Flocculation and ingress of cohesive solids in a mountainous gravel-bed river

Abstract

Gravel-bed rivers draining mountainous headwater regions are critically important for the provision of high-quality drinking water and integrity in forested landscapes. These regions, however, have been increasingly impacted by natural (especially climate change-exacerbated) and anthropogenic landscape disturbances that can increase landscape connectivity and sediment delivery from hillslopes to receiving streams. Cohesive particles (<63 μm) increasingly mobilized during landscape disturbances are key vectors for nutrient and contaminant transport across the catchment continuum. Excess delivery of fine sediment mobilized from landscape disturbances can cause deleterious physical, chemical and biological impacts on streams by increasing turbidity, compromising habitats, ecological integrity, and challenging water treatability. However, processes controlling the propagation and storage of cohesive solids in gravel-bed rivers are still poorly understood. Given the complexity of flocculation and ingress processes under turbulent flow fields and the lack of reported rigorous field observations, most hydrodynamic fine sediment transport models disregard these processes, potentially leading to inaccurate model estimations. Accordingly, the main goal of this thesis was to evaluate the processes of flocculation and ingress, and the effects of these processes on the propagation and storage of cohesive solids in a gravel-bed river through the integration of laboratory, field, and modelling approaches. Five chapters were developed to achieve this goal: Chapter 2 investigated the in-situ characteristics of suspended sediment by measuring effective particle size distributions (LISST 200x) at four different sites of the study river at a range of flow conditions; Chapter 3 evaluated methods to measure fine sediment ingress rates and directional mechanisms by deploying >250 sediment traps, consisting of triplicates of three designs of sediment traps; Chapter 4, using those sediment traps, evaluated the hydraulic (discharge, water depth, Froude number, bed shear stress, and stream power) and sedimentological (suspended sediment concentration by mass and by volume) drivers of fine sediment ingress; Chapter 5, using a rotating annular flume, characterized the processes of suspended sediment transport over different bed configurations through deposition, ingress, and erosion experiments, and; Chapter 6 calibrated and validated a semi-empirical cohesive sediment transport model (RIVFLOC) using results from the rotating annular flume, and applied the model to the observed field conditions to better understand the transport and fate of these particles in the study river. Particles <500 μm in the Crowsnest River were predominantly transported in flocculated form. While the highest observed flow energy limited the development of larger flocs due to the breakage of bigger and more loosely attached flocs, microflocs were invariably observed, and their formation correlated positively with flow energy. The characterization of effective particle sizes between ingressed and suspended sediments demonstrated that, although the particle size distributions had similar modalities for both sediment types, ingressed sediment was coarser than suspended particles – demonstrating that interstitial flocculation can occur, even though it is also likely that coarser flocs are preferentially ingressed in the channel. Field measurements of ingress demonstrated that, for flocculated particles, ingress was predominantly vertical during higher energy flows, while horizontal mechanisms of accumulation were more important during lower energy flow conditions. The model application estimated that ~60 % of the upstream suspended sediment gets trapped in the channel framework over the ~10 km study reach. Although we lack independent means to validate these findings in the field, given the nearly continuous sediment supply from diffuse sources along the catchment, the fine sediment ingress rate recorded in an annular flume was well within the range of ingress rates measured in the field, which supports the validity of the modelled estimates. Despite high accumulation rates, field observations demonstrated that no sites were saturated or clogged with fine sediments during the field campaigns. The observations reported in this study bring some novel perspectives to the understanding of flocculation and ingress processes of cohesive solids in a gravel-bed river (i.e., in-situ flocculation mechanisms of suspended and ingressed particles, floc size effects on the ingress process, and the modelling framework of the propagation and storage of cohesive particles). The reported results also indicate the potential for future investigations on the role of the interstitial load for fine sediment storage and possible exfiltration mechanisms driven by groundwater upwelling. Given the importance of cohesive sediment as a vector for contaminant and nutrient transport and the limited capacity for interstitial flushing of fines, gravel-bed rivers may delay the downstream impacts of upstream disturbances (legacy effects) for decades. Accordingly, due to the increasingly reported deleterious impacts of excess cohesive sediment in gravel-bed streams, advancing knowledge on the propagation and storage of fine sediment, as proposed in this study, is imperative for its strategic management and prevention.