Conference Agenda

Urban areas, home to the majority of the global population, significantly contribute to environmental degradation due to substantial energy consumption and greenhouse gas emissions. In the European Union, buildings account for 40% of energy use and 36% of CO2 emissions. With 35% of buildings over 50 years old and 75% considered energy-inefficient, urban retrofitting is crucial for enhancing sustainability. Compounding these challenges, current Energy Performance Certificate (EPC) databases suffer from data gaps, lack of standardization, and restricted access, hindering effective action towards the European Green Deal’s goals for a low-carbon future and net-zero emissions by 2050.

In this framework, the ESA BEE.AI (Building Energy-efficiency Estimation with Artificial Intelligence) project introduces an innovative approach to improve urban building energy performance assessments using advanced deep learning and Earth Observation (EO) data. In particular, the intended solution integrates crowd-sourced street-level optical and thermal imagery, satellite-based very high-resolution top-view visible and near-infrared imagery, high-resolution land surface temperature metrics, and urban morphology metrics. Together with existing EPCs, BEE.AI aims at generating detailed maps categorizing building energy efficiency from "A" (most efficient) to "G" (least efficient), pinpointing retrofitting opportunities. Here, besides a baseline solution implementing an end-to-end deep-learning architecture, BEE.AI also envisages an advanced system which additionally extracts and integrates features explicitly providing insights into building characteristics (i.e., building age, construction material, presence of PV panels) and thermal properties.

Pilot studies across Denmark, Austria, and Italy, engaging stakeholders from the business sector such as real estate developers, local governments, and energy companies, will showcase BEE.AI’s capabilities. These collaborations are vital for tailoring the solutions to regional architectural styles and climate conditions and ultimately setting a new benchmark for sustainable building practices, supporting a transition to a low-carbon future.

The European Green Deal sets ambitious targets emphasizing the urgent need for the ecological transition in Europe, with a specific focus on enhancing renewable energy usage. A pivotal aspect of this initiative is increasing the renewable energy share to 32% by 2030, facilitated by amending the Renewable Energy Directive to promote decentralized sources like rooftop solar installations. Numerous European cities and municipalities are advancing this goal by incentivizing solar power and streamlining approval processes.

In this framework, the DLR Solar-DE project plays a crucial role by providing detailed spatial data essential for urban planners and policymakers to scale up rooftop solar resources effectively. This project has conducted a comprehensive mapping of solar installations across the 20 million buildings in Germany using high-resolution digital orthophotos (20cm spatial resolution), surface and terrain models (1m spatial resolution), and building perimeter vector data from the Federal Agency for Cartography and Geodesy, integrated through advanced machine learning and artificial intelligence (AI) techniques. Additionally, a specialized model assesses potential roof installation sites for photovoltaic (PV) panels by calculating the possible electrical output based on peak sunshine hours, roof inclination, and orientation as well as shading effects from neighboring trees or buildings. This initiative marks the first comprehensive national survey of Germany's solar capacity and identifies significant expansion opportunities, thereby providing essential data at both the building and administrative levels to support localized and national energy policies.

The insights from Solar-DE are instrumental in facilitating targeted investments and strategies for renewable energy development, thereby contributing to the broader goals of energy transition and greenhouse gas reduction. This project not only enhances our understanding of current solar energy infrastructures but also aids in planning future expansions to meet environmental targets.

Urban decarbonisation is one of the most urgent challenges in the fight against climate change, and space technologies are emerging as critical tools to accelerate this transition. With their invaluable contributions, it is possible to better understand and manage the urban environment and promote the adoption of sustainable energy strategies.

Recent developments in digitalisation and green technologies are revolutionising the way cities can reduce emissions and optimise resource use. For example, the integration of satellite data with space-based monitoring systems allows real-time mapping of urban energy efficiency, identifying issues such as heat islands and storm runoff times. This data is essential for designing targeted interventions to reduce emissions and improve the integration of renewable energy sources.

In addition, digitalisation and the use of artificial intelligence in cities are emerging as key trends in achieving sustainability goals. Combining these technologies with advanced geospatial data not only improves energy efficiency, but also strengthens urban resilience and promotes smarter and more sustainable resource management.

In this context, the Decarbonisation Twin Support (DTS) system integrates Earth Observation (EO) data with geospatial and environmental data to create dynamic digital twins of urban environments. These digital twins are designed to monitor and optimise carbon emissions and the efficiency of renewable energy integration in urban environments. By simulating real-world conditions and analysing historical and real-time data, the DTS system provides valuable insights that can guide urban planners, architects and policy makers in reducing carbon footprints.

This approach promises to transform cities into centres of sustainability, where energy is optimally managed and emissions are significantly reduced.

Access to energy is identified as one of the basic needs towards a decent life with better chances (SDG 7). Whereas in Europe, main initiatives concentrate on implementing the European Green Deal and related national policies, in other parts of the world providing access to energy is a still ongoing task and often handled in a more pragmatic way.

The need of access to consistent and reliable energy as guarantee for economic development remains a challenge in many fast growing and rapidly densifying urban agglomerations (Africa, SE-Asia). Cities are less managed when it comes to a well-designed energy infrastructure, understanding the dissemination grid is crucial. Private initiatives and investors provide access to electricity, thus the urban fringe is often intermingled with off-grid energy units, such as mini-grids, preferably run by diesel generators. Few houses run Photovoltaic (PV) on their rooftops. Air quality is critical, space is limited and open suburban regions convert to densely populated places within few years. This is the common setting where International Financial Institutions (IFIs) engage in supporting the transition towards sustainable solutions in combination with fulfilling SDG 7, serving the ranging needs within urban agglomerations and in the rural areas.

EO can act as overarching element by providing a better understanding of the urban dynamics, and thus in tailoring the financial support accordingly. Focus is drawn to find best fitting, affordable and sustainable solutions, may this be solar rooftop solutions, hydropower, wind energy, or even biogas. This first stage does not necessarily consider commercial VHR1 satellite image data.

Within the Global Development Assistance Project on Clean Energy (GDA-CE), multi-scale approaches are sketched linked to sites if ranging extent. For Armenia, as one example, a coarse solar potential analysis was prepared on national level, benefiting from HR Sentinel-2 imagery (timeline) and most recent terrain data, going beyond the commonly used global solar atlas. A more detailed solar rooftop analysis is performed considering VHR stereo data for the capital Yerevan, emphasising on common challenges when working with spaceborne data. The latter proofs sufficient for the first stage of dimensioning potential investments. When characterising the urban structures regarding their suitability for roof-top installations itself, generic understanding of building orientation, types & sizes, distribution, and specific roof-top characteristics (obstacles, age, sub-rooftop level) is of interest. More detailed aerial flight planning is considered as far too costly and gets rather replaced by local drone flights once investment planning reaches engineering level (static).

When working with IFIs such as the World Bank, besides providing the technical solution itself, the combination of information layers of higher and lower granularity is considered valuable, as long as transferability is given. input parameters are often not optimal, project timelines are prearranged and thus creative and pragmatic solutions adapting to national specifics are necessary. This presentation provides a wrap-up on how multiple scales are reasonable within different planning stages and perspectives, showcased in multiple locations (cities and rural areas). It supports the engagements of the IFIs, being aware of limitations of regional scale vs. VHR data analyses. Benefits of receiving a most recent situational picture, linked to timeline and understanding the contextual options often surpasses this and build the base for a detailed trade-off analysis.

Smart Cities prioritise environmental impact reduction and the green economy to create and maintain healthier, more sustainable places to live and work. Almost three quarters of European citizens now live in cities and this figure is expected to reach 80% by 2050. When it comes to energy, cities are focused on a secure and sustainable supply of clean energy, as the risks of climate change and the need to reduce our carbon footprint grow ever more real. Becoming energy efficient, it means also engaging technologies to reduce emissions in the transport sector and improve infrastructure to support a green transition. The presentation will provide an overview of operational solutions -environmentally and economically sustainable-which have been developed for the needs of the cities.

As urban centres grow, the global energy landscape is undergoing a dramatic shift toward decentralization, digitization, and decarbonization to meet the 2015 Paris Agreement's stringent goals of capping the rise in global temperatures to below 2°C. This shift is even more critical in cities, where energy demand is higher.

In this ongoing urban-centric transformation, wind energy is emerging as a key player due to its efficiency and the falling costs driven by technological advances. In particular, the installation of wind turbines (WT) is proving vital for creating decentralized power systems within and around cities. This approach not only helps mitigate transmission congestion but also bolsters energy security by generating power closer to its consumption points.

However, the variable nature of wind power poses significant challenges in urban settings, where energy supply consistency is crucial. The fluctuating output of turbines can lead to periods of both surplus and insufficient energy, highlighting the need for accurate and sophisticated energy modelling. This modelling is critical to optimize urban energy grids and requires up-to-date and precise data on infrastructure locations, which is frequently lacking.

To bridge this gap, the ESA EO4Energy project employs advanced deep neural networks and leverages Sentinel-1/2 satellite imagery to map onshore WTs and identify active coal power plants (CPP). Specifically, targeting CPPs is crucial not only because they are major emitters of CO2 but also because they offer potential as heat storage solutions that could stabilize the urban grid during low wind periods by storing excess energy.

The initial results from 100 test sites are extremely promising, demonstrating the project’s effectiveness in accurately identifying WTs and CPPs. This success paves the way for applying these technologies globally to improve urban energy planning and infrastructure, ensuring cities are more sustainable and better prepared to meet future energy demands.