© PLOSI o
OPEN 3 ACCESS Freely available online
-
Low-Canopy Seagrass Beds Still Provide Im portant
Coastal Protection Services
M arjolijn J. A. Christianen1", Jim van Beizen2, Peter M . J. Herm an2, M arieke M . van Katw ijk1,
Leon P. M . Lamers3, Peter J. M . van Leent1, Tjeerd J. Bouma 2
1 D e p a rtm e n t o f Environm ental Science, Faculty o f Science, In stitu te fo r W ater an d W etlan d R esearch, R ad b o u d U niversity N ijm egen, N ijm egen, T h e N eth erlan d s, 2 Spatial
Ecology D ep artm en t, Royal N etherlands In stitu te fo r Sea Research, Y erseke, T h e N eth erlan d s, 3 D e p a rtm e n t o f A quatic Ecology a n d E nvironm ental Biology, Faculty o f
Science, In stitu te fo r W ater an d W etland R esearch, R adboud U niversity N ijm egen, N ijm egen, T h e N eth e rlan d s
Abstract
One of th e m ost frequently q u o te d ecosystem services of seagrass m eadow s is th eir value for coastal protection. Many
studies em phasize th e role o f ab o v e-ground sh o o ts In a tten u atin g waves, enhancing sedim entation and preventing erosion.
This raises th e question If short-leaved, low density (grazed) seagrass m eadow s with m ost o f their biom ass In below ground
tissues can also stabilize sedim ents. We exam ined this by com bining m anipulative field experim ents and wave
m easurem ents along a typical tropical reef flat w here green turtles Intensively graze upon th e seagrass canopy. We
experim entally m anipulated wave energy and grazing Intensity along a tran sect perpendicular to th e beach, and com pared
sed im ent bed level ch an g e b etw een v eg e ta ted and experim entally created bare plots at th ree distances from th e beach.
Our experim ents show ed th a t /') even th e short-leaved, low -blom ass and heavily-grazed seagrass veg etatio n reduced waveinduced sedim en t erosion up to threefold, and ii) th a t erosion was a function of location along th e v e g etated reef flat.
W here o th er studies stress th e Im portance of th e seagrass canopy for shoreline protection, o u r study on open, low -blom ass
and heavily grazed seagrass beds strongly su g g ests th a t below ground biom ass also has a m ajor effect on th e
Immobilization of sedim ent. These results Imply th at, com pared to shallow u n v eg etated nearshore reef flats, th e presence of
a short, low-blom ass seagrass m ead o w m aintains a higher bed level, a tten u atin g w aves before reaching th e beach and
hence lowering beach erosion rates. We p ropose th a t th e sole use o f ab o v eg ro u n d biom ass as a proxy for valuing coastal
protection services should be reconsidered.
C ita tio n : C hristianen MJA, v an Beizen J, H erm an PMJ, v an Katwijk MM, Lam ers LPM, e t al. (2013) Low -C anopy S eag rass Beds Still P rovide Im p o rta n t C oastal
P ro tec tio n Services. PLoS ONE 8(5): e62413. doi:10.1371/journal.p o n e.0 0 6 2 4 1 3
E d ito r: Richard K.F. U nsw orth, S w ansea University, U nited K ingdom
R e c e iv e d D ecem b er 28, 2012; A c c e p te d M arch 21, 2013; P u b lis h e d May 28, 2013
C o p y r ig h t: © 2013 C hristianen e t al. This is an o p en -ac cess article d istrib u te d u n d e r th e te rm s o f th e C reative C o m m o n s A ttrib u tio n License, w h ich perm its
un restricted use, distrib u tio n , a n d re p ro d u c tio n in an y m ed iu m , p ro v id ed th e original a u th o r a n d so u rc e a re cred ited .
F u n d in g : R esearch by MJAC is fu n d e d by th e N eth erlan d s O rgan izatio n fo r Scientific R esearch - S cience fo r G lobal D ev elo p m en t (NWO-WOTRO), g r a n t W 84-645
(a p p o in te d to MJAC). T he w ork o f JvB a n d TJB is s u p p o rte d by th e THESEUS p ro je c t o n in n o v ativ e te c h n o lo g ie s fo r safer E uropean co a sts in a ch a n g in g clim ate,
w hich is fu n d e d by th e E uropean U nion w ithin FP7-THEME 6 - E nvironm ent, including clim ate (co n tract no. 244104). T he fu n d ers h ad n o role in s tu d y d esig n , d a ta
collection a n d analysis, decision to publish, o r p rep ara tio n o f th e m an u scrip t.
C o m p e tin g In te re s ts : T he au th o rs h av e d ec lared th a t n o c o m p e tin g in terests exist.
* E-mail: m arjolijn.christianen@ gm ail.com
Introduction
com plexity o f such short vegetation is d eg rad ed further, e.g. due to
a high grazing intensity, it becom es un clear to w hich extent they
c an still con trib u te to coastal protection.
A lthough sedim ent stabilization is often acknow ledged as an
im p o rta n t ecosystem service o f seagrasses [19,20] a n d anecdotic
evidence points at increased erosion after a seagrass m eadow has
b e en lost (e.g. [13,21]), experim ental evidence for the exact
m echanism s involved in sedim ent stabilization rem ains scarce.
Seagrass m eadow s have b e en show n to a tte n u ate hydrodynam ic
energy from currents [22,23] a n d waves [12,24,25] a n d thereby
tra p suspended sedim ent a n d cause sedim ent accretion [14,26—
30]. H ow ever, w ith respect to sedim ent stabilization, m ost studies
only refer to the effect o f the canopy in the red u ctio n o f the
h ydrodynam ic forces th a t m ay reach the sedim ent a n d im pose a
b e d shear stress (xb) to the sedim ent [31]. It has b e en suggested
th a t below ground biom ass o f rhizom es a n d roots can stabilize
sedim ents by altering the erodability as the critical b e d shear stress
(x c rit) is increased [31]. H ow ever, the relative im p o rtan ce o f this
m echanism is generally h a rd to study w ithout d isturbing the
seagrass m eadow a n d is, therefore, generally not addressed w hen
studying the role o f these m acrophytes for coastal protection.
Biological structures located in coastal sub- a n d intertidal
ecosystem s can atte n u ate waves a n d as a result directly contribute
to coastal p ro tec tio n [1—3]. B oth re e f form ing tax a such as corals
[4], m ussels [5] a n d oysters [6] a n d m acroalgae a n d m acrophytes
such as kelp [7], seagrass [8], m angrove [9] a n d salt-m arsh
vegetation [10—12], are well know n for th eir capacity to a tten u ate
waves (see [1] for a review). As a consequence o f the red u ctio n o f
h ydrodynam ic energy, m acro p h y te vegetation typically a cc u m u ­
lates sedim ent causing the w ater above the fore- o r nearsh o re to
becom e shallow er [14,15] (but: see [16,17]). Such sedim ent
accretion also contributes to coastal p rotection, because wave
a tte n u atio n increases w ith decreasing relative w ater d e p th [18],
T h e bath y m etric w ave-atten u atin g effect o f vegetation-induced
sedim ent accretion becom es especially im p o rta n t for those
vegetation types th a t have a relatively small direct wave
a tte n u atin g effect via th eir abov eg ro u n d biom ass. T his applies
for exam ple to m eadow s o f relatively short a n d highly flexible
seagrass plants, w hich have lim ited w ave-attenuating capacity by
th eir canopy c o m p a red to stiffer vegetation [12]. If the structural
PLOS ONE I www.plosone.org
1
May 2013 | Volume 8 | Issue 5 | e62413
Grazed Seagrass Meadows Still Protect Shorelines
In the tropics, seagrass m eadow s typically occur on shallow re ef
flats in subtidal n earsh o re areas. In general, seagrass grow th is
often controlled by tem p eratu re, light availability a n d freshw ater
in p u t b u t also by physical disturbance from waves a n d associated
sedim ent m ovem ent [32,33]. T o p -d o w n effects can also drive
seagrass grow th by the foraging o f large herbivores. R ecen t studies
in B erm uda, In d ia a n d Indonesia re p o rte d intense grazing o f
green turtles o n seagrasses [34-36], w ith harvesting rates up to
100% o f the daily le a f p ro d u c tio n [36], As a consequence, these
heavily grazed m eadow s c an have a n extrem ely sparse cover w ith
a low aboveg ro u n d biom ass (± 10 g D W m _ ~) a n d short (< 5 cm)
canopy, w hile m ain tain in g a high below ground biom ass (± 5 0 g
D W m ~) [36], T his is in strong contrast to u n g razed m eadow s,
w here the aboveg ro u n d biom ass can b e at least 10 tim es higher
(e.g. biom ass 118 g D W m ", canopy height ± 2 5 cm , described in
[37]. Such grazing-induced alteration o f the can o p y structure
m akes these m eadow s interesting m odels to study the c o n tribution
o f b elow ground tissues to coastal defense.
In this study, we therefore question i) if intensively grazed
seagrass m eadow s w ith a very low -biom ass can o p y c ontribute to
coastal pro tectio n b y stabilizing the sedim ent against w ave induced
erosion, ii) if the im p o rtan ce o f the sedim ent stabilizing effect o f
seagrass changes along a cross-shore profile a n d iii) if the sedim ent
stabilization by seagrass m eadow s depends o n the height o f the
canopy. T o answ er these questions, w e experim entally m an ip u ­
lated seagrass above- a n d below -ground cover, w ave forcing a n d
grazing intensity along a transect b etw een the re e f a n d the beach.
Wave m e a s u r e m e n ts
W e m easured h ydrodynam ic forcing along the re e f flat as a
result o f waves at four stations along the d e p th profile given in
Fig. 1, using self-logging pressure sensors (W ave gauge: O S SI-10003C , O c e a n Sensor System s, C o ral Springs, ETSA). T h e
instrum ents w ere placed a t a height o f 0.1 m above the bed.
T h re e pressure sensors w ere p laced in the seagrass m eadow o n the
re e f flat, a n d one sensor m easu red the waves com ing in from the
ocean over the re e f crest a t increasing distance from the shore
(stations ‘A ’ 45 m , ‘B’ 262 m , ‘Cl’ 513 m ,‘C o ra l’ 712 m from the
shore; see Fig. lb). W ave heights w ere m easu red u n d e r a range o f
offshore w ave conditions a n d tidal elevations du rin g the w hole
e xperim ent. A total o f 3140 recording bursts w ere collected a t a
sam pling ra te o f 10 H z for 4 m inutes, every 20-m inutes.
R ecordings com prise a total o f 209 hours o f w ave m easurem ents
(over a 44 day period, du rin g ra in y season). D u rin g the
d eploym ent o f the w ave gauges w e caught a storm event (January
27, 2012, bursts 2879 to 3074), w ith peak w ind speeds reach in g
19 m s 1 from the n o rth - northw est (± 335"). W e calculated wave
a tte n u atio n values relative to the w aves com ing in a t the reef
station (Coral) for each station a t the vegetated re e f flat.
T h e o b tain e d high frequency w ave records w ere processed
according to the follow ing sequence: (1) pressure readings w ere
c onverted to w ater level fluctuations (f;), (2) erroneous spikes, shifts
a n d c o rru p te d bursts w ere rem oved from the d ata, (3) lowfrequency tidal com ponents w ere rem oved from each b u rst by
d etre n d in g the w ater level fluctuations using a polynom ial fit (4)
from the d e tre n d ed d a ta significant w ave heights (77,) (cf. e.g.
[10,39]) w ere calculated (c.f. [40]):
Methods
Field site
T h e study was c o n d u cted o n a subtidal seagrass m eadow th at
covers the fringing re e f flat o f D e raw an Island (Fig. la), Indonesia
(2"17’19’N , 118"14’53’E; see [36] for a m ap a n d m o re details).
T h e seagrass m eadow s are d o m in a ted b y Halodule uninervis
(E hrenberg, A scherson) grow ing on c arb o n a te sedim ent. T h e
c arb o n a te substrate h a d a m ed ian grain size o f 591 ± 3 0 pm (d5Q,
m ea n ± SE, M alvern L aser Particle Sizer) a n d did n o t differ
significantly b etw een stations. T h e can o p y was o f low structural
com plexity as a result o f intensive grazing by green sea turtles
(Chelonia mydas, 20.6 individuals ± 2 .2 h a *, [38]). T h e hair-like
leaves w ere short (< 5 cm), n a rro w (< 1 nini) a n d th in (< 0 .2 nini).
Shoot density was 3 3 3 5 ± 2 2 4 shoots m ~ a n d shoots only h a d
I.8 ± 0 .1 leaves p e r shoot [36], A boveground biom ass was
I I .4 ± 0 .7 g D W m ~, a n d below ground biom ass 5 2 .0 ± 4 .5 g
D W m _ ~. D u rin g the ex p erim ent (D ecem ber 2011 -F e b ru a ry
2012) spring tidal range was 2.9 m.
Hs= 4
S tf( 1)
in w hich n is the n u m b e r o f w ater level records in each burst
(h = 2400). In addition, we c o rrected the calculated significant
w ave height for the a tte n u atio n o f the wave pressure field w ith
d e p th a n d w ave p e rio d [39]. F ro m the d e tre n d ed d ata, p eak wave
periods (77) w ere com p u ted based on zero-upcrossings [41],
^
H s(StationCoral) — H s(Stationx)
H s ( StationC oral )
^QQ0/
(2)
Bottom sh ear stress calculation (xb)
Because the influence o f waves o n the sedim ent b e d strongly
depends o n w ater d epth, we calculated the w ave-related bo tto m
shear stress (ty) over tim e (c.f. [42,43]):
Survey of reef flat d e p t h profile
T6 = 0.5p/2,5Í7,„
W e m ap p e d a cross-shore d e p th profile d u rin g slack low tide
from the b eac h starting at the low w ater line, over the re e f flat, to
the coral reef. W ater d e p th along the profile was m easured by
dragging a pressure logger (Sensus ultra, R eefnet Inc., O n ta rio ,
C anada) over the seabed at a fixed speed a n d sim ultaneous logging
o f tim e a n d position using a h a n d held G P S (G P SM A P 60CSx,
G arm in , O lathe, USA). W e averaged d e p th readings (obtained
w ith a frequency o f 1 Hz), using a sliding w indow over a
60 second interval, to reduce noise as a result o f w ater level
fluctuations caused by waves a n d m ethodological errors. Finally,
the cross-shore d e p th profile is recalculated, based o n the average
b u rst re ad in g b y the w ave gauges (see below ‘W ave m easure­
m en ts’), to get the average w ater d e p th over tim e.
PLOS ONE I www.plosone.org
(3)
w here p is the w ater density, J 2 .5 is the grain roughness friction
factor calculated as 2.5d50. T h e w ave height related orbital
velocity at the b e d (Uzt>) was estim ated using [44]:
u*
= §ft \/gh
(4 )
in w hich g is the gravitational acceleration, a n d h is the m ean
w ater depth.
2
May 2013 | Volume 8 | Issue 5 | e62413
Grazed Seagrass Meadows Still Protect Shorelines
1*3
p
•
1
■ jfi|y Q
B
0
Sand
S ea g ra ss m eadow
Coral
A: 0.87 m
CL
a>
■a
*a)
->
B: 1.26 m
2
C: 1.48 m
Coral:
1.65 m
re
5
c
re
a)
E
3
200
400
600
800
4
Cross-shore distance (m)
Figure 1. Location and d ep th -p ro file o f th e experim ental site. (A) Aerial p h o to of th e field site show ing th e locations of th e stations, th e
seagrass b ed on th e reef flat in th e subtidal nearshore area (light blue), and th e coral drop off (transition to dark blue). See [36] for a m ore elab o rate
m ap. W aves are com ing predom inantly from th e north (right). (B) D epth profile at increasing distance from th e beach. Location of statio n s are
indicated including their m ean w ater d ep th s.
doi:10.1371/journal.pone.0062413.g001
o f fishing n e t a tta ch e d to the tops o f four steel poles th a t w ere
c onnected to ropes [36], a n d w ere inspected a n d cleaned trice a
week. W ave atte n u atio n by exclosures was m inim al as w eight loss
o f plaster sticks placed in a n d outside cages exposed to waves did
n o t differ significantly. T h e seagrass canopy height was d e te r­
m in ed by m easuring lengths o f 28 shoots from cores ( 0 23 cm) in
grazed {n = 35) a n d u n g razed seagrass plots {n = 15).
T h e experim ental plots w ere selected a t a location w ith
h om ogeneous seagrass substrate, w ith m in im u m distances o f
15 m betw een them . T h e plots o f each station w ere located in a
zone w ith m inim al differences in w a ter d ep th (20 cm) a n d w ere
placed at a line parallel to the shore. T rea tm e n ts w ere ran d o m ly
assigned to the plots.
Experiments to te s t a n d clarify s e d im e n t stabilization by
seag rass
T o test effects o f seagrass presence on sedim ent stabilization, we
c o m p a red the changes in sedim ent level inside b are sedim ent gaps
{i.e., 6 0 x 3 0 cm) to those inside a grazed seagrass m eadow at T 0
(n = 5). T hese m easurem ents w ere re p ea te d at 3 stations, station A,
B a n d C (for description see section “wave m easu rem en ts”) to test
for possible effects o f different hydrodynam ic forcing along the re ef
flat. G aps w ere created a t day 1 o f the ex perim ent by cutting the
roots a n d rhizom es a ro u n d a fram e (6 0 x 3 0 cm) a n d rem oving all
p lan t biom ass w ithin this fram e. T h e size o f the gaps was scaled to
p lan t size a n d to dim ensions o f turtle gaps.
T o test if waves w ere driving the erosion, we m an ip u lated wave
action entering the plots. W ave red u ctio n was achieved by
constructing bunkers o f sandbags (50 kg) w hich w ere piled up
(W :H; 5:3 bags; ± 3 :0 .7 5 m) in a sem i-circular shape to pro tect
plots th a t w ere situated 30 cm b e h in d the b u n k er (Fig. 2c). W e
c o m p a red plots w ith a n d w ithout this w ave red u ctio n trea tm e n t,
by m easuring 5 replicate plots a t th ree stations [n = 15, in total).
T o study the effect o f canopy leaf surface a rea on sedim ent
stabilization we c o m p a red sedim ent b ed level in grazed plots w ith
u n g ra ze d plots th a t w ere p ro tec te d from turtle grazing for
2 m onths (Fig. 3a). M easurem ents for b o th treatm en ts were
rep licated 5 tim es for each station. Plots com prised seagrass strips
(6 0 x 1 5 cm) b o rd e re d by 2 b are sedim ent gaps (6 0 x 3 0 cm). W e
used exclosures (1 .2 x i . 2 x 0 .3 m , 5 cm m esh, Fig. 3a) th a t were
designed to m axim ize light passage a n d m inim ize w ave a tte n u a ­
tion while excluding grazing o f green turtles. Exclosures consisted
PLOS ONE I www.plosone.org
Evaluation of se d im e n t c h a n g e
Q uan titativ e m easurem ents o f changes in b e d level w ere
o b tain ed using a sedim ent elevation b a r m eth o d (SEB, e.g.
[45,46]) a t the start a n d the en d o f the experim ent. A long m etal
pin (150 cm) was inserted into the sedim ent as a reference a t the
start o f the experim ent. A horizo n tal b a r o f 150 cm , a tta ch e d to a
second vertical p in was p laced on top o f the vertical reference pin
at each m easu rem en t until the horizo n tal b a r to u ch e d the
reference pin a n d was level. T h e distance betw een the horizontal
b a r a n d the b e d surface was m easu red at 9 points, at a diagonal
line over each experim ental plot, du rin g each m easurem ent. T h e
relative erosion du rin g the ex perim ent was d e term in ed as the
difference betw een T 0 a n d T end values. T his m eth o d was
estim ated to have an accuracy o f 5 m m .
3
May 2013 | Volume 8 | Issue 5 | e62413
Grazed Seagrass Meadows Still Protect Shorelines
Station A
? *
Station A
0
c
STORM
0)
E
B
a 4 W K A FTER
B
STORM
a> — -
a
2« E
***
0) O
> I-
ab
-5
**
CL
« ■CL) -t-1
2: t/>
I
Q . </5
re (/)
o) «
C
05
§
%-10
« .s
oZ
o S
c .2
A
cu
E
(D
(A
-15 -
•W av es
-20 J
O R educed w av es by bunker
Figure 2. The effect o f seagrass presence on sedim ent stabilization. S edim ent levels in u n v e g e ta te d g ap s co m p ared to levels in th e seagrass
m ead o w at T0 for tw o treatm en ts: g ap s exposed to w aves (black circles) or exposed to w aves reduced by w ave bunkers (white circles). Seagrass
stabilizes sed im en t b o th (A) directly after a storm and (B) 4 w eeks after a storm . The inlay show s th e se tu p of a b unker to reduce w ave en erg y to
seagrass and u n v eg etated g ap s behind (left of) th e bunkers. Significant differences b etw een stations are indicated by different letters, and b etw een
w ave ex p o sed and w ave-reduced plots by stars.
doi:10.1371/journal.pone.0062413.g002
D u rin g the experim ents the sedim ent erosion in the gaps was
also scored visually in a sem i-quantitative w ay (unchanged: ‘ —
m inim al erosion: ‘± ’, m ed iu m erosion: ‘+ ’, strong erosion '++').
T hese estim ates w ere p erfo rm ed every 3r day du rin g m ain te ­
nance checks o f all experim ental plots, a n d d a ta w ere converted to
sedim ent erosion rates using a conversion factor th a t we derived
from plots w ith b o th quantitative a n d sem i-quantitative m easu re­
m ents for the sam e day.
Evaluation of t h e w av e red uction tr e a t m e n t
T o evaluate the w ave red u cin g effect o f the sandbag bunkers,
w ithout having m o re wave loggers available, we c o m p a red w eight
loss o f plaster sticks deployed inside a n d outside a bunker, at 3
locations along the re e f flat. R elative w eight loss by dissolution o f
the plaster is considered a proxy for hydrodynam ic forcing a n d
integrates effects from tidal currents a n d waves [47,48], Sticks
w ere placed at seagrass canopy level at the seagrass - gap b o rd e r
{n —5 for each seagrass station) on a day w ith a large tidal
difference, w ith sticks staying subm erged continuously. Plaster
sticks w ere m olded using 20 m l o f m odel p laster atta ch e d to the
S tation A
B 0
B
C
r
1= T3
w
S3
2
05
re
ai
N
5
»
q
=
a) c
in —
o ü? a>
I s
in
c >
05 + 3
E JS
‘■5 £
05
in
2
Figure 3. Effect o f canopy length on sedim ent stabilization. (A) Turtle exclosure. (B) Difference In sed im en t bed level b etw een grazed and
u ngrazed seag rass strips for th e th re e stations (A, B, C) after 2 m o n th s pro tectio n by th e tu rtle exclosure. The difference In leaf length of th e can o p y
In tu rtle exclosures w as a factor 2.6 longer (117.8±16.6 mm) th an In grazed m eadow s (45.8±11.6 mm).
dol:10.1371/journal.pone.0062413.g003
PLOS ONE I www.plosone.org
4
May 2013 | Volume 8 | Issue 5 | e62413
Grazed Seagrass Meadows Still Protect Shorelines
plungers o f 60 m l syringes o f w hich tips h a d b e en cut off. T h e
sticks w ere w eighted before a n d after 24 hours o f placem ent at the
plots, after drying until co n stan t weight.
significant w ave heights. T h a t is, BSS differed significantly
betw een stations (/>< 0 .0 5 , T ab le 1). T h e relative wave height
(the significant w ave height relative to the w ater depth, H JK ) at
station A was exceptionally high c o m p a red to the o th er stations,
w hich m eans th a t the w ave height was n o t yet a cc o m m o d ated to
the local w ater d epth. As a consequence, w ave friction w ith the
seabed m ight cause w ave breaking, resulting in high turbulence
a n d (swash a n d rip) currents at station A.
T h e w ave b u n k e r trea tm e n t (Fig. 2c) was effective in th a t it
significantly red u ced w eight loss from the plaster sticks, indicatiOng
th a t h ydrodynam ic energy was significantly low er b e h in d the
sandbags c o m p a red to plots fully exposed to waves (_P = 0.01).
S ta tistic a l a n a ly s e s
A one-w ay A N O V A was used to analyze differences in wave
height b etw een stations. T w o w ay A N O V A ’s w ere used to analyze
the effect o f station a n d w ave reduction on sedim ent erosion a n d
c u rre n t velocity, a n d to analyze the effect o f can o p y length on
sedim ent b e d level. D a ta w ere log-transform ed w hen necessary to
m eet assum ptions for the A N O V A s. T o evaluate possible
differences b etw een stations, w e used T ukey H S D post hoc tests
a n d for all h y drodynam ic p a ram ete rs we used D u n n e tt’s post hoc
tests for w hich we re p o rt T-values. D ifferences at T < 0 .0 5 w ere
considered significant. R (version 2.15.1, J u n e 2012) was used for
all analyses. R esults are p re sen te d as m eans ± th eir sta n d ard
errors, unless stated otherw ise.
S e d im e n t sta b iliz a tio n
Seagrasses significantly red u ced sedim ent erosion by waves,
alth o u g h the degree o f the erosion red u ctio n strongly d e p en d e d on
the location along the re e f flat (Fig. 2a a n d b). A fter a p e rio d o f
2 m onths, stations A a n d Cl show ed significant erosive b e d level
change in artificially created b a re plots (F>< 0 .0 1 , Fig. 2b). At
station B the sedim ent was n o t significantly eroded, w hich is in line
w ith the low er h y drodynam ic forcing m easu red at this station
(T able 1). A fter a storm event, the sedim ent erosion was higher
(Fig. 2a). T h e effect o f waves o n sedim ent erosion was largest at the
n earshore, ‘sw ash’, zone a ro u n d station A a n d close to the reef
crest, ‘b re ak e r’, zone a ro u n d station CI. T h is was dem o n strated by
the m arkedly low er sedim ent b e d level at station A, th a n th a t at
station B (_P = 0.02) a n d station Cl (_P<0.001)(Fig. 2b). W h en
exposed to waves, sedim ent level in the u nvegetated gaps was
e ro d ed w ith, o n average, 5.1 cm at station A, 6.3 cm a t station Cl
a n d only 1.3 cm at station B in 66 days (Fig. 2b). R ight after the
storm event, the sedim ent erosion in wave exposed plots at station
A was a factor 2.5 h igher (—13.0 vs. —5 .1 c m , T < 0.001)
c o m p a red to erosion four weeks after the storm (Fig. 2a a n d b),
b u t erosion was n o t significantly higher for station B a n d Cl after
the storm .
Interestingly, the turtle exclosures revealed th a t grazed a n d
un g razed seagrass vegetation stabilize the sedim ent equally well.
T h a t is, excluding grazing d id n o t cause a n y difference in sedim ent
b e d level c o m p a red to the g razed trea tm e n t (Fig. 3b), even though
le a f length o f the canopy in grazing exclosures was a factor 2.6
longer
(117.8 ± 16.6 nini)
th a n
in
g razed
m eadow s
(4 5 .8 ± 11.6 nini).
T h e w ave b u n k e r trea tm e n t was effective in th a t it significantly
red u ced w eight loss from the plaster balls, indicating th at
hydrodynam ic energy was low er b e h in d the sandbags c o m p a red
to plots fully exposed to waves (P= 0.01,).
Results
H y d ro d y n a m ic fo rcin g
M e a n significant w ave heights (Hs) differed significantly betw een
stations along the re e f flat, except for stations A a n d Cl (T able 1).
D u rin g no rm al conditions (periods w ithout storms), significant
w ave height from waves com ing in from the sea onto the re e f (at
station Coral) was o n average 0.19 m w ith a n average peak p erio d
o f 6.06 s (Table 1). D u rin g the storm in Ja n u a ry , incom ing
significant w ave height increased to a n average o f 0.40 m , w ith a
p eak value o f 0.78 m (incom ing waves a t the station C oral, see
T ab le 1). T ypically, w ave height decreased from the coral, over the
vegetated re e f flat, tow ards the b eac h as is show n b y the low er
average significant w ave heights a t stations Cl to A a n d the average
relative w ave a tte n u atio n (% in T ab le 1). B ecause there was very
little standing can o p y biom ass to atte n u ate w ave energy, this m ust
be m ainly the consequence o f the decreasing w ater d e p th (Fig. lb).
H ow ever, a t certain configurations o f w ave height a n d w ater
d epth, w ave height started to increase, w hich is a typical
consequence o f shoaling o r wave breaking. T his increase in wave
height was observed a t all th ree stations o n the re e f flat (stations A,
B a n d Cl T ab le 1; shoaling is w ave a tte n u atio n < 0 ), b u t a t the
station n earest to the b eac h (A) it occu rred m ost frequently. H ere,
significant w ave height could increase up to 1.8 fold (wave
a tte n u atio n o f -88.2% in T ab le 1) relative to the incom ing wave
height. Such increase is m ost p ro b ab ly due to w ave breaking.
T h e im p act o f waves o n the reef-flat bed, estim ated as the
b o tto m shear stress (BSS), show ed roughly the sam e tren d as the
Table 1. Sum m ary of th e m easured significant wave h eight (Hs), peak wave period (Tz) and bed sh ear stress (BSS) along a cross­
shore seagrass profile (Fig. 1).
W a v e a t te n u a t io n
S ta tio n
Hs M e a n
n o rm a l
(m )
s to rm
(m )
H s M a x im u m
Tz
( n o r m a l)
n o rm a l
(m )
s to rm
(m )
(s)
m ín
BSS M e a n
BSS M a x im u m
n o rm a l
n o rm a l
m ax
n o r m a l (P a)
s to rm (Pa)
(P a)
s to rm (P a)
A
0.15a ± 0 .0 9 0.23as ± 0 .1 9
0.52
0.68
5.13 C± 1.96
-88%
18%
100%
0.046a" 0.034
0 .1 1 1abs*0.101
0.22
0.40
B
0.13b± 0 .0 7
0.24abs ± 0 .1 7 0.45
0.72
5.27b± 1 .84
-4 5 %
30%
100%
0.026b"0 .0 2 0
0 .083bs 0.077
0.14
0.36
C
0.16a± 0.07
0.30cs ± 0 .1 7
0.48
0.74
5.27b± 1 .53
-2 8 %
11%
100%
0.034c" 0.026
0 .113bs*0.086
0.19
0.38
Coral
0.19C± 0 .0 7
0.40ds± 0 .1 2
0.49
0.78
6.06 a ±1.41
0.044a*0.034
0.207cs* 0 .1 15
0.23
0.66
M eans w ith th e ir s ta n d a rd d ev iatio n s a n d m axim um significant w av e h eig h ts a re g iv en fo r no rm al co n d itio n s (n = 2945, "n o rm al" = p erio d s w ith o u t storm s) a n d d u rin g
th e storm (n = 195). W ave a tte n u a tio n values less th a n 0 in d icate w av e shoaling.
doi:10.1371 /journal.pone.0062413.t001
PLOS ONE I www.plosone.org
5
May 2013 | Volume 8 | Issue 5 | e62413
Grazed Seagrass Meadows Still Protect Shorelines
Discussion
seagrass present
coral reef
C oastal p ro tectio n a n d sedim ent stabilization b y seagrass is
often v alued as a n im p o rta n t ecosystem service, w hich generally
has
b e en
a ttrib u te d
to
seagrass
can o p y
properties
[12,24,25,29,30], T h is raises the question to w hich extent seagrass
m eadow s th a t have very little can o p y a n d have m ost o f their
biom ass in below ground tissue can still con trib u te to coastal
defense b y stabilizing sedim ents. Present results convincingly
dem onstrate th a t even intensively g razed subtidal seagrass
m eadow s, w ith a very short canopy, can still stabilize sedim ents
effectively. T his effect could b e due to the re m a in d e r o f the
canopy, b u t alth o u g h the seagrass has a relatively high density
(± 3 0 0 0 shoots m ~), the leaves are extrem ely short a n d narrow .
T h e abov eg ro u n d biom ass is m inim al (± 1 0 g m _ ~) a n d the
percentage cover o f the sedim ent is very low (< 2 5 %). It is m uch
m ore likely, therefore, th a t the difference in erosion betw een
g razed vegetation a n d b a re soil u n d e r high w ave conditions is due
to the role p layed by the relatively high below ground biom ass.
R oots a n d rhizom es can stabilize the sedim ent by reducing its
erodability. T his is a n im p o rta n t novel ad d itio n to the findings o f
previous studies, w hich identified the h ydrodynam ic effect o f the
canopy as the only essential m echanism in sedim ent stabilization
[12,22].
T h e sedim ent stabilizing effect o f g razed seagrass, w hich can
even occur by low -biom ass m eadow s, is expected to have
im p o rta n t im plications for b o th coastal p ro tectio n a n d ecosystem
functioning. W ith respect to coastal protection, b y reducing
sedim ent erodability, seagrass fields m ain tain a h igher b e d
elevation th a t will help to atte n u ate waves. W e have schem atized
these results in a conceptual d iag ram (Fig. 4). T h e sedim ent
a n ch o rin g effect by short, g razed seagrass vegetation, w hich has
m ost o f its biom ass in roots a n d rhizom es (Fig. 4c), increases the
critical b e d shear stress th a t is need ed for b e d erosion. W e
speculate th a t the presence o f a dense m at o f rhizom es a n d roots
can have sim ilar effects a t the sedim ent-w ater interface as
described for o th er b io ta th a t reduce erosion, such as biofilm s o f
m icrophytobenthos [31]. Seagrass cover causes the sedim ent level
to rem ain h igher c o m p a red to ero d ed u nvegetated gaps. In o u r
study this was u p to 13 cm , in others 18 cm [49^(Zostera marina).
O v e r longer tim e scales, this difference in erodability o f the
sedim ent is expected to seriously affect the form o f the cross-shore
height profile. T h e shallow er profile o f seagrass beds, c o m p a red to
situations w ithout seagrass, m ay im ply th a t m ore w ave energy is
abso rb ed before waves re ac h the coastal strip (Fig. 4b), because
dissipation o f w ave energy is a direct function o f w ater d e p th [1].
As a result, it is expected th a t less w ave energy c an p ro p a g ate over
the nearsh o re tow ards the b e ac h (Fig. 4b). It should how ever be
detailed how this pictu re is influenced by wave breaking. In our
study w e observed w ave break in g at the station closest to the shore,
at least d u rin g p a rt o f the tidal cycle. P referential zones o f wave
b reaking could locally experience higher b o tto m shear stress a n d
sm aller-scale variations in the profile could arise, b u t this effect will
decrease w ith the vegetation-induced stabilization o f the sedim ent.
W ith respect to ecosystem functioning, a rm o rin g o f the
sedim ent can have p ro fo u n d im plications for the subtidal seagrass
c om m unity b y the red u ctio n o f the a m o u n t o f sedim ent th a t is
resuspended. Biotic com m unities are know n to suffer from
sedim ent m ovem ent, due to processes such as direct sm othering
[50] o r b u rial [51], a n d abrasion o f tissues [52,53]. T h e prevention
o f erosion by seagrass as a fo undation species [54](Flughes et al.
2009) is fu rth er critical for bu rro w in g fauna like shrim ps th a t n eed
stable sedim ent environm ents to reinforce th eir burrow s [55].
A rm o rin g by seagrasses m ay also indirectly p ro tec t the ad jacent
PLOS ONE I www.plosone.org
Cl
Q)
Û
Swash
Surf zone
Breaker
seagrass present
03
Distance along the foreshore
/
Roots/
rhizomes
c
o
c/)
o
crit -new
LU
Bed shear stress
Figure 4. Conceptual m odel showing how erosion is decreased
along a nearshore seagrass bed w ith a m inim al canopy d ue to
the com bination o f increased critical shear stress and resulting
shallowness. S edim ent erosion occurs w hen bed shear stress (force
per unit area of th e flow acting on th e bed) exceeds a critical bed shear
stress (ib > icrit). (A) A typical d e p th g rad ien t of a n earsh o re hab itat
w here w aves break above th e coral reef, are th en fu rth er red u ced in th e
surf zone and "sw ash" o n to th e beach. S edim ent stabilization by
seagrass (green line) increases se d im e n t bed levels com p ared to a
situation with seagrass (yellow). (B) As a c o n seq u en ce o f th e reduction
of th e w ater d e p th by se d im e n t stabilization of seagrass (green line),
m ore w ave en erg y is a tte n u a te d while travelling tow ard s th e shore
com p ared to u n v e g e ta te d areas (yellow), and less w ave en erg y can
reach th e shore in th e surf zone. This highlights th e im p o rtan ce of
seagrass with resp ect to coastal defense. (C) In th e grazed seagrass
m ead o w w ith sh o rt leaves and low -biom ass, th e low structural
com plexity of sh o o ts in com bination w ith th e relative high ro o t and
rhizom e biom ass increases th e critical bed shear stress th a t is n eed ed
for erosion (icrit.).
doi:10.1371/journal.pone.0062413.g004
coral re e f c om m unity th a t can suffer critically from sedim entation,
by low ering sedim ent concentrations in the w ater colum n [4,56].
M o re generally, o u r results show the stabilizing effects o f
m acrophytes even w hen canopies are strongly reduced. T h is could
also have im p o rta n t im plications for o th er vegetated coastal
ecosystem s, such as salt m arshes a n d dunes, as well. In o u r system,
grazing b y turtles was the m ain driver m inim izing the canopy, b u t
m an y o th er processes can have a sim ilar effect, e.g. seasonal
changes in abov eg ro u n d biom ass, shedding o f leaves in a u tu m n
a n d w inter o r d eg rad atio n due to high turbidity, epiphyte cover or
eu trophication. W e show, how ever, th a t these changes in canopy
m orphology do n o t autom atically m ea n th a t seagrass beds have
com pletely lost th eir coastal p ro tectio n value. A lthough the relative
6
May 2013 | Volume 8 | Issue 5 | e62413
Grazed Seagrass Meadows Still Protect Shorelines
value o f seagrasses for coastal p ro tectio n is strongly species
dep en d en t, w ith e.g. clim ax species (e.g. Enhalus acoroides) generally
having a h igher value th a n m o re ephem eral species (e.g. Halodule
uninervis) th a t can b e highly variable in biom ass a n d cover [57],
even presence o f low -canopy sea grass beds is significant.
T h erefo re, w hen valuating seagrass habitats for coastal defense
purposes, the idea o f using abov eg ro u n d biom ass as a proxy for
w ave a tte n u atio n should be reconsidered. Such a p p ro ac h could
greatly underestim ate the coastal p ro tectio n service o f seagrass
w ith canopies o f low structural com plexity. Seem ingly insignificant
low -biom ass seagrass m eadow s th a t cover w ide re e f flats, m ay still
offer significant coastal pro tectio n services, a n d should b e valued
as such. T his ecosystem service is expected to becom e even m ore
im p o rta n t in the n e a r future, as storm frequencies are expected to
increase a n d n a tu ra l coastal pro tectio n structures like reefs are
u n d e r on-going d eg rad atio n [58],
Acknowledgments
The authors would like to thank Iris de Winter, Sabine Christianen, Hans
Wolkers, Sara Lambrecht and Jelco van Brakel for assistance with
sampling. We are grateful to Zhan H u for checking the procedure for wave
analysis and results. We are thankful to E. Koch, R. Unsworth, and an
anonymous reviewer for their constructive comments. Data are deposited
in DRYAD at http ://dx.doi.org/10.506 l/dryad.m 69 Ik.
Author Contributions
Conceived and designed the experiments: MJAC JvB TJB. Performed the
experiments: MJAC PJMvL. Analyzed the data: MJAC JvB PJMvL PMJH.
Wrote the paper: MJAC JvB PMJH MMvK LPML PJMvL TJB.
References
1. Koch EW, Barbier EB, Silliman BR, Reed DJ, Perillo GME, et al. (2009) Nonlinearity in ecosystem services: temporal and spatial variability in coastal
protection. Frontiers in Ecology and the Environment 7: 29—37.
2. Barbier EB, Hacker SD, Kennedy G, Koch EW, Stier AC, et al. (2011) The
value of estuarine and coastal ecosystem services. Ecological Monographs 81:
169-193.
3. Barbier EB, Koch EW, Silliman BR, Hacker SD, Wolanski E, et al. (2008)
Coastal ecosystem-based management with nonlinear ecological functions and
values. Science 319: 321—323.
4. Storlazzi CD, Elias E, Field ME, Presto M K (2011) Numerical modeling of the
impact o f sea-level rise on fringing coral reef hydrodynamics and sediment
transport. Coral Reefs 30: 83-96.
5. Borsje BW, van Wesenbeeck BK, Dekker F, Paalvast P, Bouma TJ, et al. (2011)
H ow ecological engineering can serve in coastal protection. Ecological
Engineering 37: 113-122.
6. Piazza BP, Banks PD, La Peyre M K (2005) T he potential for created oyster shell
reefs as a sustainable shoreline protection strategy in Louisiana. Restoration
Ecology 13: 499-506.
7. Mork M (1996) The effect of kelp in wave damping. Sarsia 80: 323-327.
8. Fonseca MS, Cahalan JA (1992) A preliminary evaluation of wave attenuation
by 4 species of seagrass. Estuarine Coastal and Shelf Science 35: 565—576.
9. Quartel S, K roon A, Augustinus P, V an Santen P, T ri N H (2007) Wave
attenuation in coastal mangroves in the R ed River Delta, Vietnam. Journal of
Asian Earth Sciences 29: 576-584.
10. Möller I, Spencer T, French J R , Leggett DJ, Dixon M (1999) Wave
transformation over salt marshes: A field and numerical modelling study from
north Norfolk, England. Estuarine Coastal and Shelf Science 49: 411—426.
11. Bouma TJ, De Vries MB, H erm an PMJ (2010) C om paring ecosystem
engineering efficiency of two plant species with contrasting growth strategies.
Ecology 91: 2696-2704.
12. Bouma TJ, De Vries MB, Low E, Peralta G, Tanczos C, et al. (2005) Trade-offs
related to ecosystem engineering: A case study on stiffness of emerging
macrophytes. Ecology 86: 2187-2199.
13. De Falco G, Ferrari S, Cancemi G, Baroli M (2000) Relationship between
sediment distribution and Posidonia oceanica seagrass. Geo-Marine Letters 20: 50—
57.
14. Madsen J D , Chambers PA, Jam es WF, Koch EW, Wesdake DF (2001) The
interaction between water movement, sediment dynamics and submersed
macrophytes. Hydrobiologia 444: 71—84.
15. Bos AR, Bouma TJ, de K ort GLJ, van Katwijk M M (2007) Ecosystem
engineering by annual intertidal seagrass beds: Sediment accretion and
modification. Estuarine Coastal and Shelf Science 74: 344—348.
16. Mellors J , M arsh H , Carruthers TJB, Waycott M (2002) Testing the sedimenttrapping paradigm of seagrass: Do seagrasses influence nutrient status and
sediment structure in tropical intertidal environments? Bulletin of Marine
Science 71: 1215-1226.
17. van Katwijk MM, Bos AR, Hermus DCR, Suykerbuyk W (2010) Sediment
modification by seagrass beds: Muddification and sandification induced by plant
cover and environmental conditions. Estuarine Coastal and Shelf Science 89:
175-181.
18. Houser C, Hill P (2010) Wave Attenuation across an Intertidal Sand Flat:
Implications for Mudflat Development. Journal o f Coastal Research 26: 403—
411.
19. Hemminga MA, D uarte CM (2000) Seagrass Ecology: An Introduction.
Cambridge, UK: Cambridge University Press.
20. O rth KJ, Carruthers TJB, Dennison W C, D uarte CM, Fourqurean JW , et al.
(2006) A global crisis for seagrass ecosystems. Bio science 56: 987—996.
21. Ramage DL, Schiel D R (1999) Patch dynamics and response to disturbance of
the seagrass postera novazelandica on intertidal platforms in southern New Zealand.
M arine Ecology-Progress Series 189: 275-288.
PLOS ONE I www.plosone.org
22. Gambi M C, Nowell ARM, Jum ars PA (1990) Flume observations on flow
dynamics in postera manna (Eelgrass) beds. M arine Ecology-Progress Series 61:
159-169.
23. Lowe KJ, KoseffJ R , Monismith S G (2005) Oscillatory flow through submerged
canopies: 1. Velocity structure. Journal of Geophysical Research-Oceans 110: 1—
17.
24. Luhar M, Coutu S, Infantes E, Fox S, N epf H (2010) Wave-induced velocities
inside a model seagrass bed. Journal of Geophysical Research-Oceans 115:
C12005.
25. Paul M, Bouma TJ, Amos CL (2012) Wave attenuation by submerged
vegetation: combining the effect of organism traits and tidal current. Marine
Ecology-Progress Series 444: 31—41.
26. W ard LG, Kemp W M , Boynton W R (1984) The influence of waves and seagrass
communities on suspended particulates in an estuarine embay ment. Marine
Geology 59: 85-103.
27. Gacia E, Duarte CM , M arba N , Terrados J , Kennedy H , et al. (2003) Sediment
deposition and production in SE-Asia seagrass meadows. Estuarine Coastal and
Shelf Science 56: 909-919.
28. Gacia E, Litder MM, Litder DS (1999) An experimental test of the capacity of
food web interactions (fish-epiphytes-seagrasses) to offset the negative conse­
quences o f eutrophication on seagrass communities. Estuarine Coastal and Shelf
Science 48: 757-766.
29. Hendriks IE, Bouma TJ, Morris EP, D uarte C M (2010) Effects of seagrasses and
algae of the Caulerpa family on hydrodynamics and par tide-trapping rates.
M arine Biology 157: 473—481.
30. Hendriks IE, Sin tes T, Bouma TJ, D uarte CM (2008) Experimental assessment
and modeling evaluation of the effects of the seagrass Posidonia oceanica on flow
and particle trapping. M arine Ecology-Progress Series 356: 163—173.
31. Le H ir P, M onbet Y, Orvain F (2007) Sediment erodability in sediment
transport modelling: C an we account for biota effects? Continental Shelf
Research 27: 1116-1142.
32. Fonseca MS, Bell SS (1998) Influence of physical setting on seagrass landscapes
near Beaufort, N orth Carolina, USA. M arine Ecology-Progress Series 171: 109—
121.
33. van der Heide T, Bouma TJ, van Nes EH, van de Koppei J , Scheffer M, et al.
(2010) Spatial self-organized patterning in seagrasses along a depth gradient of
an intertidal ecosystem. Ecology 91: 362—369.
34. Lai A, Arthur R, M arba N , Lili AWT, Alcoverro T (2010) Implications of
conserving an ecosystem modifier: Increasing green tur de {Chelonia mydas)
densities substantially alters seagrass meadows. Biological Conservation 143:
2730-2738.
35. FourqureanJW , M anue IS, Coates KA, Kenworthy WJ, Smith SR (2010) Effects
of excluding sea tur de herbivores from a seagrass bed: Overgrazing may have led
to loss of seagrass meadows in Bermuda. Marine Ecology Progress Series 419:
223-232.
36. Christianen MJA, Go vers LL, Bouma TJ, Kiswara W , RoelofsJGM, et al. (2012)
M arine megaherbivore grazing may increase seagrass tolerance to high nutrient
loads. Journal of Ecology 100: 546-560.
37. Vonk JA , Christianen MJA, Stapel J (2010) Abundance, edge effect, and
seasonality of fauna in mixed-species seagrass meadows in southwest Sulawesi,
Indonesia. M arine Biology Research 6: 282—291.
38. Christianen MJA (2013) Seagrass systems under nutrient loads, hydrodynamics
and green turde grazing. Do turdes rule the seagrass world? Nijmegen: Radboud
University Nijmegen. 138 p.
39. Callaghan DP, Bouma TJ, Klaassen P, van der W al D, Stive MJF, et al. (2010)
Hydrodynamic forcing on salt-marsh development: Distinguishing the relative
importance of waves and tidal flows. Estuarine Coastal and Shelf Science 89:
73-88.
40. Tucker MJ, Pitt EG (2001) Waves in O cean Engineering: Elsevier Science.
7
May 2013 | Volume 8 | Issue 5 | e62413
Grazed Seagrass Meadows Still Protect Shorelines
41. Möller I (2006) Quantifying saltmarsh vegetation and its effect on wave height
dissipation: Results from a U K East coast saltmarsh. Estuarine Coastal and Shelf
Science 69: 337-351.
42. V an Rijn LC, V an Rossum H , Termes P (1990) Field verification of 2-D and 3D suspended sediment models. Journal of Hydraulic Engineering-Asce 116:
1270-1288.
43. Kleinhans M G, Grasmeijer BT (2006) Bed load transport on the shoreface by
currents and waves. Coastal Engineering 53: 983—996.
44. Friedrichs C, Aubrey D G (1996) Uniform bottom shear stress and equilibrium
hypsometry of intertidal flats. In: Pattiaratchi C, editor. Mixing Processes in
Estuaries and Coastal Seas: American Geophysical Union. 405^129.
45. Austen I, Andersen TJ, Edelvang K (1999) The influence of benthic diatoms and
invertebrates on the erodibility of an intertidal a mudflat, the Danish W adden
Sea. Estuarine Coastal and Shelf Science 49: 99—111.
46. Boumans RMJ, Day JW (1993) High-precision measurements of sediment
elevation in shallow coastal areas using a sedimentation-erosion table. Estuaries
16: 375-380.
47. Porter ET, Sanford LP, Sutiles SE (2000) Gypsum dissolution is not a universal
integrator o f ‘water motion’. Limnology and Oceanography 45: 145—158.
48. EklofJS, van der Heide T , Donadi S, van der Zee EM, O ’H ara R, et al. (2011)
Habitat-M ediated Facilitation and Counteracting Ecosystem Engineering
Interactively Influence Ecosystem Responses to Disturbance. Plos O ne 6.
49. Harlin MM, Thornemiller B, BoothroydJC (1982) Seagrass sediment dynamics
of a flood-tidal delta in Rhode-island (USA). Aquatic Botany 14: 127-138.
PLOS ONE I www.plosone.org
50. Larkum AWD, O rth KJ, Duarte CM (2006) Seagrasses: biology, ecology and
conservation; Larkum AWD, O rth RJ, D uarte CM , editors. Dordrecht:
Springer. 691 p.
51. V erm aat JE , Agawin NSR, Fortes M D, U ri JS , D uarte CM , et al. (1997) The
capacity of seagrasses to survive increased turbidity and siltation: The
significance of growth form and light use. Ambio 26: 499—504.
52. Araujo R, Arenas F, Aberg P, Sousa-Pinto I, Serrao EA (2012) The role of
disturbance in differential regulation of co-occurring brown algae species:
Interactive effects of sediment deposition, abrasion and grazing on algae recruits.
Journal of Experimental M arine Biology and Ecology 422: 1—8.
Um ar MJ, McCook I J , Price IR (1998) Effects of sediment deposition on the
seaweed Sargassum on a fringing coral reef. Coral Reefs 17: 169—177.
Hughes AR, Williams SL, D uarte CM , Heck KL, W aycott M (2009)
Associations of concern: declining seagrasses and threatened dependent species.
Frontiers in Ecology and the Environment 7: 242—246.
^
Griffis RB, Suchanek T H (1991) A model of burrow architecture and trophic
modes in Thalassinidean shrimp [Decapoda, Thalassinidea). M arine Ecology-Progress
Series 79: 171-183.
Rogers C S (1990) Responses of coral reefs and reef organisms to sedimentation.
M arine Ecology-Progress Series 62: 185—202.
^
Rasheed MA, Unsworth R K F (2011) Long-term climate-associated dynamics of
a tropical seagrass meadow: implications for the future. M arine Ecology Progress
Series 422: 93-103.
5g Hoegh-Guldberg O , Mumby IJ , Ho oten AJ, Steneck RS, Greenfield P, et al.
(2007) Coral reefs under rapid climate change and ocean acidification. Science
318: 1737-1742.
May 2013 | Volume 8 | Issue 5 | e62413