Enabling the public sector to save money, innovate and make more effective policy decisions by using space technology and data

Case Study: Pixalytics – Optical and Microwave Extension for Floodwater mapping (OMEF)

General Information

  • Provider: Pixalytics
  • Technology utilised: Earth observation (EO)
  • Thematic area: Natural Hazards
  • End user(s): Environment Agency

Year

  • 2016

This project was selected through an SSGP competition run using Small Business Research Initiative (SBRI), supported by Innovate UK.

The aim of this project was to provide the Environment Agency with reliable and accurate operational flood mapping in order to improve operational flood water extent and depth mapping during flooding incidents, specifically in relation to:

  • Delineation of flood waters in urban areas
  • Improving flood mapping over a large geographical area
  • Automatic dissemination of information to staff during a flooding incident

This proof of concept project successfully demonstrated the potential for an operational system by implementing satellite remote sensing algorithms from recent scientific research to improve the accuracy of floodwater delineation, and provided a simple interface to produce the results.

The project implemented four approaches and developed an ENVI extension to allow them to be run without specialist expertise:

  • Estimation of floodwater extent based on the scientific research of Matgen et al. (2011), Giustarini et al. (2013) and Greifeneder et al. (2014)
  • Application of algorithms to a variety of satellite data in areas such as lower Severn (2007), upper Thames (2014), York (2015) and Spain (2015)
  • Evaluation of the results, compared with the EA and Copernicus Emergency Management Service remotely sensed flood extents, to determine accuracy
  • Demonstrator extension to the ENVI software package, which allows the algorithms to be run on pre-processed imagery

TerraSAR-X: © DLR e.V. 2014, Distribution Airbus DS Geo GmbH, and Environment Agency flood outlines Copyright Environment Agency (2016)

TerraSAR-X: © DLR e.V. 2014, Distribution Airbus DS Geo GmbH, and Environment Agency flood outlines Copyright Environment Agency (2016)

Lessons Learned

The project developed a potential flood mapping product for end-users such as the Environment Agency, as well as secondary stakeholders such as the insurance industry who were also interested.

Costs and Benefits

It was estimated that an operational solution could be developed for around £120,000 and there are two end-user scenarios:

  1. Monthly subscription, or cost per image, of a few hundred pounds to download flood extents from web-based system run by Pixalytics
  2. ENVI module licensed to user who runs it on their own infrastructure, with an initial license cost of a couple of thousand pounds followed by an annual maintenance fee.

As an automated web-based system, the key capability required by the user would be access to the system and some limited training in its usage.

The overall impact is that it provides a cost-effective, fast and accurate flood map that is easily disseminated to organisations without the need for specialist knowledge.

Specific benefits include:

  • Faster initial flood maps, which provide improved knowledge of flood events
  • Improved decision making on the deployment of resources in response to flood events
  • Using Sentinel-1 data as the primary data source will offer savings compared to commercial imagery
  • Disseminating derived information via a web-based interface reduces the need for specialist knowledge
  • Through ENVI, more knowledgeable end-users would have the potential to produce their own flood extents.

Next Steps

The project concludes that this can become a commercial product, but there are a number of developments required for this to happen. The development of Sentinel-1 as the main SAR dataset e.g. a non-flood archive that is ready to be used is needed.

Combining SAR data with optical and altimetry data, plus new algorithms, to enhance the estimation coverage and accuracy in urban areas is also required, as well as optimising the flood map production e.g. processing speed and conversion to vector layer.