According to a last provisional report, drawn up this Sunday, July 18, more than 175 people perished in the violent bad weather which affected part of central Europe in early July. Germany pays the heaviest price with more than 150 dead; followed by Belgium, where there are dozens of deaths. Research to find the missing is still ongoing, suggesting a heavier toll.
How can a natural disaster claim so many victims when it has been predicted with so much anticipation? What is the role of climate change in this situation? How to best anticipate this type of event? So many questions to which we will try to provide answers.
1. How to explain these sudden floods in Germany, Belgium, Luxembourg and the Netherlands?
They are explained by the very heavy rains which occurred on July 13 and 14 in these regions. For example, the Deutsche Wetterdienst (DWD, the German meteorological service) recorded 154 mm of rain in Cologne on July 14, which corresponds to 154 liters of water fallen per square meter! A real deluge which, because of its high intensity, cannot infiltrate, directly feeding the rivers by runoff.
To make matters worse, this rain followed a particularly rainy day of July 13, with soils already moistened by the rains of the last few days.
As for the meteorological situation which allowed such rains, it is relatively classic for central Europe, even if its duration is exceptional.
It is an isolated pocket of cold air - meteorologists call it a "cold drop" - which has a natural tendency to hang over a large area. This blockage favors very important rainfall accumulations in a limited space. Around the cold depression, the hot air (with a high water vapor content) condenses and causes significant waterfalls.
2. How to explain such floods in summer?
For a Frenchman, what is striking about these floods - beyond the heavy human toll - is above all the season: in fact we rather associate the floods with winter (remember that the great flood of 1910 in the North of France took place in January), possibly in the fall in the Mediterranean area.
But before exclaiming that there are no more seasons, we must accept to look beyond our borders to see that the great floods of Central Europe generally occur in summer: we can recently cite the great flood of the Elbe in June 2013, the great floods of the Danube in June 1965 and June 2013, the great flood of the Oder in July 1997. In the valley of the river Ahr, particularly affected by the floods of the latter days, the major reference floods date from July 1804, June 1910 and June 2016.
French hydrologist Maurice Pardé created a special class for these events - "Central European type floods" - which he already explained by a phenomenon of "cold drop" similar to that observed in recent days.
3. Does climate change play a role in this situation?
The phenomenon which has occurred in recent days can be said to be “classic”, both from a meteorological and hydrological point of view. At first glance, there is no need to invoke the role of climate change to explain it.
On the other hand, what can and should hold our attention is that the rainfall intensities recorded (and the accumulations) are strictly speaking “extraordinary”.
This development corresponds to what meteorologists predict as a consequence of the rise in temperatures, in under the law of Clausius Clapeyron which relates the maximum quantity of water vapor that the atmosphere can contain and the temperature of the latter; and which suggests an increase of 7% of the total quantity of water vapor in the atmosphere per additional degree Celsius: it is this increase that we can fear for heavy rains.
Other recent work on floods can also enlighten us: in an article published in July 2020 in the journal Nature, Professor Günther Blöschl (Technical University of Vienna) has shown that if the current period is not unique in the history of Europe by the abundance of flood and flood phenomena, it is unique by its temperature.
Indeed, if other relatively "rich" periods in floods existed in Europe in the past (1560–1580, 1760–1800 and 1840–1870), they were all colder than average, while the recent period was stands out with a significantly higher temperature compared to long-term averages.
4. Was this extreme event foreseen?
What will undoubtedly strike the specialists most in this event of July 2021 is that the exceptional nature of the rains which fell on the west of Germany and Belgium had been foreseen as of July 12: the Center European medium-range weather forecasting (ECMWF) had indeed announced the very high probability of exceptional rains and issued an alert transmitted to the German meteorological service (DWD).
These forecasts also seemed particularly reliable, which is quite unusual.
Here is what allows us to affirm it: because of the great difficulty of atmospheric modeling, weather forecasts are based on “sets” (we say that the forecast is probabilistic). These sets are obtained by slightly disturbing the initial conditions of the computation and by simultaneously generating several tens of forecasts, which makes it possible to evaluate their degree of certainty.
As of Monday, July 12, more than half of the scenarios indicated the possibility of extreme rains, which is rare and has certainly caught the attention of German forecasters. We can therefore ask ourselves whether the warning message (and above all, the exceptional nature of the forthcoming precipitation) has indeed arrived at the local level. Especially since it is not enough to alert, the population must be able to take shelter and the local authorities must initiate protection and relief actions. However, ensuring the proper communication of alerts and the organization of emergency services remains a local prerogative.
5. And on the French side, should we be worried?
The current floods in eastern France are of a much smaller amplitude than in Germany even if, near Belgium, we have sometimes reached historic levels. These floods were generally well predicted by the responsible flood forecasting services (SPC).
On the site vigicrues.gouv.fr, the floods were anticipated several days in advance, even if the extreme levels of the flood peaks could only be assessed 24 hours in advance in the small basins; the accumulations measured turned out to be the highest among those indicated by the ensemble forecasts.
6. Can we prevent exceptional floods?
There is no way to hold back the torrential rains, and once these rains have fallen, storing water to limit the floods becomes a technical and economic problem.
It is clearly not possible to imagine building dams or dikes to limit the floods on all the small rivers, and we must therefore resolve to have only warning systems, and to put in place preventive measures - only build outside flood-prone areas, work to reduce the vulnerability of buildings already present in flood-prone areas, better inform the population about what to do in the event of a flood, etc.
On larger rivers, upstream from larger agglomerations, it is possible to limit overflows by means of reservoir dams and dikes. These solutions have a cost, they consume space, and it is therefore necessary to reason their construction by comparing costs and benefits.
Land use planning, especially when it comes to prohibiting construction in flood-prone areas, is the “common sense” solution… which, however, comes up against special interests.
As for the so-called “soft hydraulic” solutions, advocated by nature defenders (planting hedges, grass strips, etc.), they have no effect on large-scale floods caused by exceptional rains, such as those that Germany has known in recent days.
7. How to deal with this type of event?
As we have seen, being able to predict (even several days in advance) exceptional floods is not enough: given the uncertainty that will always accompany hydrological and meteorological forecasts, the main challenge is to put in place a genuine risk culture, in order to guarantee a rapid reaction to phenomena that the population has never been confronted with.
Keeping a population mobilized and ready to react is possible if the risk is frequent (this is the case for earthquakes in Japan for example). It seems more difficult to organize for really exceptional flood events.
It therefore seems essential to pursue efforts in several directions: improvement of forecasting systems, improvement of the use of these forecasts, of crisis communication. And apart from crises, it remains essential to continue efforts on buildings in flood-prone areas.
Maria-Helena Ramos (hydrologist, Inrae) and Charles Perrin (agricultural and environmental engineer, Inrae) are co-authors of this article.
Featured image: Municipality of Erftstadt-Blessem devastated by the flooding of the river Erft, near Cologne in Germany. ©Youtube Capture France 24