Conference Agenda

The increasing frequency and severity of climate-driven hazardous events worldwide has strengthened the demand for operational Earth Observation (EO) tools capable of delivering timely, consistent, and scalable evidence for multi-hazard disaster risk management. Within this context, the ESA PREDICTOR project (PeRcEiving natural and anthropogenic Disaster conditions and assessing risks In Coastal regions Through artificial intelligence, traditional and nOvel synthetic aperture RADAR technologies; ESA Dragon 6, Project ID 95316) is designed to address this challenge by applying conventional SAR approaches and developing AI-driven methodologies for multi-temporal SAR-based change detection and Land Use/Land Cover (LULC) mapping. The datasets exploited within the project include ESA Copernicus Sentinel-1 data [1-3].

The presentation to be given at the ESA Dragon Symposium in Dublin, this year, will be mostly focused on showing and discussing some new papers recently published (e.g. [4]) by the PREDICTOR's authors and to give some insights on the research activities that are currently being conducted. The role of cooperation between Italian and Chinese institutions will also be magnified.

Overall, the ESA PREDICTOR framework provides a methodological bridge between current monostatic SAR multi-temporal processing and future distributed SAR sensing, combining physically motivated SAR indices with machine-learning fusion to support robust, scalable change detection for hazard monitoring [5-6]. The project is carried out within the ESA Dragon 6 cooperation (2024–2028) by CNR-IREA and contributes to the Dragon Programme objective of strengthening scientific and technological collaboration between European and Chinese EO research communities.

ACKNOWLEDGMENTS

This work is carried out within the ESA Dragon 6 Programme (Project ID 95316 – PREDICTOR) and within the Space It Up! project, funded by ASI and MUR – Contract n. 2024-5-E.0 – CUP n. I53D24000060005. This manuscript reflects only the views and opinions of the authors; the funding bodies cannot be held responsible for them.

REFERENCES

[1] R. Tomás and Z. Li, "Earth Observations for Geohazards: Present and Future Challenges," Remote Sens., vol. 9, no. 3, Mar. 2017, doi: 10.3390/rs9030194.

[2] D. Xue, Z. He, and D. Hu, "Application of radar remote sensing in landslide geohazard risk assessment," in Proc. Int. Symp. Lidar and Radar Mapping, SPIE, Oct. 2011, pp. 579–584, doi: 10.1117/12.912922.

[3] J. Im, H. Park, and W. Takeuchi, "Advances in Remote Sensing-Based Disaster Monitoring and Assessment," Remote Sens., vol. 11, no. 18, 2019, doi: 10.3390/rs11182181.

[4] Qing Zhao, Yifei Zhang, Antonio Pepe, Pietro Mastro, Taotao Zheng, Tianliang Yang, Coupled ground subsidence and rapid urbanization of the Red River delta region and the city of Hanoi, Vietnam, revealed through a Multi-Track InSAR analysis, International Journal of Applied Earth Observation and Geoinformation, Volume 144, 2025, https://doi.org/10.1016/j.jag.2025.104886.

[5] G. Krieger and A. Moreira, "Multistatic SAR satellite formations: potentials and challenges," in Proc. IGARSS 2005, Jul. 2005, pp. 2680–2684, doi: 10.1109/IGARSS.2005.1525618.

[6] M. Villano et al., "Potential of Multi-Static SAR Systems for Earth Monitoring and Their Demonstration Using Swarms of Drones," in Proc. IGARSS 2023, Jul. 2023, pp. 4586–4589, doi: 10.1109/IGARSS52108.2023.10282166.

Small-scale dynamic processes in the upper ocean layer are of key importance for our understanding of exchange processes between the land, ocean, and atmosphere at various scales. E.g., coastal dynamics in vulnerable areas such as intertidal flats need to be understood to assess their state, particularly in times of Global Change. Hence, a continuous monitoring of marine coastal environments is mandatory for an understanding of oceanic and atmospheric coastal processes, and eventually for the sustainable use of those vulnerable areas.

Sub-mesoscale oceanic eddies are important for the mixing of the upper ocean layer and the transport and distribution of material, including marine pollutants. The RP “Sub-mesoscale eddies” focuses on the detection and identification of multi-scale eddies on SAR imagery and on their spatio-temporal variation and coverage. Complete eddy features and associated parameters are retrieved by means of AI techniques and comparisons with numerical simulations. This study, for the first time, used global spaceborne SAR data to investigate ocean submesoscale eddies, advancing from case and regional studies to a global analysis.

Intertidal regions are particularly sensitive to natural and anthropogenic hazards, and we aim to optimize their monitoring by including multi-modal SAR data into existing monitoring schemes that are based on optical EO data and in-situ observations. Waterlines have been inferred from multi-satellite SAR data and serve as a basis for 3d topographic representations of exposed intertidal flats. Radar data acquired by SAR sensors working at different frequency bands are used to demonstrate the dependence of the radar backscattering on instrumental parameters and tidal phase.

A student excursion was carried out to investigate if patches of floating marine litter (macro plastic) can be detected by spaceborne SAR sensors. A wealth of SAR images was acquired from an experimental site in the German Bight, and the data demonstrate that the detection of plastic litter at medium to high wind speeds is difficult. In contrast, a ship-borne marine radar was able to clearly identify and track the litter patch that was floating at some distance from the research vessel.