WASDI, a cloud platform for Earth Observation and Natural Hazards

We present WASDI, an open-source, federated multi-cloud platform for Earth Observation (EO) data that has been used successfully in several impactful international projects in the field of Natural Hazards (NH), in particular with floods and wildfires. 

WASDI helps EO developers turn their algorithms into operational services in the cloud. These can also be published in a marketplace where end users can exploit them. The platform automatically interoperates with several data providers. The accessible data includes observations (such as those from the Copernicus and Landsat programmes), derived products (such as the Copernicus DEM and the ESA World Cover), and the output of simulations (such as ECMWF ERA5 reanalysis). A single interface simplifies the access and thus the development of algorithms that operate data fusion using different sources.

Thanks to these capabilities, WASDI is used to daily monitor floods in several South East Asian countries, in the context of a World Bank initiative to promote parametric reinsurance of the financial risks associated with floods. The extent of the floods on bare soil are mapped mainly using an algorithm developed by the Luxembourg Institute of Science and Technology (LIST), using Sentinel-1 data, and integrated by an algorithm that uses Sentinel-2 data developed by the CIMA Foundation. Urban floods are mapped with great accuracy using Sentinel-1 thanks to an innovative algorithm developed by LIST. These algorithms have been deployed to the platform and are now available in its marketplace.

Recently, these technologies have been used to perform an extensive assessment of the effects caused by the very large floods in Pakistan in 2022 to support the Asian Development Bank (ADB) recovery efforts in the framework of ESA’s Global Development Assistance (GDA) Disaster Resilience program.

The platform features also other algorithms, such as those for the detection of active fires based on Sentinel-3 data, which have been used for the assessment of wildfires in the Greek island of Evia in 2022, and for mapping burned areas using couples of Sentinel-2 images, developed by CIMA Foundation and used by the Civil Protection Department of Italy and other countries. This algorithm was also  used to help assess the impact of the wildfires of 2020 in Ukraine, close to the Chernobyl power plant.

These cases demonstrate the effectiveness of Earth Observation data, algorithms and cloud technology in case of natural hazards for prevention, response and assessment.