022. Enhancements and adaptations performed during the ASI project cover both operational and technological aspects, democratizing and scaling the service to a European and global level.

Adopting this operational infrastructure monitoring service signifies a significant step toward efficient and proactive maintenance strategies. By leveraging satellite-based geo-analytics information, the service ensures the safety and resilience of roads and railways infrastructure networks on a broader scale.

The Asian Development Bank (ADB) operates through resident missions in various regions, and their development missions require access to Earth Observation (EO) products and tools to enhance the understanding of climate change impacts, natural hazards, and disasters that affect the areas they are overseeing.

Planetek Italia, with the support of ESA and ADB itself, developed customized EO-derived geo-hazard indicators useful for a better understanding of the impacts of climate change, natural hazards, and disasters in developing countries like Indonesia, Bangladesh and Papua New Guinea.

The EO-developed products offer valuable support to policy and decision-makers in several ways:

1. Providing Essential Baseline Information for the implementation of projects in developing countries by the IFI like ADB: the developed satellite EO-based geo-hazards and geo-analytics represent an innovative solution that provides new ways for ADB to address challenges in developing countries.
2. Support the geo-hazards risks assessment at the country scale: The worldwide coverage and high-revising time guaranteed by the Copernicus satellites present an unprecedented opportunity for natural hazard assessment and its continuous monitoring at the country scale.
3. Empower decision-makers: The operational use of the developed satellite EO-derived products at country scale is expected to empower the decision-makers with accurate information for:
   1. Resilient strategies implementation.
   2. To support sustainable development and disaster risk reduction.
   3. Recovery support after a natural disaster.
4. Better understanding the benefit of operational use of the EO-derived products: The EO-developed products contributed to increasing the benefit awareness connected with the operational use of Earth Observation products in development projects.

To promote the operational application of the developed EO products in operational projects, the ADB supported dedicated capacity-building activities involving the national authorities of Indonesia, Papua New Guinea, and Bangladesh in different projects where Planetek Italia was involved starting from 2018 through the ESA project EO4SD.

Asphalt roads are vital for modern societies, facilitating efficient transportation of people, goods, and services. However, these roads degrade over time due to factors like temperature, oxidation, loads, and water. To address this, a project in Milan developed a methodology using very high-resolution (VHR) satellite sensors to assess road surface aging and support maintenance activities.

The reflectance spectra of asphalt roads change as they age, enabling remote sensing to monitor pavement conditions. Yet, the spatial resolution of satellite images poses challenges, often mixing road pixels with surrounding features like vegetation and buildings. To overcome this, the project used Worldview-3 satellite images to calculate indicators related to pavement color, visibility of road markings, and material composition (spectral unmixing).

These indicators were combined to create a synthesis asphalt aging index, categorizing roads into different aging classes. However, creating a precise "road mask" to isolate areas for analysis proved difficult due to various disturbances in the images such as shadows, vegetation, and vehicles. Despite advanced techniques, some elements remained, affecting subsequent spectral analyses.

Nevertheless, field evaluations of the aging index showed good agreement, proving its usefulness for the city's road maintenance department in devising timely maintenance plans. Despite limitations, the developed methodology provides actionable information on urban road aging, facilitating efficient maintenance decisions.

In essence, the project demonstrated the efficacy of utilizing VHR satellite sensors to assess asphalt road aging, offering a practical approach to support maintenance planning. Though challenges remain, the methodology provides valuable insights for maintaining urban infrastructure effectively.

The objective of Horizon 2020 JUSTNature project is the activation of nature-based solutions (NbS), based on the principle of the right to ecological space.

Ecological and socio-economic status and disparities profiles have been created for six European cities, with the objective of guiding the strategic process of NbS planning. The profiles were built as urban units' agglomerations with similar characteristics, defined by a set of indicators representing the key challenges that NbS are intended to address. They are the six (in-)justice components enabling NbS activation: air quality, carbon, thermal, spatial and temporal (in-)justices as well as flora, fauna habitat (non-)inclusiveness.

To assess the thermal (in-)justice indicators, an AI-based methodology was developed by fusing Sentinel-2 with Landsat to obtain a 10m Land Surface Temperature (LST) monthly time-series for the period 2017-2022. A clustering algorithm was applied to the LST monthly time-series to generate a “Heat Stress map”. Besides, monthly Surface Urban Heat Islands (SUHI) within the summer months were computed. Additionally, a SUHI Likelihood map was computed by averaging the maximum monthly values of SUHI. To assess the temporal (in-)justice indicators, a multi-temporal Land-Cover map, integrated by the degree of vegetation and imperviousness, was produced for the years 2018, 2020, 2022, based on Machine Learning techniques on Sentinel-2 time-series. A clustering algorithm, namely the Hierarchical Density-based spatial clustering, was then applied to the collected indicators coupled by socio-economic/demographic data, to identify common patterns stressing disparities within the city.

EO-based data, enabling the provision of spatially continuous, accurate and regular measurements of various parameters within the urban context at both local and global scales, proved to be of pivotal importance in providing city administrators and urban policymakers with the strategic supportive information they need to plan and implement JUST and appropriate NbS solutions.

The Urban Heat Island (UHI) effect is heavily affecting urban regions, especially in the Mediterranean region due to climate change. A range of Nature-Based Solutions (NBS) have been proposed to mitigate overheating in urban environments. Up to now, NBSs have been mainly applied in experimental sites at a local-scale. At the current state, NBSs are expected to only have a local effect on heat mitigation due to their limited spatial extent. Therefore, the NBS efficiency is difficult to be adequately monitored by spaceborne Earth Observation (EO) and as a result measuring the potential benefits of the NBS remains a challenging issue that hampers their widespread application. In the framework of the EU-funded project CARDIMED, an effort is being made to upscale NBS and their related benefits at a larger scale based on earth observation techniques fused with in-situ and crowdsourcing data. To fulfil the aforementioned key objective, a multi-sensory data fusion approach will be adopted, assimilating geospatial data sources collected from in-situ, crowdsourcing and EO-based sensors. More specifically, we propose a methodology for obtaining temperature data at a higher resolution, allowing this way to locally quantify the efficiency of the applied NBS solutions in the pilot regions of CARDIMED for heat mitigation. The methodology is based on a downscaling, cross-modality framework that combines measurements from ECOsystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS), Land Surface Temperature (LST) products derived from SLSTR Sentinel-3 satellite mission and in-situ data produced either from professional sensors or Citizen Science participatory campaigns. This will enable us to grasp the gap in NBS efficiency estimation, as a first step to the NBS upscaling strategy. 

Acknowledgement:

This research has been funded by European Union’s Horizon Europe research and innovation programme under CARDIMED project (Grant Agreement No. 101112731) (Climate Adaptation and Resilience Demonstrated In the MEDiterranean region).

Responding to global warming and adapting to climate change effects such as heat waves and droughts is a key priority of European and national climate change adaptation strategies. Soil sealing, covering of ground surfaces with impermeable materials or buildings, thereby preventing water infiltration into the soil, has a significant impact on the urban climate, especially in the context of urban-heat-islands (UHI). Administrations at different levels aim at reducing health risks associated with climate change and to improve human well-being through appropriate planning measures like policies, urban planning strategies but also technological solutions. The Copernicus Land Monitoring Service (CLMS) supports these activities with dedicated high-quality, pan-European data products.

The High-Resolution-Layer Non-Vegetated Land Cover Characteristics (NVLCC) portfolio from the CLMS enhance planning capabilities and support evidence-based adaptation measures to build resilience against climate impacts. The NVLCC’s raster layer focus on impervious areas, at 10m resolution across EEA38-countries aiding frequent land cover updates and serving as an early detection system for environmental changes. Derived from Copernicus Sentinel missions, the product includes components for imperviousness densities, built-up areas, (and newly permanent bare surfaces) that provide insights into artificial and bare surface cover and building constructions.

Spatially explicit information is essential for informing climate adaptation discussions, guiding zoning decisions and communicating potential climate change impacts and mitigation measures to stakeholders. The CLMS NVLCC data, when used alongside with various data sources like meteorological, demographic, and socio-economic data, can provide detailed and up-to-date information on land use and land cover, which is often lacking for effective planning. This information is crucial for assessing the impact of building development on local climate, analysing the relationship between building stock and green areas, and understanding of heat storages in urban areas.

With the increasing complexity of urban environments, there is a growing need for tools that provide precise and detailed insights into how different planning decisions might play out. Precision in planning helps in making informed decisions that can minimize risks, optimize resource use, and ensure the well- being of urban populations.

This study introduces a novel approach in urban planning by leveraging Generative Adversarial Networks (GANs) to generate multispectral synthetic satellite imagery, particularly tailored for nature-based solutions. The primary focus is on enhancing the capabilities of urban simulation tools in the context of sustainable urban development and climate resilience. The methodology extends upon existing frameworks by integrating advanced GAN architectures for the generation of high-fidelity multispectral imagery that mimics the characteristics of the Sentinel-2 satellite constellation, enabling simulation of various urban and rural scenarios like urban green spaces, water bodies, and agricultural lands.

The core of this methodology revolves around the collection of a comprehensive dataset of high-resolution multispectral images matched with urban and rural landscapes labels. An innovative aspect of this approach is the simulation of multispectral features specific to different local urban settings. By using historical Sentinel-2 images, the model gains the ability to replicate the unique ecological and urban characteristics of a particular area. This process ensures the generation of coherent multispectral data learning from past satellite images. Finally, the generation of synthetic satellite images stars from vector files representing various urban planning scenarios.

This methodology bridges the gap between theoretical planning and real-world application, offering a potential tool that enhances urban resilience, fosters sustainable development, and supports informed decision-making.

The aim of this work is to categorize well-being conditions of citizens in several urban contexts based on temperature and air pollution extreme values acquired from space. The proposed approach relies on the evaluation of the incidence of Urban Heat Island (UHI) and air pollution phenomena in the periods of the year in which they can potentially reveal their maximum intensity in specific metropolitan areas.

By considering the levels of tropospheric Nitrogen dioxide (NO2), with particular attention to winter season, and the Land Surface Temperature (LST) values, particularly in summer with possible cUHI formation,the areas in which the two phenomena overlap at a spatial level over the course of the year have been identified and categorized based on the intensity of such extreme conditions.

After a deep review of literature, different sensors and mathematical approaches have been considered to compute the UHI index, by evaluating the ability to recognize the phenomenon at varying spatial resolution, focusing on the ECOsystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS), due to its advantages in terms of spatial resolution, equal to 60m. During computations, the morphology of the city also has a crucial role: geometries of buildings and streets, presence of trees, and eventual mitigation strategies contribute to obtain the most relevant outcomes.

The air quality study has been instead settled leveraging on the results of NASA new product that provides monthly averages of tropospheric NO2 vertical column density derived from the level 2 Tropospheric Monitoring Instrument (TROPOMI) oversampled to a spatial resolution of 0.01˚ x 0.01˚(https://disc.gsfc.nasa.gov/datasets/HAQ_TROPOMI_NO2_CONUS_M_L3_2.4/summary?keywords=HAQ_TROPOMI_NO2_CONUS).

The research has been developed on metropolitan areas having high population density and consistent risks of heat waves and NO2 concentration anomalies, e.g. New York. Outcomes of the analysis are urban maps identifying different classes associated to well-being conditions for citizens considering long time period analysis.

Urban applications based on Earth Observation (EO) data are at the core of the Italian Space Agency (ASI)’s roadmap to develop downstream applications that could serve institutions to address their specific challenges and priorities in cities management. “Sustainable Cities” was indeed the theme of the first call for ideas that ASI launched in 2022 to start implementing the new “Innovation for Downstream Preparation for Science” (I4DP_SCIENCE) program.

I4DP_SCIENCE is devoted to the Scientific User Community, i.e. Italian Universities and Public Research Bodies, and is composed of joint projects with ASI demonstrating the usefulness of novel methods and algorithms to support applications of user’s interest falling within topics of national relevance, e.g. defined by the National Copernicus User Forum, and/or falling within international agendas, e.g. the UN Sustainable Development Goals (SDGs). All the demonstrations are carried out jointly with the reference users who are actively engaged since the initial user requirement consolidation and, throughout the project, via capacity building and training activities towards the user uptake.

Of the whole I4DP_SCIENCE portfolio, two projects specifically address challenges in metropolitan cities.

The LCZ-ODC project with Politecnico di Milano developed a novel workflow to produce multi-temporal and multi-resolution Local Climate Zones (LCZ) maps and assess their correlation with urban thermal comfort, while eliciting and addressing user community needs through the parallel development of open-source software tools. LCZ maps are produced and tested over the Metropolitan City of Milan by using multispectral Sentinel-2 and hyperspectral PRISMA satellite imagery, multi-source geodata (e.g. Copernicus Land Monitoring Service Imperviousness Density) and the Open Data Cube (ODC) technology.

The GEORES project with University of Bari and the Institute for Electromagnetic Sensing of the Environment (IREA) of the National Research Council of Italy (CNR) is instead developing a geospatial application meant to improve environmental sustainability and resilience to climate changes in urban areas, through a multi-risk platform composed of four main modules: (1) Sediment Connectivity; (2) Land Displacement; (3) Urban Floods; (4) Urban Wildfires. For each module, EO data (including the Sentinels, COSMO-SkyMed, SAOCOM and PRISMA), calculation models and algorithms (e.g. including interferometric Synthetic Aperture Radar techniques) are integrated to identify “hot-spots” of urban and peri-urban territory at high risk from the point of view of land degradation caused by phenomena of hydrogeological instability, sediment flow or vegetation fires. The extracted information is expressed with specific indicators (“geo-analytics”) calculated dynamically and automatically. The demonstration is undertaken in the Metropolitan City of Bari and the urban settlements in Gargano Promontory, Apulia Region, southern Italy.

In outlining the technological novelty of the algorithms and functionalities of the platforms and plugins, the paper illustrates the user-driven approach, the analysis of the user requirements, and how the novel products are outlining real perspectives for implementation by the users.

As we live in an era of big data acquisition - satellite, in-situ, wearables -, climate change and environmental risks have become much easier to map. On the other hand, domain knowledge is usually supplied by the academic sector, offering novel methodologies for hazard mapping and predictions, albeit being hard to translate those scientific-driven findings for the public administration, and society at large. Hence, public policies and public domain knowledge, including the implementation and monitoring of regulatory frameworks, often lag behind to the state-of-the-art.

As such, citizens are often ‘in the dark’ about the environmental or climatic risks surrounding them, even though about 40% of the world’s population lives within 100km of the coast, subject to sea level rise, or exposed to other weather and climate extremes such as heatwaves and droughts. Furthermore, the pressure for further urbanization and the efforts to preserve its rich natural capital are often at odds.

AtlanticSENSE builds upon these notions to leverage the state-of-the-art scientific knowledge on data acquisition, machine learning (ML) and metocean predictions to address the key environmental and climatic challenges we face, particularly in coastal settings, to deliver an user-friendly added-value information tool which content can be easily acknowledged by the society. The concept is to deploy efficient semi-automatic data science workflows on top of large arrays of freely available geospatial products (e.g., remote sensing, in-situ citizen/voluntary networks, and numerical modelling) to feed a live platform with real-time natural hazards and risks information, readily available to the community.

A preliminary proof-of-concept of the AtlanticSENSE concept has been deployed in the Greater Lisbon Area, integrating several modules already operational focusing on AIR (air temperatures, heatwaves and air quality predictions), OCEAN (ocean physics predictions, sea level rise and seawater temperature extremes) and LAND (coastal erosion, land use/land cover monitoring) domains, while the CoLAB +ATLANTIC is now seeking its validation by local stakeholders and community uptake. Further developments shall include additional layers of information, the continuous improvement of the user experience interface, and the scalability of the project to other regions.

Previous research has demonstrated the potential of combining satellite images with foundation models to predict environmental exposures, such as air pollution, and social inequalities, like house prices. Building upon these findings, we investigate the use of Sentinel-2 satellite imagery and foundation models to predict spatial signatures, which characterise urban form and function¹. We compare two machine learning approaches: a two-stage approach that does not require fine-tuning, where embeddings are first extracted from a large remote sensing foundation model and then used as input features for a separate prediction model, and a standard fine-tuned model directly predicting spatial signatures. By comparing the predictive performance of these approaches, we assess the value of fine-tuning foundation models for capturing spatial patterns related to urban form and function. Initial findings suggest that both approaches show promise. This study highlights the feasibility of combining Sentinel-2 imagery with foundation models to understand urban environments and their spatial signatures on a large scale. The results underscore the potential of this approach to provide valuable insights into urban form and function, complementing traditional data sources and offering a new perspective for urban research and policy-making.

¹ Fleischmann, M., & Arribas-Bel, D. (2022). Geographical characterisation of British urban form and function using the spatial signatures framework. Scientific Data, 9(1), 546. https://doi.org/10.1038/s41597-022-01640-8

Effective management of natural hazards and risks in urban context is essential for ensuring the safety and resilience of urban populations. The convergence of Earth Observation (EO) technologies with advanced algorithms and solutions offers unprecedented opportunities for proactive mitigation and response strategies. This paper presents the collective experiences and exemplary applications e-GEOS derived from several R&D and innovation projects focusing on different Italian cities, monitored using EO-based technology for a variety of applications.

All the projects have been instrumental in harnessing EO-based solutions to address natural hazards and risks in Italian urban areas, characterized often by ancient city centres, prone to stability issues, and suffering of temperature increase in summer season, with dangerous impact on the aged population.

Leveraging Synthetic Aperture Radar (SAR) imagery, thermal satellite data, drone-based Lidar acquisitions, and the generation of 3D city models trough MVS technique, these projects, and following activities, have pioneered innovative approaches towards the development of an Urban Digital Twin—an integrated platform for comprehensive urban monitoring and management. Through the utilization of SAR imagery, the projects have enabled the detection and monitoring of ground and structure deformation, facilitating early warning systems for geological hazards such as landslides and subsidence. Drone-based Lidar acquisitions have provided high-resolution elevation data, enhancing structural assessments and urban planning efforts. Furthermore, the generation of 3D city models has enabled the visualization and simulation of various hazard scenarios, supporting decision-making processes and community engagement initiatives. The exploitation of thermal data allowed also the definition and practical retrieval of UHI, for the benefit of citizens health. By synthesizing the experiences and lessons learned from the multi-level monitoring of Roma, Gubbio, L’Aquila, Milano, this paper underscores the transformative potential of EO-based solutions in mitigating natural hazards and reducing risks in urban environments. The development of an Urban Digital Twin represents a paradigm shift towards holistic and proactive urban management, paving the way for resilient and sustainable cities in Italy and beyond.

Adapting cities to climate change is a key challenge for humanity. Within this context, urban overheating more recently has become a high priority for cites worldwide, which are about to invest billions of Euros in measures to reduce this problem. To design sustainable urban planning strategies, detailed knowledge on the heterogeneous urban environment is urgently required. Remote Sensing in principle is a powerful means to acquire spatially explicit data for any city and thus to inform these actions and monitor their impact. However, the strong urge to use existing datasets for timely urban heat exposure monitoring and climate adaptation information must carefully consider the physical limitations of remotely sensed LST so as to not result in incomplete, wrong, or misleading indicators. Most importantly, remotely sensed LST provides a biased representation of the urban surface temperature and does not directly provide near-surface air temperature.

These pitfalls are mainly linked to the complex natures of the LST to air temperature coupling, which result in temporal and spatial mismatch of the urban surface and air temperatures respectively. Moreover, the urban LST signal requires very careful processing and interpretation due to effects of strong thermal anisotropy of heterogeneous 3D urban landscapes, emissivity assumptions, and geometrical bias towards horizontal surfaces amongst others. Readily available LST maps are therefore rather detached from the heat exposure at street level and should hence not be used directly for the support of urban planning without careful interpretation and detailed knowledge.

In this contribution we highlight the most relevant misconceptions and pitfalls with the aims to first avoid large investments based on the wrong metrics, and second to start a conversation between Urban Climate Science and Urban Remote Sensing to develop pathways towards more suitable parameters and methods that can be applied to monitor urban climate more appropriately.

The ambition of the LifeCoolCity project is to provide tools that support the management of blue-green infrastructure (BGI) in 10,000 European Union cities, aiming to strengthen their adaptive capacity to the effects of anthropogenic climate change. The project utilizes advanced technology, combining satellite imagery with high-resolution aerial data obtained through laser scanning, thermal sensors, and hyperspectral sensors. The data collected are processed using Geographic Information System (GIS) and Artificial Intelligence (AI). Moreover, our proprietary field measurements and external reference acquisition are used for data quality control and model training.

As a result, a series of analytical products to assess five factors that build the adaptive potential of cities: soil sealing, urban heat island, the quality of blue infrastructure, the quality of green infrastructure, and biodiversity level. The area of 10,000 cities is analyzed using satellite data to assess these adaptive potential factors in terms of their intensity and dynamics. Additionally, in the demonstration city of Wrocław, an intervention involving the implementation of blue-green infrastructure is being tested using the proposed decision support system.

Within the development of this system, we introduce the valuation of ecosystem services of various nature-based solutions (NBS) through consultations with residents, officials, and scientists. Next, we assess how much we can value the increase in biodiversity or the improvement of water conditions as a benefit for the residents. We propose surveys among residents with an innovative approach based on visualizations of the same place with different adopted NBS.

This allows us to consolidate the five mentioned factors into a single integrated indicator, which enables the recommendation of an appropriate nature-based solution in a given location. In cases of high natural values, the system recommends protecting the location from destruction.

As a result, city managers and residents will receive four products: CoolCity Ranking, CoolCity Report, CoolCity Design, and CoolCity Monitoring. These will help identify and enhance the adaptive needs of urbanized areas, create a strategy for managing BGI to strengthen adaptive capacities, recommend the most effective nature-based solutions, and monitor the effectiveness of their implementation.

At the conference, the system's assumptions and partial results will be presented, enabling a broad discussion on the effectiveness and applicability of the developed tools in various urban contexts. The project is co-financed by the European Union under the LIFE+ program.

The research conducts a comprehensive comparison of various European cities, emphasizing the pivotal role that nature-based solutions play in promoting sustainable urban environments. These nature-based solutions encompass a range of strategies that leverage natural processes and green infrastructure to address urban challenges. By incorporating green infrastructure, such as parks, green roofs, and urban forests, alongside ecosystem services like air and water purification, pollination, and climate regulation, cities can significantly enhance their resilience to environmental stresses.

In particular, green infrastructure helps to mitigate the adverse effects of climate change by reducing urban heat islands, managing stormwater, and sequestering carbon dioxide. This, in turn, contributes to a reduction in greenhouse gas emissions and helps cities adapt to extreme weather events. Moreover, nature-based solutions improve the overall quality of life for urban residents by providing recreational spaces, enhancing biodiversity, and fostering mental and physical well-being.

Several European cities have become exemplary models in the implementation of nature-based solutions:

Copenhagen, Denmark: Copenhagen has integrated green roofs and parks to manage stormwater and reduce flooding. The city’s Cloudburst Management Plan includes projects like the Tåsinge Plads, a multifunctional urban space that can hold excess rainwater during storms.

Vienna, Austria: Vienna has long been a leader in green urban planning. The city’s extensive green belt and urban forests are part of its strategy to enhance air quality and provide recreational areas. Projects like the Aspern Seestadt are designed to be sustainable urban districts with ample green spaces and energy-efficient buildings.

Barcelona, Spain: Barcelona has implemented a network of green corridors and parks to connect urban green spaces, promoting biodiversity and providing residents with accessible recreational areas. The city’s Green Infrastructure and Biodiversity Plan aims to increase green space per capita and enhance urban resilience.

Berlin, Germany: Berlin has successfully integrated green infrastructure into its urban landscape through initiatives like the Biotope Area Factor, which mandates a certain percentage of green space in new developments. The city’s Tempelhofer Feld, a former airport turned public park, is one of the largest urban green spaces in the world.

A crucial element in the successful implementation of these solutions is active citizen engagement. It is imperative that urban sustainability initiatives are not solely top-down but rather involve the community at every stage. When citizens are actively engaged, they are more likely to support and maintain green projects, ensuring their long-term success. Community participation can take various forms, such as public consultations, participatory planning, and citizen science projects, which empower residents to contribute to environmental monitoring and decision-making processes.

In summary, the study underscores the importance of nature-based solutions in building sustainable cities. By integrating green infrastructure and ecosystem services, European cities can not only bolster their resilience and mitigate the impacts of climate change but also enhance the quality of life for their inhabitants. The active involvement of citizens is essential in this process, ensuring that the transition to sustainable urban living is a shared endeavor driven by collective action and community spirit.

The intensity of shifting environmental conditions is expected to increase due to climate change, posing a new threat for infrastructure damage and urban resilience, resulting to infrastructure damage, commercial loss and even loss of human lives in the affected areas. A number of studies have been conducted to investigate the potential use of different Nature-Based Solutions (NBS) as a countermeasure to mitigate stormwater runoff, improving climate resilience while preserving the local biodiversity. In the current study, the impact of the traditional stone weirs on climate change-related benefit, namely storm water regulation, in more sparse populated urban areas such as the Aegean Greek islands is evaluated. The analysis for monitoring the impact of stone weirs for storm water regulation is based on a combination of space-based remote sensing observations and in-situ data, produced either from off-the-shelf sensors or citizen science participatory campaigns. Remote Sensing monitoring will also exploit very-high resolution data of Copernicus Contributing Missions analyzing different indexes, such as Normalized Difference Vegetation Index (NDVI) and ND Water Index (NDWI). Dedicated approaches and state-of-the-art models will be deployed to assess and upscale the potential of stone weirs NBS for stormwater regulation, flood mitigation and biodiversity impact at a regional scale.

Acknowledgement:

This research has been funded by European Union’s Horizon Europe research and innovation programme under CARDIMED project (Grant Agreement No. 101112731) (Climate Adaptation and Resilience Demonstrated in the MEDiterranean region).