Seasonal Patterns in Abundance of Waterfowl

Corella,2004, 28(3): 61-67

SEASONALPATTERNS IN ABUNDANCEOF WATERFOWL(ANATIDAE) AT A WASTESTABILIZATION POND IN VICTORIA

ANDREW J. HAMILTONIr and IAIN R. TAYLORI

'AppliedOrnithology Group, Johnstone Centre, School of Environmentaland InformationSciences, Charles Sturt University,PO Box 789, Albury, New SouthWales. Auslralia 2640 :Correspondingauthor (Currenl Address): Primary Industri€s Rcscarch Victoria (Knoxfield), Privatc Bag 15, FerntreeGully DeliveryCentre, Victoria, Australia3156

Received:2 Novenber2003

The seasonal abundanceof waterfowlon a waste stabilizationpond at the Western TreatmentPlant, Victoria, Australia,was studied over two years. The abundancesof species that are considered to be highly dispersive, such as the Pink'eared Duck Malacorhynchusmembranaceus and Grey Teal Anas gracilis, were erratic and inconsistentacross the two years. For other species, such as the AustralasianShoveler /nas rhynchotis,Blaak Swan Cygnus atratus, Pacific Black Duck /nas superciliosaand AustralianShelduck Tadoma tadornoides,mote consistentpatterns were observed each year. Most species used the site during what would be expectedto be their non-breedingseason. Australian Shelducks appeared to use the site as a late-spring/early-summermoulting reluoe.

INTRODUCTION Pond Nine of the WTP, and draw comparisonswith previouslypublished work. The Westem Treatment Plant (WTP) at Werribee. Victoria, is known to support large numbers of waterfowl of several species (Lane and Peake 1990), and forms part MATERIALSAND METHODS of a Wetland of International Significance (Ramsar ConventionBureau 1984). A Iargewaste stabilization pond Study site within the WTP, known as Pond Nine in the Lake Borrie The WTP occupies an area of 10851 hectaresand is situated35 system,is consideredto be of particular importancefor kilometrcs wcst of Melbourne on the shores of Por( Phillip Bay waterfowl and other waterbirds(Elliget 1980; Hamilton (38"00'5, 144"34'E).A location map is provided in Hamilron ?r dl. (2004). Observations for rhe present study were made at a wasle 2002; Hamilton et al. 20021Hamilton er dl., press). in stabilization pond known as Pond Nine, part of a series of ponds that There is little published quantitative information on make up (he Lake Bonie system. Unlike the orher pond systenrsar lhe seasonalchanges in abundanceof waterfowl at the site. WTP, the Lake Borrie ponds are inegularly shapcd and of differing Sewagetreatment changes will be implementedat the WTP sizes. Much of the Lake Borrie sysrcm was originally a natural wetland thal was in 2005, which havethe potentialto affect the usefulness divided into a series of 30 waste stabilizalion ponds through the construction of impoundments in 1936. Pond Nine is the lafgest of of the site, includingthe Lake Borrie ponds, to waterfowl lhese, covering an area of 109 hectares. It is also lhe only pond in the (see Hamilton et al. 2002), so information on seasonal sysl€m wilh a stand of dead trees (mostly Melaleuca lanceolata). me patternsof abundancecould be useful in monitoring anv averagc waler depth is 60 centimerres (Cartwright 1996, unpubl. data). chrngesand lbr developingmanagement strategies foi this Further details of the functioning of the Lake Borrie waste slabilizalion ponds Ramsarprotected wetland. with respect to sewage treatment and ecology can be found in Hamillon (2002). Seasonalmovements of Australian waterfowl are believed SanqLint proto.ol to be influencedby rhe availability and depth of inland temporarywetlands (e.9. Morgan 1954;Frith 1957;Frith Dala used here were collecled as parl of a wider study on the diumal 1959;Braithwaite and Stewart1975; Crawford 1979;pert activity budgels of waterfowl at pond Nine, the sampling protocol for 1997). It has been suggested that inland temporary which is describedin derail in Hepworrh and Hamitron (2001) and Hamilton ?/ al. (2002). In sumrnary, tloodwaters provide an abundanceof food for waterfowl surveys were made from defined observationpoints on the €mbankmenl using a Leica@ (Braithwaite and Frith 1969), Tclevid 7? and hence offer suitable telescope(20-60 x zoom magnification).Sampling, which involved (he breeding conditions for many species. Conversely, countrng total numbcr of birds on the pond, was conducted at five perrnanentwetlands, such as waste stabilizationponds, may evenly spaced times of day: sunrise, mid,morning, midday, mid- be used as non-breeding refuges (Braithwaite and Stewait afternoonand sunset.It took approximatelyone hour to sample the entlre pood: hence sampling for each time of day was 1975). Most of the ideas on waterfowl movements begun 30 minulcs are belore the midpoinr of rhe sampting period (e.g. 30 basedon either banding minutes before studiesor more commonlv. on sunrise). Sampling was undertaken on 47 dalesi lfuee times a month comparisonsof seasonrlchunges in abunJancear particular ,rom ll July 1998to 20 June 1999and then at approximatelymonthly rocauons. rntervalsfrom I Augusr 1999ro 9 Augusl 2000.

For rhe Hldhead A\!h)a au\ttaltt. The aim of this study was to quantify seasonalchanges d drring specres.ir $rj nece\sdry . to determine how long individuals stayed underwater during a dive in the abundanceof the main watefowl species so found at that no birds were misscdwhilst surveying.In a preliminarysludy, the A. Hamilton and L Taylot: Abundance of walerfowl at waste stabilization oonc, Corella 28(3)

dive times of 70 birds were measured.The maximum dive durarion was numbersto occur from late summer to early autumn found to be 25 seconds (mean = 13.4 seconds).Thus, each ficld in of view each year (Fig. 2t. Numbers was observed for at lcasr 25 secondsso thal birds surfacing from a dive, of Crey Teal and Chestnur thal would nor have been seen in the initial scan. could be observed. Teal varied greatly from month to month, with no discerniblepattern (Figs 3 and 4). The PacificBlack Duck Data p'es?, ation was most abundantfrom late summer to mid-autumnof both years (Fig. Both daily average and maxirnum values were presentedfor (he 5). The nurnbersof Hardhead(Fig. 6) seasonal abundance of waterfowl. Daily maxima are particularly followed a very similar pattem to that for the pink-eared meaningful for species rhal show diumal trends in abundance since a Duck. The largest numbers, approximately 8 000 site could still have been importanl to a species at a parricular time of individuals,occurred at the first samplingdate on ll July year even if it was only used by it at certain rimes of day_ Thcre was 1998,with all subsequentcounts some drurnal variation ir abund:lnce of certain walerfowl species (see being lessthan half this Hamiltoo e/ al. 2002) so this approach was justifier:t.it was not number.The AustralianShelduck showedclear oeaksin appropriate to present a mean rvith an eslimare o{ error (e.g. standard abundancein Novembereach year (Fig.7). At this time crror or standard deviarion) for each dale sincc rhe five valuis for each many of the birds observedwere moulting and flightless. date were pseudoreplicatcsof a day. The Black Swan was most abundantfrom mid-summerto mid-autumn(Fig. 8). RESULTS Pink-eared Ducks were presentat the site mainly Small numbersof other waterfowl were seenoccasionally betweenApril 1999and March 2000, with peaknumbers in the sludy area.One or rwo individualsof FreckledDuci between Octoberand December, but therewas no evidence Stictonetta naevosa occurred on eight sample dates, and ol a repeated seasonalpattern over two years(Fig. l). For from one to five Blue-billedDucks Oxynraoustralis were AustralasianShovelers there was a tendencyfor highest seenon 20 sampledays.

€ oooo (n 5 5000 I qooo E j sooo

2000

1000

sFflBSS3E3338o>c.=>.=R:***i 6Zsis532sSS- Date

3500

3000

2500 o d) 2000 6 o o 1500 E : z 1000

500

0r 00 cD o) o) o)oo)o) 333333 ;Ftr z:=j 3095E9= Date

Figrrc 2. Searonolabrunlance o.f Australaian Shovelerat PondNine. Dark shadingrepresents daiLy nean abundance:white reprcsents daily matinlum abu||dan.e'LabeLson,x'tLtisdenotew|irstdayoI|he,no|l|h.The1irstsanple.la|eisinline\|i|h|||e,J'd|is'and|IrlaS|

1600

1400

1200 -u '1000 m b 800 o E 600 zl 400

200

0 qFqsssqsEooo ,l doi!*9rd"!!jE=:>= Date

Figtre 3 Seannal abundanrc of Grey Teal Pond.Nke. Dark shading 't .at reprcsent^ dailJ mean abunda ce: whire represe ts daih) ,naxirum abu danu. L'lhels ott L\is denote the dav of the nnnth. Tt 'y Jirst lirst sanole date ir in line with rhe tis, and the Iasri:; at the end oJ rtt! x .\i\. 64 A. Hamilton and l. Taylot: Abundance of waterfowl at waste stabilization oond Corella28(3) 1400

1200

1000 ! o E 800 a.) o 600 E f z 400

200

0 RPPA :B i.iYY 'a ;> Date

Figwe4.Seasonalabun.lanQoJ.Ches|nutTeatatPortNine.Darhadirrepresentsdoil!n|eanabundance;whilerepresen1sdailyn|axin|un1abundance' 't' Inbels ott 'y'ajj,ir, ,x @is denote the Jirst day of the tnonth.nrc rtN sample date ii in line n'ith the and the laj i; at ttrc etd ol the aris.

100

70 ! m 60 o 50 -oQ) E 40 = z 30

20 10

0 qsqss SSFBBq 3o9E s I aoi5sg= Date

Fi9ue 5 Seasonal abu,tdanrc of Padfc Black Du.k at Powi Nine. Dark shading fpresents daily Dredn abundance: white represents daiLy naximunt abundan.e lnbeb 'x' 'y' 't, on aris denote lE fnt .lay of tlrc nonth. The jtst su pte date is in line wittt the .Ltir, and the tast is a! the en(t of the stabilizationpond 65 September,2004 A. Hamiltonand t. Taytot:Abundance ol waterfowlat waste

600

500

€ 400 d] : 300 o -o E l z 200

100

6qs$ S E E E Date

Fi}urc 6. SeasoMI obundn .e of Hardhead at Poul Nik.. Dark shadnr represe ts d!1ily nean abundance; white rprsentr dail-'r marimum abundane 'r'aris 'y'ais, 'x'arirl Lbels on denote the Jirst dat of the nonth. nv frlt sample date is i line with the and tlrc la:t is at tlv etd of tlrc

1800

1600

1400

! 1200 ro o 1000 0) -o 800 E z 600

400

200

0 o)o) o) o) o) _! {jsc: Date

Figure7. Seasonilaburdance of Australian,shelduckat PondNine. Dark shadingrepresenrs daily meanabundance: whire rcpresents daily rnaximum 'x' 'y aburdanceLabels on axis denotelhe first day of the month.The first sampledate is in line with the axis,and rhe last is at the endof thc .x' A. Hamilton and L Taylot: Abundance ol waterfowl at waste stabilization oond Corella 28(3)

€ 800 m : : 600 (D _o E j z 400

0r (Il9o)o)o)orolo)6---@@ooo)o)o)o)oooo

u)z).2,.qiEeE:.iii5ta= Date

Fi8|$e8-Sea'\onnLabundankofBlackSwanatPo|ldNine'Darkshadingrepresentsdail!nvanabudanr;'|t1i|etep|e[ents(JaiIy|| 'x' lnbels on tL\i:t 'r' 'x'.i);ls. denote the f$t dd! oI the nontt. The jirst sanple d;u is i tine wih rle oxis, an(t the last ii at tlte end oJ ttte

DISCUSSION be eratic and dependenton inland rainfall (Frith 1982), this species most commonly breeds from August to ln general, the abundance patterns observed for November, often extending through to the end of summer waterfowlar Pond Nine were u;nsistent previous with (seeMarchant and Higgins 1990). observationsof dispersaland movementof Ausrralian waterfbwl. Pink-eared Duck, Grey Teal and Hardhead are ChestnutTeals have previously been reportedleaving the all considered to be highly dispersive species, with coast around early summer for inland breeding (Frith movementsmost likely tied to inland rainfall, which is 1982). However, breeding also occurs at pond Nine in typicallyerratic (Serventy 1953; Morgan 1954;Frith 1982; artificial nestboxes (Walker 2001). Despitethis, Chestnur Pert 1997).At Pond Nine, rhey all demonstratedinegular Teal broods were rarely seen on the water over the coune abundancepatterns. However, despite the apparent effatic of the study, and no more than two on a particular date. It natureof dispersalfor thesespgcies, some fends have been appearsthat soon after alighting from the nest box, adults identified. The reporting rates, documented in the Atlas of take their young to other nearby ponds (8. Walker, pers. Vctorian Birds (Emison", al. 1987),for all threespecies comm.). It is not known if these birds later return to pond are highestin springand summer.ln rhis currentstudy Nine, and if so, in what nurnbers.However. this mav Pink-earedDuck abundanceat Pond Nine peakedin spring/ explainthe peaksobserved in aurumn 1999and 200d. summer1999/2000, but not in the spring/summerof l998/ Chestnut Teals take about 60 to 80 days to grow to full 1999. Hardheadsdisplayed a similar parrern.Substantial size (Frith 1982),and in each year a peak in abundance numbersof both specieswere also observedin lat€ winter was observedapproximately three months after the end of 1998 at.the start of sampling.These comparable patterns the breeding season. mry indicatelhat simtlar mechanrsms*ere responriblefor Pacific Black Ducks are generally consideredto be drivingthe mo\ementsof thesetwo species.In contrast, mostly sedentary in areas of permanent water (Marchant spring/summer peaks were not evident for Grey Teals at and Higgins 1990). However,Victorian Atlas data reporting PondNrne. The Ausrralasian Shoveleris alsoionsidered rates are greatestln spring and summer (Emison el nl. to be highly mobile, wirh movementspossibly influenced l987). Each year at Pond Nine numbersdid not peakuntil by wJtera\rilability in south-eastern SourhAustralia or in the end of summer/startof autumn.lt is possiblethat the Murray-Darling Basin (Frith 1982). But unlike the recruitmentlrom a late winter ro errly springbreeding sceciesdescribed above, numbers within Victoria have season(see Crome 1986) accountedfbr some of this previously been reportedas being reasonablyconsistent increase.Only fully-grown birds were counted in this study. lhroughoutthe year with no clear seasonaltrends beins ;rnd immltures lrom spring breedinguould havereacheci cvidentrsee Mlrchanr and Higginsl900,1. ln rhe preseni f'ull size by late summer (65 days to full size, see Marchant srudl therewere laresummer/aulumn peaks in both vears. and Higgins 1990), yet as many as three broods were seen ll is possrblethrr rhesebirds pond arrived ar Nine afrer at any one time. Becauseof the low numbers recordecifor rreedinginiand. While breeding is generallyconsidered to this species,recruitment through local breedingcould have September,2004 A. Hamittonand L laylorj Abundanceof waterfowlat wasts stabilizationpond

(1980). 'A of Lake Borrie,Weribee SewerageFarm had a significant effect on changes in numbers of fully- Elliget, M. study ai a waterfowl(Anatidae) refuge area' BSc (Hons)thesis' La Trobe grown birds. University,Victoria. Emison,W. 8., Beardsell,C. M., Norman,F- I ' Loyn' R H- and Australian Shelducks have previously been reported 'Ar,as S. C. (1987). of ViclorianBirds' (Depa(mentof permanent wetlands for moulting, during Bennett, migrating to Conservalion,Foresls and Lands and the Royal Australasian which they becomeflightless (Norman 1973;Frith 1982). OrnithologistsUnion: M€lbourne) It is believed that permanent water provides a refuge from Frith, H. J. (195?).BreedinS and movementsof wild ducks in inland predatorsduring this vulnerablephase. Aggregations of New SouthWales. C.S.I.R.O. wildl Re\' 2t 19 3l (1959).Thc ecologyof wild ducksin inlandN S w: ll moulting birds usually occur from around October through Frith,H. J. movemenfs.c.s.l.R.o. wildL Rer. 41 108-130. to Febrqary (Norman 1973; see Marchant and Higgins Frilh, H. J. (1982).'Waterfowl in Australia' 2nd edn.(Angus and 1990),and the samepattern was observedclearly in both Robcrtson:Sydney.) 'The yearsof this study. Many flightless birds were seenat Pond Hamillon.A. J. (2002). ccologyof waterbirdsat lhe Western Nine. Moult migrations are also well known fbr other TrcatmentPlant (Victoria),with particularreference to waterfowl'. PhD lhesis, Charles Sturl University,Wagga Wagta, New Soulh shelduck speciessuch, as the Common Shelduck ladorua tadorna (Morley 1965; Salmonsen 1968). In Australia, the Hamilton,A. J., Taylor,L R. andHepwo h, G. (2002).Activity budgets Australian Shelduck is the only species of waterfowl of waterfowl(Anatidae) on a wastestabilization pond at the Western known to form large moulting congregations(Frith 1982). TreatmentPlant, Victoria. grn! 102| 111-l?9 P (2004).Seasonal and diurnal For other species,the timing of moulting is not as well Hamilton,A. J., Taylor.L R. andRogers, patterns of waterbirdsat a wastc stabilisationpond. period, only a in abundance defined and is extended over a longer so Corclla 28. 43-54. small proportionof the populationis likely to be flightless Hepworth,G. andHamilton, A. J. (2001).Scan sampling and wate.fowl at any particulartime. This strategyhas probablyevolved activitybudgcl studies: design and analysisconsiderations S€ftdviorr' in responseto the unpredictablenature of the inland 138:l39l-1405. (1961). westNew SouthWales tn! temporary wetlands used for breeding. If these wetlands Hobbs,J. N. The birdsof south 61:21-55. were to dry up while most birds were flightless,a large Lane,B. and Peake,P (1990).'Nature conservation al the Werribee proportion of the population could die. Australian TreatmentComplex'. Rep. No. 9l1008.(Melboume and Metropolitan this risk by using permanent Boardof Works:Melbourne.) Shelducksappear to avoid 'Handbook wetlands,such as thoseat the WTP, to moult. Marchant,S. andHiggins, P J. (1990). of Ausualian,New Zealandand Antarcticbirds. Vol. I, Ratitesto Ducks'.(Oxford Black Swan populationsaround permanent wetlands are UniversityPressr Melboume.) generally believed to be mostly sedentary,and no seasonal Misscn, R. and Timms. B. (1974). Seasonalfluctuations in waterbird populationson three lakes near Camperdown,Victoria. Arrl. Bird variation in abundanceis evident in Victorian Atlas wakher 5t 128 136. reporting rates (Emison e/ at. 1987). However, increases Morgan,D. C. (1954).Seasonal changes in populationsof Anatida€at in numbers on particular wetlands in south-eastern lhe Lavcrtonsaltworks. Victoria. 1950 1953.Ile Enu 54'.263-278. Australia in late summer/early autumn have been reported Morley,J. V (1965).The moult migrationof Shelduckfrom Bridgewaler on severaloccasions (Hobbs 196l; Missen and Timms B^y. Btitislt Biftls 59: 141-14'7. Normin, F l. (1973).Movem€nt and morlalitypattems o[ Black Ducks 1974; Davis and Reid 1975), and this was also clearly and MountainDucks bandedin Victoria.Pfo.. Rotal So.. vic. 86'. apparentin both yearsin the presentstudy. The ecological t-14. 'Changes significanceof thesegatherings is not clear.Black Swans Pert, P (1997). in abundanceof GreyTeal (Anar Srdci&.r)in were only occasionallyseen breeding at Pond Nine (less two salinehabitats in southemVictoria'. Masters thesis, Charles Slurt than five breeding pairs each year), and thus recruitment Universily,Wagga Wagga, New SouthWales. RamsarConvention Bureau (1984). Conventiooon wetlandsof through breeding would have made negligible contribution intemationalimporlance especially as waterfowlhabitat. Proc. Second to the large numbersof birds seen there. It is possible Confercn(eof the Parties; Groningen,Netherlandt, 7 to 12 May though that breeding could occur elsewherewithin the ./98,1.(lnternational Union for Conservationof Natureand Natural WTP Resources:Gland, Swjtzerland.) Salmonscn,F (1968).The moulr migration.Wildfowl lgt 5-24. ACKNOWLEDGMENTS Serventy,D. L. (1953).The soulherninvasioo of nor$em btrdsduring 1952.WA. Nat.3| l'77-196. We ackoowlcdgethe linancial 'Ten support of lhe Johlstone Cenlre, Walker,E. (2001). ycarsof researchat Lake Borrie'. Workersfor CharlesSturt Univcrsity and the cornmenrsaod suggestionsfrom two Wellands:Nest Box ResearchProgram. Special Edition. Rep. No. 12. anonymousreferees, (GeclongField and Came Incorporated:Geelong.) REFERENCES

Braithwaite.L. w' and Frirh, H. J. (1969). Warerfowl in an inland swamp in N.S.W: III breeding.C.S.l.R.O. Wikjl. Rsr. 14: 65-109. Braithwaire, L. w' and Stewarr, D. A. (1975). Dynamics of water bird populations on the Alice Springs Sewage Farm, N.T. Arrl. Mldl. R?r. 2:85-90. 'lnformarion Cartwright, D. (1996). (titerarure)review of wasrewater lagoons, wilh an emph$is oD the Wesrem Trearment plant, includins Lrfte tsome. (Melboume W.rLer:werribce ) Crawford, D. N. (1979). Walerbirds: indices and ftucluarions in dry_ \e3sonrefuge rr(as. NoflhernTerriror). Ari\r. Uitdl. Rpr 6: 97.10i Crome. F H. J. (1986). Ausrralian warerfowt do nor necessarily breed on a nsrng w er te\el. Au.'t. wikL R"r. 13: 461 _480. Davis, W. A. and Reid, A- J. (19?5). The birds of rhe Somers,Sandv Poinl. HrstrnBs Djsrricls. Wrstcrnpo Bay. \iclo i, Au5lrdli; Victorian Omithological Rescarcb Group Wesrernport Reporl No. I Part 3. Vic. Nat. 92t 60-70.