Sediment resuspension by wind, waves, and currents during meteorological frontal passages in a micro-tidal lagoon

Estuarine, Coastal and Shelf Science 172 (2016) 24e33

Contents lists available at ScienceDirect

Estuarine, Coastal and Shelf Science journal homepage: www.elsevier.com/locate/ecss

Sediment resuspension by wind, waves, and currents during meteorological frontal passages in a micro-tidal lagoon Joseph A. Carlin a, *, Guan-hong Lee b, Timothy M. Dellapenna c, d, Paul Laverty d a

Department of Geological Science, California State University, Fullerton, Fullerton, CA 92831, United States Department of Oceanography, Inha University, Incheon 402-751, South Korea c Department of Marine Sciences, Texas A&M University-Galveston, Galveston, TX 77553, United States d Department of Oceanography, Texas A&M University, College Station, TX 77843, United States b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 8 February 2015 Received in revised form 21 October 2015 Accepted 17 January 2016 Available online 19 January 2016

Meteorological frontal passages are recognized as important mechanisms for remobilizing sediment in estuaries along the northern Gulf of Mexico, but few studies have addressed factors beyond wind speed as a predictor for resuspension. To better understand resuspension mechanisms during these events, this study investigated the effects of wind, waves, and currents on suspended sediment concentration near the seabed during frontal passages in the shallow, micro-tidal West Galveston Bay located along the Texas coast. In late January and early February 2013, two multi-day deployments of instrumented pods (an acoustic Doppler velocimeter, and an acoustic wave and current profiler) were conducted to capture two separate frontal passages. The results indicate that the bed shear stress under the combined effect of waves and currents showed a much stronger relationship to sediment resuspension (R2 ¼ 0.90) than wind stress alone (R2 ¼ 0.55), or currents alone (R2 ¼ 0.72). Increases in the bed shear stress due to the combined effects of waves and currents resulted from increased wave height, which is strongly related to fetch within the bay. Therefore, understanding fetch-limited wave heights as a function of wind speed and direction, in conjunction with basin geometry, may be a better way to predict sediment resuspension during meteorological frontal passages in the shallow bays of the northern Gulf of Mexico. © 2016 Elsevier Ltd. All rights reserved.

Keywords: Sediment resuspension Meteorological cold front Micro-tidal lagoon Galveston bay Gulf of Mexico

1. Introduction Sediment remobilization in shallow estuaries can have ecosystem-wide implications by redistributing sedimentassociated biogeochemical constituents throughout the system (Bianchi, 2007). Typically, this may require high-energy events such as increased fluvial discharge, or spring tides (Baumann et al., 1984; Castaing and Allen, 1981; Geyer et al., 2001; Hensel et al., 1998; Nichols, 1993). In northern Gulf of Mexico (GOM) estuaries specifically, meteorological events such as hurricanes and tropical storms can also be effective drivers in sediment remobilization (Childers and Day, 1990; Conner et al., 1989; Day et al., 2000; Hayes, 1967; Perez et al., 2000; Turner et al., 2006; Walker, 2001). Hurricanes and tropical storms however, generally affect a relatively small spatial area, and are infrequent at a specific site, thus their impact

* Corresponding author. California State University Fullerton, 800 N. State College Blvd., Fullerton, CA 92831, United States. E-mail address: [email protected] (J.A. Carlin). http://dx.doi.org/10.1016/j.ecss.2016.01.029 0272-7714/© 2016 Elsevier Ltd. All rights reserved.

may not be as significant in estuaries compared to more regular meteorological forcings. In contrast, literature has shown that seasonal meteorological frontal events (winter cold fronts), because of their frequent occurrence, and larger spatial extent, may have greater impact compared to hurricanes on estuarine and coastal environments in the northern GOM (Moeller et al., 1993; Pepper et al., 1999; Roberts et al., 1987). These cold fronts can influence sediment resuspension, and overall sediment transport within the estuary. As an example, the Atchafalaya-Vermillion Bay region receives sediment and water input from the Atchafalaya River. Annual sediment transport from cold fronts for Atchafalaya-Vermillion Bay was estimated to be about 12% of the yearly average sediment discharge from the river (Walker and Hammack, 2000). Additionally, most of the sediment transport along Louisiana's Chenier-plain coast is thought to be driven by cold fronts, resulting in regionalescale accretion for this area, located along a largely eroding section of the GOM coast (Kineke et al., 2006). The sediment transport during these events may also have critical impacts on ecosystem functioning by modifying sediment

J.A. Carlin et al. / Estuarine, Coastal and Shelf Science 172 (2016) 24e33

delivery to nearby marshes (Baumann et al., 1984; Childers and Day, 1990; Draut et al., 2005; Perez et al., 2000; Reed, 1989), nutrient transport, and other biogeochemical fluxes both within, and out of the system (Booth et al., 2000; Day et al., 2000; Perez et al., 2003). In a study investigating natural and anthropogenic mechanisms for sediment resuspension in Galveston Bay, Dellapenna et al. (2006) estimated that sediment resuspension from cold fronts annually was equivalent to ~40% of the suspended sediment load from the bay's fluvial source. While they found that shrimp trawling can resuspend the equivalent of >250% of the fluvial supply to the bay, because cold fronts occur during winter months concurrent with peaks in primary productivity, versus shrimp trawling that dominantly occurs in summer months, resuspension during cold fronts likely has a greater impact on ecosystem health. Throughout the northern GOM, cold fronts are an important meteorological event with high sediment transport potential (Armbruster et al., 1995; Chaney and Stone, 1996; Crout and Hamiter, 1981; Dingler et al., 1993; Roberts et al., 1987; Stone and Wang, 1999), but they may also be an important forcing mechanism in other mid-latitude, lowenergy coastal environments (Pepper and Stone, 2004). Northern GOM cold fronts are typically associated with large increases in wind speed, and rapid shifts in wind direction. The rapid shifts in wind speed and direction often result in dramatic changes in water level. Sometimes referred to as “wind-tides,” the changes in water level height, and the water volume flux can exceed what would be predicted from astronomical tides (Perez et al., 2000; Smith, 1977). As a result, wind stress associated with these events has often been the focus mechanism driving sediment resuspension from the shallow coastal waters in studies from the region (Huh et al., 1991, 2001; Roberts et al., 1987; Walker, 1996; Walker and Hammack, 2000). For example, research in Mobile Bay established a critical wind stress for erosion, where suspended sediment concentrations increased rapidly above background levels in the bay with increased wind stress (Ha and Park, 2012). The focus on wind speed as the primary mechanism driving resuspension during these events is likely due in part because it is easy and routinely measured, but as noted by Ha and Park (2012) the wind stress threshold for erosion observed in Mobile Bay neglected any contribution from wave-induced bed shear stress. While wind in these environments drives wave formation and sometimes current generation, for estuaries in general, both waves and currents contribute significantly to sediment resuspension and transport (see review by Green and Coco, 2014). Booth et al. (2000) noted that in shallow, micro-tidal environments, wind-driven surface waves are one of the primary mechanisms for resuspension, but they only used linear wave theory to model resuspension based on wind speeds. Understanding what is happening near the bed as a result of the wind, waves, and currents during these events may add insight to the resulting sediment dynamics, and improve our predictive capabilities for resuspension and transport during winter cold fronts. To our knowledge however, no study to date has made direct measurements in an estuary of waves and currents during a cold front event, to investigate sediment resuspension from the seabed. The purpose of this study, therefore, is to directly investigate the impacts of a cold front passage on sediment suspension in a shallow, micro-tidal estuary within the northern GOM. Two separate instrumented pod deployments were conducted in West Galveston Bay (WGB) to capture two different frontal passages in late January and early February 2013. The data presented here will correlate observations of wind stress, bed shear velocity from currents only, and bed shear velocity from the combined effect of waves and currents to sediment resuspension. Through this, we aim to better understand the predictor for sediment resuspension from cold fronts in northern GOM estuaries beyond wind speed

25

alone, by incorporating the combined effect of waves and currents near the bed during the event. 2. Study area Galveston Bay, located in the northern GOM, is the second largest estuary in Texas with a surface area of ~1400 km2 (Fig. 1). Similar to other GOM bays, Galveston Bay is shallow (average depth 2.1 m), micro-tidal (0.5e0.7 m range), and has a deep shipping channel oriented along its main axis (Armstrong, 1982; Solis and Powell, 1999). The bay is considered to be meteorologically dominated, given its small tides, shallow depths, and high susceptibility to wind forces (Solis and Powell, 1999; Ward Jr, 1980). The prevailing south-southeast winds can generate waves across the open bay resulting in water-column mixing and erosion (Ward Jr, 1980). Stratification in the open bay is rare, only occurring during freshets. The freshets are derived primarily from the Trinity and, to a lesser extent, the San Jacinto Rivers that flow into the northeastern and northwestern shores of Galveston Bay proper, respectively. These two rivers have a combined net inflow of ~390 m3/s (Armstrong, 1982; Ward Jr, 1980). WGB is a ~200 km2 back barrier lagoon sub-estuary within the Galveston Bay System (Fig. 1). Separated from the open GOM by Galveston Island, WGB is tidally serviced by the San Luis Pass (SLP) tidal inlet to the west, connected to Galveston Bay proper to the east, and is generally around 2 m in depth or less. Freshwater inflow into WGB from the Trinity and San Jacinto rivers through Galveston Bay proper is limited due to the Houston Ship Channel (HSC), Texas City Dike, and Pelican Island (Powell et al., 2003). The primary source of freshwater to WGB is from Chocolate Bayou. Chocolate Bayou has a drainage basin of ~1000 km2, and an average inflow of ~3 m3/s, with a maximum flow of ~10 m3/s and a minimum flow of ~0.5 m3/s (USGS Gage Statin 08078000 near Alvin, TX from the years 1960e2013). In open areas of WGB, wave heights are typically less than 0.2 m (Ravens et al., 2009), average current velocities are less than 0.05 m/s (Dekshenieks et al., 2000), and fair-weather suspended sediment concentrations (SSC) in the bay are typically ~20 mg/l (Ward and Armstrong, 1992). Although the astronomical tides are small, the meteorologically-

Fig. 1. Map of the West Galveston Bay study area, where the red star indicates the location of the instrument deployment site. Wind data were collected at the station at San Luis Pass (SLP, blue dot), and water level data were from the station at the Galveston Railroad Bridge (GRB, red dot). Other geographic locations mentioned in the text are also labeled on the map including: the Texas City Dike (TCD), Houston Ship Channel (HSC), and Pelican Island (PI). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

26

J.A. Carlin et al. / Estuarine, Coastal and Shelf Science 172 (2016) 24e33

dominated nature of the bay can result in changes in water level due to wind that may obscure the astronomical tides (Ward Jr, 1980). These fluctuations primarily occur from changes in wind direction (not necessarily strong winds alone) that are often associated with the passage of cold fronts. Coastal areas in the northern GOM are typically subjected to 30e40 cold fronts per year, primarily during the months of October to April (Moeller et al., 1993; Roberts et al., 1987). Cold fronts in this area generally consist of the three distinct phases: 1) the prefrontal phase where southerly winds dominate as the front approaches; 2) the frontal passage phase where winds are variable and there is a rapid shift in wind direction from the south to the north or west; 3) and the postfrontal phase where winds are out of the north and temperature and humidity falls (Moeller et al., 1993). 3. Data collection and analysis 3.1. Instrument deployments Fig. 2. Calibration curve used to estimate suspended sediment concentration from the ADV acoustic backscatter.

Two instrumented pods were deployed twice at approximately the same locations in WGB (Fig. 1) in the winter of 2013 (24 January to 2 February, and 8 February to 15 February) to capture two distinct frontal passages. One pod contained an upward-looking Acoustic Wave and Current Profiler (Nortek AWAC, 1 MHz), and the second pod contained a downward-looking Acoustic Doppler Velocimeter (Nortek Vector ADV, 6 MHz). The nominal height of the ADV measurement was 0.25 m above the bed (mab). The two pods were deployed ~100 m apart. Table 1 summarizes some of the seteup parameters for each of the instruments. Between the two deployments the instruments were retrieved for data downloading and instrument cleaning. Mean water depths during the deployments at the site were ~2.5 m. The bed sediment, collected using a grab sampler at the site during the deployment, was ~60% mud with ~40% sand (mean diameter of 66 mm).

Hourly wind speed and direction data were obtained from the Texas Coastal Ocean Observing Network System (TCOONS) San Luis Pass Station (29 40 3200 N, 95 70 2100 W) located about10 km southwest of the deployment site (SLP e Fig. 1). At the time of the study, TCOONS was managed and operated by the Conrad Blucher Institute for Surveying and Science, Texas A&M University-Corpus Christi. The wind stress (tw) was estimated by the quadratic law given as

3.2. Suspended sediment concentration calibration

tw ¼ ra CD jWjW

Suspended sediment concentrations (SSC) were derived using the backscatter intensity from the ADV. With the ADV measurement window ~0.25 mab, the derived SSC represent the near-bed concentrations for the site. Backscatter intensity was converted to SSC estimates using a calibration curve derived in the lab (Fig. 2). To perform the calibration, sediment collected from the seabed at the deployment site was continually added to a tank filled with water from the bay and the ADV, and kept in suspension using pumps. Representative water samples were collected from the tank at specific backscatter intensities and filtered using pre-weighed, and pre-dried 0.7 mm glass-fiber filters. The filters were dried overnight (at 80  C) and reweighed to determine concentration based on the

ADV

3.3. Meteorological and hydrographic data

(1)

where ra is the air density (¼ 1.2 kg/m3), CD is the drag coefficient, and W is the wind speed (m/s). The drag coefficient was calculated by following Wu (1980) such that

CD ¼ 0:5W10  103 1=2

(2)

in which W10 is the wind speed at 10 m above the seabed (m/s). Water level data were obtained from the TCOONS station at the Galveston Railroad Bridge (29 180 800 N, 94 530 4700 W) located about 23 km northeast of the deployment site (GRB e Fig. 1). 3.4. Flow data analysis and bed shear stress Burst-averaged wave depth was obtained by averaging h(t), which is the burst time series of hydrostatic depth computed from the pressure series p(t) as

Table 1 Instrument sampling parameters. Parameter

volume of water filtered.

Parameter

Deployment 1: 24 January 2013 e 2 February 2013 Sampling Rate 16 Hz Profile Interval Nominal Velocity Range 1.00 m/s Number of Cells Burst Interval 120 s Cell Size Samples Per Burst 10 Average Interval Sampling Volume 14.9 mm Blanking Distance Deployment 2: 8 February 2013 e 15 February 2013 Sampling Rate 16 Hz Profile Interval Nominal Velocity Range 1.00 m/s Number of Cells Burst Interval 1800 s Cell Size Samples Per Burst 4096 Average Interval Sampling Volume 14.9 mm Blanking Distance

AWAC

hðtÞ ¼ ½ðpðtÞ  pa Þ=rg þ zp 120 s 12 0.25 m 55 s 0.40 m 600 s 8 0.25 m 55 s 0.40 m

(3)

where pa is the atmospheric pressure obtained from the TCOONS station, r is the density, g is the gravitational acceleration, and zp is the elevation of the pressure sensor. The energy-based significant wave height, Hmo, was calculated by the following definition,

pffiffiffiffiffiffiffi Hmo ¼ 4 mo

(4)

where mo is the variance of sea surface elevation (Tucker and Pitt, 2001). The wave period Tp was determined from the peak

J.A. Carlin et al. / Estuarine, Coastal and Shelf Science 172 (2016) 24e33

frequency. Based on linear wave theory, the wave orbital velocity ub was calculated by the following equation,

ub ¼

6a sinhðkhÞ

(5)

where 6 is the wave radian frequency and a is the wave amplitude. In this study, a velocity vector u ¼ (u, v, w) is considered in a Cartesian coordinate system (x, y, z) where y is the mean flow direction (north-south), x is horizontally across the mean flow direction (east-west), and z is the vertical direction. The instantaneous flow is described in terms of mean, and fluctuating
þ u′. This approach is commonly used components such that u ¼ u in estuaries and tidal flats when waves are negligible (e.g. Andersen et al., 2007; Kim et al., 2000; Pope et al., 2006). Under the presence of waves however, each fluctuating component must be separated into wave-induced (u0 w, v0 w, w0 w) and turbulence-induced variances (u0 t, v0 t, w0 t). In the absence of waves, the bed shear stress is often estimated from the first moment (mean) statistics or from the second moment statistics. The first moment method relies on observations of time-averaged current speed using the log-profile method. In the present study, measurement of current speed at only one height above the bed precludes use of the log-profile method. The second moment method uses high-frequency measurements of turbulent flow in the constant stress layer using the following methods: 1) covariance (COV), 2) inertial dissipation (ID), and 3) turbulent kinetic energy (TKE) (e.g. Andersen et al., 2007; Kim et al., 2000; Lee et al., 2003). These methods were tested and applied in estuaries (Andersen et al., 2007; Kim et al., 2000) and shelf environments (e.g. Lee et al., 2003; Williams et al., 1999). In the presence of waves, bed shear stress is estimated by the separation of the wave-induced and turbulence-induced variances of each fluctuating velocity components. In separating waveinduced variances, there are several methods mostly relying on theoretical assumption for the nature of the wave motions (e.g. Benilov and Filyushkin, 1970; Bricker and Monismith, 2007; Shaw and Trowbridge, 2001; Thornton, 1979). Tests of these methods showed that differences are negligible as long as the orientation of the instrument is known (Bricker and Monismith, 2007). In this paper, however, we used the inertial dissipation method of wcomponent (u*idm). This method is based on the measured velocity spectral characteristics and is discussed in more detail, along with its advantages and disadvantages, in Kim et al. (2000) and Lee et al. (2003). This method is less prone to wave bias due to the probe orientation, and does not require another velocimeter, or a synchronized wave gauge. As important as the separation of wave-induced shears from turbulence is, it is also instructive to assess the combined effect of waves and currents (u*tke). The absolute intensity of velocity fluctuations was obtained from the TKE method,

E¼

u02 þ v02 þ w02 : 2

(6)

From Eq. (6) you will note that it includes both wave and turbulence variance. Near the bed, where energy production is equal to energy dissipation, the bed shear stress is proportional to the turbulent kinetic energy such that ttke ¼ CtkeE (Soulsby and Dyer, 1981; Kim et al., 2000). Various Ctke values are proposed in the literature within the range of 0.18e0.21 (e.g. Soulsby, 1983; Soulsby and Dyer, 1981; Talke and Stacey, 2008), but the constant value of 0.19 has been widely used (e.g. Kim et al., 2000; Lee et al., 2003; Verney et al., 2007). Another method for estimating the bed shear stress is to use a

27

wave-current interaction model. Wave-current interaction models are often used to predict bed shear velocity due to currents (u*c) and bed shear velocity due to the combined effect of waves and currents (u*cw) from knowledge of current at a point, near-bottom orbital velocity, and physical bottom roughness characteristics. The Grant and Madsen wave-current interaction model (Glenn and Grant, 1987; Grant and Madsen, 1986; hereinafter reffered to as GM) was applied because of its simplicity and wide usage in the literature. In the application of the GM model, burst-averaged current velocity (uc), wave orbital velocity (ub), and mean grain size of the bed sediment were used as input. The bed roughness was defined as the sum of grain roughness, drag roughness and movable bed roughness. Here, the grain roughness is on the order of grain diameter (2.5ds, where ds ¼ 66 mm is the mean grain size). The drag roughness was calculated using the relationship given by Nielsen (1992) with the ripple model of Wiberg and Harris (1994), the movable bed roughness was estimated by Xu and Wright (1995). The GM model and its application are described in more detail and has been successfully applied to inner shelf conditions (e.g., Lee et al., 2002). 4. Results 4.1. Meteorological and water-column observations 4.1.1. Event 1 The first deployment was conducted between 24 January 2013 and 2 February 2013 (Fig. 3). The frontal passage sequence during this deployment began around January 29th with the prefrontal phase as the front approached (yellow shade on Fig. 3), followed by the frontal phase early on January 30th (gray shade on Fig. 3), and the postfrontal phase late January 30th to early January 31st (blue shade on Fig. 3). During the prefrontal phase, winds were initially from the south-southeast (Fig. 3a). Over the course of this phase the winds gradually increased in speed, and changed direction to the south. Water levels exhibited a typical tidal pattern prior to, and during, the onset of the prefrontal phase (Fig. 3b). Toward the end of the prefrontal phase, when the winds shifted to the south and approached their maximum speeds, water levels increased beyond the astronomical tides as seen with the minimal low tide and elevated high tide on the 29th. During this same time, at the end of the prefrontal phase, there were peaks in both wave heights (~0.1 m) and wave periods (~2.5 s) in the bay (Fig. 3c). Current speeds were also the highest at the end of the prefrontal phase, ~0.2 m/s flowing from southwest to northeast (Fig. 3d). The onset of the frontal phase was marked by a dramatic change in wind direction. Winds rapidly shifted from the south to a northnorthwesterly direction. As the frontal phase continued, the wind speed gradually increased, and the wind direction became more westerly. Through the frontal phase water levels precipitously dropped, from more than 0.3 m above mean sea level (MSL) to about 0.5 m below MSL. This was a nearly 1 m change in water level in less than 24 h. During the frontal phase there were also peaks in wave height and wave period in the bay. Maximum wave heights were >0.4 m, and maximum wave periods were >3 s. Southerly flowing currents dominated during the frontal phase. In the postfrontal phase the winds were initially from the north, but as this phase continued, the winds gradually decreased in speed, and shifted back to a west-northwesterly direction. Water levels remained low during this phase, with a high tide on January 30th at ~0.3 m below MSL. Wave height and wave periods remained near background levels, <0.1 m and 2 s respectively, throughout this phase. Current speeds decreased during this phase. Following the postfrontal phase the winds increased in speed and returned to a southerly direction as water levels increased to pre-event levels.

28

J.A. Carlin et al. / Estuarine, Coastal and Shelf Science 172 (2016) 24e33

Fig. 3. Data collected from the first deployment. The data include: a) wind speed and direction, b) water level, c) wave height (red) and wave period (black), d) near-bed current speed and direction, e) calculated wind stress, f) modeled bed shear velocities using currents only (u*c, blue) and the combination or waves and currents (u*cw, red), and g) estimated suspended sediment concentrations. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

4.1.2. Event 2 The second deployment was conducted from 8 February 2013 to 15 February 2013 (Fig. 4). A relatively short prefrontal phase (yellow shade on Fig. 4) began late February 9th, and was followed by the frontal phase (gray shade on Fig. 4) after mid-day on the 10th. In contrast to Event 1, the frontal phase was relatively prolonged, lasting through the early part of February 12th. This was the result of the front stalling offshore proximal to the study area, shortly after its passage. The postfrontal phase (blue shade on Fig. 4) began after mid-day on February 12th and persisted through February 14th. The winds during the relatively short-lived prefrontal phase initially increased rapidly in speed, and changed direction from dominantly east to south (Fig. 4a). Following this shift in direction, the wind speed remained fairly consistent at ~6 m/s out of the south for the first half of February 10th. Water levels increased during the prefrontal phase, reaching a peak >0.4 m at high tide

Fig. 4. Data collected from the second deployment. The data include: a) wind speed and direction, b) water level, c) wave height (red) and wave period (black), d) near-bed current speed and direction, e) calculated wind stress, f) modeled bed shear velocities using currents only (u*c, blue solid line) and the combination of waves and currents (u*cw, red solid line) with estimated bed shear velocities using currents only (u*idm, blue dashed line) and the combination of waves and currents (u*tke, red dashed line), and g) estimated suspended sediment concentrations. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

(Fig. 4b). Waves increased slightly in both wave height and wave period in the bay during this phase (Fig. 4c), and current speeds were generally low (Fig. 4d). The transition from prefrontal to frontal phase was marked by an abrupt change in wind direction similar to Event 1. At the beginning of this phase the winds rapidly shifted from the south to the north. As the frontal phase progressed, the wind speeds gradually increased, and the direction shifted to the northeast. Midday on February 12th, at the end of the frontal phase, the winds briefly shifted back to the south and increased in speed. Water levels during this phase remained elevated, although water levels at high tide were lower than the high tide during the prefrontal phase (~0.3e0.4 m); water levels during low tides were higher than low tide level during the prefrontal phase. Overall, during the frontal phase water levels were elevated, and there was a reduction in tidal range as water levels dropped below MSL only once during this

J.A. Carlin et al. / Estuarine, Coastal and Shelf Science 172 (2016) 24e33

phase. Three significant peaks in wave height and period were observed during the frontal phase. The first peak was the largest, occurring prior to mid-day on February 11th. Wave heights rapidly increased to >0.4 m in the bay, with >3 s periods. Shortly thereafter, wave heights decreased to ~0.2 m, then increased again late on February 11th to ~0.3 m for the second peak during the phase. Wave period decreased to <3 s, but remained elevated during this time. The final peak was small (0.1 m wave heights and a slight increase in wave periods >2 s). Current speeds also peaked during the frontal phase. The speeds were the greatest from February 11th to early February 12th. Flow was to the southwest during these peaks, with speeds approaching 0.2 m/s during the first peak, and 0.1 m/s during the second. A third peak in current speeds was smaller in magnitude, flowed northward, and occurred at the end of the frontal phase. The initiation of the postfrontal phase occurred late in the day on February 12th, and was characterized by a shift in wind direction from the northwest to the north. The winds increased in speed through the first half of the postfrontal phase. In the latter half of the phase, there was a decrease in wind speed, and wind direction shifted to the northeast. Water levels rapidly decreased similar to Event 1. In less than 24 h, the water levels decreased by ~0.6 m. In the latter of half of the postfrontal phase water levels increased, and the astronomical tidal signal returned to dominance. A small peak in wave height (~0.1 m) was observed midday on February 13th, while wave periods were variable for this phase and much of the rest of the deployment. Current speeds were low during postfrontal phase and varied in direction.

29

peak in wind stress (~1.5 Pa) was observed at the end of the frontal phase, concurrent with the brief period when the winds shifted back to south. During the postfrontal phase, an additional peak in wind stress (~0.2 Pa) was observed prior to midday on February 13th. As mentioned above, for this deployment we were able to estimate as well as model bed shear velocities. Throughout the deployment there was a good agreement between estimated and modeled bed shear velocities, but with only a slight overestimation of modeled bed shear velocities (Fig. 5). The R2 value was 0.68 between estimated u*idm and modeled u*c representing currentonly velocities, while it was 0.95 between estimated u*tke and modeled u*cw representing combined wave-current velocities. Bed

4.2. Wind stress and bed shear velocity 4.2.1. Event 1 Calculated wind stress and modeled bed shear velocities, u*c and u*cw, for the first deployment are also shown in Fig. 3. Due to the low sampling frequency of the ADV during this deployment, we were unable to estimate bed shear velocity from flow data, and therefore can only present the modeled bed shear velocities. During the prefrontal phase, the wind stress follows wind speed as expected (Fig. 3e). A brief peak in wind stress was observed at the end of the prefrontal phase. During the frontal phase wind stress rapidly increased, peaking at ~0.5 Pa, before also rapidly decreasing to near 0 Pa at the end of the frontal phase. Wind stress remained minimal throughout the postfrontal phase as well. The modeled bed shear velocities increased midway through the prefrontal phase (Fig. 3f). Two distinct peaks at the end of the prefrontal phase corresponded to similar peaks observed in wind events as well as wave heights. As with the wind stress, bed shear velocities would have also rapidly increased through the frontal phase, peaking near 5 and 2 cm/s for u*cw (red line, Fig. 3f) and u*c (blue line, Fig. 3f) respectively, before also rapidly decreasing at the end of the phase. Estimated bed shear velocities were also minimal during the postfrontal phase similar to wind stress. 4.2.2. Event 2 Wind stress and bed shear velocities for the second deployment are shown in Fig. 4. During this deployment we were able to calculate bed shear velocities from flow data. During the prefrontal phase wind stress is generally low (<0.1 Pa), and there were two brief peaks (<0.2 Pa) that were observed during this phase (Fig. 4e). During the frontal phase, wind stress is initially low, but rapidly increased starting early on February 11th, and peaking around midday at >0.3 Pa. Following this peak, wind stress rapidly decreased, but remained elevated, fluctuating around 0.2 Pa until midday on February 12th were it decreased to near 0 Pa. A second

Fig. 5. Comparisons of estimated bed shear velocities to modeled bed shear velocities for a) currents only, u*c; and b) combined waves and currents, u*cw. The least squares linear regression is shown in red, and a 1:1 line is shown in black. Data used in these plots were from the second deployment only; see text for more detailed discussion. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

30

J.A. Carlin et al. / Estuarine, Coastal and Shelf Science 172 (2016) 24e33

shear velocities were generally low during the prefrontal phase, and began to increase after the beginning of the frontal phase (Fig. 4f). Three peaks were observed during the frontal phase. The first and largest peak occurred before midday on February 11th, and corresponded to peaks in wind stress, current speed and wave heights. Bed shear velocities decreased rapidly following this peak, but subsequently increased after midday, concurrent with the second peak in waves and currents during the frontal phase. The third peak occurred at the end of the frontal phase, also concurrent with peaks in wind stress, and wave height. Similar to wind stress, the small peak during the postfrontal phase, prior to midday on February 13th, is also concurrent with a small peak in wave height. 4.3. Suspended sediment observations 4.3.1. Event 1 The SSC data for the first deployment are shown in Fig. 3g. The near-bed SSC increased during the prefrontal phase from background levels of 100 s mg/l to a peak of 100e200 mg/l toward the end of the prefrontal phase. This first peak in SSC corresponded to the peaks in current speeds, wave heights, wind stress, and estimated bed shear velocities near the end of the prefrontal phase. SSC decreased during the transition to the frontal phase, but rapidly increased up to ~1000 mg/l during the frontal phase. This SSC maximum corresponded to the peak in wind stress, the maximum wave heights, and peak in modeled bed shear velocity during this phase. During the postfrontal phase SSC dropped below 100 mg/l, and remained consistent through the postfrontal phase. 4.3.2. Event 2 Near-bed SSC remained low (100 s mg/l) during the prefrontal phase, and increased on February 11th during the frontal phase. At this time SSC increased rapidly to about 1000 mg/l corresponding to the peaks in wind stress, current speeds, wave heights, and bed shear velocities. The concentrations remained high (>500 mg/l) throughout February 11th corresponding to the elevated bed shear velocities. These concentrations decreased towards the end of the frontal phase, with a slight increase at the end of the phase. During the postfrontal phase, SSC increased near mid-day February 13th corresponding to the peak in wind stress, wave height, and bed shear velocity. 5. Discussion As with the previous studies, these results show a relationship between wind stress and sediment resuspension. For both events (Figs. 3 and 4), peaks in near-bed SSC correspond to peaks in wind stress. Moreover, in Fig. 6a, where SSC are plotted against wind stress, the plot shows a resuspension threshold at ~0.1 Pa, similar to the wind-induced critical erosion threshold (0.08e0.1 Pa) observed in Mobile Bay (Ha and Park, 2012). Beyond this threshold, SSC increased linearly with increasing stress (R2 ¼ 0.55). While there is an apparent relationship between wind stress and resuspension during these events, wind stress alone is obviously not the best predictor for resuspension. To improve on predictions of resuspension during these events beyond a wind-based

Fig. 6. Plots comparing wind stress a), current-only bed shear velocities b), and waveand-current bed shear velocities c) to suspended sediment concentrations. Data in a) incorporates both deployments, while data in b) and c) are for the second deployments only because shear velocity estimates were not available for the first deployment. The blue lines in the plots are regression lines, for a) and c) the regressions represent those data above an observed critical threshold (0.1 Pa and 1.1 cm/s, respectively). See text for a more detailed discussion. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

J.A. Carlin et al. / Estuarine, Coastal and Shelf Science 172 (2016) 24e33

critical erosion threshold, we looked at the impact of waves and currents near the bed during the events. The theoretical underpinning of sediment resuspension has advanced by the combined effects of waves and currents in estuarine and coastal settings (e.g., Dyer, 1986; Nielsen, 1992), and therefore to address the implications of such mechanisms at the bed, we looked at both the effect from currents alone and the combined effects of waves and currents. In Fig. 6b and c, u*idm and u*tke values are compared to SSC, respectively. The u*idm is most like the u*c and therefore represents the currents-only induced bed shear stress, while u*tke is most like u*cw that incorporates the total stress from waves and currents. Using u*idm as u*c, and u*tke as u*cw is reasonable given the relatively strong relationships between these components as shown in Fig. 5. From Fig. 6b, we do not observe a similar clear “threshold” as we did with the wind stress relationship from Fig. 6a, but we also do not observe significant improvement in the relationship to SSC. Overall, the R2 value between u*idm and SSC is 0.72, and thus only slightly better than SSC and wind stress. Combining waves and currents near the bed however; we observed the strongest relationship of bed shear stress to SSC (R2 ¼ 0.90, Fig. 6c). Unlike u*idm, there appears to be a break in the data below a SSC value of ~200 mg/l. The boundary of this concentration is centered at the bed shear velocity of ~1.1 cm/s, quite close to the threshold value (u*cr ¼ 1.09 cm/s) for the initiation of motion for the mean grain size (~66 mm) of the study site (Soulsby and Whitehorse, 1997), and this threshold for initiation of motion appears to correspond to the threshold value for wind stress (Fig. 6a). Overall, these results indicate that the wave-current induced stress near the bed is the better predictor of sediment resuspension during a cold front compared to wind stress and currents alone. To apply wave-current bed shear stresses as a predictor for sediment resuspension during these events, one would need to know parameters such as wave height, period, and the water depth. For any given estuary, the best prediction of the wave parameters will result by accounting for fetch. The results from this study also demonstrate the importance of fetch on generating waves that increase SSC. For example, during the second deployment (Fig. 4) we observed times when similar wind stress resulted in variable wave heights. Early on February 12th wind stress was ~0.2 Pa and wave heights were ~0.3 m. Midday on February 13th when wind stress was again ~0.2 Pa, wave heights were only ~0.1 m. The difference during these two periods was wind direction. On February 12th, winds were out of the northeast, while on February 13th, winds were out of the north. For WGB, winds out of the northeast are oriented along the main axis of the bay, thereby maximizing fetch, while north winds are oriented across the main axis of the bay. Fig. 7 illustrates the relationship between wind direction and wave height from observations during both deployments, and highlights that the waves generated from northerly winds are typically smaller than winds that maximize fetch in the bay, with more northeasterly or northwesterly components. The data also show that during these times when fetch is maximized, SSC increases. On February 12th with the maximized fetch SSC were ~500 mg/l compared to ~100 mg/l on February 13th (Fig. 4). As expected the bed shear velocity at this time on February 12thexceeded the 1.1 cm/s threshold, while on February 13th bed shear velocity for the most part remained below this level. Therefore, the best predictor for resuspension during cold fronts in northern GOM estuaries may be waves as a function of fetch. When and where wave heights are maximized due to fetch, the bed shear stresses will be maximized, thereby maximizing resuspension. Fetch-limited wave heights as a function of wind speed and direction when combined with the basin geometry for the

31

Fig. 7. Plot comparing wind speed to wave height across different wind directions: North (320 e15 , blue), East-Northeast (15 e50 , red), and West-Northwest (230 e320 , green). Wind and wave data were from both deployments. The regression lines for each directional group are shown (dashed lines) with R2 values, and the relative location of the deployment site (yellow star) to the directional groups is shown in the inset. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

estuary not only predict resuspension from the event overall, but can also be used to predict when resuspension is most-likely to occur within the event. For example, in WGB we can predict that resuspension is most likely to occur during the frontal phase when winds are typically more northwesterly or northeasterly, rather than the postfrontal phase when winds are expected to be out of the north. The data support this prediction, as resuspension was most pronounced during the frontal phase for both events. In other estuaries it may occur during other phases based on the orientation of the specific estuary relative to the path of the front, thus this general predictive relationship should be applicable in other systems. For any specific estuary, when and where fetch is maximized generating the largest wave energies that are translated to the bed during the event, is when and where resuspension is likely to be maximized.

6. Conclusions This study built upon previous research that showed the relationship between sediment resuspension and wind stress in estuaries during a cold front passage, by concurrently investigating the impact of waves and currents near the bed during the event. Results from two separate instrument deployments that captured two distinct cold front events showed that bed shear stresses under the combined effect of waves and currents provide a better predictor for sediment resuspension. The data showed that periods of high SSC near the bed corresponded to periods of increased wave height and wave periods, which was found to be related to the fetch based on the estuarine geometry. Thus, this study showed that better prediction of sediment resuspension due to cold fronts in northern GOM estuaries requires increased waves heights due to fetch (wind speed and direction). This improved predictor for resuspension during these events also allows for predicting when and where resuspension may occur within the event by understanding typical wind patterns versus the orientation of the estuary. Given the orientation of WGB, we found that resuspension was maximized during the frontal phase when winds were most likely to be oriented along the main axis of the bay maximizing fetch. These results may be applicable for other

32

J.A. Carlin et al. / Estuarine, Coastal and Shelf Science 172 (2016) 24e33

estuaries too, and predicting relative sediment resuspension timing and intensities during cold front events is important for understanding sediment and associated sedimentary material (e.g. nutrients, sedimentary organic matter, contaminants) transport during these relatively frequent mid-latitude meteorological events. Acknowledgments The authors would like to thank Ryan Gay, Judson Crouch and other students at Texas A&M University e Galveston for their assistance in the field, and in the lab. This research was supported in part by the project, “Development of integrated estuarine management system”, funded by the Ministry of Oceans and Fisheries of Korea to GL, and by Inha University Research Grant (INHA-5190301) to GL. The authors would also like to thank Terri Patchen for help during the writing processes, and Eric Wolanski, Steve Mitchell, and two anonymous reviewers for their constructive comments on earlier versions of the manuscript. References Andersen, T.J., Fredsoe, J., Pejrup, M., 2007. In situ estimation of erosion and deposition thresholds by Acoustic Doppler Velocimeter (ADV). Estuar. Coast. Shelf Sci. 75, 327e336. Armbruster, C.K., Stone, G.W., Xu, J., 1995. Episodic atmospheric forcing and bayside foreshore erosion: Santa Rosa Island, Florida. Trans. Gulf Coast Assoc. Geol. Soc. 45, 31e37. Armstrong, N.E., 1982. Response of Texas estuaries to freshwater inflows. In: Kennedy, V.S. (Ed.), Estuarine Comparisons. Academic Press, New York, pp. 103e120. Baumann, R.H., Day, J.W., Miller, C.A., 1984. Mississippi deltaic wetland survival: sedimentation versus coastal submergence. Science 224, 1093e1095. Benilov, A.Y., Filyushkin, B., 1970. Application of methods of linear filtration to an analysis of fluctuations in the surface layer of the sea. Atmos. Ocean. Phys. 6, 810e819. Bianchi, T.S., 2007. Biogeochemistry of Estuaries. Oxford University Press, New York, 706 pp. Booth, J., Miller, R., McKee, B., Leathers, R., 2000. Wind-induced bottom sediment resuspension in a microtidal coastal environment. Cont. Shelf Res. 20, 785e806. Bricker, J.D., Monismith, S.G., 2007. Spectral wave-turbulence decomposition. J. Atmos. Ocean. Technol. 24, 1479e1487. Castaing, P., Allen, G.P., 1981. Mechanisms controlling seaward escape of suspended sediment from the Gironde: a macrotidal estuary in France. Mar. Geol. 40, 101e118. Chaney, P., Stone, G., 1996. Soundside erosion of a nourished beach and implications for winter cold front forcing: west Ship Island, Mississippi. Oceanogr. Lit. Rev. 43. Childers, D.L., Day, J.W., 1990. Marsh-water column interactions in two Louisiana estuaries. I. Sediment dynamics. Estuaries 13, 393e403. Conner, W.H., Day Jr., J.W., Baumann, R.H., Randall, J.M., 1989. Influence of hurricanes on coastal ecosystems along the northern Gulf of Mexico. Wetl. Ecol. Manag. 1, 45e56. Crout, R.L., Hamiter, R.D., 1981. Response of bottom waters on the west Louisiana shelf to transient wind events and resulting sediment transport. Trans. Gulf Coast Assoc. Geol. Soc. 31, 273e277. Day Jr., J.W., Psuty, N.P., Perez, B.C., 2000. The Role of Pulsing Events in the Functioning of Coastal Barriers and Wetlands: Implications for Human Impact, Management and the Response to Sea Level Rise, Concepts and Controversies in Tidal Marsh Ecology. Springer, pp. 633e659. Day, J.W., Britsch, L.D., Hawes, S.R., Shaffer, G.P., Reed, D.J., Cahoon, D., 2000. Pattern and process of land loss in the Mississippi Delta: a spatial and temporal analysis of wetland habitat change. Estuaries 23, 425e438. Dellapenna, T.M., Allison, M.A., Gill, G.A., Lehman, R.D., Warnken, K.W., 2006. The impact of shrimp trawling and associated sediment resuspension in mud dominated, shallow estuaries. Estuar. Coast. Shelf Sci. 69, 519e530. Dekshenieks, M.M., Hofmann, E.E., Klinck, J.M., Powell, E.N., 2000. Quantifying the effects of environmental change on an oyster population: a modeling study. Estuaries 23, 593e610. Dingler, J.R., Reiss, T.E., Plant, N.G., 1993. Erosional patterns of the Isles Dernieres, Louisiana, in relation to meteorological influences. J. Coast. Res. 9, 112e125. Draut, A.E., Kineke, G.C., Huh, O.K., Grymes III, J.M., Westphal, K.A., Moeller, C.C., 2005. Coastal mudflat accretion under energetic conditions, Louisiana chenierplain coast, USA. Mar. Geol. 214, 27e47. Dyer, K.R., 1986. Coastal and Estuarine Sediment Dynamics. Wiley, p. 342. Geyer, W.R., Woodruff, J.D., Traykovski, P., 2001. Sediment transport and trapping in the Hudson River estuary. Estuaries 24, 670e679. Glenn, S.M., Grant, W.D., 1987. A suspended sediment stratification correction for

combined wave and current flows. J. Geophys. Res. Oceans 92, 8244e8264. Grant, W.D., Madsen, O.S., 1986. The continental-shelf bottom boundary layer. Annu. Rev. Fluid Mech. 18, 265e305. Green, M.O., Coco, G., 2014. Review of wave-driven sediment resuspension and transport in estuaries. Rev. Geophys. 52, 77e117. Ha, H.K., Park, K., 2012. High-resolution comparison of sediment dynamics under different forcing conditions in the bottom boundary layer of a shallow, microtidal estuary. J. Geophys. Res. Oceans 117, C06020. http://dx.doi.org/10.1029/ 2012JC007878. Hayes, M.O., 1967. Hurricanes as Geological Agents: Case Studies of Hurricanes Carla, 1961, and Cindy, 1963-Report of Investigations. Texas Bureau of Economic Geology. University of Texas, Austin, TX, p. p. 54. ^ ne Hensel, P., Pont, D., Day, J., 1998. Short-term sedimentation dynamics in the Rho River Delta. France: the importance of riverine pulsing. Estuaries 21, 52e65. Huh, O.K., Roberts, H.H., Rouse, L., Rickman, D.A., 1991. Fine grain sediment transport and deposition in the Atchafalaya and Chenier Plain sedimentary system, Coastal Sediments '91. ASCE 817e830. Huh, O.K., Walker, N.D., Moeller, C., 2001. Sedimentation along the eastern chenier plain coast: down drift impact of a delta complex shift. J. Coast. Res. 17 (1), 72e81. Kim, S.C., Friedrichs, C., Maa, J.-Y., Wright, L., 2000. Estimating bottom stress in tidal boundary layer from acoustic doppler velocimeter data. J. Hydraul. Eng. 126, 399e406. Kineke, G., Higgins, E., Hart, K., Velasco, D., 2006. Fine-sediment transport associated with cold-front passages on the shallow shelf, Gulf of Mexico. Cont. Shelf Res. 26, 2073e2091. Lee, G., Dade, W.B., Friedrichs, C.T., Vincent, C.E., 2003. Spectral estimates of bed shear stress using suspended-sediment concentrations in a wave-current boundary layer. J. Geophys. Res. Oceans 108 (C7), 3208. http://dx.doi.org/ 10.1029/2001JC001279. Lee, G., Friedrichs, C.T., Vincent, C.E., 2002. Examination of diffusion versus advection dominated sediment suspension on the inner shelf under storm and swell conditions, Duck, North Carolina. J. Geophys. Res. Oceans 107 (C7), 3084. http:// dx.doi.org/10.1029/2001JC000918. Moeller, C.C., Huh, O.K., Roberts, H.H., Gumley, L.E., Menzel, W.P., 1993. Response of Louisiana coastal environments to a cold front passage. J. Coast. Res. 9 (2), 434e447. Nichols, M.M., 1993. Response of coastal plain estuaries to episodic events in the Chesapeake Bay region. In: Mehta, A.J. (Ed.), Nearshore and Estuarine Cohesive Sediment Studies, vol. 42. American Geophysical Union, Washington DC, 581 pp. Nielsen, P., 1992. Coastal Bottom Boundary Layers and Sediment Transport. World Scientific, p. 324. Pepper, D.A., Stone, G.W., Wang, P., 1999. Bottom boundary layer parameters and sediment transport on the Louisiana inner-shelf during cold front passages. Gulf Coast Assoc. Geol. Soc. Trans. 49, 432e439. Pepper, D.A., Stone, G.W., 2004. Hydrodynamic and sedimentary responses to two contrasting winter storms on the inner shelf of the northern Gulf of Mexico. Mar. Geol. 210, 43e62. Perez, B., Day, J., Rouse, L., Shaw, R., Wang, M., 2000. Influence of Atchafalaya River discharge and winter frontal passage on suspended sediment concentration and flux in Fourleague Bay, Louisiana. Estuar. Coast. Shelf Sci. 50, 271e290. Perez, B.C., Day, J.W., Justic, D., Twilley, R.R., 2003. Nitrogen and phosphorus transport between Fourleague Bay, LA, and the Gulf of Mexico: the role of winter cold fronts and Atchafalaya River discharge. Estuar. Coast. Shelf Sci. 57, 1065e1078. Pope, N., Widdows, J., Brinsley, M., 2006. Estimation of bed shear stress using the turbulent kinetic energy approachda comparison of annular flume and field data. Cont. Shelf Res. 26, 959e970. Powell, E.N., Klinck, J.M., Hofmann, E.E., Mcmanus, M.A., 2003. Influence of water allocation and freshwater inflow on oyster production: a HydrodynamiceOyster population model for Galveston Bay, Texas, USA. Environ. Manag. 31, 0100e0121. Ravens, T.M., Thomas, R.C., Roberts, K.A., Santschi, P.H., 2009. Causes of salt marsh erosion in Galveston Bay, Texas. J. Coast. Res. 25 (2), 265e272. Reed, D.J., 1989. Patterns of sediment deposition in subsiding coastal salt marshes, Terrebonne Bay, Louisiana: the role of winter storms. Estuaries 12, 222e227. Roberts, H., Huh, O., Hsu, S., Rouse, L., Rickman, D., 1987. Impact of cold-front passages on geomorphic evolution and sediment dynamics of the complex Louisiana coast, Coastal Sediments '87. ASCE 1950e1963. Shaw, W.J., Trowbridge, J.H., 2001. The direct estimation of near-bottom turbulent fluxes in the presence of energetic wave motions. J. Atmos. Ocean. Technol. 18, 1540e1557. Smith, N.P., 1977. Meteorological and tidal exchanges between Corpus Christi Bay, Texas, and the northwestern Gulf of Mexico. Estuar. Coast. Mar. Sci. 5, 511e520. Solis, R., Powell, G., 1999. Hydrography, mixing characteristics, and residence times of Gulf of Mexico estuaries. Biogeochem. Gulf Mexico Estuaries 29e61. Soulsby, R., 1983. The bottom boundary layer of shelf seas. In: Johns, B. (Ed.), Physical Oceanography of Coastal and Shelf Seas, Elsevier Oceanography Series, vol. 35, pp. 189e266. Amsterdam. Soulsby, R., Dyer, K., 1981. The form of the near-bed velocity profile in a tidally accelerating flow. J. Geophys. Res. Oceans 86, 8067e8074. Soulsby, R., Whitehorse, R., 1997. Threshold of sediment motion in coastal environment. In: Proc. Pacific Coasts and Ports '97 Conf. University of Canterbury, Christchurch, New Zealand, pp. 149e154.

J.A. Carlin et al. / Estuarine, Coastal and Shelf Science 172 (2016) 24e33 Stone, G.W., Wang, P., 1999. The importance of cyclogenesis on the short-term evolution of Gulf coast barriers. Trans. Gulf Coast Assoc. Geol. Soc. 49, 47e48. Talke, S.A., Stacey, M.T., 2008. Suspended sediment fluxes at an intertidal flat: the shifting influence of wave, wind, tidal, and freshwater forcing. Cont. Shelf Res. 28, 710e725. Thornton, E.B., 1979. Energetics of breaking waves within the surf zone. J. Geophys. Res. Oceans 84, 4931e4938. Tucker, M.J., Pitt, E.G., 2001. Waves in Ocean Engineeringvol. 5. Elsevier, Amsterdam, 550 pp. Turner, R.E., Baustian, J.J., Swenson, E.M., Spicer, J.S., 2006. Wetland sedimentation from hurricanes Katrina and Rita. Science 314, 449e452. Verney, R., Deloffre, J., Brun-Cottan, J.-C., Lafite, R., 2007. The effect of wave-induced turbulence on intertidal mudflats: impact of boat traffic and wind. Cont. Shelf Res. 27, 594e612. Walker, N.D., 1996. Satellite assessment of Mississippi River plume variability: causes and predictability. Remote Sens. Environ. 58, 21e35. Walker, N.D., 2001. Tropical storm and hurricane wind effects on water level, salinity, and sediment transport in the river-influenced Atchafalaya-Vermilion Bay system, Louisiana, USA. Estuaries 24, 498e508.

33

Walker, N.D., Hammack, A.B., 2000. Impacts of winter storms on circulation and sediment transport: Atchafalaya-Vermilion Bay region, Louisiana, USA. J. Coast. Res. 16 (4), 996e1010. Ward, G., Armstrong, N., 1992. Ambient Water and Sediment Quality of Galveston Bay: Present Status and His- Torical Trends, Report vol. 1. Galveston Bay National Estuary Program, Webster, Texas, p. 22. Ward Jr., G.H., 1980. Hydrography and Circulation Processes of Gulf Estuaries, Estuarine and Wetland Processes. Springer, pp. 183e215. Wiberg, P.L., Harris, C.K., 1994. Ripple geometry in wave dominated environments and symmetric ripples. J. Geophys. Res. 99 (C4), 775e789. Williams, J., Rose, C., Thorne, P., O'Connor, B., Humphery, J., Hardcastle, P., Moores, S., Cooke, J., Wilson, D., 1999. Field observations and predictions of bed shear stressesand vertical suspended sediment concentration profiles in wavecurrent conditions. Cont. Shelf Res. 19, 507e536. Wu, J., 1980. Wind-stress coefficients over sea surface near neutral conditions e a revisit. J. Phys. Oceanogr. 10, 727e740. Xu, J.P., Wright, L.D., 1995. Tests of bed roughness models using field data from the Middle Atlantic Bight. Cont. Shelf Res. 15, 1409e1434.