The flood-proof city, multi-layer safety approach applied to the Amsterdam region

In view of climate change and spatial developments, how can Amsterdam maintain and increase its flood resistance? This was the central question asked in an regional pilot project in which research was conducted into general-purpose strategies for the application of multi-layer flood risk management in urban areas.

The study explored:

  • the current safety risks in the Amsterdam region;
  • the potential impact of climate change and raising the flood risk management standards;
  • potential solution strategies and choices for increasing flood risk management in different districts of the city.


The Flood-resistant City is one of the six area pilot projects in the Sub-programme on Flood Risk Management within the Delta Programme, in which the necessity and consequences of new flood risk management standards and the applicability of the so-called multi-layer flood risk management approach are studied. This approach means that, through a combination of measures in different layers, the desired level of flood risk management (limiting damage/loss and the risk of victims) is achieved. The measures fall into three categories: the reinforcement of flood defence systems (layer 1), sustainable spatial planning (layer 2), and crisis management (layer 3). The joint findings of the regional pilot projects have been set down in a synthesis report.

The flood risk management task in the Amsterdam region is complex. This densely populated urban area near an industrial port area is situated within three dyke rings providing different levels of protection. There are risks of flooding coming from different sources (the North Sea, the Lek river, the Markermeer lake, and the regional water system). Two primary flood defence systems also crisscross the city; they are completely penned in by urban development. The city has both elevated sections and low-lying polder areas (reclaimed areas), as well as a large spatial planning dynamism (as seen along the Northern shore of the IJ lake). Because of the spatial developments, there is a need for clear directives on flood protection in the near future.

In the pilot project, it was found that layer 1 (prevention by means of dykes) by far provides the best and most cost-effective guarantee of good protection against flooding. It is recommended that the protection level of the Lek dyke be increased. Furthermore, measures in layer 2 (spatial planning) and layer 3 (disaster management) provide possibilities to limit the risks further. In new build and restructuring projects, there are opportunities to link measures in a cost-effective manner to vital infrastructure. The Security Regions appear to be unequipped for operational actions and management in the event of large-scale flooding. The action perspective for the inhabitants and companies there is limited.

For the successful application of the multi-layer flood risk management approach, it is necessary for the national government to form a vision for a number of areas because it requires considerations that are supra-regional:

  • It is recommended that a national vision be developed for evacuation routes stipulating that when motorways are widened or the rail infrastructure is expanded, the authorities need to consider whether below-grade construction or the construction of tunnels would be desirable from the perspective of flood risks and evacuation strategies.
  • The construction of utilities in a manner that permits compartmentalisation for each polder or district will limit domino effects on surrounding regions during a flood. This requires management from the national government.
  • A vision on the usefulness and necessity of increasing the action perspective of crisis management during large-scale floods.
  • Form a single Randstad dyke ring within which Amsterdam and Utrecht are entirely encompassed. The idea behind this is that it provides the possibility in the future to further increase the protection level in this region when its population and economic value have further increased. It is also expected that flooding in the Randstad will be considered increasingly less acceptable in view of the large economic impact it would have, social disruption it would cause, etc.

The status of the primary category C dykes is still up for discussion in the flood risk management policy, even though they are crucial for a number of regions. In the new flood risk management policy, the status should clearly describe this, particularly when the tolerable risk level of the present primary A dykes is adapted or dyke rings are merged.

Results appeared on
Project type
Municipality Amsterdam
Urban flooding
Type of project