Exkurs: Unsicherheiten bei der Analyse und Attribution von Hochwasserereignissen

Statistics is a subject of many uses and surprisingly few effective practitioners. The traditional road to statistical knowledge is blocked, for most, by a formidable wall of mathematics. The approach in An Introduction to the Bootstrap avoids that wall. It arms scientists and engineers, as well as statisticians, with the computational techniques they need to analyze and understand complicated data sets.

What can we say about changes in the hydrologic cycle on 50-year timescales when we cannot predict rainfall next week? Eventually, perhaps, a great deal: the overall climate response to increasing atmospheric concentrations of greenhouse gases may prove much simpler and more predictable than the chaos of short-term weather. Quantifying the diversity of possible responses is essential for any objective, probability-based climate forecast, and this task will require a new generation of climate modelling experiments, systematically exploring the range of model behaviour that is consistent with observations. It will be substantially harder to quantify the range of possible changes in the hydrologic cycle than in global-mean temperature, both because the observations are less complete and because the physical constraints are weaker.

Extreme river floods have been a substantial natural hazard in Europe over the past centuries, and radiative effects of recent anthropogenic changes in atmospheric composition are expected to cause climate changes, especially enhancement of the hydrological cycle, leading to an increased flood risk. For the past few decades, however, observations from Europe do not show a clear increase in flood occurrence rate. Here we present longer-term records of winter and summer floods in two of the largest rivers in central Europe, the Elbe and Oder rivers. For the past 80 to 150 yr, we find a decrease in winter flood occurrence in both rivers, while summer floods show no trend, consistent with trends in extreme precipitation occurrence. The reduction in winter flood occurrence can partly be attributed to fewer events of strong freezing-following such events, breaking river ice at the end of the winter may function as a water barrier and enhance floods severely. Additionally, we detect significant long-term changes in flood occurrence rates in the sixteenth to nineteenth centuries, and conclude that reductions in river length, construction of reservoirs and deforestation have had minor effects on flood frequency.