rasdaman newsletter 05/2022


DynAWI: Revolution of Extreme Weather Forecasting

The DynAWI project (Dynamic Agricultural Weather Indicators for Extreme Weather Forecasting) combines Artificial Intelligence and Machine Learning methods with geospatial data delivery and processing systems to produce up-to-date, high spatial and temporal resolution maps of extreme weather hazards for agriculture.

Coupling datacube technology with Artificial Intelligence methods opens up new opportunities for analyzing large amounts of data, e.g. on phenology and weather, quickly and accurately for agricultural issues. In DynAWI, web services for crop-specific assessment and identification of extreme weather situations are established based on standardized initial data. The resulting datacube services will be integrated into the European data infrastructure GAIA-X.

The consortium partners all contribute their specific expertise in the field of indicator derivation (Julius-Kühn-Institut, Germany, and Vereinigte Hagelversicherung, Germany), datacube technology (rasdaman GmbH, Germany), Artificial Intelligence (Soilution, Germany) and soil erosion (University of Augsburg, Germany), as well as operational extreme weather risk assessment (Vereinigte Hagelversicherung, Germany).

The agricultural weather indices (AWIs) are validated through continuously measured field data to determine and represent the uncertainties of the indices. The large amount of data from various Earth observation data and the complexity of the analyses requires datacube technology for implementation. The rasdaman datacube technology provides the underlying core functionality here.

With three selected extreme weather situations (drought, late frost and soil erosion by water) as use cases the project will first identify optimal AWIs adapted to local site conditions. From this, models for region-specific spatio-temporal forecasts of extreme weather situations will be developed.

DynAWI is in part funded bei German Federal Ministry of Agriculture and Food (BMEL).