TCS Daily

A Flood of Information

By Willie Soon - September 16, 2003 12:00 AM

During the July 1997 flooding of the Oder River, tens of thousands were evacuated and about 100 died in Poland and the Czech Republic (Figure 1). In August 2002, severe flooding of the Elbe River in central Europe also took an enormous human and financial toll (estimated to be US$18.5 billion). During the 1990s, heavy and costly floods also occurred on the Rhine (1993), Meuse (1995) and Morava (1997) rivers.



map of flood area

Figure 1: Location of major rivers in north central Europe. From



Proponents of the theory of catastrophic, human-induced global warming have been quick to point out that these summer flooding events could have resulted from increasing man-made greenhouse gases in the atmosphere -- a frequent claim made after most extreme weather events. Computer model predictions, such as the one shown in Figure 2 from the Danish model, show increasing rainfall over most of Europe, but at the same time show increased summer dryness. Christensen and Christensen, writing in the February 20, 2003 issue of the journal Nature, explained that despite the broad area of summer drying (the large area marked in red across Europe in panel a), incidences of heavy rainfalls (pockets of green area in panel b) also can be expected to increase significantly in various locations around Europe. They wrote, "we find that CO2-induced warming can lead to a shift towards heavier intensive summertime precipitation over large parts of Europe."



Figure 2: Computer predicted (a) large summer drying (reduction in rainfall in % relative to present-day average) but (b) possibility of increase in large severe rainfall events in Europe under a CO2-global warming scenario by 2100. Note that increasing trend in very severe summer rainfall is especially predicted for regions around Germany, Poland and the Czech Republic or roughly covering the Elbe and Oder flooding disaster events of the late 1990s and 2002 highlighted in this ChartiFact. [Chart from Christensen and Christensen, Nature, vol. 421, 805-806, February 20, 2003].


How do the flooding events of the last decade rank with previous events in history?


For a historical perspective, Professor Rudolf Brázdil of the Masaryk University in the Czech Republic (in PAGES news, vol. 10, no. 3, 21-23, December, 2002, remarked that the first reliable report of a flood in Prague came from the chronicler Kosmas:


In the year of our Lord 1118 in the month of September there was such a flood as, I think, it has not been on the Earth since the Deluge. This river of ours, the Vitava, suddenly broke out of its bed -- how many villages, how many houses in the suburbs, huts and churches did it take away! At other times, although it happens rarely, the water reaches only the floor of the bridge, but this flood rose to a height of ten ells [i.e., approximately 6 meter] over the bridge.


One can only gain a relative sense of the severity and human cost of this 1118 flooding event of the Elbe river compared to the 2002 disaster (or the 1997 shock in Oder) by folding in the human developmental factors over time like population growth, land-use changes (including deforestation and urbanization), reductions in river length and the construction of reservoirs and channelization. In addition, Otto Malek, a flood-protection expert at the German Federal Environment Ministry, warned that "the real problem is that there are too many people building houses, and amassing material assets, in areas known to be in danger of flooding."


But what does the flooding data record in Europe really indicate?


Figure 3 shows newly and carefully assembled results by a group of German hydrologists and statisticians published last week (September 11, 2003) in the journal Nature. Significantly and confidently, these researchers confirm that the modern disastrous flooding events on the Elbe and Oder rivers were no more extreme nor the most unusual when compared with other historical events of the past 1000 years. This fact alone is enough to dispel the drive to suggest that the 1990s and 2002 flooding events of central Europe can be uniquely tied to the increasing levels of man-made greenhouse gases.



Figure 3: Rate of occurrence of heavy floods (of magnitude 2 and 3) and intensity of floods for the Elbe and Oder Rivers of central Europe for the last 1000 years. The data show that extreme flooding events are either decreasing (winter and summer events for the Elbe and winter events for the Oder) or remaining constant (summer events for the Oder), in direct contradiction to unsubstantiated attributions that the 1990s and 2002 large flooding events in Europe were increasing activity related CO2 global warming. [Chart from Mudelsee et al., Nature, vol. 425, 166-169, September 11, 2003]


Instead of confirming the tendency for more cases of extreme flooding events -- predicted by the computer models under the warming by increasing level of atmospheric CO2 especially in the last part of 20th century -- Figure 3 suggests the opposite. It appears that the rate of occurrence of extreme flooding events has been either decreasing or remaining constant for all summer and winter data for the Elbe and Oder Rivers. Figure 3 also shows that a relatively higher rate of extreme floods, attributable to increased precipitation, occurred during the 16th century for the Elbe River.


Through a proper statistical analysis (with results indicated by the tendency pointed by the direction of the heavy arrows in Figure 3), Mudelsee et al. were able to find a consistent link between climate variation and the downward trend in winter floods and the absence of a trend in summer floods in the 20th century record. The reduction in the winter flooding events for both rivers was credited to fewer and fewer strong freezing events (that are potentially flood enhancing upon breaking of the river ice), which "may have been caused by warming or increasing pollution of river waters."


Complicating this issue is that flooding events are amplified by urbanization. Even if rainfall held constant, an increase in urbanized areas would result in less infiltration to the soil and more surface runoff, increasing flood peaks and frequencies. And this region of Europe has seen significant urbanization over the last hundred years. River control structures serve to further constrain the river, prohibiting it from spreading out through the floodplain and increasing both the peak flow and its speed. Thus, it is even more surprising to see that the flood events for these major rivers of central Europe have not been increasing over the last decade.


If the model results are correct in that a CO2-warming should yield more heavy summer rainfall for regions around Germany, Poland and the Czech Republic as shown in Figure 2, then an increasing trend in extreme flood events should begin to occur. But no such expected trends can be confirmed from real-world data.


TCS Daily Archives