Comparison of hydrosedimentary signatures of catchments from the French Critical Zone network
Résumé
Soil erosion and sediment transport understanding has important implications for the management of soil and water resources in the critical zone. Water erosion is considered the main threat to soils, causing irreversible losses. The transfer of suspended sediments (SS) downstream poses problems of degradation of the quality of rivers by clogging the bottom and is also a privileged vector for the transport of nutrients, contaminants or pathogens. Particulate transfers arealso very sensitive to changes in land use and in the occurrence and severity of rainfall.Most studies that have sought to synthesise measurements made in different contexts focused on the spatial variability of erosion rates on hillslopes (Cerdan et al., 2010) or on the comparison of integrative variables such as sediment fluxes in different hydrosystems (Milliman and Syvitski, 1992; De Vente et al. 2011; Vanmaercke et al., 2011). However, very few studies have been interested in comparing the relationships between the variables that allow to establish the fluxes (i.e. iquid discharge, turbidity, SS concentrations) whereas their analysis can shed light on hydrosedimentary processes, particularly by exploring their temporal variability (Vercruysse et al., 2017). Within the OZCAR and RZA French networks, many observatories carry out measurements of discharges, turbidities and/or point measurements of SS concentrations at catchment outlets. These datasets are used to address the following questions:
i) Are there similar temporal hydrosedimentary signatures at some sites, ii) Is it possible to identify control factors responsible for these signatures? Results show a large diversity of the temporal hydrosedimentary behaviours of the 22 studied catchments as illustrated by some striking features. Indeed, 90% of the SS are transported during less than 0.1% of the time for some sites and during 40% for others. Half of the SS are transported at SS concentrations exceeding 34 g/l for some sites and less than 0.1 g/l for others.
First attempts have shown that some of these indicators of temporal variability of SS fluxes can be predicted by site (e.g. area, slope, pedo-geo-morphology, soil use, meteorological forcing) or hydrological characteristics.