Multivariate statistical assessment of flood scenarios in large river basins under consideration of tributary impacts and flood types

Abstract

<p>Despite the assumption of a homogeneous sample in many flood frequency analyses, flood events can be generated by highly different meteorological and catchment conditions. In large river basins, the diversity of flood generation is increased even more due to spatial and temporal variability of meteorological loads and hydrological processes within the basin and the resulting event-specific flood superposition of main river and tributary floods. Any consideration of additional attributes increases the complexity of statistical characterisation of flood events and scenarios. However, consideration of such differences is imperative if the effectiveness of flood control measures is to be assessed in a spatial context. This is mostly done by flood scenarios, which are usually derived using individual historical floods along with deterministic model-based simulations. We, instead, performed hydrograph-based flood-type classification and volume-based runoff analyses to estimate the contributions of sub-basins to floods in large basins. Using this information, we generated long synthetic samples of peak-volume-pairs to apply a multivariate statistical flood-frequency model that yields a type-specific conditional probability of a flood peak given the peaks in tributary stations. The results show that only some combinations of flood types may result in extreme peaks downstream of confluences. They also highlight the need to distinguish runoff-generation mechanisms for the larger floods from ones that drive smaller, more frequent events. Finally, design floods for different scenarios of flood-type combinations and assigned probabilities are derived, an approach that can be used to assess possible climate impacts to flood frequency. Case studies in several large river basins in Central Europe demonstrate the importance of distinguishing between flood types and flood scenarios, especially for extreme flood events.</p>