Pre-event conditions and rainfall–runoff characteristics drive suspended sediment source variability

Most of the total sediment load transported in river systems is carried in suspension. Therefore, if we are to reduce soil erosion and sediment export, it is essential to determine suspended sediment sources and the drivers of its mobilisation into the river network. In this study, we combined the monitoring of suspended sediment fluxes and the sediment fingerprinting approach to test if pre-event conditions and rainfall-runoff characteristics drive suspended sediment source variability in catchments under a semi-oceanic climate. The sedimentological response to storm runoff events was studied in seven nested sub-catchments of the Attert River basin (0.4 - 245 km²), which have contrasting geological bedrock (sandstone, marls and shale) and land uses (forest, cropland and grassland). We collected stream water samples during storm runoff events (~30 events per catchment) using automatic water samplers to measure suspended sediment fluxes. In parallel, time-integrated suspended sediment and sediment sources samples (i.e., from different land use types) were collected and analysed in the laboratory (geochemistry, colour and organic properties) to determine the sediment origin using the sediment fingerprinting approach. Each sampled event was parameterized to describe rainfall, runoff, sediment transport and the relative contribution of each land use type to the sampled suspended sediment. Next, we assessed the relationships between variables. We found higher significant correlations between suspended sediment loads and runoff parameters (i.e., peak discharge and event runoff) than between suspended sediment loads and rainfall parameters (i.e., event precipitation, antecedent rainfall, and maximum rainfall intensity). Peak discharge for single events was found to be the best predictor of sediment loads in the studied catchments. We show that most events exhibit clockwise hysteretic loops between discharge and suspended sediment concentration in all studied catchments. We attribute this finding to the erosion or remobilization of sediment previously deposited on the channel bed or an adjacent area. During most of these events with clockwise hysteretic loops, sediment source apportionment presented a consistent pattern.