Fig 1.
Settlement of coral larvae onto flat and ridged substrates in oscillatory flow.
(a) Schematic of the oscillatory flume tank and particle tracking system. Major components include (1) laser sheet, (2) acrylic viewing section, (3) PVC T-socket, (4) custom 3D-printed PVC-to-acrylic connectors, (5) PVC elbow, (6) motor and piston assembly, and (7) high-speed camera. (b) Spatially-averaged fluid velocity in the central viewing section of the flume obtained from particle tracking velocimetry measurements (PTV; circles) over a full oscillation period. The grey highlighted sections correspond to the phases of peak flow (straight arrow) from left to right and the turning point (curved arrow) from rightward to leftward flow. (c) Photographs of the flat and ridged CaCO3-based settlement substrates and 3D laser confocal maps showing the micro-scale topography of both substrate types (top) and the millimeter-scale ridges of the ridged substrates (bottom). (d) Larval settlement data on flat and ridged substrates in static and oscillatory flow conditions (**: p < 0.001; post-hoc Tukey HSD). (e) Top view of a ridged substrate illustrating larval settlement locations in static (top, n = 40) and flow (bottom, n = 11) conditions overlaid onto a single ridged section. (e) Tracer particle pathlines during peak flow above flat (top) and ridged substrates (bottom). A region of flow recirculation above the ridged substrate is identified (yellow arrow).
Fig 2.
Millimeter-scale ridges generate regions of flow recirculation that influence larval settlement.
(a) Regions of flow recirculation identified over flat (left) and ridged (right) substrates during peak (top) and turning point (bottom) flow using the Q-criterion metric. (b) Range plot of the maximum Q-criterion values over flat and ridged substrates and gap regions during a full period of oscillation. The dotted line is the average standard deviation in Q-criterion across all measurements, which was used as the threshold for the identification of vortex structures (Qthresh). (c) Larval settlement data under a flow region with a Q-criterion greater than or less than Qthresh for D. labyrinthiformis (left) and C. natans (right). Settlement is presented as the mean percent of total settlers in each experimental run and the error bars represent the standard error of the mean. There was a significant difference in the settlement between regions above and below the Qthresh for both coral species (*: p < 0.01; post-hoc Tukey HSD).
Fig 3.
Millimeter-scale ridges increase the duration of larval settling windows.
(a) Visualization of the relative fluid speed over flat (left) and ridged (right) substrates during peak (top) and turning point (bottom) flow. The fluid velocity (U) is normalized by the average larval swimming velocity (uℓ). Dotted black boxes represent regions ≤1.5 mm above the substrate surface that were used to calculate duration of settlement windows. Black arrows near the bottoms of the ridges indicate regions where the velocity remains low even at the turning points. (b) The average relative flow speed within the dotted black regions plotted over an average period for flat (top) and ridged (bottom) substrates. The yellow regions highlight the settling windows during which the local flow speed drops below uℓ (black line) plus one standard deviation (grey band).
Fig 4.
Millimeter-scale ridges modify the boundary layer flow to facilitate larval transport and settlement.
(a) Results of larval settlement simulations on flat substrates and ridged substrates with millimeter-scale (2.5 mm) and sub-millimeter-scale (0.25 mm) ridges in oscillatory flow. There were significant differences in settlement between substrates with millimeter-scale ridges and flat and sub-millimeter-scale ridged substrates (**: p < 0.001; post-hoc Tukey HSD). (b) Simulation of a larva trajectory (colored dots) over a surface with millimeter-scale ridges. The color of the dot indicates the instantaneous Q-criterion experienced by the larva at each location. The spacing between dots increases with larval speed. The red line shows the boundary between regions of U > 5uℓ and U < 5uℓ during the peak rightward flow phase. (c) The instantaneous Q-criterion (top) and relative flow speed (bottom) experienced by the simulated larva in (b). After encountering multiple regions with high-Q values (grey band), the larva experiences several intervals of low relative fluid velocity, or active windows (yellow bands), before settling.