A historical flood record of the Meuse derived from flood deposits in the abandoned meander of Hedikhuizen

A multidisciplinary study covering sedimentology, geochemistry and palynology has been carried out on a sediment core, retrieved from an abandoned Meuse meander. Abandoned meanders have been proven to be excellent sediment archives and aid for reconstructing paleoflood records. Recent severe floods as that in July 2021 remind us of the reality of flood risks and the predicted increase due to global climate warming. Instrumental hydrological data spans only the past 200 years. Additional data from paleofloods are required in order to understand long-term flood dynamics and to make predictions for the future. A 6-meter-long core, from the paleomeander of Hedikhuizen reflects the end of the medieval period (1474 AD) up to the early 20th century and almost comprises the full Little Ice Age (1340-1850 AD). Grain-size data combined with XRF data produced a flood event record, which was detrended and normalized by calculating the z-score, in order to filter out the geomorphological influences. In addition to these methods, pollen analysis and thermogravimetric analysis have been
used to create a flood chronology. Using historical records, flood events recognized in the core have been correlated to an age, resulting in an age-depth model. Apart from the natural hydrosere, anthropogenic forces are found to have influenced the infill of the meander. Inundation works around 1774 increased the connectivity to the Meuse and influenced the sedimentation. Based on this study, in combination with that of a similar project that considered the Late roman and Early Medieval, it is likely that the Little Ice Age (LIA) has increased the flood frequency of the Meuse. However, the magnitude of flood events seemed lower in de LIA. Within the LIA, the magnitude of the flood events is at a maximum during the cold maximum of the LIA. Deforestation, being of high scale during the 16th and 17th centuries is not reflected in an increase in flooding in the record (Buis, 1993) It is known
that river engineering elevates the frequency and magnitude of flooding (Hall et al., 2014; Munoz et al., 2018). Dike building started around 1200 AD and river engineering kept increasing until now, which is likely to have influenced the flood risk too. Probably reflected by the highest frequency and average magnitude at the youngest phase of the record. Lastly, the flood of July 2021 broke the Meuse peak discharge winter record with 3260 m3
/s. This study shows that the floods of 1643 and 1740 had a similar peak discharge (according to their calculated recurrence time). However, the peak discharge of the flood in 1643 was estimated to be 3600 m3 /s (van der Broeck, 2012). It is likely that the events of 1643 and 1740 have a higher actual recurrence time and discharge due to an error in the calculation of the recurrence time.