Sie verwenden einen veralteten Browser mit Sicherheitsschwachstellen und können die Funktionen dieser Webseite nicht nutzen.
Timescale: Oct.2015 – Sept.2018
Prof. Dr. Jürgen Kurths, Potsdam Institute for Climate Impact Research (PIK)
Prof. Dr. Bruno Merz, Helmholtz Centre Potsdam GFZ German Research Centre For Geosciences
Traditionally, floods are seen as essentially random events, and extreme flood occurrence and magnitude are assumed to correspond to an independent, identically distributed random process. However, unusually large floods in drainage basins of all sizes seem to be related to large-scale atmospheric circulation anomalies. Recent work has shown that 20 major flood events over the last 100 years had nearly identical storm tracks, moisture source and delivery patterns. Such persistent anomalies may in turn reflect modulation by climate phenomena such as the El-Niño Southern-Oscillation (ENSO) and North-Atlantic Oscillation (NAO). For regions where floods are tightly linked to meteorological conditions (e. g., strong influence of ENSO on flooding in Peru) traditional regression-based approaches may be sufficiently reliable. In most other regions new methods are needed to decipher the more subtle linkages between climate and floods (e. g., Western Europe winters being only weakly linked to NAO).
Objectives and Methods
The PhD-project aims at understanding space-time dynamics of extreme natural events by using complex networks. Complex networks based on event synchronization are a novel approach for studying space-time dynamics of strong natural events. By analysing the concurrent or lagged occurrence of extreme events at different locations, this technique has been successfully applied to new phenomena of the spatial organisation of extreme monsoon rainfall in Asia and South America. In this project, the technique will be generalised to include more physical parameters. A special challenge will be to combine synchronisation of event-based networks with networks obtained from other climate and flood-influencing variables, such as temperature or catchment wetness, leading to a network of networks. This includes the development of appropriate measures of causality to quantify possible interactions between the networks. The methods will be adapted for spatio-temporal climate and flood data, with the ultimate goal of better understanding how extreme river floods in Central Europe and in the Mekong River Basin are linked to climate.
Flood times series of up to ten river gauges in Central Europe and in the Mekong River Basin of different magnitude (e.g. peak-over-threshold with on average three events per year) and associated meteorological and catchment variables (sea surface temperature, atmospheric pressure, rainfall, snow cover, soil moisture, etc.) will be derived. From the data a network of networks is constructed where the interactions within each network and, more challenging, between the networks should provide new insights in the underlying mechanisms. Time-dependent analysis techniques for non-stationary data and statistical methods for testing the significance of transient structural properties in this network of networks will be further developed.
Ankit Argawal is based at the research domain “Transdisciplinary Concepts and Methods” of Potsdam Institute for Climate Impact Research (PIK).