Search

In many places worldwide, dams plays an important role in maintaining stability for societies by providing essential services such as drinking water, crop irrigation, energy production, and mitigating the impact of floods. Data-based and simulation studies have tried to attribute the impact of dams on downstream flood peaks, but the impact of dams on flood damage and risk remains weakly assessed. This study focusses on the most intensive dam construction period (1950-2002) in the Rhine river basins, one of the most regulated river basins in Germany. To understand the impacts of dams on flood damage, we have developed the Simple Basin Model (SBM) composed of a catchment model (HBV model) and a routing component based on the Streamflow and Synthesis & Reservoir Regulation (SSARR). In order to simulate dam operations, we have developed a dam module that activates water storage based on threshold levels of the hydrograph and releases water back to the river once the flood wave has passed. Flood damages are estimated by developing relationships between flood peaks and the associated damage. The Parameter Set Shuffling (PASS) is used to calibrate the catchment model using observed discharge data and catchment descriptors. The model chain is then used for estimating flood damages for the scenarios with and without dams. This research provides valuable insights into the impact of dams on flood damage and risk., The 28th IUGG General Assembly (IUGG2023) (Berlin 2023)

As our climate system climbs through its current warming path, temperature and precipitation are greatly affected also in their extremes. There is a general concern that climate change may affect also the magnitude and frequency of river floods and that existing and planned hydraulic structures and flood defenses may become inadequate to provide the required protection level in the future. At the same time, land-use changes and river training also have affected, are affecting, and will affect river floods in the future.Deterministic rainfall-runoff models are used in nearly all hydrologic planning, design, and management activities, and in theory can account for changes in forcing and processes due to climate and environmental change. Yet they are not easily transferred to ungauged locations, and they cannot properly reproduce extreme design quantiles obtained from commonly accepted statistical methods. We aim at coupling a newly developed methodology for regional model parameter calibration based on machine learning techniques, which allow for tuning models in a distributed way over large areas, to a post-processing approach that adds a stochastic error to the deterministic model predictions and reduces the bias in the estimation of extreme discharges associated to low exceedance probabilities.The research questions are:How well can the stochastic post-processing of the outputs of rainfall-runoff models, designed to reproduce everyday river flow dynamics, reproduce the statistics of extreme events obtainable through flood frequency analysis?How well regionally distributed rainfall-runoff models with post-processed outputs compare to regional flood frequency analysis outcomes?Preliminary results will be discussed during the presentation., The 28th IUGG General Assembly (IUGG2023) (Berlin 2023)

Subwavelength focusing of far-infrared light (λ=392 μm) is introduced to resolve spatially the photoconductivity of a two-dimensional electron gas. An increased cyclotron resonance amplitude is observed at one of the sample edges which proves that it is at the edge where states with nonequilibrium population can exist

Publication in the conference proceedings of EUSIPCO, Bucharest, Romania, 2012