|Location: Hörsaal A (1.3.14)
Time: Tuesday 15.06.2010, 14h c.t.
This talk is on long-term memory in climate fluctuations and is divided into 3 parts. In the first part, long-term memory will be introduced and its relevance in climate is shown. It is explained how long-term memory can be detected also in the presence of slowly varying external trends. In the second part, the occurrence of extreme events is discussed. It is shown how the conventional laws (derived for uncorrelated systems) are changed in the presence of long-term memory, and how this leads, in a natural way, to a clustering of extreme events. The third part of the talk is devoted to the detection problem, i. e. the detection of trends in records with long-term memory. This problem is relevant, e. g., for detecting the amount of an anthropogenic trend in a temperature record which, as null hypothesis, can be viewed as long-term correlated. Analytical equations for the relevant quantities are given, that enable trend estimation in any long-term correlated record., including local, regional and global temperature records as well as river run-off data.