|Keywords:||flood; drought; large scale; uncertainty; satellite observations; hydrological extremes|
|Full text PDF:||http://dspace.library.uu.nl:8080/handle/1874/310177|
Hydrological extremes regularly occur in all regions of the world and as such have large impacts on society. Floods and drought are the most severe hydrological extremes, in terms of their societal impact and potential economic damage. These events are amongst the most costly natural disasters, due to their often large spatial extent and high societal impact. The main objective of this thesis is: To reduce uncertainty in simulations, reanalysis, monitoring, forecasting and projections of hydrological extremes for large river basins. The first part of this thesis focusses on the uncertainty in hydrological simulation and short-term flood forecasting. I try to reduce the uncertainty in, (i) precipitation, (ii) historic hydrological simulations and state estimates and (iii) short term flood forecasting. Thereafter, I focus on the vagueness in drought terminology by studying the definitions that are used to identify and quantify drought. I present an intercomparison between frequently used drought indicators to study their differences and provide a solution to define drought under a changing climate. In the final part of the thesis, the hydrological drought projections are studied in more detail. Magnitude and directionality of changes in hydrological drought characteristics are largely unknown. I evaluated the impact of the climatology and catchment characteristics as well as the impact of human water use on future hydrological drought. The results from this thesis show that the inclusion of data assimilation, satellite data and model simulation is beneficial in the field of flood forecasting and hydrological modelling in general. This is shown by the results obtained in the first part of this thesis, where I successfully integrated different sources of observations to reduce uncertainty in hydrological model simulations. It was also proven in the second part of this work that there are multiple ways to identify drought events and that the impact of climate change on drought identification is significant. This impact of a changing climate contributes to the current vagueness in drought terminology and should be taken into account when studying the phenomena drought. Finally, I showed that the impact of climate change on future hydrological drought is highly dependent on the local climatology and the impact of human influence. This thesis showed that it is possible to reduce the uncertainty in the simulations of hydrological extremes by combining existing data, models and frameworks. These findings can bring major advances in the field of hydrological modelling to support monitoring and forecasting of hydrological extremes in large river basins.