Picture courtesy of Art Rock (Hennie)
If you have been around any Dutch people for the last couple of days, you might have noticed something strange. Whenever the temperatures drop below zero degrees, the Dutch tend to get a bit excited. After all, we are known to be a nation of ice skaters. But it’s not until it gets really cold, and for a longer period of time that we tend to get really aroused. Other nations observe us in horror as we switch into gear: it might be the year of another Elfstedentocht.
The Elfstedentocht, or eleven city tour, is the stuff of Dutch legends. Where the British have King Arthur and ‘1966’, we have this – despite the Americans’ best efforts to sell us Hans Brinker and his dike-plugging abilities. The tour consists of 200km, or 125 miles, of ice skating across the Northern region of the Netherlands known as Friesland, visiting 11 different towns and cities, often in gruelling weather conditions. As the ice has to be at least 15cm thick along the entire route, the Tour of Tours doesn’t happen very often: a mere 15 times over the last century, with the last rendition on 4 January 1997.
With the ice mega-marathon being so rare, speculation on whether there will be another one, and when exactly it will happen has become a sport in its own right. The Royal Dutch Meteorological Institute (KNMI) prides itself on being the only institution in the world with an official ice growth model. An example of the model’s predictions is shown in Figure 1. On the Y-axis, numbers above 0 represent water temperatures, while sub-zero figures show predicted ice thickness. The bottom graph shows predicted air temperatures. The model uses data of a high resolution (specific ‘weather’ data for a 16×16km grid), represented by the red line, and data of a lower resolution (using a 32×32km grid, the blue line) to predict ice thickness given 50 slightly different scenarios (the green lines).
Figure 1: Predicted ice thickness in the Netherlands, as of 8 February 2012.
Looking at these graphs, things are starting to look pretty good: ice thickness is where it should be, and the big freeze seems to have settled for at least the end of the week. Reason enough for the ‘rayonhoofden’, 22 regional ice managers, to gather for a meeting for the first time in 15 years last Sunday. However, as George E. P. Box put it: all models are wrong, the practical question is how wrong do they have to be to not be useful. Although the Dutch ice growth model is pretty sophisticated, taking factors such as wind chill and measurement uncertainty into account, it only models ice growth for 2-meter deep stagnant water in central Holland. As you might imagine with a 200km long route, not all of the waters and canals abide by that assumption.
Although a large part of the route is covered by the much desired ‘zwart ijs’ (black ice or smooth, hard ice), there are still holes in some places, caused by anything from fierce winds, to geese and snow. But with still half a week of freezing temperatures to go, insufficient ice quality can be dealt with. Ice transplantations, in which ice gets dumped into ice holes to encourage them to close are in full swing, and even the army has been called in to sweep snow, which can act as insulation, from the ice on the route.
The Dutch are desperate for an Elfstedentocht. The tour is already a rare event, and with average temperatures predicted to climb over the 21st century, the prospect of seeing thousands of ice skaters race across the canals of Friesland seems ever further off. The next couple of days will be vital. The Dutch will try and keep their cool, but whether it will be enough for the Elfstedentocht only time will tell.
Figure 2: Model for relative probability for an ‘Elfstedentocht’ (ice thickness > 15cm) in the 21st century. ‘Relatieve kans’ = relative probability Red line: IPCC highest predicted temperature increase White line: IPCC mean predicted temperature increase Black line: IPCC lowest predicted temperature increase