Delft-FIAT: An open-source flood impact analysis tool

Presenter : J. Sala

Type: Oral

Flood impacts, such as damage and casualties are increasingly important in flood risk management. Estimation of flood impacts enables policy makers to decide on the type of measures to reduce flood risk, and how much investment is warranted. A flexible, generic tool (Delft-FIAT) has been developed using Python, NumPy and GDAL to enable the quick set up a flood impact model in conjunction with the OGC WMS to visualize the results in maps. Delft-FIAT uses the unit loss method, which requires a combination of different data layers: (i) Flood hazard variables; (ii) land-use or objects information (location, type and maximum damage); and (iii) impact functions (usually relating some hazard variable e.g. water depth to an impact or damage fraction).
The main advantage of this open-source tool is that it is flexible, as it can take up many different types of impact categories and these can be easily adjusted by the user. Moreover, it is efficient because the calculation engine can handle very large and high resolution input maps thanks to block division techniques in the input/output rasters and vector files.
The presented tool is a good example of how open source geospatial software can contribute to environmental monitoring and in particular to flood risk simulations. The tool has already been applied in more than 10 flood risk studies around the world on different spatial scales. These studies show that Delft-FIAT enables researchers to focus on the data and impact knowledge rather than the calculation technology.

Keywords: flood risk, flood impact, simulation software, water, open geospatial standards, data visualization, disaster risk management