Turbulent flows and riparian vegetation induced drag for river flooding

Floods account for almost a third of all natural disasters. They are responsible for more than half of the fatalities and approximately a third of all economic losses are a direct result of natural catastrophes (Berz, 2000). At present, about 5 million people in 2 million properties live in flood risk areas in England and Wales. In order to mitigate flood risk, it is essential that a sufficient understanding of flow characteristics in rivers should be achieved, to enable accurate models and therefore accurate estimation of water level to provide accurate flood warnings and inundation maps of towns and cities. A river, however, cannot be considered in isolation from its wider environment so that the modern management of rivers and their floodplains is necessarily directly concerned with the role of vegetation. Trees and shrubs can be commonly found growing at the edge of rivers in relatively straight lines. In rural areas, this pattern is the result of common agricultural practice, where the riverbank marks the field boundary and the trees and shrubs form part of the hedgerow system. Despite their ecological functions in riparian environment, trees and shrubs have a large impact on flow resistance. The presence of trees and shrubs generates turbulence owing to wakes which reduces the conveyance capacity of a river and its floodplain, with consequent raising of local flood levels. As yet no one has paid attention to how riparian vegetation such as trees and shrubs along the edges of floodplains influence the flow behaviour, resistance and water level though many studies on aquatic plants have been undertaken. Trees and shrubs offer more flow resistance than aquatic plants, leading to higher water level. This project is therefore to study the effects of riparian vegetation through laboratory experiments in order to mitigate flood risk and to provide an appropriate guidance on the management of riparian vegetation. The experiments will be conducted at Loughborough University and will involve measuring riparian vegetation drag force, turbulence, bed friction, and water surface variation with a measuring drag device, Acoustic Doppler Velocimetry (ADV), a pitot tube, a Preston tube, a 3-D digital scanner, and Photogrammetry. These measured flow parameters will be used to investigate effects on the key parameters such as drag coefficient, flow resistance and water level for the management of riparian vegetation and flood risk assessment.

As computing resources become more efficient, two dimensional analyses are becoming more commonplace in the world of commercial modelling. Although the effects of vegetation are still commonly accounted for by changes in roughness parameters (e.g. Manning coefficient) within these models, the additional dimension of vegetal drag, from this project provides the opportunity of using more sophisticated approaches in compound channels with one-line or partial vegetation along the floodplain edges. This research proposal particularly focuses on the impact of tree and shrubs as the key topics of boundary shear stress and turbulent flow structure. The area of influence of individual vegetal elements and vegetal drag coefficients will be generated for a better representation of roughness parameters during river floods. Hence, this will provide immediate benefits to software developers and practitioners who are currently involved in 2/3D CFD software such as TELMAC, FLUENT, CFX, PHEONIC, etc for flood risk management, bank erosion, sediment/pollutant transport assessment, restoration and riverine biodiversity projects. The turbulence data collected with this project can be also used for validation and development of advanced 3-D models such as LES, DES etc in commercial software.

From this project, a better understanding of roughness from vegetal drag will also provide an upgrade to the Roughness Adviser in the Conveyance and Afflux Estimation System (CAES) for one-line vegetation. In addition to this, the fundamental analysis of vegetated flow will also lead to a contribution of secondary currents to flow resistance, the development of shear layer width, the effect of vegetation on boundary shear stress and the parameters used for SKM modelling. The main engine SKM within the CAES currently used can be further rigorously validated and possibly modified by a better presentation of physics accommodating the shear layer width and a drag force concept than the recent one on bed generated turbulence. This will provide immediate benefits to practitioners who are currently using CAES for river management and flood risk assessment.

The project will also allow more accurate flood inundation mapping as a result of a better presentation of roughness parameters currently used by practitioners and modellers. This will benefit the general public, for example, via house insurance for those who live in high risk flood plain and at the border of flood inundation areas. Other improvements are to the flood warning and flood evacuation route, as well a contribution to guidance on the incorporation of flood water, i.e. capacity and space for water and car park design.

To ensure that benefits outlined above are realised, a number of dissemination activities are proposed:
We will invite five CFD and CAES users and developers, and compound channel flow researchers to attend yearly meetings of the project. We will show our research output and discuss and exchange ideas concerning practical issues such as how to add new findings of vegetal drag roughness from this project to RA in CAES, how to use the modified SKM and how to create user friendly formation of turbulence data on a website for end users.

One workshop at LU will also be organized for practical end users and blue sky academics at the end of the project. This will introduce project outputs including vegetal drag coefficients and turbulence data. The workshop will primarily provide the following:
1) Roughness parameters for vegetation along the edge of floodplains in particular, to users of CG and RA in CAES.
2) Turbulent data on a website at LU to 2/3D CFD users and developers.
3) Instantaneous time and space turbulent flow structures obtained PIV to blue sky researchers.