Geospatial Artificial Intelligence (GeoAI) — the convergence of satellite-derived data with modern machine-learning techniques—has become a decisive factor in global disaster-risk governance. In little more than five years, the active Earth-observation satellite fleet has tripled, and open-weight vision models now convert petabytes of imagery into decision-ready layers in minutes.
Vienna, 23 June 2025 – The Living Planet Symposium (LPS 2025), organized every three years by the European Space Agency (ESA), took place from 23 to 27 June 2025 at the Austria Center in Vienna, Austria. LPS brought together thousands of Earth observation scientists, policymakers, and industry leaders to explore how satellite data can drive climate action and sustainable development.
This is event is available for participation on an ongoing basis
AI for Good is the United Nations’ leading platform for harnessing artificial intelligence to address global challenges. It brings together a diverse network of stakeholders, including policymakers, researchers, industry leaders, and civil society, to explore and promote the responsible and impactful use of AI technologies.
Recommended Practice: Flood Mapping Practice Using Sentinel-1 and Sentinel-2 Imagery
UN-SPIDER has published a new Recommended Practice that leverages both Sentinel-1 Synthetic Aperture Radar (SAR) and Sentinel-2 optical imagery to improve flood detection and mapping. Developed by the Regional Centre for Mapping of Resources for Development (RCMRD), this method offers a multispectral and radar-based approach to identifying flood-affected areas with greater precision, especially in regions with persistent cloud cover or challenging terrain.
On 15–16 May 2025, the Ministry of Emergency Situations (MES) of Kyrgyzstan hosted a two-day training workshop titled “Multi-hazard Risk Assessment for Risk-Reduction Planning.” Organised by UNOOSA/UN-SPIDER in partnership with ESCAP-APCICT, the Geoinformatics Center (Asian Institute of Technology) and ITC–University of Twente, the course addressed the country’s overlapping threats from earthquakes, landslides, floods and climate-related extremes.
Worldwide, storm surges pose a threat to coastal communities. Particularly in low-lying African coastal cities, the impact of climate change induced storms and sea level rise is a threat to the vulnerable community (Nhantumbo et al. 2023). Especially with climate change, this hazard is likely to intensify (IPCC 2022), calling for urgent action. To identify the exposed groups, assets, and infrastructure, and to implement effective storm surge risk management, an understanding of the hazard, its timing, and geographic patterns is crucial. Hazard maps can serve as a first approximation for estimating the spatial extent of storm surges and sea level rise, as well as the exposed population and assets.
For detailed hazard maps, hydrodynamic models are often used due to the complexity of storm surges. They are influenced by factors such as storm intensity, the storm's track, bathymetry, and more. Hydrodynamic modeling therefore requires high-quality and complex parameters, such as historical data sets or detailed coastal formation data for a specific area, which are not always available or may be costly to obtain. This makes free, detailed modeling of storm surges very challenging.
Other methods, such as the bathtub approach, require fewer datasets and can serve as a first approximation for modeling the potential geographical extent of coastal flooding. This Recommended Practice outlines the steps to visualize the geographical extent of coastal flooding or sea level rise using just two datasets: a digital elevation model (DEM) and a shoreline dataset. The choice of a DEM is crucial; preferably, a high-resolution DEM should be used, as it is more likely to represent the surface accurately. However, high-resolution datasets are not always available for free. In such cases, a mid-resolution DEM, such as the Copernicus 30m DEM, which is globally accessible, should be used when open-access DEMs are the only option.
To perform the step-by-step procedure, it is required to install QGIS (preferred 3.28.15 Firenze or newer version).
Required Datasets
Digital Elevation Model (e.g. Copernicus 30m DEM, WorldDEM neo)
Coastline vector layer (e.g. OpenStreetMap water-body layer, WorldDEM Ocean shoreline a product of Airbus)
You can freely download the Copernicus 30m DEM from the OpenTopography, CODE-DE or the Copernicus Data Space Ecosystem website as described on the step-by-step page.
Develop scenarios of impacts of storm surges in specific coastal areas;
Conduct an initial assessment of risk associated with storm surges;
Improve storm surge early warning systems using scenarios of the extent of coastal flooding.
Strengths:
Can be used in all coastal areas of the world, the COP30 DEM and other commercial DEMs are available globally
Free and open-source software can be used (e.g. Quantum GIS)
COP30 DEM:
Resolution: 1 arcsec (approx. 30m)
Absolute vertical accuracy: < 4m (90% linear error)
The globally accessible Copernicus DEM has a resolution of 30 m. When using this dataset, finer details may be omitted due to its resolution.
To address this limitation, alternative datasets with higher resolution can be considered. One such dataset is the WorldDEMneo from Airbus. However, for some regions there are restrictions regarding geographic availability. Consult with Airbus to determine whether data can be supplied for your region of interest.
IPCC (2021): Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. V. Masson-Delmotte, P. Zhai, A. Pirani, S. L. Connors, C. Péan, S. Berger et al.United Kingdom and New York, NY, USA.
Nhantumbo, B., Dada, O. A. and f. E.K. Ghomsi (2023): Sea Level rise and Climate Change – Impacts on Africal Coastal Systems and Cities. DOI: 10.5772/intechopen.113083
Geospatial Artificial Intelligence (GeoAI) — the convergence of satellite-derived data with modern machine-learning techniques—has become a decisive factor in global disaster-risk governance. In little more than five years, the active Earth-observation satellite fleet has tripled, and open-weight vision models now convert petabytes of imagery into decision-ready layers in minutes.
