Investigation of Riverbed Roughness in Channels with Flexible Vegetation Using Image Processing Techniques
Year: 2025
Type: article
Abstract: Flexible vegetation is common in river channels, in which it enhances water quality but also complicates flow velocity distributions, impacting channel roughness. This study explored the changes in channel roughness caused by the presence of flexible vegetation. By combining theoretical analysis with flume experiments, we enhanced the resistance calculation formula to encompass both the deflection angle and the degree of submergence of plants. Furthermore, threshold segmentation and a Gaussian mixture model were employed to mitigate 60%–75% of the width affected by trailing artifacts. This method allows automatic and accurate reading of the deflection angle, and, ultimately, a Manning’s coefficient calculation formula was derived that incorporates the characteristic parameters of flexible vegetation. The formula’s accuracy was investigated using experiments, and changes in roughness under different vegetation arrangements were analyzed systematically. The results show that the error in the calculated values of the formula developed in this study was within 10% of the measured values, which indicates the high practicality and applicability of the formula in rivers with flexible vegetation. Additionally, with the same planting density and flow conditions, the clumped arrangement exhibited the most significant increase in roughness, nearly 1.4 times that of the uniformly arranged roughness, followed by the random and quincuncial arrangements, whereas the uniform arrangement had the least variation. This study successfully enabled the accurate calculation of roughness in channels containing flexible vegetation, providing a foundational data basis for numerical simulations of such environments.
Source: Journal of Hydraulic Engineering
Authors: Hong Chen, Bangguo Song, Huiming Zhang, Jing Yan, Yue Zhao, Xu Zhao
Institution: Hohai University
Language: English
Subfield: Ecology
Field: Environmental Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4415401435