GEO Disaster SBA S&T review – Ankara, Turkey, 25 March 2010 EUCENTRE proposal for GEO The Remote Sensing Section of EUCENTRE, after careful consideration of the GEO Workplan, has submitted a proposal, supported by the Seismic Risk section, to address seismic vulnerability mapping based on Earth Observation (EO) The EUCENTRE proposal was endorsed by UN- SPIDER and included into the WP as the leading theme of the subtask DI-09-01a, “Vulnerability and Risk Mapping” Fabio Dell’Acqua, Diego Polli EUCENTRE – University of Pavia

GEO Disaster SBA S&T review – Ankara, Turkey, 25 March 2010 Proposal pillars Framework with a combination of methods Addressing different aspects of vulnerability (physical, human, ease of emergency intervention and recovery, …) EO-based, but not exclusively (ancillary, GIS …) Initially focussed on urban areas Extraction of features of interest from satellite images (e.g. number of floors, ratio between window and façade areas, width of vertical walls between windows, …) Fabio Dell’Acqua, Diego Polli EUCENTRE – University of Pavia

GEO Disaster SBA S&T review – Ankara, Turkey, 25 March 2010 The processing chain Fabio Dell’Acqua, Diego Polli EUCENTRE – University of Pavia

GEO Disaster SBA S&T review – Ankara, Turkey, 25 March 2010 Case study: Messina (Italy) Fairly complete dataset, including satellite optical and airborne radar We have proposed Messina as a possible supersite (ISPRA is willing to contribute relevant data) optical image & radar image of buildingvulnerability curves Fabio Dell’Acqua, Diego Polli EUCENTRE – University of Pavia

GEO Disaster SBA S&T review – Ankara, Turkey, 25 March 2010 Case study: Priolo (Italy) Vulnerability map Detection of critical features Seismic vulnerability model High-resolution satellite image Fabio Dell’Acqua, Diego Polli EUCENTRE – University of Pavia

GEO Disaster SBA S&T review – Ankara, Turkey, 25 March 2010 Case study: Caribbean islands Optical QB images + TerraSAR-X acquisitions Output in a GIS-compatible format test sites (in red)sample vulnerability map Fabio Dell’Acqua, Diego Polli EUCENTRE – University of Pavia

GEO Disaster SBA S&T review – Ankara, Turkey, 25 March 2010 S&T issues some parameters relevant to vulnerability estimation: either can not be extracted from remotely sensed data at all, or they can, but only by relying on weak correlations limited availability of VHR radar data integration of: proximity sensing “community remote sensing” (voluntary contribution of information under various forms) Fabio Dell’Acqua, Diego Polli EUCENTRE – University of Pavia

GEO Disaster SBA S&T review – Ankara, Turkey, 25 March 2010 Conclusions The following goals were reached: A vulnerability model has been chosen A processing chain has been set up and experimented The chain is not yet automatised, although some of its component are We lack accurate ground truth to assess the quality and reliability of results Ideas for funding a currently underfunded activity? Fabio Dell’Acqua, Diego Polli EUCENTRE – University of Pavia