Experimental flume and numerical studies into the influence of floodplain vegetation on river-corridor hydrodynamic processes
Date
2023
Journal Title
Journal ISSN
Volume Title
Abstract
The active channel has historically been the primary focus of river hydrodynamic process studies and river engineering. However, increased global flood risk and awareness of ecosystem services provided by floodplains has encouraged managers to broaden their perspective beyond the banks. As water exits and reenters the channel during floods, water, nutrients, and sediment are exchanged with the floodplain. This flux is heavily influenced by both channel-floodplain hydrologic connectivity, or the ability of water to access the floodplain, and by floodplain land cover types. River and hydrologic modifications that result in disconnected floodplains include channel planform and cross-section geometry alterations, diversions and dams, levees, land cover change, and river sediment mining. As river managers, land-use managers, and landowners acknowledge the benefits of functional, laterally connected river corridors, more river restoration projects are undertaken with a primary goal of reconnecting a river channel to the adjacent floodplain. A major component of large river restoration and river engineering projects includes designing for and predicting future flow scenarios using hydraulic models and other analytical methods. Developing a hydraulic model for river restoration design relies on the theory and science of fluvial morphodynamic processes as well as the parameterization of hydraulic roughness coefficients. Because of the historical emphasis on in-channel processes, the scientific literature related to channel-floodplain hydrodynamics and floodplain roughness parameterization is sparse. Specifically, there are limited studies investigating the influence of vegetation on channel-floodplain exchange flow, lateral connectivity, and resulting channel topography. To address this knowledge gap, I conducted a series of physical and numerical modeling experiments where floodplain vegetation and flow parameters were varied. In Chapters 2 and 3, I present the results of flume experiments where I measured bedform topography and the flow field under varied floodplain vegetation conditions at two overbank flow depths. The experiments were conducted in a 1-m wide meandering compound channel inset in a 15.4-m long, 4.9-m wide basin. The channel bed was a mobile sand-and-gravel mixture with a median sediment size of 3.3 mm, and sediment transport occurred only within the channel. I tested bare and vegetated floodplain conditions with 2.7-cm diameter rigid emergent vegetation elements at spacings of 3.0 units m-2 and 12.1 units m-2. My observations of the flow field indicate that high density vegetation enhances secondary circular flow through the meander bend and reduces momentum exchange at the channel-floodplain interface. At a low relative depth, flow through high density vegetation was deflected away from the down-valley direction and forced to reenter the channel at a steep angle with respect to the channel centerline. However, at a high relative depth, dense vegetation steered in-channel surface flows more closely following the channel centerline. These observations shed light on the hydrodynamic processes leading to flood wave attenuation, enhanced nutrient cycling, and channel altering stresses, and these results may inform river restoration riparian management best practices. To investigate bedform response, I performed a moving-window analysis of topographic surface metrics including skewness, coefficient of variation, and standard deviation, as well as topographic patch analysis of area and contagion to measure changes in bedform heterogeneity as flow depth and vegetation density were varied. My results show that both greater density vegetation and larger flows can increase bedform topographic heterogeneity. These findings suggest that floodplain vegetation and natural hydrologic regimes that include overbank flows can enhance stream habitat complexity. Designing for the effects of established vegetation conditions and prioritizing floodplain vegetation planting may be useful for river managers striving to achieve successful biomic river restoration. Expanding on the observations made in the flume, I explored the ability of a 2D hydraulic model to predict the effects of vegetation on meandering channel flow dynamics. I used the TreeLS point cloud processing tool to automatically extract woody floodplain vegetation characteristics and estimate Manning's roughness coefficients for vegetation from aerial LiDAR. I investigated the influence of varied vegetation densities on channel-floodplain exchange flows in HEC-RAS 2D. I developed hydraulic models for three reaches along the Butokamabetsu River in the Hokkaido University Uryu Experimental Forest in Northern Japan where each reach had distinct biogeomorphic characteristics including channel width, slope, sinuosity, and floodplain vegetation density. I found that in the lower gradient, higher sinuosity reaches, floodplain vegetation density had more influence on channel-floodplain exchange flow attenuation. These results highlight the importance of planning for the presence and density of vegetation in river restoration projects particularly in lower gradient, more sinuous stretches of river. The results and analysis presented in this dissertation suggest that biological drivers such as rigid emergent floodplain vegetation play an important role in river form and function particularly in conjunction with floods that occasionally access the floodplain. These detailed observations of flow, sediment, and resulting bed morphology as well as analysis of innovative remote sensing techniques provide a basis for an improved understanding of morphodynamic processes in meandering rivers.