Flow in seagrass canopies: The influence of patch width

Seagrass beds and the communities they form are well known for their ability to alter their local hydrodynamic environment, reducing current velocities and altering turbulent structure in and around the canopy. Much of the quantitative information that has been published on the interaction of seagrass canopies with flowing water has been derived from laboratory flume studies. The few studies that have been conducted all point to similar patterns of flow alteration around the seagrass canopy. Differences among the results of the study are likely primarily due to different experimental configurations. Some studies have used seagrass beds much narrower than the width of the flume while others have used seagrass beds extending the full width of the flume. The validity of the latter design has often been called into question because of scaling issues. In this study, artificial seagrass was used to examine the effects of bed width in a laboratory flume on the spatial pattern of water velocity and turbulence intensity within the bed. As seagrass bed width was increased, blocking more of the cross-sectional area of the flume, the seagrass became less effective at reducing within-canopy current velocities while over-canopy flow was increased. Narrow patches (0.3 m in a flume, 1.0 m wide) were significantly more effective at reducing current velocity within the canopy than were wider patches, but experienced higher turbulence intensity. Using laboratory findings from experiments to predict field flow conditions when patch geometry differs substantially from that of a flume may either over- or under-estimate flow reduction and turbulence intensity. This is particularly the case within the first meter of horizontal distance as flow enters the canopy. Therefore, flume conditions where the bed width equals the flume width may be more appropriate for mimicking flow interaction with broad and shallow seagrass beds. Use of bed widths narrower than the flume width are likely more accurate for modeling small, developmentally arrested patches, or recently established patches such as those arising from restoration projects.