Storm surges and tidal waves are global phenomena that considerably affect human populations in coastal and island regions. According to the Guide to Storm Surge Forecasting published by the World Meteorological Organization in 2011, storm surges can be defined as “oscillations of the water level in a coastal or inland body of water in the time range of a few minutes to a few days, resulting from forcing from atmospheric weather systems. According to this definition, the so-called wind waves, which have durations on the order of several seconds, are excluded”. Storm surges are a coastal phenomenon triggered by strong winds in the oceans and seas due to tropical cyclones and other similar weather systems at sea.

Tsunami modelling, sea-level rise studies and storm surge hazard mapping have been done using deterministic and probabilistic models. However, deterministic models require precise oceanographic data, as well as data on bathymetry in the coast, coastal geometry and high-resolution digital elevation models in the coastal area and ancillary data on surface roughness in coastal areas. In many developing countries these data sets are not available.


Computer Requirements

To perform the step-by-step procedure, it is required to install Quantum GIS (3.6.0-Noosa or newer version).

Required Datasets

  1. Digital Elevation Model (e.g. WorldDEMTM a product of Airbus Defence and Space, please refer to this website for the study area
  2. Coastline vector (e.g. WorldDEMTM Ocean Shoreline a product of Airbus Defence & Space, please refer to this website for further information

In case you want to test the Recommended Practice with WorldDEMTM data, please get in contact with Airbus through the contacts mentioned on the overview page.

The precision and reliability of the results of modelling strongly depend on the quality and resolution of the Digital Elevation Model (DEM) used as basis for the analysis. The high resolution and level of detail provided by WorldDEMTM data allows an accurate delineation of the water distribution and of potential water streams between structures and buildings.

The Airbus Defence and Space elevation data WorldDEMTM is more sensitive to the distribution of water than the SRTM DEM: the use of the WorldDEMTM can lead to a more precise estimation of the effect of coastal flooding caused by e.g. storm surges or tidal waves.

Comparison between WorldDEMTM and SRTM results
Figure 1: Comparison between WorldDEMTM results vs SRTM results. SRTM shows completely different results


  • Develop scenarios of impacts of storm surges in specific coastal areas;
  • Conduct an initial assessment of risk associated with storm surges;
  • Improve storm surge early warning systems using scenarios of the extent of coastal flooding.

Strengths and Limitations: 


  • Can be used in all coastal areas of the world, the WorldDEMTM data is available globally
  • Free and open-source software can be used (e.g. Quantum GIS)


  • Resolution: 0.4 arcsec (approx. 12m)
  • Absolute vertical accuracy: < 4m (90% linear error)
  • Relative vertical accuracy: < 2m (slope <= 20%) / < 4m (slope <= 40%)

Limitations and Restrictions:

The WorldDEMTM data is globally available. However, for some regions there are restrictions regarding geographic availability. Consult with Airbus Defence and Space to determine whether data can be supplied for your region of interest.

Figure 2: Workflow of using elevation data for storm surge coastal flood modelling


This Recommended Practice allows users to visualize the geographical extent of coastal flooding or sea level rise on local, regional or global scale (depending on the resolution and accuracy of the incoming digital elevation model). It can be used exclusively as a first approximation to determine areas that are prone to inundation and can serve as a first assessment for further, more in-depth analysis of coastal flood and sea level rise assessment. The Recommended Practice is developed using the World Digital Elevation Model (WorldDEMTM) product of Airbus Defence and Space. For the sake of clarity - the Recommended Practice has not been developed for any other use and purpose than the above described one and is consequently not usable for and in navigation, any hazardous environment requiring error free performance.

Disaster type: Flood and Severe Storm

Disaster Cycle Phase: Preparedness

Test Site: 

Larger Accra region, Ghana


The coastal region of Ghana was heavily affected by tidal waves in June 2017. Many people have been displaced and houses, infrastructure and fishing gear (boats, nets) have been destroyed. This Recommended Practice can be a first assessment to apply further analysis to identify safer ground for relocation of exposed communities. For more information please refer to following link provided by the National Disaster Management Organization of Ghana (NADMO): http://nadmo.gov.gh/index.php/archive/13-nadmo-articles/71-nadmo-tours-….


The model can be applied to any coastal region of the world.

Data Access: 

This Recommended Practice requires the use of commercial product from Airbus Defence Systems. 

The procedure makes use of the free and open Quantum GIS (QGIS) desktop software.

Data Preparation/Pre-processing: 

1. Load data in QGIS

First all data relevant for coastal flood modelling has to be loaded in QGIS: The Digital Elevation Model raster file and the Ocean extent vector file. Figure 3 shows an example of the coastal region in Accra Region in Ghana, where the WorldDEMTM elevation data and the ocean extent of the WorldDEMTM Ocean Shoreline have been used.

WorldDEMTM ocean shoreline
Figure 3: WorldDEMTM and WorldDEMTM Ocean Shoreline for N05W001, Accra region, Ghana

2.Raster calculation

In the next step, all elevation pixels below a certain elevation threshold will be detected. Opening the Raster Calculator through Raster > Raster Calculator the window shown in Figure 4 will pop up.

First, select your DEM layer and click on the ‘less or equal’ math operator to create a Raster Calculator Expression. Then set the threshold for the flood level you would like to simulate. In this sample case a value of 5m has been set. Furthermore the Output layer as well as Output format has to be set.

The result is a binary layer classifying all potential flooded areas by 1. An example is shown in Figure 5, where all pixels below or equal 5m are indicated in blue.

QGIS Raster Calculator
Figure 4: QGIS Raster Calculator


Potentially flooded areas
Figure 5: Potentially flooded areas (flood level of 5m) indicated by blue for N05W001, Accra region, Ghana

3. Define NoData value

All dry areas in the classification mask are now set to NoData using the QGIS Translate tool. The tool is accessed via Raster > Conversion > Translate (Convert Format). Besides setting the Input layer (Classification from previous step) and Converted output file, the NoData value has to be assigned to 0.

QGIS Translate tool
Figure 6: QGIS Translate tool

4. Polygonize

After defining the NoData value for the classification layer, the file is going to be converted to a vector file using the Polygonize tool. After opening the tool through Raster > Conversion > Polygonize (Raster to Vector) the input and output files have to be defined. In addition the user can decide if the 8-way-connectivity should be considered for the conversion.

QGIS Polygonize tool
Figure 7: QGIS Polygonize tool

Processing Steps: 

5. Filter inland water patches

To filter inland water patches, which fulfill the elevation threshold but are not connected with the ocean, the Select by Location tool (Vector > Research Tools > Select by Location) in QGIS is applied (see Figure 8). Therefore the polygonized potential water extent has to be intersected with a coastline vector to filter inland patches. The features of the polygonized layer from the previous step have to be chosen as Select features from and will be compared with features from a costal vector file. The geometric predicate should be set to ‘intersect’ and a new selection should be created.

After the new selection has been created the polygons will be saved as a new water extent layer for the modelled flood height through a right-click on the polygonised layer (Export > Save Selected Features As), see Figure 9. Take care that the option Save only selected features is activated.

Figure 10 and 11 show the flooded areas in case of an event with a flood of 5m compared to the old ocean extent for the Accra region, Ghana.

QGIS Select by Location tool
Figure 8: QGIS Select by Location tool


QGIS Saving of Vector Layer
Figure 9: QGIS Saving of Vector Layer


Flooded regions
Figure 10: Flooded regions in case of a flood level of 5m for N05W001, Accra region, Ghana


Flooded regions
Figure 11: Zoom-In: Flooded regions in case of a flood level 5m for N05W001, Accra region, Ghana


6. Example map

Potential coastal flooding
Figure 12: Example of a map for Ghana displaying different levels of coastal flooding.