Author(s): E. K. Jang
Linked Author(s):
Keywords: Riparian vegetation; Flow resistance; Natural channels; Large-scale experiments; 2-D numerical modeling
Abstract: Vegetation in rivers is pivotal in the hydraulic processes of natural channels, influencing flow, sediment transport, and the overall ecosystem. Accurate estimation of this vegetation flow effect is crucial for predicting flow velocity, shear stress, sediment transport, and ecological habitats in the rivers and streams. However, it varies based on species, growth period, and group size configuration. In particular, vegetation resistance, which affects flood level variation, is dynamic, depending on distribution, physical properties, and hydraulic conditions. Recent studies have explored flow resistance variation based on plant density, flexibility, and hydraulic conditions. Models for estimating the flow resistance of general vegetation sometimes simplify the shape of complex vegetation into a rigid cylinder or ignore leaves and flexibility. Alternatively, models based on data measured in laboratory-scale experiments are often used, but it is still being determined whether they can be applied to large-scale sites. One of the reasons is that the vegetation survival patterns in natural rivers, vegetation with finite width and length, are generally clustered in a patchy form. Also, a composition woody plant is a complex of stems, branches, twigs, and leaves. Because the physical properties of natural leafy vegetation are challenging to make on a laboratory-scale, the natural characteristics and scale of riparian vegetation and patches should be considered to improve understanding of nature. This study aims to secure reliable data on vegetation flow resistance through full-scale experiments to complement the simplification of existing vegetation and the limitations of the laboratory-scale. The accurate water level measurement data, trusted based on full-scale experiments, is implemented through a two-dimensional numerical model. Implementation as a two-dimensional numerical model will address the limitations of existing models, such as the limited definition of the characteristic domain in experimental studies and the use of a single drag coefficient. The results of this thesis are expected to improve our understanding of vegetation flow resistance and contribute to the development of more accurate and reliable methods for estimating vegetation flow resistance in natural channels. It will have important implications for the management and restoration of rivers and streams and for the design of hydraulic structures that interact with vegetation.
Year: 2024