River flow prediction through rainfall–runoff modelling with a probability-distributed model (PDM) in Flanders, Belgium

The hydrological probability-distributed model (PDM) is widely used all over the world and its applicability has also been investigated in Flanders, Belgium. Rainfall–discharge relations for all gauging stations installed on non-navigable watercourses were modelled over a long succession of monitoring years. In all, 1456 years were modelled. Typical characteristics (peak flow, volumes) of modelled series are compared with observations. Based on the relatively long time series, reliable discharge values can be generated with the PDM. Water volumes and peak characteristics are very close to the observed values. The set of 98 PDMs was analysed and clustered. Three cluster approaches were considered: a single-parameter approach, a parameter set approach and an approach with known cluster zones, delineated on hydrological flow characteristics. The single-parameter approach, the parameter set approach and the combination of both gave less detailed regional information than the clustering on hydrological characteristics.