Flood Control and Risk in the Netherlands

January 4th, 2009

Posted by: admin

The January issue of Wired also has an article on Dutch efforts currently underway to expand their flood control plans.  With most of the country below sea level, the Dutch started expanding their dam and flood control systems in the early part of the 20th century.  The Wired article describes the next big phase in the expansion of this system, where the annual chance of flood in high-risk areas will be reduced from one in 10,000 to one in 100,000.  By comparison, New Orleans protection has risk levels where the chance of flood in a year one in 100 (but for more severe hurricanes compared to the storms that spark flooding in the Netherlands).

This is a particularly daunting public works project, expected to take a century.  As the Dutch have experience with long term projects of this scope, it’s not likely to be as much of a shock as it would if this was their first attempt at long range geographic modification.  However, it’s hard to see a 100-year long project surviving the rough-and-tumble politics of most democracies.  While there are certainly climate change implications for what they are doing, the adaptation practices of the Dutch will set an example for engineers and urban planners around the world.  It’s worth taking a look.

2 Responses to “Flood Control and Risk in the Netherlands”

  1. Jos Says:

    This article is a typical example of how the media in the Netherlands has responded to last year’s Delta-committee’s report. Misrepresenting the science.

    The Delta-committee’s goal was to come with a worst-case scenario, not with an estimate of the most likely sea level rise, and came up with numbers between 65 and 130 cm – as the article states. However, this article states that sea levels WILL rise by that amount.

    Although the fact that the Delta-numbers are a worse-case scenario, which is clearly formulated in the report if one would take time to read it carefully, many politicians, environmentalists and journalists have misrepresented this, taking the 65-130 cm as a guarantee. The Delta-committee was heavily criticized for this – in particular the lack of mentioning what IPCC or KNMI have been estimating as best guess for sea level rise, which is more closely to somewhere around 30 cm for the next century. Furthermore, sea level has risen gradually during the 20th century (about 15-20 cm) without any sign of acceleration, so for the moment there is no immediate reason to start worrying about the worst-case. Rather, since this is a 100-year plan, and many have argued that it would be more clever to closely monitor how the science (and the real world) continue to develop and adjust the plans accordingly, rather than already planning 100 years ahead.

    In addition, I don’t know about most people, but looking at changes of ‘1 in 10.000 or 1 in 100.000 years’ has always appeared to me as a little absurd. In part because at those timescales we are for example looking at potentially totally different climates like a new ice ages, in part because we really do not know what the distribution of weather extremes is over such long periods, in part because we are talking about the long tails of a probability distribution, and statisticians will tell you that it is a very complicated game to derive meaningful changes from long tails of real world distributions, and in part because we have to no way to estimate to reliable those chances are.

    However, this is a very, very ambitious plan comes down to a guaranteed investment of 1.5 billion euros for the coming 100 years! Any engineer or bureaucrat would lick his chops – so to speak – when looking at that number and the planning involved …

    Oh, and by the way, the committee did not take the self-gravity effect into account. Probably most people have never heard of it, but we definitely will the coming years. Popular belief states that for example due to the melting of Greenland globally sea levels will rise by about 6-7 meters. However, the self-gravity effect – also known and the Glacial Isostatic Adjustment – states that because of their vast amounts of ice mass, the ice caps of Greenland and Antarctica cause a gravitational pull on the sea level which locally and regionally cause a sea level rise up to several meters. If then for example the Greenland ice cap would vanish, sea levels around Greenland will drop within a radius of around 2000 km. And up to about 6000 km sea levels will not rise by as much as can be expected from a global uniform redistribition of the melted ice water. For the Netherlands this would mean in case of a melt of Greenland that sea levels are to rise only by about 2 meters, not 6-7 meters. This is thus a particularly important issue for the Netherlands and western Europe, but unfortunately has not gotten much attention, also because it was discovered halfway the 19th century and then ‘forgotten’ for near 150 years! Only very recently it was ‘re-discovered’. As a result, IPCC 4AR only spends on small paragraph on the issue in the main report, stating that it is only a regional effect. Well, in a way it is, but it is very, very relevant for those regions and countries where it occurs.

    Anyhow, expect some papers about this topic in Nature and Science the coming years.


    Unfortunately, since this is a little known topic, I could not find a good English website explaining this in more detail. You can google for ‘GIA’ or ’self gravity’ or ‘Mitrovica’ or ‘Vermeersen’ and you will find some information. Jerry Mitrovica (U. Toronto) and Bert Vermeersen (U. Delft, the Netherlands) are two of the leading scientists in this field.


  2. 2
  3. Mark Bahner Says:

    This is exactly why a portable world-wide storm surge protection system is the best way to go.

    Here are some dates for important floods in the Netherlands. The dates are in U.S. convention, with month first. Where the flooding occurred over several days, the approximate middle day is given. Where exact dates are not known, the middle of the month (the 15th) is given.

    1) 1/31/1953
    2) 1/14/1916
    3) 11/15/1836, 12/25/1836
    4) 2/4/1825
    5) 1/24/1820
    6) 12/24/1717

    Notice how all the dates are from roughly November to February. Therefore, if a portable storm surge protection system were developed for the U.S., it is conceivable that the system could be sent over to the Netherlands (i.e., the North Sea area) after the U.S. hurricane season ends in November, to protect against storm surge during the North Sea storm season.

    Obviously, there are significant differences in storm surge for the U.S. Gulf and East coasts versus the North Sea (e.g. water temperature differences, coastal terrain differences). So the designs for the two systems might be so different that it wouldn’t be possible to use the exact same system in both places. But it’s possible that the systems could have many similar components.