Findings of the Intergovernmental Panel on Climatic Change (IPPC) suggest that climate change is largely due to human influences, particularly the release of greenhouse gases and aerosols. On a global scale, the surface temperature rose by 0.6 °C from 1860 to 1990. This was probably the largest temperature increase in the last 1000 years. The decade from 1990 to 1999 was the warmest of the 20th century. 

Higher temperatures are indicative of an intensification of the hydrological cycle, in particular due to increased evaporation. This makes it quite possible that more intensive rainfall and larger or more frequent floods will occur. However, the relationships between climate changes and floods are not completely understood, and data from the past present a very inconsistent picture. A statistical analysis of extreme runoffs (annuality > 100 years) in 29 large catchment areas (area > 200,000 km²) has indeed proven a significant increase in extreme flood runoffs during the 20th century. Nevertheless, the extent to which such changes can be attributed to climate change caused by humans and other factors (natural climate variability, changes to rivers and drainage areas, etc.) is still uncertain. One single event or a small number of events taking place in quick succession can in principle hardly be assigned to the cause “climate change”. Climate change can only influence the probability of the occurrence of such events.

Large regional differences

However, global changes have very different regional effects. For example, a decrease in the number of winter floods has been ascertained for the Elbe and Oder in the last 150 years. Such events are frequently connected to dangerous ice jams. No trend could be proved for summer floods. The perceptible effects of global climate change on a regional scale must therefore be viewed with caution. Regional statements concerning changes in the flood situation are still very uncertain. This applies in particular to extreme events.

For the Rhine area, seasonal changes are predicted. The yearly runoff cycle will become more pronounced, i.e. stronger runoffs are expected in winter, weaker runoffs in summer. This is on the one hand a consequence of the seasonal redistribution of precipitation. For the winter half-year increasing, for the summer half-year decreasing precipitation is expected. On the other hand, winter precipitation will in future progressively take the form of rain instead of snow, so that the risk of winter flooding will increase. The predicted tendency towards summers with lower precipitation in the Rhine area does not automatically mean a smaller flood hazard. There are indications that more intense convective rain storms such as thunderstorms are occurring, which can lead to flash floods.

Similar findings were obtained by an investigation of river levels, restricted to Germany, with regard to flood trends (frequency, seasonality and magnitude) for the years 1951–2002 (Petrow et al., 2009). This study identified significant upward trends in a number of catchment areas, in other words an increased flood hazard, and downward trends at only a few measurement levels and without an identifiable pattern. Levels with significant trends are spatially ”clustered”, in other words are more frequent in certain regions, and show seasonally differentiated changes in flood hazard patterns (e.g. an increase in winter floods). Both are indicative of climatic influences as the cause of the changes. The investigation shows that there are hardly any changes in flood hazard in northeastern Germany (Elbe, Oder). On the other hand, marked changes occur in the south, the west and the centre of the country. There are also more changes in the winter half-year than in the summer.