If a flood wave of a river coincides with flood waves of tributaries, extreme floods can occur, like the floods of 1993,1995,2002 and 2003 in the Meuse basin. These observed floods are analysed to see if patterns can be derived with regard to meteorological conditions, initial conditions and peak coincidence. The dataset used in this study is collected on initiative of the International Meuse Commission (IMC) and consists of data of the entire Meuse basin. The gained knowledge is besides input for the discussion on flood management also important for the improvement of flood forecasting systems. The Meuse is a rainfed lowland river and geographically its catchment can be divided in 3 zones. In zone 1 and 3 the river flows through a wide, mainly permeable river valley; in zone 2 the river valley is narrow and steep and soils are mainly impermeable. The most important tributaries are: Chiers, Semois, Viroin, Lesse, Sambre, Ourthe, Arnbleve, Vesdre, Jeker, Geul, Roer, Niers, Aa and Dommel. The analyses are performed by means of meteorological and hourly discharge data of zone 1, 2 and 3. The observed floods had a few characteristics in common. The precipitation was uniformly distributed over the basin. Precipitation amounts were very large during a relatively short period and low precipitation amounts were accompanied by snow. If temperatures dropped below freezing point, it was not enough to decrease the infiltration capacity. For all seven events the basin was saturated. The run-off coefficient varied for each event. As a result of these conditions, high to extreme discharges occurred in the Meuse basin. For the different discharge measuring stations, the time of occurrence of peak discharge was determined; the peak time. As some stations were not located at confluences of the Meuse and tributaries, the peak time was corrected with the propagation time of a flood wave from station to confluence. In case data was lacking, this time was estimated. The discharge data show that a flood wave in the Meuse basin is generated at different locations and at different times; this is due to differences in precipitation patterns and geographical features of the area. The Meuse discharge at Borgharen often peaks before the discharge at Ampsin and Chooz, which are both located upstream of Borgharen. The most important tributaries for the timing of the flood peak at Borgharen are the Ourthe, Sambre, Lesse and Arnbleve. For the peak at Chooz, the Semois is important and for the peak at Ampsin the Sambre and Lesse are important. Finally, it is examined if a relation between run-off and peak timing existed, but due to insufficient data a clear relation could not be given. It is recommended to examine this relation in a next research project. Due to occurrence of different flood waves during a flood event, measures taken to reduce peaks at one place along the Meuse can have negative effects on the peak discharge at other places along the Meuse. Hence, flood management should be done at basin level. Hydrological/hydraulic model computations are recommended to select the most effective flood measures. For calibration of the model other flood events should be included. The extra flood events can present information with regard to propagation times and occurring discharges. By quick analysis of the precipitation event, the precipitation amount can be compared with one of the seven events of this study, which allows for a global indication of the expected discharge event. The data used in this study only allow for an outline of the functioning of the Meuse catchment during flood events, which can be used as a directive for the next research step.
All data in the Integrated Marine Information System (IMIS) is subject to the VLIZ privacy policy
