Bart Schultz
The value of public and private property per unit area is many times higher in urban areas than in rural areas. Therefore, the design standards for flood protection for urban areas should normally be considerably higher than those for rural areas, although this is not always possible in densely populated areas. In such cases, the measures need to be based on the urban situation.
In the Netherlands, the physical measures for protection against flooding for polders that could be flooded from the sea or the IJsselmeer are between 0.001 and 0.025% per year, from the rivers at 0.08% per year and from the canals at 0.1 to 0.5% per year. There is a major programme to increase these levels even further. In other countries, the levels are much lower with generally a probability of flooding of 0.5 to 5% per year. In such cases, when an extreme event really occurs, there is a serious risk of loss of a large number of human lives, large-scale evacuations of people from the flooded areas and enormous damage.
In particular, rapid urbanisation in flood-prone areas in emerging countries has led to flood protection levels generally being well below the economic optimum. Developments in the least developed countries are characterised by similar processes, albeit at a slower pace. Flood damage in rural areas mainly concerns crop yield losses and in coastal polders the risk of inundation by saline water. Design standards would have to be such that these flood risks are reduced to a level of normally in the order of magnitude of a probability of occurrence of 5 to 10% per year.
Inadequate flood protection is even prevalent in some developed countries. For example, the flood protection measures implemented after the flooding caused in 2005 by Hurricane Katrina in the New Orleans area have only a level of safety with a probability of failure of 1% per year. This means that if a hurricane of the same strength as Katrina were to hit New Orleans, flooding could again occur. This almost happened in 2008 when Hurricane Gustav approached New Orleans but fortunately changed its course at the last moment, and in 2021 when Hurricane Ida followed more or less the same path.
Some characteristic design standards for flood protection:
|
City/country |
Chance of failure in percent per year |
|
Dhaka, Bangladesh Australia, England and United States (including New Orleans after Katrina) India: cities and industry Rural area China: large cities cities rural area British Columbia, Canada |
2 1
1 4 0.5 1 5 0.5 |
Design standards would have to be linked to the economic value and potential damage to public and private property in a polder, as well as to the risk of loss of life and evacuation. It is generally difficult to value the risk of loss of life. A well-known approach is therefore to determine the economically optimal level of protection, take into account the population in the polder and finally make a political decision on the design standard for the flood protection measures.
The effect that the increase in the value of buildings can have on design standards for flood protection measures can be shown graphically, showing that as the value of buildings in an area increases, flood damage increases proportionally. As a result, the optimum level of safety will also increase. For example, in the attached theoretical graph, from a probability of occurrence of approximately 4% per year in 2020 to a probability of occurrence of 0.1% per year in 2070.
At this probability, the total cost of flood protection and damage is minimal. This would require significant investments in flood protection just to maintain the economically optimal level. This theoretical example does not even take into account the increase in the number of people in the flood-prone area.
