1

MECHANICS OF SEDIMENT SUSPENSION AND TRANSPORT WITHIN A FRINGING REEF

ANDREW W.M. POMEROY1, RYAN J. LOWE2, MARCO GHISALBERTI3, CURT D. STORLAZZI4, MICHAEL CUTTLER5 and GRAHAM SYMONDS6

1. The UWA Oceans Institute and The School of Earth and Environment, ARC Centre of Excellence for Coral Reef Studies, The University of Western Australia, Australia, Andrew.Pomeroy@research.uwa.edu.au

2. The UWA Oceans Institute and The School of Earth and Environment, ARC Centre of Excellence for Coral Reef Studies, The University of Western Australia, Australia, Ryan.Lowe@uwa.edu.au

3. The School of Civil and Mining Engineering, The University of Western Australia, Australia. Marco.Ghisalberti@uwa.edu.au

4. U.S. Geological Survey, Pacific Coastal and Marine Science Center, Santa Cruz, California, USA, cstorlazzi@usgs.gov

5. The UWA Oceans Institute and The School of Earth and Environment, ARC Centre of Excellence for Coral Reef Studies, The University of Western Australia, Australia, Mike.Cutter@resarch.uwa.edu.au

6. Centre for Australian Weather and Climate Research, CSIRO Wealth from Oceans Flagship, Wembley, Australia, Graham.Symonds@csiro.au

Abstract: Large bottom roughness is a characteristic of most coral reef environments and this has been shown to have a substantial impact on hydrodynamic processes in these environments. In this paper, we evaluate suspended sediment concentration (SSC) data as well detailed hydrodynamic data over a coral reef flat in Ningaloo Reef, Western Australia, to understand how this bottom roughness affects these processes. A well-developed logarithmic velocity layer consistently developed above a canopy layer during the experiment. Estimates of bottom stresses from these logarithmic profiles were comparable with estimates obtained directly from turbulent Reynolds stresses, and an order of magnitude greater than those typically reported for sandy beach environments having similar flow. Nevertheless, the sediment grain size distribution of the suspended load was very fine relative to what should be mobilized by these stresses, indicated the large roughness substantially suppressed sediment transport.

Introduction

The presence of large and complex bottom roughness (or canopies) imposes substantial drag forces on flows generated by waves and currents. These forces have been shown to be substantially larger in canopy environments than in sandy beach environments (e.g., Lowe et al., 2007; Rosman and Hench, 2011; Pomeroy et al., 2012). Although canopies are well known to influence bottom boundary layer hydrodynamics, which are strongly coupled to the sediment dynamics, the physical processes controlling the suspension and transport of sediment in coral reef environments are much less well understood.

086 20/3/2015

2

Consequently, knowledge gained from the study of sediment dynamics in beach environments is often translated into reef environments and applied to infer sediment suspension and transport processes, often through the use of numerical models (e.g., Storlazzi et al., 2011). In this case, emphasis is usually placed on correct representation of the hydrodynamic processes (as these can be readily measured) with friction parameters often extended until the flow conditions match those measured in the field (e.g., Van Dongeren et al., 2013). While this approach yields ‘correct’ representation of the hydrodynamic processes, this approach is likely to have consequences for the prediction of sediment dynamics in these environments. A detailed and quantitative study of sediment suspension and transport processes in reef environments is therefore required to elucidate these processes and is the focus of this study.

For coral reef environments, much of the insight into reef sediment dynamics has been derived from qualitative studies undertaken to assess the negative role that suspended sediments may play on reef organisms (e.g., Rogers, 1990; Umar et al., 1998). Consequently, knowledge of the physical processes involved in sediment transport is primarily limited to studies that focus on fine terrestrially derived sediment (e.g., Ogston et al., 2004; Storlazzi et al., 2009). These sediments typically discharge into the reef environment and have long suspension times (due to very low fall velocities) that enable their transport pathways to be described, at least to first order, by mean currents.

In addition to fine-grained sediment, coarser carbonate sediment produced by the reef is an important source of sediment. The production and reworking of carbonate sediment in reef environments is a balance between constructive and destructive physical, biological and chemical processes (e.g., Scoffin, 1992) as well as differences in the ecological state of a reef (Done, 1992). Calcium carbonate that is continually incorporated into reef frameworks by coral calcification (Vecsei, 2004), as well as other production sources such as crustose coralline algae (Perry and Hepburn, 2008), becomes mobilized as sediment by destructive processes (e.g., by bio- and mechanical erosion) that degrade these frameworks over time (Scoffin, 1992). The carbonate sediment that is released can then be exported away from active growth regions of the reef to other parts of the reef or lagoon, deposited to form coastline features or transported offshore into deeper water.

It is generally accepted (and demonstrated in beach and riverine environments) that the initiation of sediment motion can be directly predicted from bottom stresses that represent momentum transfer from the water column to the seabed. In particular, when a critical shear stress threshold is exceeded sediment is assumed to be mobilized, e.g. per various forms of Shield’s equation (Shields,

086 20/3/2015

3

1936) relating sediment properties (i.e., median grain size and density) to the bottom stress τ, or equivalently through a friction velocity u

* . However,

these bottom stresses can be difficult to directly measure and as a consequence, an estimation of this stress is often obtained from the near-bed velocity profile. For a variety of coastal environments with smooth to moderately rough seabeds, this profile is usually well described by the ‘law of the wall’, where there is a constant stress region near the bed in which the velocity increases logarithmically. Under steady uniform flow (i.e., currents) this model is frequently used to estimate *u and consequently a bottom drag coefficient CD. The addition of waves increases the complexity and have been incorporated in a number of ways such as by the addition of a wave related friction velocity (e.g., Madsen and Grant, 1976).

For reef environments subjected to current-dominated flow, a logarithmic boundary layer profile has been demonstrated to exist in the region above the coral canopy (e.g., Reidenbach et al., 2006). Thus while the flow structure may deviate considerably from logarithmic within a canopy, above the canopy (i.e., in a constant stress layer) a logarithmic profile is usually observed. For wave-dominated reefs, however the applicability of a logarithmic boundary layer for the time-averaged flow across the reef flat remains unclear. Furthermore, the importance of such a velocity profile to the suspension and transportation of sediment in a reef environment is unknown.

In this paper, we describe results obtained on a coral reef flat during a 3 week field study conducted at Tantabiddi (Ningaloo Reef, Western Australia). The objective of this component of the study was to quantify suspended sediment concentration within and above the reef canopy, evaluate the applicability of a ‘law of the wall’ profile on a wave dominated coral reef flat and determine how bottom stress imposed by the canopy relates to the sediment suspended within the water column.

Site Description

A 3 week field experiment (27 July – Aug 14 2014) was conducted in the northern region of Ningaloo Reef (Western Australia), focusing on an ~5 km section of reef near Tantabiddi (21°52'6.03"S, 113°58'58.26"E, Figure 1). The specific area was centered on a section of reef bounded to the north and south by channels that cut into the reef crest and outer reef flat.

086 20/3/2015

4

Figure 1. (a) A detailed plan view of the site with the locations of instruments indicated. (b) Photograph of the suspended sediment pump sampling scaffold on the reef flat located at S2.

At this site, the forereef slope rises at ~1:20 to the reef crest, which is located 2.0 – 2.7 km from the coastline. The reef flat is ~0.6 – 1.5 m below mean sea level and typically ranges between ~300 to 1200 m in width. The lagoon is generally ~3 m deep and is connected with deeper channels that are up to ~6 m deep; it also varies in width along the coast. In contrast to many parts of southern Ningaloo Reef that typically have near 100% coral coverage, this site was specifically chosen as it contained a mix of coral, algal, and other reef communities (Cuttler et al., 2015) that were deemed more similar to reef systems observed worldwide.

Field Study

The field study consisted of two main components: (1) a detailed study of the hydrodynamics and sediment transport within the bottom boundary layer of the reef flat, and (2) a broad scale hydrodynamic and sediment transport study throughout the reef and lagoon. The intensive sampling site focused on the reef flat to quantify fine-scale sediment fluxes in the reef bottom boundary layer. This consisted of a sawhorse instrument frame with 3 vertically-distributed acoustic Doppler velocimeters (ADVs), 3 co-located OBS loggers (FLNTUs), a high-resolution acoustic Doppler profiler (ADP-HR), and a pump sampling array with 6 intakes logarithmically distributed above the seabed that was pumped to a scaffold platform (Figure 1b). In a 20 m region surrounding the intensive sampling site, detailed fine-scale topographic surveys were conducted to quantify the bottom roughness, along with photo-transects to characterize benthic habitat composition. The broad scale sampling strategy involved the collection of hydrodynamic (waves, currents, water levels) and suspended

086 20/3/2015

5

sediment concentration data at 12 sites. The instrument array consisted of an acoustic wave and current meter (AWAC) that measured wave conditions on the forereef, 2 acoustic Doppler current profilers (ADCPs) measuring the water-column current profiles in the channels, 4 acoustic Doppler profilers (ADPs) that measured the current profiles within the lagoon, and 5 wave gauges distributed across the reef. Most sites had co-located optical backscatter sensors (OBSs) that were deployed to estimate suspended-sediment concentrations (7 in total) near the bed. Lagrangian measurements obtained by use of 14 GPS tracked drifters supplemented the fixed hydrodynamic instrumentation. Table 1 lists the instrumentation and sampling regime relevant to this paper.

Table 1: Instruments deployed during experiment

Site and Depth Instrument Sampling Information#

S1 (forereef ~9.5 m) Nortek AWAC 1 Hz with 2048 s burst every 3600 s; current profile every 5 min, 30 bins at 0.5 m; velocity sample height: 1.04 m

RBRVirtuoso D Continuous sampling at 1 Hz; pressure sample height: 0.2 m

S2 (reef flat ~1.4 m) Nortek ADV 8 Hz with 2048 s burst every 3600 s; velocity sample height: 0.23 m, 0.53 m and 0.85 m.

Nortek ADP-HR Continuous 1 Hz current profile, 31 cells at 25 mm, velocity sample height: 0.22 m; pressure sample height: 0.07 m

Pump samples Hourly during daylight; SSC sample height: 0.224, 0.269, 0.339, 0.509, 0.764, 1.016 m.

# Samples heights are relative to the seabed

Methods

On the reef flat, the suspended sediment concentration was determined from pump samples with intake ports located perpendicular to the cross-reef flow (c.f., Bosman et al., 1987). During daylight hours, water samples were collected at hourly intervals using peristaltic pumps and stored in 2 L bottles. The samples were vacuum filtered onto pre-weighed membrane filters (Whatman ME27, 0.8 μm), dried (75°C for 24 hrs) and weighed. The filters were analyzed under a microscope to identify grain properties, which were then measured from ~50 images of the filter under a microscope. The discrete pump samples were then used to calibrate time-averaged 30-min bursts of ADP-HR backscatter,

086 20/3/2015

6

which had been corrected for sonar decay (e.g., Ha et al., 2011), to obtain a profile time series of concentration.

For a well-developed turbulent boundary layer, an inertial sublayer region exists where mean velocities exhibit a logarithmic profile and is related to the generation of turbulence by shear at the bed. We evaluated if such a region develops for currents on reefs dominated by waves. 15-min bursts of instantaneous velocity data obtained from the ADP-HR and ADVs were rotated into the major axis of flow for that burst to define the flow speed. We chose to use this varying directional frame of reference as a well-developed boundary layer can be expected to develop in the major direction of the flow. The use of a 15-min data ensemble minimized the influence of the tidal flow. The bursts of time-averaged velocity data obtained by the ADP-HR were then evaluated to determine if a region of the profile conforms to a ‘law of the wall’ profile:

U u*

log

z d

z0

(1)

where u* is the friction velocity; κ = 0.41 is Von Karman’s constant; z is the elevation above the bed at the base of the roughness; d is the zero displacement elevation (the mean elevation where the moment is dissipated); and z0 is a hydraulic roughness parameter (m). While the logarithmic profile will breakdown inside a canopy, a logarithmic profile usually occurs within a constant stress layer above a canopy (Rosman and Hench, 2011) such as those observed above sea-grass beds in flumes (e.g., Nepf and Vivoni, 2000). Values of u*, d and z0 were adjusted to obtain a least-squares best fit and was conducted over a elevation range of 0.5 m – 0.8 m with 16 data points. This region was identified as the most consistent region where a logarithmic form was expected to develop. Above the upper limit the flow profile deviated from a logarithmic form as it approached a free-stream velocity, while below the lower limit a defined inflection point varied between 0.3 – 0.4 m and indicated a transition region to the within-canopy flow. To ensure a well-defined logarithmic region, we retained estimates of u* from profiles with a R2 > 0.95.

In an intertial sublayer, the Reynolds stresses are constant and linearly decrease to zero at the outer edge of the boundary. Using a second independent approach, we compared u* determined from the log-law fit with an estimate of u* evaluated from the ADV Reynolds stress data that were obtained at 8 Hz. The wave contribution to 15-min data ensembles (coincident with the ADP-HR data) were removed following Shaw and Trowbridge (2001) and the friction velocity ''2

* wuu determined from the turbulent contribution (u’ and w’) that

086 20/3/2015

7

remained. We retained estimates of u* that were obtained when the log-law fit criteria was satisfied (those bursts where R2 > 0.95 for the ADP-HR). In order to compare our bed stress results with other experiments, we recast u* in terms of a bottom drag coefficient = ∗/ where U0 is the free-stream velocity that we define to be above the identified logarithmic layer region in the ADP-HR profile.

Figure 2. Hydrodynamic wave conditions measured at S1 (on the fore reef) and at the high-resolution reef flat sample site (S2). (a) The tide (b) The significant wave height on the fore reef (c) High frequency (solid) and low frequency (dashed) waves observed on the reef flat. (d) Suspended sediment concentration at z = 0.5 m (red) and 0.2 m (blue) relative to the bed.

086 20/3/2015

8

Results

Key hydrodynamic observations

During the experiment, the incident significant wave heights Hm0 measured offshore on the forereef at S1 ranged from 0.7 - 2.4 m (Figure 2b). These waves generally approached the reef at an angle 15 – 25° off the cross-reef direction, and had peak periods (Tp) ranging from 12 s for lower wave heights to 19 s for the swell events (not shown). A large reduction in the short-wave heights (and hence short-wave energy) was observed on the reef flat at the location of the high-resolution analysis S2 (Figure 2c), located approximately mid-way across the reef flat. These waves were larger than the infragravity waves at this location during high tide but were similar in height at low tide. Notably, the short-wave heights at this location were strongly modulated by changes in the tidal depth over the reef, due to tidal variations to the depth-limited wave height with minimal influence from the larger swell conditions. These results are consistent with observations made on a section of Ningaloo Reef further south at Sandy Bay (Pomeroy et al., 2012). Consistent mean wave-driven flows generated by the wave breaking were oriented across the reef at this location (not shown).

Suspended-sediment concentrations and composition

On the reef flat, the suspended-sediment concentration (SSC) was consistently within the range ~0.5 mg/L to ~3 mg/L. Near the bed (z = 0.2 m), as well as higher in the water column (z = 0.5 m), the SSC was strongly modulated by the tide. Early in the experiment when the waves were low, the SSC at the two elevations were similar in magnitude. However, during the swell events that span 5 – 8 Aug and 9 – 12 Aug, the SSC concentrations at the two elevations differed substantially. While the near bed SSC increased in magnitude, this increase was consistently smaller than the increased SSC observed higher in the water column.

The median grain size (D50) from the analysis of the pump sample filters was ~70 μm in the samples analyzed to date (e.g. Figure 3a). Although a small number of larger grains were identified, ~40% of the grains were within the range of 60 – 85 μm. From visual classification, the grains were overwhelmingly coral-derived sediment with a small quantity of quartz grains. This is consistent with the composition of the bed sediment at this site, with the exception that the grain sizes were smaller than those contained in bed surface samples (Figure 3b, Cuttler et al., 2015). Grains sizes < 60 μm that represent

086 20/3/2015

9

clay or silt fraction were not considered and are assumed to be well mixed in the water column due to their long settling times.

Figure 3. (a) Grain size distribution from near bed pump sample filter at 14:00 on 8 August 2014. (b) Grain size distribution from bed surface sample at the sawhorse location. The vertical red line indicates the equivalent diameter of the bed stress determined from the ADV Reynolds stresses.

Boundary layer observations and bottom stress estimates

Using a least-squares fit, the friction velocities were estimated using the log-law technique initially holding d = 0 constant and then by allowing d to be a free parameter. The former approach assumes that the zero plane momentum sink is located at the seabed, whereas the later assumes that this sink is located higher in the canopy. The estimated u* was not substantially different between the two log-law fits (Figure 4a) and varied throughout the experiment with the direction of the flow (Figure 4b). On the reef flat, benthic coverage by reef communities varies, and hence roughness characteristics, is somewhat patchy. The variation of u*/U0 with flow direction would likely be due variations in upstream roughness, that will affect the development of the boundary layer at the measurement location.

086 20/3/2015

10

Figure 4. (a) u* estimated by a fixed displacement of d = 0 versus d as a free parameter in the log-law fit. (b) u* normalized by the free stream velocity (U0) defined at the top of the log-low fit as a function of the flow direction (relative to cross-reef flow).

To gain confidence that the log-law fit does in fact represent a logarithmic boundary layer, we evaluated the friction velocity from the Reynolds stresses obtained by the ADVs. There is good agreement between the friction velocity u* obtained by the two techniques, with a slight over prediction by the log-law technique for d = 0 (Figure 5a) as well as for d = free (Figure 5b). This has also been observed for current-dominated flows over other reefs (e.g., Reidenbach et al., 2006). An improved fit with the ADV data could be obtained if we forced d = 0.2, approximately equivalent the 1/2 – 2/3 of the height of reef communities in the local region where the measurements were made (Figure 5c).

Of particular interest is the stress imparted directly on the bed by the overlying flow. Throughout the experiment, u* varied according to the prevailing conditions over a range of ~0.001 – 0.07 m s-1. When recast in terms of a bottom drag coefficient CD, with the free stream flow defined higher in the water column, this equates to CD ~0.06 and 0.10 (2 = 0.02 and 0.07) for d = 0 and d = 0.2, respectively. This estimate of CD is within the range reported for other reef flats (e.g., Lowe et al., 2005; Lugo-Fernández et al., 1998; Reidenbach et al., 2006) but we note that comparison of CD with other field values is problematic due to varying definitions of the free stream velocity (Rosman and Hench, 2011). However our values are consistently 1 – 2 order of magnitude larger than those values typically observed on sandy beach environments O(0.003).

Implications for sediment dynamics

The results from this study provide quantitative insight into suspended sediment on a coral reef flat as well as the structure of the prevailing boundary layer

086 20/3/2015

11

Figure 5. Comparison of u* estimated from log-law fit (black) and Reynolds stresses (red) for different zero displacement (d) parameters.

above the reef canopy. The suspended sediment was dominated by the very fine sand fraction – coarse silt fraction, although larger sediment sizes in much smaller numbers were also observed. The SSC was tidally modulated with higher concentrations observed higher in the water column (i.e., above the canopy) than those within the canopy, particularly during the two swell events.

The dominance of the fine sediment fraction within the water column was a notable observation. Despite the seemingly high bed stresses, this indicates that sediment suspended experienced a much lower bed stress than that experienced by the flow above the canopy. This is demonstrated by a comparison of the bed stress required to mobilize the sediment against the bed stress calculated directly from the hydrodynamic measurements (Figure 3). This is consistent with two-layered velocity structures that have been proposed for coral reef environments where reef flat bottom stresses, estimated via log-law fits or Reynolds stresses, reflect the stress applied by the canopy to the flow above the canopy while within the canopy (where sediment is initially suspended) the flow is reduced and may possess a very different form (Lowe et al., 2005). We note that the addition of a wave-induced bed stress would only act to further enhance the bed stress and therefore including this additional stress would only further enhance

086 20/3/2015

12

the discrepancy. Consequently, the use of the bed stress values required to calibrate reef hydrodynamics as source terms in sediment transport equations are likely to overestimate sediment suspension and transport processes.

Future directions

This study is ongoing with the overarching aim to determine the mechanisms that drive sediment fluxes across this coral reef flat. Future research will investigate in greater detail the structure of suspended sediment concentration and velocity profiles on a coral reef flat as well as how a two-layered velocity structure affects sediment suspension and transport processes. More broadly, future research will evaluate sediment transport processes throughout a coral reef environment. The study will evaluate the suitability of existing (beach-derived) transport mechanisms to describe sediment transport over the reef flat and within the lagoon noting that low-frequency wave motions play an important role in this process (Pomeroy et al., 2015).

Conclusions

The importance of sediment suspension and transport to coral reef ecosystems is well established, however, data that specifically quantifies sediment suspension and transport processes are limited. Consequently, knowledge of the physics associated with these processes is primarily based upon principles developed for sandy beach environments that are extended to complex canopy environments without a firm theoretical basis. This has implications for the results from numerical models applied in these environments, which are likely to incorrectly quantify the order of magnitude of sediment suspended and transport.

In this study we conducted a detailed field study to obtain sediment suspension, transport and geometric characteristics, as well as the boundary layer structure in a coral reef environment dominated by waves with the following key results:

1. Suspended sediment concentrations were greater higher in the water column, were tidally modulated and dominated by the fine sand fraction, although larger sediment sizes were also observed.

2. A clear logarithmic velocity layer develops above the reef canopy but did not extend into the canopy; instead a transition layer was observed.

3. The shear stress on the reef flat is at least an order of magnitude larger than the shear stress estimated for sandy beach environments however this shear

086 20/3/2015

13

stress is due to the canopy drag on the flow above and does not represent the actual shear stress imparted on the bed sediment.

Acknowledgements

A.W.P. is grateful for support by a Robert and Maude Gledden Postgraduate Research Award and The Gowrie Trust Fund (2013, 2014). MC is supported by a Scholarship for International Research Fees (SIRF) and a University International Stipend (UIS). This project was funded by the Western Australia Marine Science Institute (WAMSI) Dredging Science Node (Theme 2/3), an Australian Research Council Future Fellowship (FT110100201) and ARC Discovery Project grant (DP140102026) to RJL, as well as the U.S. Geological Survey’s Coastal and Marine Geology Program. The authors thank Sana Dandan, Jim Falter, Jeff Hansen, Malcolm McCulloch, Leonardo Ruiz Montoya, and Gundula Winter for their assistance during the experiment.

References

Cuttler, M., Lowe, R., Hansen, J., Falter, J. and Pomeroy, A.W.M., 2015. Grainsize, composition and beform patterns in a fringing reef system. The Proceedings of Coastal Sediments.

Done, T.J., 1992. Phase shifts in coral reef communities and their ecological significance. Hydrobiologia, 247(1-3): 121-132.

Ha, H., Maa, J.-Y., Park, K. and Kim, Y., 2011. Estimation of high-resolution sediment concentration profiles in bottom boundary layer using pulse-coherent acoustic Doppler current profilers. Mar. Geol., 279(1): 199-209.

Lowe, R.J., Falter, J.L., Koseff, J.R., Monismith, S.G. and Atkinson, M.J., 2007. Spectral wave flow attenuation within submerged canopies: Implications for wave energy dissipation. J. Geophys. Res. Oceans, 112(C05018).

Lowe, R.J., Koseff, J. and Monismith, S.G., 2005. Oscillatory flow through submerged canopies: 1. Velocity structure. J. Geophys. Res. Oceans, 110(C10016): 1-17.

Lugo-Fernández, A., Roberts, H.H., Wiseman Jr, W.J. and Carter, B.L., 1998. Water level and currents of tidal and infragravity periods at Tague Reef, St. Croix (USVI). Coral Reefs, 17(4): 343-349.

Madsen, O.S. and Grant, W.D., 1976. Quantitative description of sediment transport by waves. Coastal Engineering Proceedings, Honolulu, Hawaii.

086 20/3/2015

14

Nepf, H.M. and Vivoni, E.R., 2000. Flow structure in depth-limited, vegetated flow. Journal of Geophysical Research, 105(C12): 28547-28557.ton, A.S., Storlazzi, C.D., Field, M.E. and Presto, M.K., 2004. Sediment resuspension and transport patterns on a fringing reef flat, Molokai, Hawaii. Coral Reefs.

Perry, C.T. and Hepburn, L.J., 2008. Syn-depositional alteration of coral reef framework through bioerosion, encrustation and cementation: Taphonomic signatures of reef accretion and reef depositional events. Earth-Science Reviews, 86(1–4): 106-144.

Pomeroy, A.W.M. et al., 2015. Spectral wave-driven sediment transport across a fringing reef. Coast. Eng.

Pomeroy, A.W.M., Lowe, R.J., Symonds, G., Van Dongeren, A. and Moore, C., 2012. The dynamics of infragravity wave transformation over a fringing reef. J. Geophys. Res. Oceans, 117(C11022).

Reidenbach, M., Monismith, S. and Koseff, J., 2006. Boundary layer turbulence and flow structure over a frining coral reef. Limnology and oceanography, 51: 1956-1968.

Rogers, C.S., 1990. Responses of coral reefs and reef organisms to sedimentation. Marine Ecology Progress Series.

Rosman, J.H. and Hench, J.L., 2011. A framework for understanding drag parameterizations for coral reefs. J. Geophys. Res. Oceans, 116(C08025).

Scoffin, T.P., 1992. Taphonomy of coral reefs: a review. Coral Reefs, 11(2): 57-77.

Shaw, W.J. and Trowbridge, J.H., 2001. The direct estimation of near-bottom turbulent fluxes in the presence of energetic wave motions. Journal of Atmospheric and Oceanic Technology, 18(9): 1540-1557.

Shields, A., 1936. Application of similarity principles and turbulence research to bed-load movement.

Storlazzi, C.D., Elias, E., Field, M.E. and Presto, M.K., 2011. Numerical modeling of the impact of sea-level rise on fringing coral reef hydrodynamics and sediment transport. Coral Reefs, 30: 83-96.

Storlazzi, C.D., Field, M.E., Bothner, M.H., Presto, M.K. and Draut, A.E., 2009. Sedimentation processes in a coral reef embayment: Hanalei Bay, Kauai. Mar. Geol., 264: 140-151.

Umar, M.J., McCook, L.J. and Price, I.R., 1998. Effects of sediment deposition on the seaweed Sargassum on a fringing coral reef. Coral Reefs, 17: 169-177.

Van Dongeren, A. et al., 2013. Numerical modeling of low-frequency wave dynamics over a fringing coral reef. Coast. Eng., 73: 178-190.

Vecsei, A., 2004. A new estimate of global reefal carbonate production including the fore-reefs. Global and Planetary Change, 43(1–2): 1-18.

086 20/3/2015