PhD-Project Q5 by Abhirup Banerjee (GFZ/PIK): Recurrence analysis of event-like hydrological data with uncertainties
Timescale: Oct. 2018 – March 2022
Prof. Dr. Jürgen Kurths, PIK Potsdam
Dr. Norbert Marwan, PIK Potsdam
Prof. Dr. Bruno Merz, GFZ Potsdam
Temporal changes of flood hazard are difficult to detect and to attribute to the underlying drivers, because multiple factors influence and mask each other. Recurrence analysis will be used to compare the recurrence properties of different potential drivers and to identify potential couplings. Further methodological developments of this technique will extend its capabilities to study event-like data (extreme events), data with uncertainties, and spatio-temporal recurrences.
Objectives and Methods
The project will investigate flood behaviour with respect to local effects, e.g. implementation of flood retention basins, and external controls by using recurrence analysis. It will focus on the flood event scale and study the dynamic interactions of different variables, such as precipitation, temperature and catchment wetness. From the methodological point of view, this project will extend the available recurrence analysis framework to event-like data. Potential candidates of such an extension are the TACTS approach and the multiscale event synchronization approach. Explicit consideration of uncertainties will contribute to the current development of the recurrence analysis by using a probabilistic approach. The available candidates for recurrence based coupling analysis will be adapted correspondingly and applied to time series of river discharge, snow cover, temperature, soil moisture, precipitation, etc. of catchments in Central Europe. Their dynamic relationships will be analysed by studying and comparing their recurrence behaviour on the event scale.
KEMTER, M., FISCHER, M., LUNA, L.V., Schönfeldt, E., VOGEL, J., BANERJEE, A., Korup, O., Thonicke, K. (2021): Cascading hazards in the aftermath of australia's 2019/2020 black summer wildfires, Earth's Future, 9, https://doi.org/10.1029/2020EF001884.