A seagrass shading experiment to determine the effects of a dredge plume

2012·3.5 MB·English

Contributions
New TR, Field RP and Sands DPA (2007) Victoria’s but- YenAL(2011)Melbourne’sterrestrialinvertebratebiodiver-
terflies in a national conservation context. The Victorian sity: losses, gainsandthenewperspective. The Victorian
Naturalist124 243-249. Naturalist128 201-208.
NewTRandSan,dsDPA (2002)Conservation concernsfor ,
butterflies in urban areas ofAustralia. Journal ofInsect
Conservation6,207-215.
RelfMC and New TR (2009) Conservation needs ofthe
Altona skipper butterfly, Hesperillaflavescensflavescens Received4November2011;accepted23February2012
Waterhouse(Lepidoptera: Hesperiidae),near Melbourne,
SaVnidcstoDriPa.AJaonudrnNaleowfITnRsec(t20C0o2n)seTrhveatAicotnio1n3P,l1a4n3f-o1r49A.ustral-
ianButterflies.(EnvironmentAustralia:Canberra)
A seagrass shading experiment
to determine the effects ofa dredge plume
Hugh Kirkman1 Adam Cohen, and Harry Houridis,
,
'MarineScienceandEcology,5aGardenGrove,Seaholme,Victoria3018
Email:[email protected]
Abstract
TheenvironmentaleffectonseagrassofsedimentplumesfromdredgingisoftenquestionedinEnvironmental
ImpactStatements. Seagrassinsouthern PortPhillipBay,Victoria,Australia,wasshadedinwinterandspring
to the same level as the worst scenario ofshadingby an expected dredge plume. After 90 days ofshading,
Heterozosteranigricaulisshootnumbersreducedby61%.After 134daysofshading,theshootdensityreduced
by84%.Someoftheshadeswereremovedafter71daysofshading,butshootdensitycontinuedtodeclinefora
further40daysintheseplots,andnorecoverywasobservedthroughoutthistime.Intheseplots,whereshades
hadbeenremoved, shootdensities then stabilisedand no furtherlosswas reportedatday 134. A minimum
lightrequirementof12.5%-25.6%appearstobesuitableforsustainingH.nigricaulisbeds.(TheVictorianNatural-
ist129(3),2012,97-108)
Keywords: Heterozostera nigricaulis shading,dredging, shootdensity, light requirement
,
Introduction
Oneofthe major impacts ofdredging is an in- fectsofdredgingplumes.Recently,Mackeyetal.
creaseinturbidityandsuspendedsedimentsin (2007) studied manyparameters ofAmphibolis
thewater column and thesubsequent decrease griffithii biology under shade conditions, in ar-
in light availability to benthic plants. This re- eas adjacent to harbour dredging programs in
searchdescribessiteandseagrassspecificshad- theirregion.Althoughawiderangeofmorpho-
ingexperiments to determine the potential ef- logicalandphysiologicalvariablesrespondedto
fectofa dredgeplume. reduced light availability, the majority ofvari-
Much research has been carried out on the ablesshowed substantial recoveryafter42 days.
susceptibilityofseagrassesto shading (Bulthuis Thiswasoneofthefirstexperimentsinthepeer
andWoelkerling1983;GoldsboroughandKemp reviewed literatureto match sitespecificdredg-
1988;Peraltaetal.2002;Gaciaetal. 2005).Little ingactivitieswith responsebyseagrassto those
ofthisresearchhasbeenappliedtounderstand- activities.Previousworkusuallyappearedinen-
ing the effects of reduced light conditions on vironmental effects and impact statements and
seagrass under a dredge plume. In an excellent internal government or corporate reports. Pre-
review ofthe environmental impacts ofdredg- viously accepted methods used in research on
ingon seagrasses, ErftemeijerandLewis (2006) seagrass, with the modelled reduction in light
looked at 45 case studies globally, accounting causedbyadredgingplumearebroughttogeth-
fortheloss ofabout21000haofseagrass. They erandthewaythatsite-specificevaluationscan
recommend site specific evaluations ofthe ef- bemadeisshown.
Vol 129 (3) 2012 97
Research Reports
Unlike Mackeyet al. (2007), this studytested for specific time periods. The degree ofshad-
only changes in shoot density ofseagrass, be- ingwasequivalenttotheexpectedreductionin
cause this is the most likely and most reliably lightfromadredgingplume.TherecoveryofH.
consistent parameter to be affectedbyshading. nigricaulis upon return to naturallight intensi-
Fundingwas not available for measuring many ties was also examined for a short period. The
morphologicalandphysiologicalvariables.This compensation depth ofthisspecieswasusedto
study demonstrated a method of determining estimatetheminimumamountoflightthespe-
shadingeffects on seagrassin themost efficient ciesrequiresforsurvival.
way. Recently, Bite et al (2007) measured the This studywas undertaken during thewinter
photosynthetic efficiency ofZostera capricorni and spring months, and provides information
and Halophila ovalis using Pulse Amplitude on seagrass response to reduced light intensi-
Modulated fluorometryin short-term shading. ties only during those seasons. Some discus-
Theyfound thatphotosyntheticefficienciesand sion has been provided in the text regarding
effective yields increased significantly in both previous seagrass shading studies undertaken
species for shaded plants. Both species showed insummer.
a strong degree ofphoto-adaptation to shading
Methods
that may allow them to tolerate and adapt to
SiteSelection
short-termshading.
Theseagrassspeciesexaminedduringthepresent Sites were selected in consultation with the
studywas Heterozostera nigricaulis Kuo, family dredging proponent, prior to undertaking the
J. shadingstudy,which consideredthefollowing:
Zosteraceae, a narrow-leafed plant that inhabits
subtidal,shelteredand moderatelyexposedsandy • Plume modelling data to determine the key
sites where the dredge plumes will reduce
bottomenvironmentsinPortPhillipBay(Bulthuis
lightavailablefortheseagrassH. nigricaulis
1983). Priorto2005,Heterozosteranigricauliswas • Sites where data have been previousl;y
named Heterozostera tasmanica but Kuo (2005) collected. There was a considerable amount
distinguished between these species, and identi- ofuseful dataavailable from previous studies
fiedtheHeterozosteragrowingin PortPhillip Bay in Western Port Bay (Bulthius 1983; Bulthius
asnigricaulis. andWoerlkerling 1983);
The variability of seagrass morphological • Whether the seagrass meadows were
measurements is large and often prevents sta- permanent or ephemeral, as the monitored
tistically-sound statements being made about communities needed to be present for the
measurablechangesinlessthanfiveyears. Pro- duration of the environmental monitoring
ductivity measurements, i.e. of growth using program.Thiswasdeterminedfromprevious
hole punching (Short and Duarte 2001), were knowledge;
initially identified as an appropriate tool for • Shading experiments were not undertaken
determining seagrass health during this shad- within marineprotectedareas;
ing study. However, during the initial stages of • Sites were checked for accessibility and
fieldwork, it was evident that these measure- logistical constraints, because of the need
ments could not be undertaken in an accurate to work in readily accessible areas and in
low wave energy environments suitable for
ortimeefficientmanner.Shootdensityhasbeen
conductingshadingexperiments, and
demonstrated to be a useful tool to assess sea-
• Information on the physical characteristics
grass population status and it has been exten- ofeach potential study site was sought from
sivelyusedoverthelastdecade (forareviewsee initial siteinspections.
Marbaetal. 2005). A single location was selected for the shading
This study documents an experiment that experiment,approximately2kmeastnorth-east
aided in determining how Heterozostera nigri- ofSorrento (Fig. 1). There are numerous places
caulisrespondstothelowlevelsoflightexpect- in Port Phillip Bay where H. nigricaulis grows
ed to be caused by dredging in southern Port (BlakeandBall2001);however,mostoftheseare
Phillip Bay, Victoria, Australia. The objective eithersparselyvegetated,orgrowwithinmarine
oftheexperimentwas to determinethe impact protectedareas. Thelocationselectedneeded to
of reduced light on H. nigricaulis by shading
98 TheVictorianNaturalist
Research Reports
of seagrass, from which the expected light
attenuationandbenthiclight levelswerethen
estimatedusingthefollowingrelationship:TSS
vslight attenuationcoefficient derived froma
laboratory based experiment undertaken by
Longmoreetal. (2004).
• Incidental TSS measurements taken inside
the dredge plumes during the trial dredging
program;and
• ThePAR(photosyntheticactiveradiation)data
fromthefixedbenthiclightmetersitesduring
thetrial,i.e.MudIsland,Camerons Bightetc.
Light intensity was measured at Sites 4 and
6 of the six replicate sites (Fig. 2), to record
incident light intensities at the sea-bed under
the shades and without shade. Light attenua-
Fig. 1. Port Phillip Bay. The shading experiment was tioncoefficientswerecalculatedforthelocation
at Sorrento and proposeddredging in South Channel
andatQueenscliff. by comparing sea-bed and mid-water read-
ings. Light was measured using 2% Odyssey®
beattherepresentativedepthof4-5 mandonly Photosynthetic Irradiance Loggers (with built
afewplacesexistwhichcontaincontinuousbeds in sensors), programmed to take light readings
at this depth. This nominal depth was chosen at 10 minute intervals. These loggers measure
cminonssoiudtehreirnngtPhoarttHP.hinlilgirpicBaaulyisbectowmemeonn2lymoacncudr6s dcoomwinngweilnlifnrgo-mlitghhetsaindedofwitlhlensohtadedse.tect light
depth (BlakeandBall2001). The light loggers measured light in PAR
Lightmeasurementsandjustification expressed in pmole/m2/s and were calibrated
To predict the amount ofincident light likely against the Primary Industries Research
tooccurunderthedredgeplumesandestablish Victoria (PIRVic) Licor® meters. The sensors
the light intensities for future shading experi- on the loggers were cleaned every 10-14
mentsthefollowingdatawerereviewed: days at the same time as the light data were
• The initial model was based on turbidity downloaded.
modelling data obtained from continuous Theloggersweresecuredbystarpicketstothe
dredging in the South Channel for three sediment,withthesensorprotrudingjustabove
months (March-May2005) (Cardno,Lawson the top of the picket, and run continuously
and Treloar 2006). The turbidity modelling for two weeks, at which time the data were
wasusedtoestablishthelikelytotalsuspended downloaded and the logger redeployed. One
solid(TSS)concentrationsoverrelevantareas logger was placed underthe shade in the mid-
Butterflyclipand
Cabletiepoints
SeagrassShadeScreens(4rti) A
(70%shading,30%transmission) Ij\XS.tee/lsupports
2m _ . L"!
1.2ft
Fig.2. Shadescreensoverseagrass.
Vol 129 (3) 2012 99
Research Reports
die ofthe plot, one mid-water 1.3 m above the posed for experimentation purposes, i.e. light
seabedandoneon theseabed (0.3 mabovethe intensityundertheshadesat4 m, wasat6% of
bottom). surfaceirradiance.Theminimumlightlimitre-
The data from the turbidity modelling and portedforH. nigricauliswas 5% ofsub-surface
trial dredge program were used to predict the light (Bulthuis 1983).
average light reduction likely to occur in the Undernaturalconditions, withabackground
vicinityofseagrassbeds. Theequation usedfor light attenuation coefficient of0.2 nr1 the in-
estimating light attenuation was based on the cident light intensity on the seabed at, 4-5 m
Bougert-BeerLaw: depth isbetween 33% and41% ofsurface irra-
AC=In(IZi/IZ2)/Z Equation 1 diinagnicnet.eTnshietryefwoarse,cahossheandefocrloutshewiinthth7e0s%hasdhaidn-g
where AC is the light attenuation coefficient study, as this resulted in a theoretical incident
(nr1), Izi is the light at depth 1 (seabed), IZ2 is light intensity ofbetween 9% and 12% surface
thelightatdepth2(mid-water)andZisthedif- irradiance under the shades. This was similar
ferencebetween depths 1 and2. to the averageleveloflightreduction expected
The modelled TSS concentrations at selected under the dredge plumes in thevicinityofthe
seagrass sites, during 11 weeks ofdredging in seagrassmeadows.
South Channel East, rarely exceeded 5 mg/L Irradiance ontheseabed (IZi) was used along
(Cardno, Lawson andTreloar 2006). Since this withthe irradiance mid water (IZ2) to calculate
concentration isexpectedforlessthanapproxi- thelightattenuation coefficient (AC) usingthe
mately 10% ofthe time, a more realistic con- Bougert-Beer Law (Equation 1). The irradi-
centration likely to be regularly encountered ance just below the surface (I0) was then cal-
during dredging is 3 mg/L. This concentration culated (Equation 3). The surface irradiance
ofmaterialsuspendedbydredgingwouldbe in was calculated for average light measurements
additiontoambient(background)TSSconcen- between 1200 hand 1300 h for theduration of
trationsofapproximately2mg/L(Longmoreet thestudy.
alTh2e00r4e)l.ationshipbetweenTSSandlightatten- Io=IZi xeACxd Equation3
uation (Longmore et al 2004), was then com- Thepercentagesurfaceirradianceontheseabed
,
pared to approximatebenthic lightlevelslikely wascalculatedusingEquation4;thepercentage
during dredging. This laboratory-based study ofsurface irradianceunderthe shades was cal-
derived relationships for one sediment sample culated usingEquation 5 andthepercentageof
from the South Channel that found that, for a light under the shade as compared to the sea-
given plume TSS concentration, the resultant bedwasdeterminedusingEquation 6.
aalmoonue,nti.e.ofexlcilghutdiantgtebnaucaktgiroonudnud,ewtaos:the plume % Surfaceirradianceonseabed=Izl/I0x 100
Equation4
AC=0.115 xTSS (r2=0.89) Equation2 %Surfaceirradianceundershade=Ishade/Iox 100
where AC is the light attenuation coefficient Equation5
due to the plume from South Channel (nr1) % Light under shade compared to seabed =
tainodnTiSnSthiestphleutmoeta(lmsgu/sLp);enrd2eisdtshoelicdoecfofnicceinenttrao-f Ishade/Lix 100 Equation6
determination. Therefore,foraplumeTSS of3
ExperimentalDesign
mg/L,theplumerelatedlightattenuation coef-
ficient would be 0.345 m 1 Ifbackground light The methods of Bulthuis (1984), Kirkman
(1989), and Longstaff and Dennison (1999)
attenuationisaround0.2nr1 inthesouthofthe
were adopted for designing the work. Shade
Bay(Longmore etal 1996), then, using Equa-
., screens were put in a dense and continuous
tdirieropantdhi1a,ontcfhee4.rmeTshuielstaainpntpcrliiodgxehnittmiantltieeglnhystit1iy0nat%tenaosfintosyumripfnraaocl-e sSehaogortasdsenmseitaydaotwthaitsasinteawvaesrabgeetdweepetnh3o0f0-48m0.0
100 TheVictorianNaturalist
Research Reports
shoots nr2 Shades were steel frames each of Treatments
2 m x 2 m. and 60 cm high, on which were As depicted in Fig. 2, there was a treatment
connected shadecloth screens (Bulthuis 1984), where the seagrass was shaded for the entire
using plastic shade cloth clips and cable ties. durationoftheexperiment(shaded treatment);
The frame was held in place by star pickets. and a treatment where seagrass was shaded
Thesidesoftheshadeswereapproximately200 until seagrass health had visually diminished,
mm above the sea-bed (Fig. 3). Non-shaded at which point, the shades were removed
control treatments using steel frames (of the (recoverytreatment).
samedimensions)werealsoestablishedateach Eighteen replicates were used for each
ofthe sites (Fig. 2). Our shades were smaller treatment for the shoot density counts, to
than those of Mackey et al. (2007), but had reduce the level of variation within each
skirts around them to reduce lateral intrusion treatment.Experiencewithrandomorhaphaz-
oflight. ardly thrown quadrats showed that 18, 0.0625
Although LongstafF and Dennison (1999) m2quadrats, in H. nigricaulis was a reasonable
recorded minimal exchange of rhizomes number and size to reduce the coefficient of
between transplanted and non-transplanted variationtoaminimum.Tofurtherreducevar-
seagrassfrom biomass sampling under shades, iation,fixedquadratswereused(Austin, 1981).
the rhizome mat was cut to an approximate Due to time constraints, a power analysis was
depth of 10 cm around the boundary ofeach not undertaken to determine the appropriate
shadeandthecontrolsto preventtranslocation levelofreplication andquadratsize.
of material between non-shaded seagrass bed The shading experiment consisted ofa total
andshadedsites.Theeffectsofrhizomecutting ofsixreplicate sitesat onelocation (see Fig. 2).
could havebeen investigated byalso including Shade screensfor eachexperimental treatment
controls with rhizomes that were not cut. (shadedandrecovery),plusacontroltreatment,
Flowever, these controls were not established weredeployedateachofthesixsites.
becauseoflogisticalconstraintsassociatedwith Placing clear plastic shades atthecontrol sites,
the need to effectively sample all of the sites in accordance with the methods described in
in a single day and because ofthe findings of Kirkman (1989) was considered. Previously,
Fitzpatrick and Kirkman (1995) and LongstafF clear plastic shades have been used during
and Dennison (1999). shading studies to mimic the shade cloth, by
Fig. 3. Experimental
design for the Hetero-
zostera nigricaulisshad-
ingexperiment.
0.0625m1 fixedquadrat
withineachslie
Vol 129 (3) 2012 101
Research Reports
reducing the amount of water flow over the for seagrass meadow-wide health, or for
seagrasspresentundertheshades.Clearplastic detecting changes during dredging and, so,
shades were not deployed, however, as they was discounted (Dr Peter Ralph, University
would have had significant fouling, making of Technology, North Sydney pers. comm.
them unsuitable as controls and failure to 2006). The use ofphysiological or biochemical
accountforshade clotheffects, otherthanlight monitoring tools during a dredging program
reduction, wasconsidered to be lesssignificant alsohaslimitations,i.e.practicalityofsampling
than the confounding effects of using plastic and time to analyse results. The focus was to
shade cloth. Kirkman (1989) also reported document the integrated response ofthe sea-
that there was no difference in productivity grass using an easilyquantifiable and rapid as-
ofEcklonia radiata (a kelp) under clear plastic sessmenttechnique,i.e. shoot density.
controls, compared to the unshaded controls, Shoots were defined as black, lignified stems
whichsuggeststhatthereisnoimpactfromthe with green shoots attached or small, usually
shadesthemselves, in relationto limitingwater single, leaves that grew directly from under-
flowand reducingthe settlementofsuspended ground rhizomes (Kuo 2005). When the term
particulates on benthic plants. Mackey et al. ‘newshoots’isuseditreferstothesesinglegreen
(2007) also faced this problem, but eventually leaveswhichgrewfromthesandysubstratum.
didnotuseproceduralcontrols. Heterozostera nigricaulisshootswerecounted
within the shaded, recovery (following shade
Countsandvisualobservations
removal) and control treatments using 25
rSehsopootndseensviatryiiasbltehselaeansdtvaisriaablkeeoyfalplasreaamgertaesrs Tchmrexe2q5uacdrmatfsixweedre(0.p0l6a2c5edmu2n)dsetreeleaqcuhadsrhaatds.e
included in any monitoring program (Duarte
arenldiabKlierkpamraanme2t0e0r1)s.(ICtoilsliaelrsoetonaelof20t0h7e).moNsot sacnrdeens.ecTohnedatrotyalsnhuomotbserwoefreshocootnssi(dperriemdarya
single shoot in the counts) present within the
biomass sampling via coring was undertaken m
0.0625 2 quadrat areawas recorded, with the
because ofits destructive nature and was also
exception of dead shoots. Dead shoots were
not considered suitable because fixed quadrats
thosethathadnoleavesora fewdeadleaves.
were used to monitor changes in shoot density
(a much more reliable parameter). Fixed Qualitative observations were also made
throughout the experimental period, which
quadrats in terrestrial plants have been used
included documenting any morphological
successfullyto reducevariability (Austin 1981)
changes in seagrass health, such as changes in
and their use is argued by Hartnoll (1998) in
leafwidth, changes in shoot, leafand rhizome
marine ecosystems. Marba et al. (2005) used colourandpresence/absenceofepiphytes.
shoot density in fixed quadrats as a means
of measuring seagrass population dynamics TimingandFrequencyofSampling
very successfully in the Mediterranean. Fixed The experiment began during the week
quadratsremovetheproblemofenvironmental commencing5June2006andranforfourmonths,
heterogeneity and permit the detection of fromJunetoOctober2006. Shadingexperiments
relatively small changes. There are problems wereundertakenonH.nigricaulisatapproximately
with the representativeness of the quadrats thesametimeofyearasdredgingisproposed,i.e.
and changes within them may be artefacts of primarily winter and spring. Shoot density was
biological processes rather than community reportedatbetween 14and24dayintervalswithin
changes. There are also constraints on the theshadedandcontroltreatmentsovertheexperi-
statistical analysis oftime seriesbasedon fixed mentalperiodondays0, 19,35,49,61,78,95, 110
quadrats. and 134and intherecoverytreatmentondays0,
More subtle responses such as changes to 35, 61,78, 95, 110 and 134.Shadeswereremoved
seagrassphysiologywereconsidered, including fromthe‘recoverysites’onday71.
the measurement of in situ photosynthesis
Statisticaldesign andanalysis
via Pulse Amplitude Modulated (PAM)
For analysing the percentage change in shoot
Fluoremetry. This technique, however,
previously has not been applied to monitoring density, compared to Day 0, avalue of100 was
102 TheVictorianNaturalist
Research Reports
addedtothepercentagechangereportedineach Results
quadrat,toensurethatallofthevaluesreported Percentagechangeinshootdensity
a positive number for the statistical analysis. The percentage change in shoot density in the
Average shoot density for the three quadrats shaded treatments, compared to the control
presentundereach shadescreenwascalculated treatment over the four month experiment
forundertakingthestatisticalanalysis. fromJunetoOctober2006,isillustrated inFig.
For the purpose of the data analysis, the 4a. A gradual decline in the number ofshoots
recovery treatment was excluded, to avoid an was evident after approximately two months
unbalanced dataset, i.e. eight sampling events ofshading. A significant decline of61% in the
in the shaded treatment and six sampling number ofshoots was detected in the shaded
eventsintherecoverytreatmentand,therefore, treatmentsafterthreemonths,at Day95.
the shaded treatment was directly compared Percentagechangeinseagrassshootdensityin
with the control treatment. Compositing of the controls was near zero during winter then
thetwo datasets, i.e. shaded and recovery, also increased during spring from Day 78. Seagrass
would have meant that the main effects and shootdensityundertheshadeclothscontinued
interactions on Days 18 and 52, whereno data to decline until the last sampling event (Day
werecollectedfortherecoverytreatment,could 134), where therewas an 84% reduction in the
nothavebeenfullyexamined. numberofshootsreportedafteralmostfourand
a halfmonths ofshading, this was significantly
differentfromthecontrols (p=0.002).
Fig. 4a. Percentage change
in shoot density for control
and shaded Heterozostera
nigricaulis. Vertical lines are
onestandarderroraboutthe
mean.
0 20 40 60 100 120 140
Day
Fig. 4b. Percentage
change in shoot den-
sity for control and
recovery treatments
forHeterozostera nig-
ricaulis. Verticallines
are one standard er-
roraboutthemean.
Vol 129 (3) 2012 103
Research Reports
Shoot density did not significantly change in evident from the large standard error bars de-
the recovery treatment, compared to the con- pictingvariabilitybetweenthequadrats,within
trols during the two month shading period sites. There was a high degree ofvariability in
(p>0.05).Shadeswereremovedfromtherecov- seagrass response observed between the sites.
erysites,however,becauseanumberofthesites Notethat,withrandomorhaphazardlythrown
wereshowingaconsiderabledeclineinseagrass quadrats, the variance would have been much
health based on qualitative observations (see greater.Inoneinstance,withintheshadedsites,
Observed Changes below). Following the re- theseagrassdensitiesincreased,i.e.Site4shad-
moval ofthe shades in the recovery treatment edtreatment,whereasatalloftheothershaded
(Fig.4b), on Day71 oftheexperiment, seagrass sites, seagrass densities decreased. Similarly, in
shoot densities continued to decline signifi- some instances seagrass densities in the con-
cantlyuntilDay 110, comparedtothecontrols. trols decreased, e.g. Site 5, whereas shoot den-
Shootdensityhadreducedtomorethanhalf(60 sity at other control sites increased, e.g. Site 2
± 7%) since the beginning ofthe experiment, andSite 3, after day78 (seeFig. 5). Despite the
whereasshoot densityin thecontrols increased level ofnatural variability, there was sufficient
54% (p=0.012). The significant reduction in replicationwithinthetreatments (n=6) to ob-
shoot density in the recovery treatment, com- serve significant differences in shoot densities
pared to controls was partly due to seagrass in between theshadedsitesandthecontrols.
the controls responding to a change ofseason,
Observedchanges
fromwintertospring,wherebynewshootswere
Leaves in the shaded treatment became paler
produced. This significant difference, however,
was also partly due to shoot densities in the after one month ofshading, but had relatively
recovery treatment reducing by a further 40% similar phenotypic characteristics compared
to the controls. The main difference detected
sincetheshadeswereremoved,whichsuggested
after shading was the reduced overall epiphyte
alag effect in seagrass response to the removal
cover.
ofshading.Thecontinueddeclineinshootden-
sityupto Day 110 mayhavebeenduetothein- After two months of shading, no obvious
morphologicalchanges were apparent inshad-
ability ofplants to recover before that because
ofalackofstored material (Fig. 4b). No recov- ed seagrass, except that the leaves were now
ery was observed, but two months after shade clean ofepiphytes. The shoots and leaves still
removal, at Day 134, shoot densities appeared appearedto containchlorophyllpigment.
to stabilise (59 ± 9%), where no further loss in Afteralmostthreemonthsofshading,byDay
shoot density was reported (Fig. 4b). This sug- 83, dead, blackish-brown seagrass leaves were
geststhatseagrassinthepresentstudymaytake apparent. Average leafwidth was also measur-
manymonthstofullyrecoverfromshading. ablylowerintheshadedtreatmentcomparedto
Newshootswerefirstdocumentedinthecon- thecontroltreatmentfromDay61 untiltheend
trol treatment on Day 95, at the beginning of oftheexperiment (Table 1).
spring,whereasverylittlenewshootgrowthwas Adventitious roots (Cambridge et al. 1983)
reported in the recovery treatment. Fewer new were reported in the recovery treatment two
months after shade removal. Some flowering
shoots were reported within the shaded treat-
ment than in the controls. There was also new also was reported in the recovery treatment at
shoot growth throughout the control treatment this time; however, there was considerablyless
onDay 110,butonlyafewshootswerereported floweringcomparedto controls.
in the recovery treatment. Two months after Incidentlightintensity
shaderemoval,however,onDay134,largenum- Light intensities measured at Sites 4 and 6 in
bers ofnew shoots atall sixsiteswere reported the H. nigricaulis experiment are presented in
intherecoverytreatmentresultinginnonetloss Table 2. Light attenuation coefficients (AC)
ofshoots,comparedtoDay 110 (Fig. 5). were calculated at each site and the amount of
The changes reported in shoot density were light available to the plants under the shades
quite variable over time, both within sites and was determined. The average light intensity
between sites (Fig. 5). This was particularly measured from beneath the shades at Sites 4
104 TheVictorianNaturalist
Research Reports
and 6 was considerably lower than expected, undertheshades between Juneand September
with PAR concentrations reported between 5 may be attributed, in part, to seasonal and
and 15 pmol/m2/s. The 70% shade cloth was diurnal variations in the amounts of sub-
intended to reduce light intensity to between surfacelightavailable.
9% and 12% of sub-surface irradiance under The reduction in sub-surface irradiance
the shades, at 4-5 m depth. As indicated in compared to predicted levels was due to large
Table 2, however, light intensities beneath the amountsofsiltandparticulatematerialsettling
shadeswerebetween 1%and3% ofsub-surface ontheshadesbetweencleaningevents.Natural
irradiance. The variability in light intensity incidentlight intensity reported on the seabed
MeanPercentageChangeinShootDensity-Site1 MeanPercentageChangeinShootDensity-Site2
Days Days
MeanPercentageChangeinShootDensity-Site3 MeanPercentageChangeinShootDensity-Site4
Days Days
MeanPercentageChangeInShootDensity-Site6 MeanPercentageChangeinShootDensity-Site6
Days Days
Fig.5.Percentagechangeinshootdensityatindividualsites.Verticalbarsareonestandarderroraboutthemean.
Vol 129 (3) 2012 105
ResearchReports
(U, mid-water (Iz2) and sub-surface (I0) allin- reducedby61%.Shadingofthesamespeciesin
creased over the experiment, probably largely Western Portinwintercausedasimilardecline
due to increasing light from winter to early in shoot density after two and a half months
spring. The amount of sub-surface irradiance (Bulthuis 1983). Bulthuis (1983) reported that
reaching the seabed varied between 68% in seagrass took longer to reduce in density in
Juneand34%inSeptember.Thehigheramount winter, compared to the summer months. In
ofsub-surface light reported on the seabed in contrast with a spring and summer shading
June compared with July was due to low light study undertaken on H. nigricaulis in Corio
attenuation and good water clarity during Bay(Fig. 1),wherea 100%lossofshootswasre-
early winter. AC reported from June 2006 to ported after threemonths ofshadingwith 70%
September2006variedfrom 0.10 to 0.27nr1. reducedincidentlight(Bulthuis 1984),thecur-
The AC values determined may have been rent experiment was carried out in winter and
affected somewhat by the small distance, i.e. 1 earlyspring. All shoots were not lost, although
m,betweenthelightloggers,whichwereplaced four and a halfmonths ofshading did reduce
mid-water and on the seabed. A longer path shoot density by 84%. The difference between
length, i.e. >1 m, would have been preferable, the rapid loss ofshoots in the Bulthius (1984)
but theelevatedloggercouldnotbeplaced any experimentandtheslowerreductionin density
m
higher than 1 due to the shallow nature of inthisoneisthathisplantsmayhaveshoweda
thesites;anyhigherandthelightprobeswould greater tolerance to reduced light intensities at
m
havepresentedahazard toboating. 4 depth compared with 2 m, and the differ-
enceinseason.
Discussion
The changes in shoot density were variable
Heterozostera nigricaulis shoot density under
shading significantly decreased over time. Af- over time, both within sites and between sites.
ter three months of shading, shoot numbers There was also a high degree ofvariability in
observedseagrassresponseto shadingbetween
Table 1.Leafwidthchangesundershadeandatcontrol.
Date DayNo. Treatment Leafwidth SE Number
(mm) ofleaves
5Aug2006 61 shade 1.6 ±0.2 52
61 control 2.0 ±0.2 52
8Sept2006 95 shade 1.6 ±0.1 57
95 control 2.1 ±0.3 53
17Oct2006 134 shade 1.6 ±0.2 43
134 control 1.8 ±0.2 52
Table 2. Average light intensity taken at Sites 4 and 6, between 1200 h and 1300 h for the months ofJune
throughSeptember2006.
HeterozosteranigricaulisSites4and6 AverageLightfrom 1200--1300h(|imol/m2/s)
Month June July August Sept
LightUnderShades(IShade) 7.4 4.6 15 5.7
LightIntensityattheSeabed(Izi) 168 162 264 288
LightMidWater((Iz2) 185 203 330 342
Lightattenuationcoefficient (AC) 0.10 0.21 0.22 0.27
L%igohftSSuubb--SSuurrffaacceeI(rIr0a)dianceonSeabed 26848 43379 46229 38454
%ofSub-SurfaceIrradianceunderShade 3.0 1.2 2.3 0.7
%ofLightundershadecomparedtoSeabed 4.4 2.8 5.5 2.0
106 TheVictorianNaturalist