Effects of Woody Plants on Dune Erosion and Overwash

Nobuhisa Kobayashi, M.ASCE1; Christine Gralher

2; Kideok Do

3

Abstract: A laboratory experiment consisting of 5 tests was conducted to examine the effects of

woody plants on erosion and overwash of high and low dunes. Scarping occurred on the

foreslope for three high dune tests and no scarping occurred for two low dune tests. A narrow

vegetation zone on a steep backslope of a high dune did not reduce wave overtopping and

overwash in comparison to the corresponding bare dune. A wide vegetation covering the high

dune reduced foreslope scarping, prevented wave overtopping initially, and reduced sand

overwash after the initiation of wave overtopping. A wide vegetation zone covering an entire

low dune reduced dune erosion by retarding wave uprush and reducing wave overtopping and

overwash.

DOI: 10.1061/(ASCE)WW.

CE Database Subject Headings: Vegetation; Wave Overtopping; Dune; Erosion; Overwash;

Sediment Transport.

1 Professor and Director, Center for Applied Coastal Research, Univ. of Delaware, Newark, DE 19716

(corresponding author), E-mail: nk@udel.edu. 2 Master’s student, Dept. of Civil and Environmental Engineering, Univ. of Delaware, Newark, DE 19716.

3 Ph.D. Student, Dept. of Civil and Environmental Engineering, Seoul National Univ., 1 Gwanak-ro, Gwanak-gu,

Seoul, 151-744, Republic of Korea.

2

Introduction

Approximately 3% of the U.S. population presently lives in areas subjected to the 1%

annual chance (100 yr) coastal flood (Crowell et al. 2010). This percentage will increase with

the rise of mean sea level even without any morphological change. The mean sea level rise will

increase the likelihood of shoreline erosion and dune overwash (Gutierrez et al. 2007). One

possible countermeasure to reduce the severity of dune overwash and flooding is the active

planting of vegetation (e.g., Rosati and Stone 2009). The field data by Hayashi et al. (2010)

indicated that woody plants on a natural dune appeared to have prevented overwash during a

severe storm. However, no predictive model exists to assess the effectiveness and limitation of

vegetation as the countermeasure against dune overwash.

Dune erosion and overwash have been investigated mostly by field data obtained after

storms (e.g., Wang et al. 2006). The field data do not reveal the progression of dune erosion and

overwash during a storm. Laboratory experiments were conducted to measure the dune profile

evolution (e.g., Figlus et al. 2011) but no accepted similitude exists for coastal sediment

transport. The effect of vegetation on overwash has been discussed (e.g., Donnelly et al. 2006)

but has never been measured. Grass and woody plants are described for dune stabilization

against windblown sand transport (e.g., U.S. Army Engineer Research and Development Center

2003) but have not been used against wave overtopping and overwash.

Wave attenuation by vegetation has been investigated in relation to prediction of waves

propagating across flooded, vegetated lands and for the possible use of natural or artificial kelp

in promoting beach accretion. Dalrymple et al. (1984) examined wave diffraction due to

localized areas of wave energy dissipation, such as dense stands of kelp, pile clusters, or

submerged trees. Asano et al. (1992), Kobayashi et al. (1993), and Méndez et al. (1999)

3

analyzed the wave-induced flow field in areas of fixed and flexible vegetation. Løvas and

Tørum (2001) conducted a laboratory experiment on wave propagation above submerged kelp in

the surf zone and its effect on dune erosion. The kelp caused significant wave attenuation but

had only a minor effect on the dune erosion. Mangrove forests were reported to have reduced

the damage by the 2004 Indian Ocean Tsunami (e.g., Asano 2008). The effect of coastal

marshes on nearshore waves has been investigated numerically and experimentally (Bender et al.

2008; Augustin et al. 2009) after Hurricane Katrina in 2005. The effect of coastal marshes on

storm surge was discussed by Resio and Westerink (2008). These studies indicate the varied

effects of vegetation on waves and storm surge. However, the effect of vegetation on wave

overtopping and overwash has not been investigated previously.

An experiment was performed in a wave flume to measure the profile evolutions of bare

and vegetated dunes in the presence of wave overtopping and overwash. Cylindrical wooden

dowels were used to represent woody plants that may reduce overtopping flow more than grass.

First, three high dune tests were conducted for a bare dune and two vegetated dunes with narrow

and wide vegetation zones. Second, two low dune tests were conducted for a bare dune and a

dune with a wide vegetation zone. The following sections present the experimental setup and

measurements, the three high dune tests, and the two low dune tests.

Experiment

The experiment was conducted in the wave tank of the University of Delaware which is

30 m long, 2.5 m wide and 1.5 m high. A dividing wall along the length of the middle of the

wave tank was installed to reduce the amount of fine sand, the water level change due to wave

overtopping, and seiching development in the wave tank. Fig. 1 depicts the experimental setup

4

in the 1.15-m wide flume which is similar to the dune overwash experiment by Figlus et al.

(2011). No nearshore bar was formed in this experiment. A piston-type wave maker in 1-m

water depth generates a 400-s burst of irregular wages corresponding to a TMA spectrum. The

wave maker has no capability of absorbing waves reflected from the dune. The spectral

significant wave height and peak period were approximately 19 cm and 2.6 s, respectively. The

sand beach placed on a plywood bottom with a slope of 1/30 consisted of well-sorted fine sand

with a median diameter of 0.18 mm. The placed sand was moistened and compacted before each

test. The measured specific gravity, porosity and fall velocity were 2.6, 0.4 and 2.0 cm/s. Eight

capitance wave gauges were installed for the measurement of the free surface elevation along the

wave tank. Three acoustic Doppler velocimeters (one 2D ADV and two Vectrinos) were used to

measure fluid velocities at an elevation of 1/3 of the local water depth above the bottom. A laser

line scanner mounted on a motorized cart was used to record alongshore transects at 2 cm cross-

shore intervals with an accuracy of ± 1 mm, yielding 3D bathymetry of the entire subaerial (after

lowering the water level) portion of the bed. An array of three submerged ultrasonic transducers

are used to record three cross-shore transects for the submerged portion of the bed.

Water and sand transported over the impermeable vertical wall in Fig. 1 during each

400-s run were collected in a basin. The elevation of the wall crest was 6 cm above the still

water level (SWL). A sand trap made from polyester fabric mesh retained grain diameters

exceeding 0.074 mm and allowed water to pass through. The trapped sand is analyzed to obtain

the water content and dry sand mass. The collection basin included a water recirculation system

consisting of a pump, a flow meter, pipes, and a valve to maintain a constant still water level in

the 2.5-m wide tank. The water volume change in the collection basin was measured using a

wave gauge (WG) 9 in Fig. 1 and a mechanical float to ensure data accuracy. This experimental

5

setup allowed the accurate measurement of the water overtopping rate and sand overwash rate

averaged over the 400-s run.

Table 1 summarizes the five tests conducted in sequence. Each test comprised a number

of runs of the same 400-s bursts of irregular waves impinging on a dune. The initial profile of

the high (H) and bare (B) dune is depicted in Fig. 1 where the crest elevation was 21 cm above

SWL and the foreslope and backslope of the dune were 1/2 and 1/3, respectively. HB test was

terminated after 6 runs when the dune crest was lowered to the elevation of the wall crest. The

high dune was rebuilt for two high dune tests with vegetation. Cylindrical wooden dowels were

used to represent woody plants. The diameter and length of each dowel were 0.9 cm and 30 cm,

respectively. Each dowel was placed vertically with a burial depth of 20 cm. The cross-shore

and alongshore spacings of the dowels were 4 cm and each dowel was in the center of the square

whose area was 16 cm2. The burial depth of each dowel was checked after each run and adjusted

to 20 cm at the beginning of all runs to avoid uprooting. The 10 cm height of the dowel above

the sand surface ensured the emergence of the dowel top above uprushing water. The cross-

shore distance of the vegetation zone from the vertical wall was 40 cm and 80 cm for the narrow

(N) and wide (W) vegetation, respectively. The dowels were aligned in the alongshore direction

so that the laser scanner could measure the alongshore transects at 1 cm cross-shore intervals in

the vegetation zone. Photos depicting the dowel arrangements were presented by Gralher et al.

(2012).

The high and narrowly vegetated dune for HN test was eroded to the level of the wall

crest after 6 runs. The high and widely vegetated dune for HW test was much more resilient and

terminated after 28 runs because of the alongshore variability of the scarped dune profile. Photos

of the scarped profiles were presented by Gralher et al. (2012). The dowels were removed after

6

HW test and the scarped dune profile was smoothed and made uniform alongshore to create the

initial profile of the low (L) and bare dune. The dune for LB test was eroded up to the vertical

wall after 3 runs. The low dune was rebuilt and the dowels were planted in the wide vegetation

zone. LW test was continued for 20 runs to examine the temporal variations of the wave

overtopping and overwash rates after the vertical wall was exposed to wave action.

For each run, the free surface elevation η above SWL was measured at WG1 – WG8

located at the onshore coordinate x = 0.0, 0.25, 0.95, 8.3, 12.9, 15.5, 17.1 and 18.6 m along the

centerline of the flume where the vertical wall was located at x = 19.9 m. The fluid velocities

were measured at x = 12.9, 15.5 and 17.1 m in the vicinity of the flume centerline. The

measured 400-s time series sampled at 20 Hz were reduced by removing the initial 20-s

transition period before the data analysis. The 380-s time series from WG1 – WG 3 were used to

separate incident and reflected waves at the location of WG1. The incident waves are

represented by the spectral significant wave height moH and peak period

pT as well as the

significant wave height sH and period

sT . The reflection coefficient R is defined as the ratio

between the values of moH for the reflected and incident waves. Table 2 lists the average value

of these wave parameters at 0x = for all the runs in each test. The value of R decreased

slightly with the decrease of the foreslope slope during each test.

The measurements for all the runs in the five tests were analyzed to examine the effects

of the vegetation on the dune and beach profile evolution, wave hydrodynamics, and overwash

for the high and low dune tests. Scarping and slumping occurred on the foreslope of the high

dune but no scarping occurred for the low dune. The vegetation effects on the dune profile

evolution may be separated into flow resistance and sand reinforcement attributable to the

exposed and buried parts of the dowels, respectively. The sand reinforcement effect by the

7

buried parts (idealized roots) of the dowels were observed to be negligible in the absence of

scarping.

High Dune Tests

The measured bottom elevations were averaged alongshore to obtain the beach and dune

profile as a function of x at time t with 0t = at the beginning of each test. The vertical

coordinate z is positive upward with 0z = at SWL. Each profile is identified by its run number

starting from run number 0 for the initial profile. The run number is affixed to the test name.

Fig. 2 shows the measured dune profiles for HB, HN and HW tests in the zone of x = 16 – 19.9

m of noticeable profile changes in front of the vertical wall. The sediment budget in this zone is

examined at the end of this section. The profiles seaward of x = 16 m did not change much

probably because a number of preliminary runs were conducted to optimize the experimental

setup for HB, HN and HW tests. Three thick lines are used in Fig. 2 to differentiate the initial,

intermediate, and final profiles in each test. The wave overtopping rate oq started to increase

rapidly after HB3 and HN3, whereas 0oq = up to HW20 and oq remained small until the end of

HW test which was terminated due to alongshore variability. The zone of the vegetation is

indicated in Fig. 2 for HN and HW tests. For HB and HN tests, the dune crest was lowered

initially by the seaward sand transport and foreslope scarping and subsequently by wave

overtopping and overwash over the vertical wall. The narrow vegetation on the steep backslope

did not reduce the backslope erosion caused by the overtopping flow. The wide vegetation

covering the entire dune protected the foreslope against direct wave attack and reduced dune

scarping. The backslope erosion caused by wave overtopping and overwash started after HW20

and progressed slowly because of the small oq during HW21 – HW28. It should be noted that

8

the profiles were not measured after runs 11-13, 15-16, 18-19, 21-22 and 24-25 in HW test

because of the slow profile evolution.

Fig. 3 shows the mean η and standard deviation ησ of the measured free surface

elevation η at the eight wave gauges for all the runs for HB, HN and HW tests. The measured

values of η were negative (wave setdown) at WG1 – WG3 outside the surf zone and WG4 near

the breaker zone. The values of η were positive (wave setup) at WG5 – WG 7 in the inner surf

zone and at WG8 in the swash zone where the averaging for WG8 buried partially in the sand

was performed for the wet duration only. The decrease of η at WG8 during each test was

caused by the decrease of the bottom elevation associated with the berm erosion in Fig. 2. The

standard deviation ησ is proportional to the spectral significant wave height 4moH ησ= . The

measured values of ησ varied little from WG1 to WG4 and decreased in the surf and swash

zones. The values of ησ at WG8 increased during each test because the berm erosion lowered

the bottom elevation. The wet probability wP is defined as the ratio between the wet and total

durations where 1.0wP = at WG1 – WG7 and wP at WG8 was approximately 0.9 initially and

increased to 1.0 after the berm erosion. Fig. 3 indicates that the wave conditions at WG1 – WG7

varied little during each test because the bottom elevation change in the zone of x = 0.0 – 17.1 m

was small.

The mean u and standard deviation uσ of the measured cross-shore velocity u in the

inner surf zone ( x = 12.9 – 17.1 m) did not change much during each test. Table 3 lists the

average value of u and uσ for all the runs in each test. The measured values of u were negative

and the undertow current decreased from x = 12.9 m to x = 15.5 m before its increase at x =

9

17.1 m where some broken waves in the inner surf zone broke again on the steeper bottom slope

(see Fig. 2). The standard deviation uσ varied little between x = 12.9 – 15.5 m and increased at

x =17.1 m probably because of the increased wave breaking on the steeper bottom slope. It is

noted that the measured alongshore and vertical velocities were small in comparison to the cross-

shore velocities in this experiment.

Fig. 4 shows the temporal variations of the measured wave (water) overtopping rate oq ,

sand overwash rate bsq , and ratio /bs oq q for HB, HN, and HW tests. The ratio represents the

volumetric sand concentration in the overtopping flow over the vertical wall. The measured

rates are plotted at time t corresponding to the middle of each run. HB and HN tests were

terminated when oq and

bsq reached the upper limit of approximately 18 cm2/s and 0.5 cm

2/s,

respectively, in the previous experiment by Figlus et al. (2011). The upper limit occurred when

the dune crest was at the same elevation as the vertical wall crest as shown in Fig. 2. The narrow

vegetation on the steep backslope did not reduce oq and

bsq at all. As for HW test, wave

overtopping and overwash did not occur until HW21 ( t = 8,000 – 8,400 s) and qbs/qo = 0 is

plotted for qo = 0 and qbs = 0. The measured rates after HW21 increased slowly and oq = 2.7

cm2/s and

bsq = 0.1 cm2/s for HW28. The wide vegetation was very effective in reducing

oq

and bsq . The vegetation on the foreslope appeared to have retarded wave uprush on the upward

slope. The ratio /bs oq q was almost 0.2 at the beginning of wave overtopping and overwash,

possibly because loose sand particles on the surface were transported at the beginning. The ratio

decreased at a slower rate for HW test with the progression of wave overtopping and overwash.

The vegetation may have generated additional turbulence and increased the sand concentration.

10

The sediment budget in the zone of x = 16 – 19.9 m is examined using the measured

profiles shown in Fig. 2. The volume change ( )cV t per unit alongshore length is obtained by

computing the area change of the measured profile at time t from the initial profile at 0t = and

cV is positive for erosion. The temporal variation of cV is presented in the top panel of Fig. 5 for

HB, HN, and HW tests. The sand volume ( )oV t per unit alongshore length associated with

overwash is obtained by integrating the sand overwash rate bsq from 0t = to time t at the end of

each run where the sand porosity of 0.4 is included in oV for the comparison of

cV and oV . The

overwash volume oV is plotted in the bottom panel of Fig. 5. The value of

oV is zero or positive

and for HW test, 0oV = for t = 0 – 8,000 s. The sand volume transported offshore from the zone

of x = 16 – 19.9 m is the offshore loss LV given by ( )L c o

V V V= − . The sand transported

offshore from x = 16 m was deposited in the zone of x = 0 – 16 m. The accurate measurement

of this deposited sand volume was not possible because of the error of about 1 mm for the

bottom elevation measurement. For example, the vertical elevation difference of 1 mm over the

horizontal distance of 16 m corresponds to 160 cm3/cm. It was more accurate to estimate

LV

using the measured values of cV and

oV along with the conservation of sand volume.

The small negative (deposition) values of ( )c o LV V V= + in Fig. 5 occurred when the

onshore sand transport volume ( )0L

V < at x = 16 m exceeded the small overwash volume oV .

The large positive values of cV and

oV occurred at the same time and the large sand overwash

volume resulted in the large erosion volume in the zone of x = 16 – 19.9 m. The ratio /o cV V was

0.63, 0.79 and 0.17 at the end of HB, HN, and HW tests, respectively. The majority of the dune

erosion for HB and HN tests was caused by sand overwash over the vertical wall. The wide

11

vegetation reduced the wave overtopping and overwash rates but increased the offshore sand

loss. This increased offshore loss may be related to the increased offshore return flow resulting

from the decreased wave overtopping rate. The measured undertow current at x = 17.1 m was

somewhat larger for HW test as listed in Table 3. Nevertheless, the sediment dynamics on the

vegetated dune is not clear for lack of depth and velocity measurements in the vegetation zone.

Low Dune Tests

Fig. 6 shows the measured dune profiles for LB and LW tests where no scarping occurred

in these tests. Thick lines are used for LB0, LB2, and LB3 as well as LW0, LW10, and LW20 in

order to indicate the sequence of the profile evolution. The low dune crest in LB test retreated to

the vertical wall after 3 runs. The vegetated dune in LW test was eroded slowly and exposed to

20 runs where the measured profiles after LW1, LW14-15, and LW18 were unreliable and

excluded in Fig. 6. The vegetation zone for HW test was located landward of the still water

shoreline in Fig. 2. The vegetation zone for LW test became exposed to broken waves in the

inner surf zone as the still water shoreline moved landward during LW test. A small hump

associated with local scour was formed at the seaward edge of the vegetation zone. The vertical

wall was exposed to direct wave action toward the end of LW test.

Fig. 7 shows the mean η and standard deviation ησ of the free surface elevation at WG1

– WG8 for all the runs for LB and LW tests. The wet probability at WG8 was unity because its

location x = 18.6 m was seaward of the still water shoreline from the beginning of LB and LW

tests. The cross-shore variations of η and ησ from WG1 to WG7 were very similar to those in

Fig. 3 because the incident waves and the bottom profile in the zone of x = 0 – 17.1 m were

almost the same for the five tests. The decrease of η and the increase of ησ at WG8 with the

12

progression of LB and LW tests were smaller in Fig. 7 because the vertical erosion at WG8 was

smaller for these tests. The mean u and standard deviation uσ of the measured horizontal

velocity listed in Table 3 were also similar except for u and uσ at x = 17.1 m for HW test with

28 runs of no or little wave overtopping.

Fig. 8 shows the temporal variations of oq ,

bsq and /bs oq q for LB and LW tests. The

wide vegetation reduced oq and

bsq by a factor of about 3 and 2, respectively. The wave

overtopping rate oq remained almost constant throughout LW test. The wide vegetation was

effective in reducing oq even after its seaward segment was situated in the inner surf zone. In

reality, woody plants could be destroyed by the wave force or uprooted due to erosion. These

factors were not considered in the present experiment. The sand overwash rate bsq decreased

with the progression of erosion in front of the vertical wall possibly because of the reduced

availability of sand in the overtopping flow. The ratio /bs oq q was larger for LW test probably

because of the increased turbulence in the vegetation zone. The temporal trends in Fig. 8 follow

those in Fig. 4 because LB and LW tests were essentially the continuation of HB and HW tests,

respectively.

Fig. 9 shows the temporal variations of cV and

oV for LB and LW tests. The volume

change cV per unit alongshore length in the zone of x = 16 – 19.9 m increased almost linearly

with time. The rate of increase of this erosion volume was larger for LB test because of the

larger overwash rate bsq in Fig. 8. The cumulative overwash volume

oV attributable to bsq

increased with time but the rate of increase of oV for LW test decreased with time because of the

temporal decrease of bsq for LW test. Erosion in front of the vertical wall was caused mostly by

13

wave overtopping and overwash. The ratios of /o cV V at the end of LB and LW tests were 0.82

and 0.73, respectively. In short, the wide vegetation reduced the wave overtopping and

overwash rates and the volumetric erosion rate in front of the vertical wall. The results of the

low dune tests were influenced by the vertical wall. Additional tests are required for low dunes

without scarping in the absence of the vertical wall.

Conclusions

A laboratory experiment consisting of 5 tests was conducted to examine the effects of

woody plants on dune erosion and overwash. Three tests for a high dune examined the

vegetation effects on erosion and scarping on the foreslope as well as wave overtopping and

overwash on the backslope. The narrow vegetation on the steep backslope did not reduce the

measured wave overtopping and overwash rates. The wide vegetation covering the foreslope and

backslope reduced scarping, prevented wave overtopping initially, and reduced the overtopping

and overwash rates after the initiation of wave overtopping. However, the reduced wave

overtopping resulted in the increase of offshore sand transport from the eroded dune. Two tests

for a low dune examined the vegetation effects in the absence of foreslope scarping. The wide

vegetation covering the entire low dune reduced the dune erosion by decreasing the wave

overtopping and overwash rates. The vegetation retarded wave uprush on the upward slope in

the swash and inner surf zones but increased the sand mobilization in the vegetation zone.

The experimental results may be useful in designing a vegetation zone to reduce dune

overwash. However, additional tests are required because the present experiment was limited to

the specific diameter, height, spacing, alignment, and burial depth of rigid wooden dowels

without uprooting. A large-scale experiment will also be necessary to quantify scale effects in

14

the present small-scale experiment. Depth and velocity measurements in the vegetation zone

will be required to elucidate the interaction processes among waves, vegetation and sand.

Nevertheless, the present experimental results will be useful for the initial development of a

numerical model for vegetated dune overwash.

Acknowledgments

This study was supported partly by the EU THESEUS Project and the U.S. Army Corps

of Engineers Coastal and Hydraulics Laboratory under Contract No. W912HZ-11-P-0173. The

third writer was supported by the Basic Research Program (400-20100155) of the National

Research Foundation, Ministry of Education, Science and Technology of Republic of Korea.

References

Asano, T. (2008). “Evaluation of tsunami attenuation performance of coastal forests based on

tree morphology.” Proc. 31st Coastal Engineering Conf., World Scientific, Singapore,

1446-1457.

Asano, T., Deguchi, H., and Kobayashi, N. (1992). “Interaction between water waves and

vegetation.” Proc. 23rd Coastal Engineering Conf., ASCE, Reston, Va., 2710-2723.

Augustin, L.N., Irish, J.L., and Lynett, P. (2009). “Laboratory and numerical studies of wave

damping by emergent and near-emergent wetland vegetation.” Coastal Eng., 56, 332-340.

Bender, C.J., Atkinson, J., Cialone, M.A., Grzegorzewski, A., and Smith, J.M. (2008).

“Numerical simulation of the effects of Louisiana coastal marshes on nearshore waves.”

Proc. 31st Coastal Engineering Conf., World Scientific, Singapore, 1012-1023.

15

Crowell, M., Coulton, K., Johnson, C., Westcott, J., Bellomo, D., Edelman, C., and Hirsch, E.

(2010). “An estimate of the U.S. population living in 100-year coastal flood hazard areas.”

J. Coastal Res., 26(2), 201-211.

Dalrymple, R.A., Kirby, J.T., and Hwang, P.A. (1984). “Wave diffraction due to areas of energy

dissipation.” J. Waterway, Port, Coastal, Ocean Eng., 110(1), 67-79.

Donnelly, C., Kraus, N., and Larson, M. (2006). “State of knowledge on measurement and

modeling of coastal overwash.” J. Coastal Res., 22(4), 965-991.

Figlus, J., Kobayashi, N., Gralher, C. and Iranzo, V. (2011). “Wave overtopping and overwash

of dunes.” J. Waterway, Port, Coastal, Ocean Eng., 137(1), 26-33.

Gralher, C., Kobayashi, N., and Do, K. (2012). “Wave overwash of vegetated dunes.” Proc.

33rd Coastal Engineering Conf., Sediments, 1-7, http://journals.tdl.org/ICCE.

Gutierrez, B.T., Williams, S.J., and Thieler, E.R. (2007). “Potential for shoreline changes due to

sea-level rise along the U.S. Mid-Atlantic Region.” Open-File Report 2007-1278, U.S.

Geological Survey, http://pubs.usgs.gov/of/2007/1278.

Hayashi, K., Hashimoto, K., Yagisawa, K., and Kobayashi, N. (2010). “Beach morphologies at

Notsukezaki sand spit, Japan.” Proc. 32nd Coastal Engineering Conf., Sediments 48, 1-12,

http://journals.tdl.org/ICCE.

Kobayashi, N., Raichle, A., and Asano, T. (1993). “Wave attenuation by vegetation.” J.

Waterway, Port, Coastal, Ocean Eng., 119(1), 30-48.

Løvas, S.M., and Tørum, A. (2001). “Effect of the kelp Laminaria hyperborean upon sand

dune erosion and water particle velocities.” Coastal Eng., 44, 37-63.

Méndez, F., and Losada, I.J., and Losada, M.A. (1999). “Hydrodynamics induced by wind

waves in a vegetation field.” J. Geophys. Res., 104(C8), 18,383-18,396.

16

Resio, D.T., and Westerink, J.J. (2008). “Modeling the physics of storm surges.” Physics

Today, 61(9), 33-38.

Rosati, J.D., and Stone, G.W. (2009). “Geomorphologic evolution of barrier islands along the

northern U.S. Gulf of Mexico and implications for engineering design in barrier restoration.”

J. Coastal Res., 25(1), 8-22.

U.S. Army Engineer Research and Development Center (2003). Coastal Engineering Manual.

Coastal and Hydraulics Lab., Vicksburg, Miss.

Wang, P., Kirby, J.H., Haber, J.D., Horwitz, M.H., Knorr, P.O., and Krock, J.R. (2006).

“Morphological and sedimentological impacts of Hurricane Ivan and immediate poststorm

beach recovery along the northwestern Florida barrier-island coasts.” J. Coastal Res., 22(6),

1382-1402.

17

List of Figures

Fig. 1. Experimental setup for high bare dune test.

Fig. 2. Dune profile evolution for HB (top), HN (middle), and HW (bottom) tests.

Fig. 3. Cross-shore variations of mean (top) and standard deviation (middle) of free surface

elevation η and wet probability wP (bottom) for HB (left), HN (middle), and HW

(right) tests.

Fig 4. Temporal variations of wave overtopping rate oq (top), sand overwash rate

bsq

(middle), and ratio /bs oq q (bottom) for HB, HN, and HW tests.

Fig. 5. Temporal variations of cumulative sand volume change cV (top) and overwash volume

oV (bottom) per unit width for HB, HN, and HW tests.

Fig. 6. Dune profile evolution for LB (top) and LW (bottom) tests.

Fig. 7. Cross-shore variations of η (top) and ησ (middle) for LB (left), and LW (right) tests.

Fig. 8. Temporal variations of oq (top),

bsq (middle) and ratio /bs oq q (bottom) for LB and

LW tests.

Fig. 9. Temporal variations of cV (top) and

oV (bottom) for LB and LW tests.

18

Table 1. Summary of Five Dune Tests

Test Dune Vegetation Number of Runs Duration (s)

HB

HN

HW

LB

LW

High

High

High

Low

Low

Bare

Narrow

Wide

Bare

Wide

6

6

28

3

20

2,400

2,400

11,200

1,200

8,000

19

Table 2. Incident Wave Characteristics at 0x = for Five Tests

Test moH

(cm)

sH

(cm)

pT

(s)

sT

(s)

R

HB

HN

HW

LB

LW

18.6

18.7

18.5

18.3

18.8

18.3

18.4

18.3

18.0

18.6

2.65

2.57

2.57

2.57

2.57

2.27

2.29

2.29

2.30

2.30

0.16

0.16

0.15

0.11

0.11

20

Table 3. Mean and Standard Deviation of Horizontal Velocity u

Test

x = 12.9 (m) x = 15.5 (m) x = 17.1 (m)

u

(cm/s)

uσ

(cm/s)

u

(cm/s)

uσ

(cm/s)

u

(cm/s)

uσ

(cm/s)

HB

HN

HW

LB

LW

- 5.0

- 5.2

- 5.2

- 4.7

- 5.4

17.4

18.2

17.5

16.7

16.9

- 3.3

- 4.3

- 4.0

- 3.4

- 3.5

17.0

18.1

17.8

16.8

17.0

- 4.1

- 4.5

- 5.2

- 4.0

- 4.4

20.1

20.4

22.0

20.2

19.9

SWL

collectionbasin

sandtrap

returnflow

1 m

ADVimpermeablevertical wall

motorizedcart

laser linescanner

-3 0 5 10 15 20 m

Vectrino

−0.2

−0.1

0

0.1

0.2 HB0

HB3

HB6

−0.2

−0.1

0

0.1

0.2

Ele

vation

(m) HN0

HN3

HN6

16 17 18 19 20−0.2

−0.1

0

0.1

0.2

x (m)

HW0

HW20

HW28

SWL

SWL

SWL

vegetation

vegetation

−1

0

1

2

3

4

5

η(c

m)

1

2

3

4

5

ση

(cm

)

0 5 10 15 200.8

1

Pw

0 5 10 15 20x (m)

0 5 10 15 20

HB HN HW

0

5

10

15

20

qo

(cm

2

s) HB

HNHW

0

0.2

0.4

0.6

qbs

(cm

2

s)

200 2400 4600 6800 9000 112000

0.05

0.1

0.15

0.2

qbs/qo

t (s)

−400

0

400

800

1200

Vc

(cm

3

cm

)

HBHNHW

400 3100 5800 8500 112000

200

400

600

800

1000

Vo

(cm

3

cm

)

t (s)

−0.2

−0.1

0

0.1

Ele

vation

(m)

LB0

LB2

LB3

16 17 18 19 20

−0.2

−0.1

0

0.1

x (m)

LW0

LW10

LW20

SWL

SWL

vegetation

−1

0

1

η(c

m)

0 5 10 15 202

3

4

5

ση

(cm

)

x (m)0 5 10 15 20

LB LW

0

4

8

12

16

20

qo

(cm

2

s)

LBLW

0

0.1

0.2

0.3

0.4

qbs

(cm

2

s)

200 1800 3400 5000 6600 80000

0.01

0.02

0.03

0.04

qbs/qo

t (s)

0

350

700

1050

1400

Vc

(cm

3

cm

)

LBLW

400 2000 4000 6000 80000

200

400

600

800

1000

Vo

(cm

3

cm

)

t (s)