Use of Satellite Remote Sensing Data in Flood Analysis

Back

Client Details:

Satellite Remote Sensing Flood Analysis Athlone

Customer Name:Use of Satellite Remote Sensing Data in Flood Analysis

Location:Athlone, Ireland

Athlone, Co. Westmeath: A Case Study

Floods and flooding impact more people globally than any other type of natural disaster! Floods have the potential to devastate entire communities and cities, destroy homes, take lives and even spread disease. With global warming having a greater effect on our climate than ever before, an increasing global population and a growing amount of people living in flood-prone regions, worldwide vulnerability to flooding is set to increase further in the near future.

Detecting and monitoring flood events globally is crucial to our understanding of the processes, how it happened, how it can be prevented, how it can be managed, how can we prepare better in the future and how can we best warn people situated in at risk areas. In the midst of a disaster, information is needed as quickly as possible to provide an overview of the situation, to improve crisis management and to help response efforts.

With an unprecedented amount of satellites monitoring the Earth on a constant basis, there has never been more data available to help us assess, monitor, analyse, model and view floods around the globe as and when they happen. The data and imagery which satellites provide have proved invaluable as countries around the globe seek to identify at risk areas and inform planning to help mitigate the impact of future flooding.

The risk and impact of floods on local communities has never been more stark, with the cost of cleaning up from the severe floods which hit various parts of Ireland at the end of 2015 and the beginning of 2016 estimated to be upwards of €100 million. As a result of those and other floods across Europe, the demand for and access to satellite remote sensing data and imagery has increased greatly.

What We Did

Following on from the devastating floods that hit Ireland at the beginning of 2016, Mallon Technology decided to look at satellite data and imagery of flood hit areas to examine the impact of the floods and to highlight just how such data can be used in the planning of future flood prevention measures. We decided to focus on Athlone, Co. Westmeath for this case study as it is an area of the country which has been affected by a number of destructive floods in recent years.

To analyse the floods that hit Athlone in January 2016 we used data collected from the European Space Agency's Sentinel-1 satellite. Sentinel-1 is an Earth Observation satellite, offering key information services to help improve our understanding of the effects of climate change. Carrying a 12m long advanced synthetic aperture radar (SAR), working in C-band, Sentinel-1 is able to capture images of the Earth's surface through rain and cloud, regardless of whether it is day or night.

We chose to analyse imagery gathered from Sentinel-1 for two reasons. Firstly, the radar data provided by the satellite is particularly effective and useful in assessing the impact of flood impacted areas as unlike other remote sensing satellites it is able to penetrate rain and cloud cover, issues which generally affect flood hit areas. This also means that a flood can be monitored in real-time as it progresses, with a clear picture of the height and extent of it emerging. Secondly, Sentinel-1 offers regular coverage of the Earth's surface, with a revisit period of only 6 days. This allows us to capture an affected area before a flood to help visualise and identify any changes to the landscape.

For our analysis of the flooding in Athlone we used two Sentinel-1 radar images. One captured before the flooding and the other captured at the height of the floods. The first image was captured on 17/10/2015, pre-flooding. Athlone town, the River Shannon and Lough Ree are all clearly visible in the image. The image is composed of three light bands; the VH band (projected in red), the VV band (projected in green) and the VV band+VH band (projected in blue).

The second image was captured on 09/01/2016. This shows the same area around Athlone after a sustained period of heavy and persistent rainfall. There is a marked difference between the two images. The second image clearly details the extent of the flood with vast swathes of area along the River Shannon and around Lough Ree clearly under water. The image is composed of the same three bands as the first image.

By merging the VH bands from both of the radar images above, we were able to produce a third, change detection image. The VH band for the first image is projected in red and the VH band for the second image is projected in green. When merged, the red areas in the change detection image are a result of very low signal returns in the VH band of the first radar image compared to the much higher signal returns for these same areas in the VH band of the second image. Low signal returns indicate areas with low levels of texture and in this case, these areas clearly indicate the presence of water.

Applications

Following the production of the change detection image, we were able to use the data to create a simple vector layer of the flooding. Utilsing advanced spatial analysis tools, we were able to produce an image composed of pixel values, showing the areas of maximum and minimum change. An unsupervised classification was then carried out on the image to group those pixels of maximum change, with the group exported as a vector layer. The layer was further refined using various geo-processing tools to produce a vector layer of the flooding which can be overlaid on various basemaps and other imagery.

The image above shows the flooding overlaid on an OCM Landscape basemap. The usefulness of creating a vector layer of the flooding really becomes apparent when it is overlaid with other GI data. This enables us to analyse areas and features which intersect with the flooding layer and which could be impacted by the flooding. For example; a database of residential and commercial properties could be overlaid on the vector layer to quickly detect properties which are directly affected by or are within a certain distance of the flooding area identified. Similarly, census data could be used to pinpoint vulnerable populations within the flooded area. This could help inform decisions made for response plans to put measures and resources in place which ensures those populations that require extra assistance are attended to first.

The Benefits

The study of flooding in the Athlone area, using satellite remote sensing data, has highlighted a number of potential benefits for using such techniques. Benefits include:

  • First responders are able to quickly and easily access data which can help assess the situation and inform any rescue efforts
  • Sentinel-1 data is usable even during adverse weather conditions
  • Flood-prone and at risk areas are easily identifiable
  • Such studies can help inform future decisions on flood planning and prevention measures, including important areas to protect, such as densely populated regions
  • There are a range of satellites which provide free and easily accessible data, including the Landsat and Sentinel series of satellites
  • Historic analysis can be carried out on particular areas to help identify trends and patterns in flooding
  • Frequent revisit times ensure that there is a rich amount of data available for a wide range of areas
  • Sentinel series of satellites collect a range of different data all of which can be used to help visualise and analyse any particular area of coverage
  • Analysed data can be overlaid onto various basemaps and other imagery

Conclusions

As the Athlone case study has shown, the use of satellite remote sensing data is clearly able to visualise the extent of flooding in the area in January 2016. Such information is invaluable as it is able to help inform a range of decisions across a host of businesses and organisations.

The mapping and remote sensing services seen in this case study can be used to provide value to organisations across multiple industries. Industries which can benefit from these services include:

  • Agriculture - to help in the verification or analysis of agricultural land usage to inform drainage and and flood planning
  • Insurance - to determine properties at risk from flooding and to inform premium rates
  • Governmental - to identify key areas in need of protection and to help inform measures to prevent or mitigate future floods
  • Transport - to help identify infrastructure which could be at risk of floods or which requires extra protection and prevention measures

Why us:

  • Geospatial and remote sensing expertise
  • Large flexible team of geospatial and remote sensing experts
  • Experience handling large and complex datasets
  • Ability to process and analyse satellite imagery
  • Experience of advanced spatial analysis techniques
  • Experience in a range of GIS software applications