I t I NYl (zrt6l 166I requeldos v I '86S

luortreEBuatr^Ipus.I Tr uollB^JosuoJIo luoqradeo v \ 'CY '{'C 'eua'J TVAIHJTIV t |rB!J T, uosrBed )I'Y'f fq rrodou t t t I t t t t t t IUOdgU SSflUC0Ud 'rrudv t ?66I l\tr.TSNNVHJU.rlrASAnAV( flHr dO 9NrNSdO cNIAtO'rTOd ,\UVnrSf, ?\SAUYIT-.ISOddO flSn (UrSUflJyA\ I vnvutsnvNulJ_gt^/r 1N11l0vNVt1oNlri t nluulst€c l0 "tNty,{luvdto t AUVUSnlHl

{* t I w01 HOdlglr I ?trtrrluYlwustl $flrc t 9268t0 llllllilllfrtflilililililililt1 L I ,).,'! ;\'* :r

Coverphotograph of AustralianPelicans (Peleunus conspiciltalr..t at Mandurahby A.G. Wells, I AFIAP.ARPS. I I I I I IHFtIPfiAI]Y Dl|.!,.1ir"ji:.:., i il.: ::rltii:finnf,l l CONTENTS ,! lirrrr,rri,..., , rr,.,t VtF.liitirN,\t Jti i iliAt.iA SUMMARY I 1. BACKGROUND ..2 2. STUDYAREA. I 3. DESIGNCONSIDERATION ,'.2 I 4. PELICANNESTING ACTIVITY AND SUCCESS ...... 4 T i 5. PELICANNUMBERS AND DISTRIBUTION., ...... 7 I t 6. WATERBIRDSPECIES, NUMBERS AND DISTRIBUTION I I I 8. WATERBIRDSAND SALINITIES OF HARVEYRIVER DELTA...... ,. 13 I t 9. BANDEDSTILT NUMBERS AND DISTRIBUTION 15 t 10.WATERBIRD ROOST SITES.,...... 18 I 11. ACKNOWLEDGEMENTS ...... 19 I 12.REFERENCES...... "...... 19

t tlitterhird ( i.te ol Peel-Ea'vey Eshtary Opening ol the D. ftsviltu ('hannel Jblowing I T TABLES

l. Summaryof pelicannesting activity and success at BoundaryIsland, July 1994- July 1997...... --.---.6 2. Black Swannumbers on Peel-HarveyEstuary in commonsurvey months of 1975-77 zrd1996-97...... -.... 13 3. NumbersofBanded Stilt on CreeryLagoon and the whol€ofPeel-Harvey Estuary in 1976-77znd1996-97...... 17 4. ElevationsofNirimbaCayandNorthofPointBirchroostsiteEMarch1994..-...... 19

FIGURES

1. Peel-HarveyEstuary ...... 3 2. NumbersofPelicans on Peel-Hawey:Feb 1975 - May 1977and Oct 1995- May 1997 ...... 8 3. NumbersofBlack Swans on Peel-Harvey:Feb 1975 -June 1977andOct 1996-Feb 1997...... 12

APPENDIX

l. Predictedand observed effects of DawesvilleChannel on water levelsand salinity inPeel-Harvey Estuary...... 21

It;ateftnd L:se of Peel-Hatuzj, Estuary folo\ring Opening of the D&'esvilc Chatuel SUMMARY

Peel-Harveyestuary, 70 km south of Perth Westem Australi4 is listed as a Wetland of Intemational Importanceunder the RamsarConvention. Surveys in the mid 1970sshowed that ai least 100 000 birds and80 speciesat timesmade use ofthe estuary'sexlensive, shallow and highly productive waters.

During the mid to late 1900sPeel-Harvey has becomeincreasingly eutrophic due to leachingof agricultural fertilisers from the catchment.Public objectionsto odours emanatingfrom decafng macro- algae and the increasingoccurrence of blue-greenalgae blooms have resultedin govemmentand communityaction to alleviatethese problems.

Onecomponent ofthe government'sstrategy has been to increaseflushing ofthe estuaryby constructinga secondentrance to the ocean.This entrance- the DawesvilleChannel - was completedin April 1994.A conditionofgovemment approval ofthe Channelwas that monitoring be undertakenso that impactscould be assessedand, where desirable and practicable, managed. This report describes progress in monitoringof waterbirduse ofPeel-Harvey following opening ofthe Channel.

The watertird program has a number of components.These include surveys of the numbers and distribution of all watertird species;monitoring of pelicannesting activity and successand an assessment of current suitability and use of threeformerly significantroost sites.Salinities in the lowest reachesofthe Harvey River and the healthofvegetation on the HarveyRiver Delta are also beingmonitored.

Data collectionfor the waterbirdprogram began between 1994 and 1996.Previously unpublished data from waterbirdsuweys undertaken in the mid 1970sare being used to providea baseline.

Somepreliminary conclusions can be drawnfrom the datacollected to date.For example,pelican nesting successappears to be sufficientat presentfor total pelicannumbers on the estuaryto be maintained.There hasbeen no declinein the numberofEastem Cudews using the estuary.Curlew Sandpipers, however, are far lessabundant. This may be dueto a numberoffactors quite unrelated to constructionofthe Channel.

Thehealth ofvegetation along the lowerreaches ofthe HarveyRiver Delta appears to be declining.This is probablydue to increasedsalinity and tidal amplitudein thispart ofthe estuarydue to the Chamel.

Furthermonitoring is neededto adequatelyassess post-Channel use of Peel-Haweyby waterbirds.It is proposedthat field work continueuntil 1999.

LliterhirLl Use ol Peel-Haflry Eshlot),Jbtlouitry Apeni,rg ofthe Duwesvitlt (-hotlel '1. BACKGROUND Peel-Harveyestuary, 70 km south of Perth, WesternAustrali4 is the largest estuary in south-westem Australia and is popular for recreatior\ principallycrabbing prawning,fishing, boating and enjoymentof nature.The estuaryalso supportsa significantcommercial fishing industry.

Peel-Harveyis an intemationallysignificant habitat for waterbirds.Tens of thousandsregularly gather on the estuary. In some years the total numbermay exceed lOO 000. More than eighty specieshave treen recordedincluding many migrants from breedinggrounds in northemEurasia and SiberiaThe estuaryis also of great significancefor speciesresident in Australia such as pelicans,swans, ducks, grebes,stilts, avocetsand many others (ANCA 1996,pp.893-5). In recognitionof its significance,Peel-Harvey has been listed, together with the Yalgorup lakes, as a Wetland of International Importance under the Ramsar Convention(Government of WestemAustralia 1990).

There is substantialurban development along lhe northern(Mardurah) and western(Dawesville) shores of the estuary.Land adjoiningthe eastemand southemshores is predominantlyfarmland or nature reserve, wilh somerecent small holding subdivisions.

Clearingof nativevegetation, drainage works anduse of agriculturalfertilisers has resultedin Peel-Hawey estuarybecoming eutrophic. Complaints during the 1970sabout the accumulationof decayingalgae and the associatedsmell led to efforts being made to solve the problem. Following detailed and lengthy investigationit was decidedthat a multi-facetedapproach should be taken. Key elementswere to be modificationof fertilizer use in the catchmentand excavationof a channel- the DawesvilleChannel - to providea secondconnection to the seaand thereby increase tidal exchangewith the ocean.

The proposalto construct a channelat Dawesvillewas approvedin principle by the Westem Australian '1989. Govemmentin January Worksbegan in February1992 and the Channel was opened in April 1994.

A conditionof approvalof the Channelwas a requirementthat monitoringbe undertakenso that impacts could be assessedand, where desirable and practicable, managed

One componentof the monitoring program was to be an assessmentof waterbird use of the estuary followingcompletion ofthe Channel.This work wasto be doneby the WesternAustralian Department of Conservationand Land Management(CALM). Funding arrangementsfor the DawesvilleChannel monitoringprogram were determinedby Govemmentin February1994. CALM was to undertakeits componentsusing existing departmentalresources. Funds were to be allocatedby CAIM ftom 1994/95 onwards.

This progressreport describes the waterbirdmonitoring wolk undertakenby CALM sinceopening of the Channel.Some preliminary findings are presented. A morecomprehensive report will be preparedwhen datacompilation and analysishave been completed.

2. STUDYAREA The study area is the eniire Peel-Harveyestuary ( area c. 136 km2. shorelinelength c. 75 km) including open waters, shaltows, tidal flats, shorelines and fringing marshes (Figure 1). The Hawey River Delta and the lower reachesofthe Harvey River are also included.

3. DESIGNCONSIDERATION Waterbirds are highly mobile animalsand most speciesusing Peel-Harveyand other estuariesof south- westem Austtalia have their breeding grounds elsewhere, from vegetated Swamps a few hundred metres away to the arctic tundras ofnorthem Russia. Waterbird numbers on south-west estuaries therefore change dramaticallyfrom seasonto seasonand - dependingupon breedingsuccess, climatic conditions elsewhere

Iljatefiiftl u'se ofPeel-]Ian'e| Esn&^ foltou'hry Opettntg oJ the Dave$vilte Ch.mtl I

in Australiaand other factors - from vearto vear.

Thesechanges in numbersoccur independentlyof ary chatges broughtabout by alteredconditions on the estuz-riesthemselves. For this reason,in order to determinewhelier or not a particular malagement action has had a significant effect on waterbird numbers,one would ideally monitor waterbird numbers on a "baseline,' seasonalbasis for at least two years prior to the action being t ken, so that a rlay be established. F lt F F ! ! t I I \o I I Figure l. Peel,Harveyestuary locality nap. In the caseof the Dawewille Channel,monitoring could not be commenceduntil after the channel$,as opened-It is thereforedifficult to assessthe exlent to which lhe chamel has or will affect \vaterbird use oF "post-channel" I the esnrary.Nonetheless it is importantto determine useby $'aterbirdsas tiel, constitute oneof the primaryvalues ofthe estuaryand are the basis upon $hich it is intemationallvrecognised.

urban and recreationalpressures on the estuaryand its marginsare increasingand an up-to{ate I krou{edge of how birds use the estuaryis essentialto minin se adverseinpacts resultingfrom these t Waterbird Use ofPeeLHaneyEstudryfollowine OpeninS afthe Davesyirre Channet pressures.Advetse impacts resulting irom the Channel may also become evident - though perhaps unproveablefrom a strict scientificviewpoint - as a result of ongoing post-Channelmonitoring Someof theseimpacts may also prove to be manageable.

The waterbirdmonitoring project hasseveral components. Progress with eachis describedbelow.

4. PELICANNESTING ACTIVITY AND SUCCESS

4.1 Background The AustralianPeli can (Pelecanus conspicillatus) rcgularly aestsat 8-9 locationsin WesternAustralia (J. Lane, unpub. data). Most of theselocations are in the north of the State,between Shark Bay and the east Kimberley.The only breedingsites south of SharkBay are Peel-Harveyestuary and Oyster Harbour near A1bany.ielican nestingon peel-Hawey was first observedin December1962, at Nirimba Cay (Shugg 1964).Nesting on oyster Harbour(Green Island) was not recordeduntil 1985(collins, P.J. 1985,pers. comm.:Harris 1987). pelicansnest colonially, tlpicaily on small low islandsnear the coast and occasionallyinland. On Peel- Harvey they are known to have nestedon Nirimba cay and creery, channel, Boodalanand Boundary Islands(AIJCA 1996,p.894) and an un-named island in BoggyBay (Fuller,P.J. undated, pers. comm ) creery is a long (c. 1.8 km), narrow island on the north side of Peel Inlet, to the east of the natural entrancechannel. Pelicans are known to havenested here on four occasions- 1963(Anon 1963, Serventy 1964),1964, 1965 and 1967 (FWD, undated). The breedingattempt in 1963(August) was unsuccesstul. All nests,thought to numberbetween 40 and 50, were abandonedduring incubation. The reason for abandonmentwas unclear,however two possibilitieswere put forward (Sewenty 1964). Somefishermen believeda fox Vulpeswlpes may havebeen responsiblq as one was seento swim to the island from the closely adjacentmainland. Disturbance was also considereda possiblecontributory factor, as the nesting sitewas near a boatingchannel. Of the otherthree nesting attempts at Creery only one(1964) is thought to havebeen successful. The number ofyoung produced is not known.

At least one of the pelican nestingattempts previously attributed to Creery Island in fact occurred on channelIsland, immediately to thewest. This attempt, thought to be around1967 (Easton, K. 1975,pers. comm.), apparentlyfailed. The areachosen for nestingwas also close to the boating channelthat passes betweenCreery and Channel Islands and disturbance was probably excessive

NirimbaCay is a small(c. 150m x c. 50 m), low, sandistand on the southside ofPeel Inlet,in AustinBay It is relativ;ly free from disturbanceas it is surroundedby an extensivearea of shallowsand is far from shore.Pelicans are known to havenesred on Nirimbain the 1960s(shugg 1964;FWD, undated),1970s (Fuller,P.J. 1973, pers. comm.; FWD, undated; J. Lane'unpub. data) and 1980s (Jaenscb R P 1986'pers comm.;Daw t986). Nestingprobably continued into the i990s, howeverthis is uncertain.Nesting attemptson Nirimba during winter were neversuccessful, due to flooding of nestsand eggs(Anon 1975,J Lane,unpub. data). Nesting attempts during summer,however, were succ€ssfulas watel levelswere lower andthere was little disturbance.From the l96osto the 1980s,and possibly into the earlyi990s, Nirimba Cay wasthe major,successful pelican nesting site in Peel-Ifurveyestuary'

In Septemb€r1973, pelicans nested on a smatlisland c. l/2 mile(0.8 km) to the eastof Nirimbacay. This colony was deserted,probably because the islandwas in shallowwater closeto the mainlandand accessby predatorswas possible when tides were low (Fuller,P.J 1973,pers. comm.).

BoodalanIsland is a small(c. 150m x c. 125m) sandyisland near the entranceto the canalestate at South Yunderup. It was created in 1972 with material dredgedto provide a navigablechannel to the estate (Serventy1975). Pelicans are knownto havenested on Boodalanin 1974(Serventy, D.L. 1974,pels.

u'.)terbird Lise ofPeet-Ha ey Est ('ry fotlo*nu Opening oJthe Dawesvi eChatnel I

comm.; Anon 1975, Serventy 1975, I. Larle,unpub. data), 1975(1. Lzte, unpub. data; Bodeker 1975) and 1988 (Love, M.R. 1988, pers. comm.). Nesting successhas always been low, however, due to frequent disturbanceand evenvandalism (Seventy, D.L. 1975,pers. comm.; Bodeker 1975).

Boundary Island is a large (c. 25 ha), sandy island at the northem end of Peel Inlet, near the main navigationchannel. It was created in 1987 (Pszczolq J. 1998, pers comm.) with material dredged from Peel-Harvey's natural outlet in order to increase tidal exchange with the ocean. Pelicans are known to have nestedhere in 1989 @avis, G. 1989, pers. comm., 16 November) and might also have attempted in other years.The outcome ofthe 1989 attempt is not known. Disturbanceis a concem as the island is adjacentto the main navigation channel, is easily accessible (it has deep water access and a jetty) and is formally reserved for recreation. Without controls on acc€ss,Boundary Island cannot be considered a secure nestingsite for the pelicansof Peel-Harvey.

4.2 Rationale Opening of Dawesville Channel had the potential to adverselyaffect pelican nesting successon Peel- Harvey by causing an increasein tidal amplitude. As mentionedabove, the main pelican nesting site in Peel-Harvey,Nrimba Cay, is very low lying (0.22 m AHDr in March 1994).Even a slight increasein tide height during summer could result in drowning of all eggs or newly-hatchedchicks in the colony. The potentialfor highertides to make boat accesspossible to Nirimba Cay was also a concem.

Given the above, and the possibility that nesting on Boundary Island was being adversely affected by disturbancg it was decided that pelican nesting successshould be a focus of the Dawesville Channel waterbird monitoring program. The fact that adverseimpacts might also be ableto be rectified was anoth€r importantconsideration.

4.3 Methodology Boundary Island was visit€d at fortnightly intervals or thereaboutslrom July 1994 onwards to monitor pelicannesting activity. When€verpossible, counts or estimateswere made ofthe number of adults in the nest area, the number on nests,nest contents(eggs or young, usuallydetermined from a distance)and the numberof flightlessyoung. Evidence of nest area desertionor mortality was also recorded,together with the probable cause where this was apparent The amount of inlormation obtained during visits varied dependingupon how closely the nest area could be approachedwithout causingan unacceptablelevel of disturbance.Great care was taken to prevent monitoring visits from prejudicing the survival of eggs or chicks.

Creery Island, Boodalan lsland and Nirimba Cay were monitored infrequently because of the lower probabilityofnesting occuning on theseislands and the needto containcosts. Ground visits were made to all three in July 1994. They were also inspectedby plane at approximatelytwo monthly intervals lrom October 1995 onwards during aerial censusesof total pelicannumbers on the estuary.Ground inspections I u,erealso madein October 1996,December 1996 and February 1997,during generalwaterbird surveys.

4.4 Results I FromJuly 1994to July 1997there were four periodsofcontinuous pelican breeding activity on Boundary Island,varying in lengthfrom 4-11 months(Table l). Thusbreeding activity (any stagefrom adultson I neststo fledgingofyoung) occurred in 28 of37 months. The estimatednumber of youngfledged during each breeding period varied from 35-45to 150-250.The first breedingperiod consisted ofa singlenesting effort. Subsequent breeding periods each contained two I or moreoverlapping efforts.

I I ,t}tr) = AusrralianHeight l)arum.

: (itcol|>eel- Uatefii an'ev Eshutr, /b owitlg Openuryol the Dd'rcsville Chantrcl The total numberof chicksfledged fiom BoundaryIsland in the past 3 yearsis in the vicinity of 500-700. This figure is a rough estimatg ratherthan a precisecount, becauseof the considerablecare taken by the authorsnot to excessivelydisturb breeding birds and their young.The figUreis, however,considered to be of adequateprecision to meet the objectives of the monitoring prograrq especially when used in combinationwith dataon total pelicannumbers on Peel-Harvey(see section below).

Somedisturbance occurred at BoundaryIsland during the monitoringperiod. An inspectionon 7 October 1994revealed that atl 34 unfledgedchicks still alive on the islandat that time had desertedthe nestingarea andwalked to the shore.Six downy chicks,all lessthan three weeksold, were found dead in and around the nes!area. One smashedegg was found in a nestscrape. The ageofthe chicks,the distancemoved, the presenceof small dead chicks and the smashedegg together gave a strong indication that major dirfutbun"" had occurred.The most likely causewas a carelessvisit by a personor people. Significantly' schoolholidays had begun on Friday30 Septemberand there were manymore peopleon the €stuaryin the first week of Octoberthan there had been during the precedingtwo months. I Table l. Summaryofpelican nestingactivity andsuccess at BoundaryIsland' July 1994- Jluly1997 '

Period Max # of adults Max # of adults Estimated# of in nestarea on nesls young fledged T

Jul94 - Oct 94 4 mths 100 65 o/ 35-45 ,'| I Dec94 - Jun95 7 mths 300 235 150-200 Jan96 - Nov 96 I I mths 250 100 zto 150-250 Nov 96 - Apr 97 6 mths 300 80 240 150-200 I

On 16 April 1997a family party carryingbinoculars was observedto approachwithin 20m of a group of 23 one week to four week old pelicanchicks in the nest area.They stayedfor t€n minutes,their presence I causingthe birdssome distress. On one occasionduring summer1996-97, tents were seenon the northem end of Boundary Island some I distance&om the nest area.Various boats, including hired house-boats,have also been seenon the beach or tying up at the jetty at the northernend ofthe island.

No signsofnesting activity have been found on CreeryIsland, Boodalan Island or NirimbaCay duringthe I pastthree years. During this period,Boodalan Island has been visited frequently by people,sometimes with dogs. Nirimba Cay hasbeen inundated on manyoccasions tkoughout the year (DoT tidal data neededto confirm), makingsuccessful nesting impossible. I 4.5 Discussion During the past three years,disturbance has caused cessation of one pelicanbreeding event on Boundary I Island-andsome associated mortality of chicks.Over most ofthis period, however,there has beenlittle or no disturbanceofthe colony and most breedingattempts have been successful. This situation may change in the firture as the numberof peopleusing the estuaryand visiting BoundaryIsland increases.There is a growing awarenessof the existenceof this breedingcolony and this may also result in more disturbance. I Furthermonitoring ofnesting successand disturbance levels is thereforewarranted. The lack of breedingon Creeryand BoodalarIslands is neithersurprising nor significant,for reasons I indicatedearlier in thisteport. The lack of breedingactivity on Nirimba Cay is not unexpectedas it is now ftequently inundated.The T

It/aterbir.l Use ol Peel-Hanq, Enua4 fot[o1a'ing Opening oJ he Dawesvile Channel I I exent to which increasedinundation has beendue to the smallrise in sealevel that hasoccurred in recent decades,to an increasein tide heightsdue to openingofthe DawesvilleChannel, or to erosion(if this has occurred)ofthe cay hasyet to be determined.An examinationofrelevart data(to be obtainedfrom DoT) is needed.The loss of Nirimba Cay as a breedingsite now leavesBoundary Island as the only successful F site on Peel-Haweyand makesmonitoring of breedingsuccess and disturbanceon the latter island all the I morelmporra . Breedingsuccess on BoundaryIsland over the past thf,eeyears appears to havebeen sufficient to maintain total pelicannumbers (see sectionbelow) on Peel-t{awey. However, as disturbancelevels are likely to l increase,further monitoringofpelican numbersis warranted. I 5. PELICANNUMBERS AND DISTRIBUTION 5.1 Background During the mid 1970s,numerous surveys were undertakenoftotal pelicannumbers on Peel-Harveyestuary I (J. Lane,unpub. data). The populationranged from 540to 2100individuals and on mostoccasions was in excessof 1000 (Figure 2). since that timq waterbird surveyshave been undertakenon parts of p€el- tlawey and on many other waterbodiesthroughout western Australia (e.g. Jaenscll vervest & Hewish 1988;Jaensch & vervest 1990a"b; || Halseet al. 1992i'storey et al 1993).It is evidentfrom thesesurveys that, in the mid 1970s,Peel-Hawey estuary regularly supportedlarger numbersof pelicansthan any other waterbodyin the southernhalf ofthe State.The highestever countsfrom other waterbodiesofthis region I are Vasse-Wonnerup750, Indoon l^ake500 and Lake Guraga120 (Ja€nscL Menifield & Raines1993). 5.2 Rationale Openingof the Dawesville Channelhad the potential to adverselyimpact on pelican numbersin Peel- I Harvey by reducing nesting success(see section 4.2). It was therefor€ considereddesirable to monitor pelicannumbers following Channel completion. : The numbersof pelicansfrequenting the estuaryin yearsimmediately preceding opening of the channel werenot known.However, precise count data were available from the mid 1970s.These could be usedto t provide a baselinefor comparisonwith results of posrChannel counts. It was therefore considered worthwhileto repeatthe 1975-77surveyq {iom October1995 onwards. Counts were not undertakenin the l8 monthsfollowing opening ofthe Channelbecause of otherdemands on personneland the needto I containcosts. 5.3 Methodology The numberand distributionof pelicanson Peel-Haweyhas been det€rmined at two month intervalssince t october 1995using a single-engine,high wing aircraft(cessna 172) with a pilot plusone observer. Each surveytakes l-1.5 hoursdepending upon bird numbersand flying conditions. Individuals and small groups are counted from the air. Large flocks are photographedand munted subsequently.The survey I methodology,route and timing (both day of year and time of day) closely follow those of the pelican countsconducted duirLg 1975-77. I 5.4 Resuns Total numbersof pelicanscounted on Peel-Harveyfrom october 1995to May 1997are shown in Figure 1.Numbers counted during the period1975-77 are also shown, for comparison.

I Thenumber inhabiting Peel-Harvey in the pasttwo yearshas ranged fiom 268 in october 1995to 706 in February1996. There are two peaksin numbers;in February1996 (706) and February l99Z (630).There I is somesuggestion that numberson the estuaryare rising, howeverit is too earlyto be definiteabout this.

I II'atethird ('se ofPeel,Hdney Esruarylollovittg Openi gofthe Dovesvi eChqnet t Numbers from october 1995 to May 1997 are lower than in the samemonths from october 1975 to May 1977, when the range was Sg5to ilyZ. Numbers since December 1995 (range 415 to 705) are similar, however,to those ofFebruary to November 1975 (range540 to 619) I

5.5 Drscussio'| Pelican numbers on south-west estua.iesvary seasonally,with greatest numbers in summer-autumnand I fewest in winter-spring. Leg-banding studies and monitoring of breeding colonies indicate that this seasonal variation is due to ttre departure of aclults to breeding sites in northem WA in winter and the arrival ofpost-breeding adults and young in summer(J. Lane, unpub. data). Birds remainingin the south- I west all yiar round are either non-breedersor, in the case of Peel-Harvey and Oyster Harbour, non- breedersplus residentbreeders.

- Pelicannumbers on Peel-Harveyduring the past nineteenmonths (october 1995 May 1997) have varied I in accordancewith this seasonalpattern of immigrationand emigration,with many residentbreeding birds (and non-breeders)remaining during winter-spring 1996.Numbers counted during this period are rougltly comparablewith t.hoseof February 1975 to November 1975 and on this basisit could be concludedthat I little changehas occurred. Recent numbersare, however, very much lower than those of 1976 and early monthsof 1977.This invitesexplanation.

e rcts - tstt

* tggs - lggz

1,500 t...... ,' r'a rtt-'

J F M AM J J A SO N D J F l\'lAl\,I

- Figure2. Numbersof Pelicanson Peel-Harvey:Feb 1975 - May 1977and Oct 1995 Ma1 1997'

(Figure 2) The dramatic increase in pelican numbers that occurred from December 1975 to December 1976 or might havebeen due in pirt to highly successfulepisodes ofbreeding, ifthese occurred,on Peel-Hawey of coioniesin the north ofthe State.lt is probable,however, that much ofthe increasewas due to an influx pelican birds from eastem Austmlia. Lake Eyre, in South Australia, filled in 1974 and supported a large breedingcolony that year, and probably in 1975-76 (Blakers, Davies & Reilly 1984, p.34). As its waters recededlnd saiinityincreased, the lake's fish populationdied and many thousandsofpelicans flew to othel 'the p.6'l). parts ofthe continent and to New Guineain searchoffood (ibid.; Sunday zrnres20 Nov. 19'17,

l{atefiid Lise of PeeLHarve}, Enuary.foltovng Opening ofthe Dares' e Ch@nl?t Someofthese birds are likely to havereached waterbodies on thewest coast, including Peel-Harvey. 1976 was also a drier year than averagein the south west and this would have led to further concentrationof birds on permanentcoastal waters. Further investigationof probatrlesources is proposed.This work will enablePeel-Harvey's role in the maintenanceofAustralia's pelicanpopulation to be better understoodand I will assistin the interpretationof resultsfrom the currentmonitoring program.

There is no clear relationshipbetween the total number of pelicanscounted on Peel-Flarveybetween I October 1995 and May 1997 and the numberof young fledging from Boundary Island during the same period. This may be due to a numberof factors including dispersalof recently fledged young to other waterbodies,movement of older pelicansbetween Peei-Harvey and other waterbodies,migration of adult I pelicansto distantbreeding grounds, and high naturalmortality offledglings. I 6. WATERBIRDSPECIES, NUMBERS AND DISTRIBUTION : 6.1 Background Peel-Harveyestuary is recognisedas one of the most important watedird habitatsin Westem Australia. More than80 specieshave been recorded and many are present in greatabundance. In February1977, a 'll0 I surveyofthe entireestuary revealed more 000 birds(J. Lane,unpub. data). In November1982 a surveyofthe eastempart ofPeel Inlet aloneproduced c. 41 000 (ANCA 1996,p.894). No otherwetland in south-westernAustralia is knownto supportas many birds.

I Specieswith highestnumbers are Balded Stilt (60 000,February 1977), Grey Teal (15 400,March 1987), EurasianCoot (15 000, October1976), Black Swan(8100), Red-necked Stint (8100),pied Cormorant (4000),Little PiedCormorant (4000), Curlew Sandpiper (2800), Black-winged Stilt (2700),Sharp-tailed I Sandpiper(2400), Australian Pelican (2100), Avocet (1500), Grey Plover (600) (ANCA 1996,pp.894-5).

ln June1990, Peel-Harvey estuary, together with YalgonipNational ParL wasdesignated as a Wetlandof IntemationalImportance under the RamsarConventio . Peel-Harvey'slisting was primarily on the basisof I its intemationallysignificant use by waterbirds(Govemment of WestemAustralia l99O). Many of the speciesusing the estuaryare also listedunder the Japan-Australiaand China-AustraliaMigratory Birds I Agreements. 6.2 Rationale Openingof the DawesvilleChannel had the potentialto afect the suitabilityof Peel-Harveyfor use by : waterbirds.The numberof high tide roost sitescould be reducedby an increasein tide heights.Higher tidescould also cause flooding of pelicannesting grounds (see other sections of this report)and the death of somefringing vegetationused by waterbirdsfor feeding,refuge, roosting and breeding.Changes in the I estuary'ssalinity regime, when combinedwith an increasein tide heightq could also contribute to the deathof vegetation.Increased tide heights,particularly during summer, could result in an increasein disturbance,by permittingaccess to importantwaterbird feeding and loafing areas that werepreviously too shallow for boats. Altered water levels, salinities and circulation pattems could also have largely I unpredictableeffects on the food webssupporting the estuary'swaterbird populations. Clearly most if not all speciescould be affectedin someway (someperhaps positively) by creationofthe Channel.For this reasonit u,asdecided that comprehensivesurveys to determineuse ofthe estuaryby all speciesshould be I conducted. As indicatedin Section3, in orderto establisha baselineit wouldhave been preferable for surveysto have beenconducted for at leasttwo yearsimmediately prior to the Channelbeing opened. This, however, was not achievableThe only baselinedata on waterbird use of the entire estuarycome from comprehensive suweystaken by the principalauthor (JL), with assistancefrom one ofthe co-authors(GP), in the mid- I970s.These surveys could be repeatedprecisely as fulI detailsof surveymethodologies, routes, times etc.

Itakrbi Use ofPeeLHavq, Est arvfo o\|l,g Openng ofthe Dauesvi cChan el I I the 19'76-77 had been recorded and the same personnel were available. It was therefore decided to tepeat surveysduring two ofthe five yearsprogrammed lor posfChannelmonitoring I in Two years of surveys have been planned in order to obtain an indication of year-to-year variability chosenas wateriird numbersand distribution.Post-Channel years two and five (1996-97 & 1998-99)were Whereas it is likely that at least some ofthe effectsofthe Channelwill take severalyears to be expressed. I in 1996-97 \n 1976-77six surveyswere conductedat two monthly intervalsflom August 1976 onwards, For this this was not possible,due to other dernandson staff time. It will also not be possiblein 1998-99. reasonthe 1696-97 surveyswere limited, and the 1998-99suweys will be limited, to October, December and the number of species, I and February.These months were chosenbecause both total waterbirdnumbers particula y oftransequatorialmigrants, are expectedto be greatestat this time ofthe year'

6.3 MethodologY I 1996' Suweys of all species ofwaterbirds throughout the Peel-Harvey estuary wer€ conducted in October foot), December l99i and February 1997. The areas surveyed,modes of transport (plane, boat and survey tim:: equipment used (binoculars, audio cassetterecorder, notebook, maps), survey routes' 11d I personnel were the same as, or very similar to, those of surveys conducted in the sarne months of 1976-77

aerial On the first day ofeach suwey, pelican numbers and distribution on Peel-Harvey were determined by this I census.Black Swan numbersand distribution on Harvey Estuary alone were also determinedduring aerial survey. The swans ofPeel Inlet were surveyed by boat.

species on the second day, a survey was made (by boat) of the numbers and distribution of all waterbird Peel Inlet downstream from the old Mandurah Traffic Bridge. waterbirds on the north-€astem side of upstream from the same bridge to the mouth of the serpentine River and including creery and channel day Islands and the creery marshes were surveyed by boat and on foot. In october 1996, the day one and two surveyswere all done in the one day, as in October 1976'

Bridge on day three, all waterbirds on the north-w€stem side ofPeel Inlet, from the old Mandurah Traffic Island, were to ward Point, including Boundary Island and the samphireislands to the west of channel shorelineof surveyedby boat and o; foot. ln iebruary 1997' a separatesurvey by boat of the western Harvey Estuary from Island Point to Ward Point was also conducted

On day four, all waterbirds on the shoreline and the inner and middte shallows of the eastem and southem separate side oipeel Inlet, from yunderup Canalsto the eastside ofPoint Grey, were surveyedby foot. A Grey to foot suwey of the east side of Point Crrey and the eastem side of Harvey Estuary from Point Herron Point, was also conducted.

and The shoreline and shallows between the mouths ofthe serpentine and Murray Rivers, plus the middle Bay were outer shallows on the eastern and southern sides of Peel Inlet fiom Yunderup Canals to Robert surveyed by boat and on foot on day five

A suwey of all waterbirds at the southem end of Hawey Estuary (from Herron Point Ford south)' six' includini the lower reachesand delta ofthe Hawey River, was conductedby boat and on foot on day the last dav ofthe surveys

6.4 Resulfs and Discus.sion numbers The processoftranscribing the 1996-97survey data (17 daysofrecordings ofthe distribution and partially completed of an estimatedOO species=) from audio cassettesto maps and to a databasehas been that results The information gatheredduring the 1975-76surveys also needto be addedto this databaseso may be compared.This work is proposedfor a subsequentreport'

Althoughsurveydatahaveyettobeassessed,somestrongimplessionshaveemergeddufingthe1996.9T

Estuary ngOpening althe Daues'i eCh'nrcl IO I;aterbinl Use of Peel-Hat-ve1' foltot surveys.

Little E9Iets Egreltq gqrzettq are undoubtedlymore numerousand more widespreadaround the estuarv ! now thanin 1976-77.This reflects a generalincrease in numbersof Little Egretin south-westemAustralia over the pasttwenty years.

Therehas been no declinein the numberof Eastern curlews Numeniusmadagascarier*is on the estuary. I Thoughthey are still rarely obsewedelsewhere in the southemhalf of WestemAustralia. thev continueio regularlyfrequent the nonhem end ofpeel Inlet in smallnumbers (10 or so) as before. ! curlew SandpiperscalTris Jerruginea were far lessabundant in 1996-97than h l9i6-77. This misht be due to year-to-yearvariations in total numbersmigating to Australia from arctic breedinggroundf, to a I long term decline in the abundanceof this speciesin Australia or throughout its range-,or to altered conditionson Peel-Harvey.waterbird suweys proposedfor l99g-99, plui informarion-to be sought on I trendsin the abundanceofthis speciesat othersites in Australia,will assistin inter?retingthis observ;ion. 7. BLACKSWAN NUMBERS AND DISTRIBUTION

I 7.1 Background Peel-Harveyis one ofthe most significanthabitats for Black SwansCygnus atralr/.r in WesternAustralia. I Countsconducted during the mid 1970sregularly produced between 2O0O and 6000 individualsand on one occasiontheir numbersexceeded 8000. (J. Lane & G. pearsoq unpub. data). Few other wetlandsin WestemAustralia support so manyswans.

I Greatestnumbers are found spreadout acrossthe extensiveshallows of Peel Inlet, particularly over the sezgrass(Ruppia sp.) beds on the eastern and southern sides of the Inlet. In some vears thev also congregatein large numbersat the southemend of Harvey Estuaryduring ratesummer ani auturnn.Here I they havehad access to fresh drinking water flowing all year round from the llarvey River. Small numbersof swans breed on nearby freshwaterwetlands and lead their young to the permanent watersof the estuarywhere they are reared.other birds depart at the beginningof winter to breed on I wetlandselsewhere on the Swan CoastalPlain or further in-land,retuming in summeras inland waters recede.

I Whengathered on openwater, Black Swansare sensitiveto disturbanceby boatsand peopleor dogs wadingin the shallows.It is panicularlyimportant for wild swansto havelarge areas of undisturb;d perman€ntopen water, with assuredfood suppliesand freshwaterfor drinking, when they undergo their annualmoult of flight fe3thers.At this time they are unableto fly, and thereforevery vulnerabli, for a I periodof lour to six weeks.

7.2 Rationale Histodcally, Peel-Harvey has been an important habitat fior the Black Swan. one of Western Australia's largest and most appealing waterbirds. Black Swans are also an important component ofthe estuary food web and have a significant role in the cycling of nutrients. They graze extensively on submerged iquatic plants of the system, particularly the seagrass Ruppia sp. (J. Lane pers. obs.). changes in ihe nirient status ofthe estuary, and in the species composition, distribution or productivity ofaquatic plants have the potential to afect swan numbers, Increases in tide heights also had the potential to affect swan numbers, I for exampleby allowing increasedboat accessto feeding, loafing and moulting areas.In the past these areashave been little disturbed.

Due to their large size and preferencefor open waters, swanscan be readily surveyed.The total number :

l|/ate rbird Use of Peel-Harve,r' Estuary fotto$,ing Openi'lg of the Dawesvi Ie Ch.xDrel ll I

(albeit the mid a.nddistribution of swanson Peel-Harveyfor a period prior to constructionof the channel I 1970s) were known with some precision.For ihese and the above reasonsit was decided to institute a program of monitoring of swan numbers following opening of the Channel Because there can be that the coniiderable year-to-year vadation in numbers on Peel-Harvey (see section 7-4), it was decided In order to contain costs, and because I monitoring should extend over at least two and possibly three years. February) ofother iork demands, the monitoring will be limited to tkee surveys (October, December and peak each year. These months have been chosen because swan numbers are expected to be at or near their the in spring-rummer. The swan censuseswill also coincide with surveys of all other waterbird species on ! estuary(see section 6.3). 7.3 Methodology ! Surveysof swan numbersand distribution on Peel-Harveywere undertakenin October 1996, December 1996 and February 1997. Surv€ydates were chosento coincide,within a day or so, with thoseof 1976-77

and too ! The swans on Peel Inlet were surveyedby boat as they were expected to be too numerous a single dispersedfor precise counts to be uniertaken from an aircrafl. Each survey was conducted by entire obierver (Gp or AC) working from a small dinghy travelling in a clockwise direction around the or anchored and short forays made on foot to survey Creery Lagoon and ! inlet. The boat was beached field Boggy Bay. Binoculars and telescopeplus tripod were used and data were recorded on maps and notebooks.Each suwey took severalhours to complete. ! I Thousands I | $ts - tstt I * tggo- l ggz -'- t I I

j I a o. ,..1 I \,Ja

o.....-_---.....-a -- \-,,,r

J FMA[I1J JASON

- Figue 3. Numbersof BlackSsans on Peel-Han'ey:Feb 1975- June1977 and Oct 1996 Feb1997

on Swan numbers and distribution on Harvey Estuary were determined during the aerial survey ofpelicans large Peel-Hawey(see section5.3 for details).Iridividuals and small gtoups were counted from the ail. No of flocks were encounteredon Harvey Estuary so it was not necessaryto take photographs Surveys Harvey Estuary took approximately30 mins to complete

Enuaryfotlowing Opening oftha Dlwcsvi eCho|tel 12 l+akrbird Lise ofPed-Hawry t

The swan survey methodology, route and timing closely matchedthose of the swan counts conducted I durj,')'g1976-77 . 7.4 Resulls t The total numbersof swanscounted on Peel-Harveyin 1975-'17and 199G97 are shown in Figure 3.

There were many more swans on Peel-Harveyfrom late 1975 to mid 1977 (and possibly beyond) than in I the survey months of 1996-97. There was less difference between numbers counted in the first three censusesof 1975 (February,July and August) than in the surveymonths of 1996-97.Notq however, that in February1975 there were more than twice as many swans(l120) on Peel-Harveythan in February 1997 I H70\. To facilitate comparison,the number of swanscounted in the samemonths of 1975-7'1and 1996-97 are I provided in Table 2. Table2. BlackSwan numbers on Peel-Harveyestuary in commonsurvey I monthsof 1975-77and 1996-9'1. Year October December February 1975-'t 6 2643 2188 2t62 I 1976-77 8057 4881 37s6 1996-97 1052 i56 472 I The numberscounted in October,December and February 1975-76 were 2.5 to 6 timesgreater than in the samemonths of 1996-97Numbers counted in 19'16-'l'lwere 8 to 14times sreater than in 1996-97

I 7.5 Dlbcussion The reasonor reasonsfor the dramatic differencein swan numberson Peel-Harveybetween late 1975 to 1977 and October 1996 to February 1997 are not known. Possiblecauses include differencesin breeding I seasonsin preceding years, differences in habitat availability (wetlands with water) in south-western Australia(and possiblyelsewhere in the continent) in the yearsof survey,or differencesin food availability or disturbancelevels in Peel-Harveybetween the two surveyperiods. : The fact that numbers counted during the first three surveysof 1975 were similar to those of 1996-97 suggeststhat the huge differencebetween subsequent mid 70s counts and thoseof 1996-97 may be due to short term factors rather than long term ones. Further surveys are needed to determine whether or not a I long term changehas occurred and, if so, the likely reasonslor it. Some investigationof possiblereasons for the dramatic variations in swan numbers observed in the mid 1970s also needs to be undertaken. Information on trends in macro-algae abundance and distribution also needs to be examined in relation to I changesin swan numbers.This work is proposedto be undertakendudng the next two years.

I 8. WATERBIRDSAND SALINITIES OF HARVEYRIVER DELTA 8.1 Background I For manyyears the upper(southem) end ofthe HarveyEstuary has been considered an importantrefuge areafor waterfowl. Many thousandsof ducks and swanshave at times been observedcongregated at the mouthofthe HarveyRiver delta and on the lowerreaches ofthe HarveyRiver where it passesthrough the I deltaformation (T. Riggertpers obs., J. Lanepers. obs.).

: lLatefiiftl Ure of Peel-lla*t1., l):nury follorntg O# ing aj th. Dr'\tesvil? (.hdt\'tel l3 "bird's river channel The Hawey River Delta is known as a foot" delta becauseofthe resemtrlanceofthe protrudesinlo the and its disiributariesto a bird's loot when viewed from the air. The delta formation levelis verY HarveyEstuary a distanceofc. 1.5km andhas an averagewidth ofc. 0'3 km The ground ground may be loq in ttre viclnity of9.5 to 1m aboveaverage water levil. Whenthe river is in flood the delta is SheoakCasurtrina utd completely inundated.The dominant vegetationof the -Swamp -obesq Eucalypns rudis SwampPaperbark Me lateuca rhaphiophllla, with increasingnumbers ofFlooded C;um places are patches u. on"'p-*"d, up"tream.The undeiorey is dominatedby introducedspecies. In there I of bulrushlTplu sp. fringing the main river channel I TheHarveyRiverisbelievedtoflowallyearround,Mostflowoccursduringwinter,decliningduring . throughout .p.ine Ho*"u"', a small flow of freshwater- at times no more than a trickle continues "spent" lrom the "u,nri". "nd autumq apparentlyeven in dry years. This is due to dischargeof water European Harvey Irrigation area. Summer-autumnflow may also have occurred naturally before settlement,that is, prior to clearingofnatural veg€tationard drainingof swamps I AlthoughtheupperHarveyEstuaryhasyettobeofrciallyprotectedasawaterfowlrefugearea,itseastem narrow reserve sho.esLd the outer end of the Delta are reservedfor natureconservation. There is also a privatelyowned and for ,,drainageand conservation' on bothbanks ofthe lowerfuver. Land to the westis nver. or population"pressuresare increasing,however these have not yet had a serious impact on the I to Island Point' !"iuu.y. ff",,.on point Ford, u *nJ b'. extendingacross the estuaryfrom Herron Point and preventsall but the smallestboats from pro"""ding further south. shallow water, sand bars, logs submergedrocks restrict boat access to the Deltaand River' I year-roundsupply The relative seclusionof the upperHarvey Estuaryand lower reachesof the River, the channel and- of freshwater for drinking, thi protection from strong winds offered by the tree-lined proximity of I distributariesof the Delb, the sale roost sites at the mouth of the Delt4 and the close a near-idealdry extensiveshallows for feedingand loafing, all combineto makethis part of Peel-Harvey seasonrefuge for waterfowl. I 8.2 Rationale water in the Constructionof the Dawesville Channelwas intendedinter alia ro iflcreasethe salinity of (Nodtlaria) blooms'^ Harvey Estuaryin spring and therebyto preventthe occurrenceof blue-greenalgae I cessationof In adiition to reducing the time taken for the Harvey Estuary to becomesaline following in the winter river flows, construction of the Channelwould also causean incteasein tidal amplitude low tideslower' estuary(see eppendi* l). Thus,during summer-autumrL high tides.would be higherand Harvey Theseiwofaciois in combinationhad ihe potentialto significantlyalter the hydrologyofthe upper I Estuary.

Thepotentia,leistedforanincreaseinsalinitiessouthofHerronPointFordanduptheHarveyRiver.This. I mightdirectlyaffectthefieshwaterdrinkingsuppliesofwaterfowlusingthisarea.Acombinationof winds blow in inc"reasedsatinity and increasedtide heighis, particularly when strong north-westerly vegetation.was autumn,also hai the potentialto adverselyaffect the HarveyRiver Delt. sincemuch of its wouldbe ofconcerndue to both its intrinsic I likelyto be at or nearits tolerancelimits. bamage to the Delta be madeof- value and its significancefor waterfowl. For thise reasonsit was decidedthat a study should Estuary and' in the salinity regilre, use by waterlowl and trendsin vegetationhealth ofthe upper Ha.rvey particular,the Delta and lower reachesofthe River. I 8.3 MethodologY to the limit of The salinityofPeel-Hafiey watersfrom severalhundred metres north of Herron Point Ford I intervalsfrom mic salt water intrusion up the Hawey River (liom the estuary)was profiled at fortnightly profiling was 1994 to mid 1996, a;d at monthly intervals subsequently(except mid winter). Additional periodsof strong undertakenon the openingday of annualmarron fishing seasons (see below) and during I

(/se Eshta\'Io1os'insOpenxls ofthe Duwesville Chatmel 14 Ulaterbir(t ofPeel-Hayq T I north-westerlywinds in autumnor at the beginningofwinter, beforestrong river flow.

t Salinitywas measuredat l-2 monitoringpoints north ofHerron Point For4 6-7 monitoringpoints between the Ford and the mouth of the Delt4 and at 0.5 km intervalsupstream from the mouth ofihe Delta. The lower Harvey River is very shallow(mainly 0.5-l m in summer),with numerousshallow sandand rock barsand spitsmaking navigation difrcult. Thereare, however,a numberof"deep" (2-5 m) holesscattered alongits length.These were also profiled routinely.

At eachmonitoring point salinity was measured 0.1 m and0.2 m belowthe surfaceand at 0.2 m intervals thereafteruntil the bottom was reached.Water temperaturewas also recordedroutinely. Dissolved oxygen levelswere measuredin the river, at severalmonitoring points on either side of the limit of salt water intrusion.

A Hamonsalinity bridge was useduntil June 1995and a Tps 90 FLMV microprocessorfield analyser from June1995 onwards. These instruments were calibratedagainst standard solutions at the beginning and end of profiling runs. A NATA-tested thermometerwas used for calibrationof water temoiraturi readings.water samplesat both endsof the recordedsalinity rangewere taken during profiling runs and sentto the WA ChemistryCentre for salinityand conductivitydetermination.

I The water level in the estuarywas measuredat the staff gaugeon the south side of Island Point at the beginningand endofeach profiling run. The river level was measuredat a post 4-45 km uostreamfiom the t mouthofthe Delta. Vegetationmonitoring sites were establishedalong the banksof the Delta and the River (seeGibson et al 1997).General observations of tree and shrubvigour were also madewhile conductingsalinity profiling. Colourphotographs were taken on occasrons. : Waterbirdspecies and numbersat the mouthof the Delta and on the lower reachesof the river were recordedon each visit. Additional data were collected during aerial surveysof the pelican and swan I populations(october 1995 onwards;see sections5 and z) and during the generalwaterbird surveysof October1996, December 1996 and February 1997 (see section 6). I Thesuccess or otherwiseofattempts by recreationalfishermen to catchmarron at variouspoints along the lowerreaches ofthe HaweyRiver on marronseason opening day (the first or secondsaturday in January eachyear) was also monitored.Marron are sensitiveto salinity and oxygenlevels and, in the absenceof comprehensivepre-Channel salinity data for the river, were consideredpotentially useful bio-indicators of I changein waterquality. Information on marrondistribution and abundatce in the riverprior to openingof theChannel could potentially be obtained from marronersand compared with post-Channelobservationi. I t 8.4 Results and Drscussion Resultsof the salinity profiling are being processed.Some historical data have been located and further att€mptswill be madeto accessmor€. The data that havebeen obtained suggest a significantincrease in : river salinityduring summer-autumn dry seasonssince construction of the DawesvilleChannel. This is supportedby reportsfrom recreationalmarroners who saythat mafion are no longer found in the lowest reachesofthe riveras they used to be prior to openingofthe Channel.

I Someofthe treesand shrubs ofthe Deltaare showing signs of stress.The foliage ofmany melaleucashas becomepaler and more yellowish in late summer-autumnthan in otherseasons, with somebranches and a few individualtrees having died completely.This might be a normal occunence,but could also be due to I an increasein salinityand tide heightduring the annualdry season.Lack of pre-Channelinformation on the healthofthe Delta'svegetation and seasonalchanges in this hindersinterpretation. Further monitoring is proposedto seewhether the observedchanges persist or worsen.Profiling of waterlevel and salinity will I alsocontinue.

Ilaterhird Ltse of P?al-Han'ey Estuaryfo owing Apeni ga/the Da',esvi e Channet 15 yet Data gatheredconceming waterbird species and numbersin the vicinity ofthe Delta and river have to comparisonscan be made.This work is I be ffised. Historicaliiformation needs to be collatedbefore proposedfor a subsequentrePofi I 9. BANDEDSTILT NUMBERS AND DISTRIBUTION

9.1 Background I Duringthe mid 1970s,Peel-Harvey estuary supported very large numbersof waterbirds,the most were an abund-ant being the Banded Stllt ctadorhynchts leucocephalus. In December 1976, there c. 60 000 estimated 8750 individuals of this species.By February 1976, this number had glo\l'n to I (J. Lane' Numbers declinedjust as rapidly to 6350 in Aptil 1977 and by June 1977 only 25 remained Peel Inlet. unpub. data). Virtually all ofthe BandedStilt recordedon the estuary during 1976-7"1were on rnuinlyontheeastemandsouthemsides,andintheCreeryMarshes.Veryfew(15inApril1977)were found on Harvey Estuary.

"east l98l There are also publishedrepo.ts of6394 BandedStilt at Peellnlet & south" sometimebetween period (Jaensch, alld 1g92, und-qooo on th" ct."ry Marshes on anolher occasion during the same Menifield & Raines 1993). Craig & Moore (1994) reported 9350 on the Creery Marsheson 28 December 1993. shortly pdor to openingofthe Channel

It is very unusualto find large concentrationsof Banded Stilt on tidal waters, they are more comnronll (in found on non-tidal wetlands,particularly salinelakes of the coast o. interiof. A possibleexplanation part at least)for the large numters found on Peellnlet in 1976-77is the exceptionallysmall tidal amplitude- combined with the very extensive area of isee Appendix l) prior to construction of the channel, proOu"iiu" shallows availablefor feeding and roosting flocks. In the four years prior to opening of the (Maloney, 1998' bhannel the averagedaily tidal rangein PeelInlet during summer-autumnwas c' 9 cm B pers. comm.). This smali range was partly due to geographiclocation (the astronomictide range on this being part of the coast is small), bul was primarily due to tidal exchangebetween the lnlet and the ocean greatly restrictedby the narrownessand lengthofthe single,natural entrance channel to Peel-Harvey.

9.2 Rationale oneoftheprimeobjectiveso|constructingtheDawesvilleChannelwastoincleasetidalexchange the between peei-Harvey and the ocean. An unavoidableconsequence was that tidal anplihtde within estuarywould also increase.though by how much was uncertaln'

range' rt was Given the general absenceof Banded Stilt |Iom tidal waters elsewherewithin the species lt was consideredlhatan increasein tidal range might adverselyimpact upon stilt numbersin Peel-Hawey thereforedecided that Banded Stilt numbersshould be monitored following construction of the Channel' Ideally they would also have been monitored for 2-3 years immediately prior to opening the Channel. However this was not possible,for reasonsgiven elsewherein this report

(oI Peel Due to staff and budget limitations,frequent monitoring of total numberson the entire estuary therefore decided to restrict Inlet alone) over a f,rolonged period iould not be undertaken. It was December monitoring of total stilt numbersto rhe surveysof all waterbird speciesplanned for october, and Februaryof 1996-97and 1998-99

the The whole-of-estuarysurveys would be supplementedby more frequent counts of stilt numberson an estuallne creery Marshes,an extensivearea of pools, intertidal flats and samphiremarsh surrounding Creery Island and the mainland' Large lagoon 1cr""ry Lugoon) on the north side of Peel lnlet, between were nu'*u"ri ofganOed Stilt had previouslybeen recorded on Creery Lagoon (see above) The Marshes

Estuuv Opennq of thc DutrPtvitt'Cha ll l6 tlitzthiftl L'sea[Peet'tlan'av lolou'tng I I alsoreadily accessible and close io Boundarylsland, where monitoring ofpelican nesting activity was to be undertakenon a fortnightlybasis (see section 4.3). 9.3 Methodology Visils were made to the Creery Marshes at fortnightly intervalsto count the total number of Banded Stilt present.Binoculars and,/ortelescope plus tripod were usedand numberswere recordedin field notebooks. With experience it was found that two or three vantage points on the northern side were sufficient to determinenumbers within the entire area. on occasionsit was necessaryto approachflocks more closely in order to obtain accuratefigures.

9.4 Resurfs Resultsofthe October1996 and February 1997 counts oftotal numbersofBanded Stilt on Peel-Harvey arenot yet available(the audiotapes oflhese suweys have yet to be fullytranscribed).

Countdata from the October1996, December 1996 and February1997 surveys of stilt numberson the CreeryMarshes and the December1996 survey of the entireestuary are presentedin Table3, together wlrn oata Irom lv /o- / / I

Table 3. Numbers ofBanded Stilt on Creery Lagoon and the whole ofPeel-Harvey t estuaryin 1976-77 and 1996-9'L Location Year Aug Oct Dec Feb Apr Jun "t60 CreeryMarshes t976-77 3 240 13000 2 600 ll I t996-97 0 0 6

Peel-flarvey 1976-'7'l 800 1500 8 800 60 000 6 400 I 1996-9't \25 ,|

I In summer-autumn of 1976-71 there was a dramatic increasein stilt numbers(to 13 000) on the Creery Marshes.This paralleleda huge increasein total numberson Peel-Harvey(to 60 000). A similar increasein numberson the Marshes did not occur in 1996-97,though thousandsofBanded Stilt were seenelsewhere I on the estuaryin February 1997 (precisenumbers are not yet available). "waders A paper by Singor (1997) on the of the Creery wetlands and adjacent mudflats', includes a I discussionofthe numbersofBanded Stilt he counted on the Creery Marshesduring 29 visits to the area betweenApril 1994 and october 1996. In summary,singor said thousands(>2000) of Banded Stilt were presentprior to the heary rains which fell in inland regions in Feb-March 1994 (this should be 1995) and I that following their departure(in February)only two BandedStilt had beenseen (on I I November). I 9.5 Drbcussion It is prematureto discussthe significanceof the results presentedabove in relation to opening of the I Channel.Data not presentedhere need to be trarscribedand collated. All dataneed to be consideredin the contextofthis species'unusual breeding cycle (Banded Stilt breed only on salinelakes ofthe arid interior, followingexceptionally heary, episodic rains). Tide height data from 1976-77al'd 1996-97 also need to be I obtained.Further monitoring ofBanded Stilt numbers,preferably for two moreyears, is needed.This work will be reportedon at a laterdate. t

I ltdterhid Use o.f Peel-ltdrve), Eshurylbllovng Opening ol !tu Dawesville Channet t7 I I 10. WATERBIRDROOST SITES

10.1 Backgrcund Severalspecies ofwaterbirds inhabitingPeel-Harvey typically roost in very large flocks on a small number of sand bars and cays. Surveys during the mid 1970s showed that the Little Black Cormorant Phalacrocore salcirost J, for example,roosted in singleflocks of severalthousand birds on sandbars of Herron Point Ford (length c. 2 km), either side of Egg Island. Thousandsof Pied Cormorants Phqlacrocorqxydrirs frequently roosted on high points of a low sandcay (c. 400 m x c. 75 m) to the northofPoint Birch.Ntrimba Cay (c. 150m x c. 50m)was also a significa.ntroost site

1O.2 Rationale The three roost sites referred to above, and a number of other Smallerroosts, had the potential to be adverselyaffected by an increasein tide heightsbrought about by construction of the Dawesville Channel (see Appendix 1). Each site was barely exposedduring high tides of the mid 1970s(i.e. pre-channel) and even a small increase in tides could have greatly diminished their suitability for roosting. Increased tide heightcould also have causederosion ofthe sitesthrough increasedwave action

The potential significanceof a loss of preferredroost sitesfor the above speciesis unclear. Use of other, non-preferred locations such as stretches of island or estuary shoreline could result in higher energy costs due to longer distancesto travel, more frequent disturbanceand even an increasein mortality due to terrestrial predators such as the introduced Red Fox having access. Significant negative impacts on waterbird populations would, however, be hard to prove. On the other hand, it would not be technically "topped diffcult or particularly expensivefor new caysor sandbars to be cfeated,or old ones up", should this seemdesirable.

It was thereforedecided that the elevationofthe three existingroost sitesshould be measuredprior to and five yearsa-fter opening ofthe Channel.This would enablesite elevationsto be comparedwith water level ,""o.d, pr"- and post-Channel.Any lossesdue to erosion could also be calculated.Post-Channel use of roost siteswould also be assessed,as part ofthe waterbird surveysto be conductedin 1996-97& 1998-99 (section6). Observationsliom thesetwo surveyperiods could then be comparedwith those of 19'16-'71.

1O.3 Methodology Suweyors from the Departmentof Transport (DoT) measuredthe elevationofthe three major roost sites in March 1994, one month prior to opening of the Channel. Estuary water levels were monitored continuously(pre- and post-Channel)by DoT. The speciesand numbersof waterbirds roosting on these and other sites of Peel-Hawey were recorded during general waterbird surveys conducted in October 1996, December 1996 and February 1997 (see section 6.3). These data may be compared with those collectedin th€ samemaruter and in the samemonths of 1976-77.Data from the pelican suweys conducted from 1975-77and 1995-97will also be examinedfor changesin roost site selectionby this species.

1O.4 Resuns and Discussron Suwey maps showing spot elevations ofthe three maior roost sites in March 1994 have been prepared by DoT. The elevationsofNirimba Cay and the roost site north ofPoint Birch are shown in Table 4. Vertical accuracyis SEz 0.02nq horizontal accuracySE 0.3m.

Large scale aerial photographs need to be examined before comparable data can be derived aom the suwey mapsofHerron Point Ford. This work is proposed

pre- and post-Channeltidal data are being sought.These will enablecomparisons to be made with roost site elevations.

2 SE = Srandard Enor - a slatistical measue ofenor.

ing Opening of the Dm'esvilte Channel I8 llatefiird Use of Peel-Harvq Estuary fo olt Post-Channeldata concerningselection and use of roost sites by waterbirdswill be compared with pre- Channeldata following completion ofthe 1998-99whole-of-estuary waterbird suweys.

Table4. Elevationsof Nirimba Cav and North ofPoint Birch roost sites-March 1994

Avge horizontal Average Roost 10 highestspot heights(m AHD) separation of elevation spot heights(m) (mAHD) Nirimba Cay 0.22,0 22.0 22,0.22,0.22, 0.22. 0.22 0.22,0.22,0.20,0.20 North ofPoint -o.20.-0.20, -0.21 , -0.24.-0.24. t25 -0.24 Birch -o 24,-0.24.-0.27,-O.28. -0.28

Meanwhile, the waterbird monitoring program (pelican nesting successand general waterbird survey components)has alreadyrevealed that substantialuse is being made of Boundary lsland as a roost site by pelicansand cormorants. This large island on the northem side of Peel Inlet was created in 1987 by the dumping of material dredged from the natural entranceto the estuary. Even if some loss of mid 1970s roost siteshas occurred, our observationsindicate that this has beenat least panly offset by constniction of Boundary Island. Boodalan Island, constructedin 1972 at the entranceto Yunderup Canalson the east sideofthe Iniet, also providesan additionalsite, however this is subjectto somedisturbance.

'11. ACKNOWLEDGEMENTS The foliowing are thanked for their assistancewith this project.

Lois Cade, Romeny L1'nch, Ruth Morgan, RaeleneHick, Beth MacArthur, Neville Watts and Andrew Williams have provided voluntary assistancein the field with salinitv profiling. Phil Fuller kindly provided pelicannesting records flom 1973.John Blyth made usefulcomments on an earlierdraft ofthis report.

Nello Siragusaofthe DepartmentofTransport (DoT) arrangedthe elevationalsurvevs of Nirimba Cay and tlvo other waterbird roost sites.John Muilally (DoT) took a personalinterest in undertakingand conveying the resultsofthis work. Ben Maloney (DoT) was most helpful in providing tidal and other inlormation

12. REFERENCES ANCA (1996). A Direclery oJ Impo'tant l|etktnds in Australia, Secon

Itatcfii'dtt:?dPtctllanavEstuaD,lb ov,ingOpeningoj ic Dufcj\.illc (:hunncl 19 Davis, G. (1989). Information regardingpelican nesting at Boundary Island conveyedby telephoneto J Lane,WA Departmentof Consewation& Land Managementon 16 November1989' Daw, A. (1986). Visit to Nifimba cay on 28 Nov 1986.Record sheetsent to Royal Australasian OrnithologistsUnion WA Branch. & Land Easton,K. ltOzS;. L"n"t of May-June1975 to J. Lane, WA Departmentof Conservation Managementreporting details of nestingattempt around 1967 on ChannelIsland. Map included.Refers t.l nh^tnsrrnhs recording futtei, e.f. lteZ:y. Note to J. Lane,WA Departmentof Conservation& Land Management detailsofvisit with N. Kolichisto Nirimbacay on l7 oct 1973and island further east on 30 Sep1973 FWD (undated).Extracts (ftom Department ofFisheries & wildlife files?) for creeryand channel Islands Reserveand Nirimba cay Reserve.Extracts h€ld by J. Lane, wA Departmentof conservation& Land Management. Govemmentof WestemAustralia ( 1990).Wetlands nominated by the Govemmentof WestemAustralia for inclusionon the List of Wetlandsof IntemationalImportance, Ramsar Convention Nominating documentprepared by WA DepartmentofConservation and Land Management, Perth 43pp Halse,S.A., Vervest, R.M., Munro,D.R., Pearson, G.B. & Yung,F H (1992) Annualwaterfowl counts in south-westemAustralia - 19s9/90.WA Departmentof conservationand Land Management TechnicalReport 29. .pelican Harris,A. (l9fi7). familiesfind a new homein the deep south','lhe WestAuslraliqn,24 Jzn.. p.16. JaenscbR.P. (1986). tetter of 25 Nov 86 to J. Lane, WA Departmentof Conservation& Land Managementreporting visit by A. Dawon 20 October1986 JaenschiR.p.& Veweit, R.M. (1990a).Waterbirds at remotewetlands in WestemAustralia, 1986-88, Part 1: LakeArgyle and Lake Gregory. Royal Australasian Omithologists Union Report 32 Jaensch.R.P. & Vervest,R.M. (1990b).Waterbirds at remotewetlands in westem Australia,1986-88, Part 2: Lake Macleod, SharkBay, CamballinFloodplain and Parry Floodplain.Royal Australasian OrnithologistsUnion Report 69. JaenschR.F., Merrifield, j. & Ruin"r,J. (1993).Waterbirds of south-westemAustralia: highest numbers countedl98l-92. Supplementto \\rA BirdNotes No 68 Jaensch,R.P., Vervest,R.M. & Hewish,M.J. (1983).Waterbirds in naturereserves of south-western Australial98l-1985: reserveaccounts. Royal Australasian Omithologists Union Report 30. 290pp. Love, M.R. (1988).Report of 8 June1988 conceming visit to BoodalanIsland on 16 May 1988 WA DepartmentofConservation and Land Management Dwellingup District file 24105' s"*enty, D.L. (1964).Nesting of pelicansat Mandurah.I. An abortivenesting attempt at creery lsland. WesternAustralian Naturalist. 9(4)180- 1 Sewenty,D.L. (1974).Letter of 9 July 1974to WA Departmentof Fisheries& Faunareporting visit to BoodalanIsland on 7 lvly 1974. Sewenty,D. L. (1975).Aihird nestingstarion ofthe pelicanin PeelInlet. Westem Australian Naturalist. 13 (213):67-8 . serventy,D.L. (1975).Letter ol7 oct 1975to wA Departmentof Fisheries& Faunareporting vandalism ofpelicancolony on BoodalanIsland shugg,H.B. (1lo+). Nestingof pelicansat Mandurah.II. Successfulnesting of pelicansin Peel Inlet. WesternAustralian Naturalist. 9(4\.81-4. Singor,M. ( 1996).Australian pelicans breeding on Boundarylsland. West. Aust. Bird NotesNo 78:l-3' Singo.,l',.1. (199i). Wadersofthe Creerywetlands and adjacentmudflats, Westem Australia. The Stilt 30:39-48. storey,A.w., vervesr,R.M., Pearso4G.B. & Halse,s.A. (1993).Wetlands of the Swancoastal Plain. Voi 7, Waterbirdusage of wetlandson the SwanCoastal Plain. Water Authority of WesternAustralia andEnvironmental Protection Authonty.

;oterhirct Openin4 of the Daw"svillc Chmnet 20 Lise of P€el'Har,el E:nta4: folloving APPENDX 1. Predicted and observed effects of Dawesville Channel on water levels and salinity in Peel-Earvey Estuaryr.

General Water levels in Peel-Harvey Estuary are influenced by several factors including astronomical tides, river inflows, atmosphericpressure changes, wind forces, storm surgesand longrerm fluctuations in mean sea- level.

The Dawesville Channel was expected to cause considerablechanges in the influencesof astronomical tides, river inflows and storm surges on the water level regime within peel-Hawey.

Tidal Range The averagetidal range (MHIM to MLLW') in the oceanadjacent to the estuary,is c. 57.5 cm.

Prior to opening ofthe Dawesville Channel,the averagetidal range ofPeel Inlet was c. l0 cm and that of Harvey Estuary was c. 9 cm. Following openingofthe channel, the averagetidal range increasedto 32 cm for PeelInlet and 45 cm for Harvey Estuary.

peel Prior to the Channelthe averagewater level in Inlet was -0.3 cm AHD3, the MHHW (..springtide,') +5.0 level was cm AHD and the maximum level was +53.1 cm AHD. These,nay b" "ompur"d *ith the +0.22 m AHD elevationof Nirimba Cay (seepelican nesting section ofreport).

Extreme Water Levels Prior to opening ofthe DawesvilleChannel, most ofthe extremewater levelsin Peel-Harvevresulted from river flows. The I in 100 year river flood level was estimatedto be 1.8 m AHD in peel tnlet ard l.i m AHD in Harvey Estuary. opening ofthe channel was predictedto lower the 100 year flood level to L I m AHD in Peeland L0 m in Harvey

The Channelwas Predictedto changethe likely source of extreme water levels from floodine eoisodes (typically during winter, when salinitiesare low) to storm surges(which may occur at any time o-fthe year. including summer-autumnwhen salinitiesare high) with the channel, ocean surgeswould be more readily transmittedinto peel-Harvey. A storm surse of similar magnitudeto that induced by tropical cyclone Alby (April 197g) was predicted to causethe i,ater level in the estuaryto rise c. 20 cm higherthan previously.

Salinity in the Harvey Estuary H}persalineconditions usedto developin Peel-Harveyeach autumn. It was predictedthat this would cease following opening ofthe channel. This has not yet occurred.ln autumn 19t5, one year after opening, the salinitynear the southernend ofHaruey Estuary exceeded46 parts per thousand(approx. 1.5 " sea*u-t"r).

' "Da\€svitle Much ofthe inforrnation Fesented has b€en extracted from th€ Channel Mo torurc pro|n.arnme:'Iwo year Tecbnicai Review. Status Report' O4a.ch 1997) p.€pared under contract by D.e. Lord & Assocratcs pn Ltd for the watcr & Ri\ers Commission. That document quotes other, authodtative sources. ' MHtrw = Mean HiSh High water (tlrc long term average of the highest tid€ each month) MLLW = Mean Low 1,o*, water (the lonS telm average ofthe lorvesr tide eaoh month) " AlIl) : Australian Heisht Datum.

fiaterbird U:e ofPee!-Haney Estuarylo otrirrg Apenng o/ rhe DatLesvi e Chonel 2l