Table Of ContentADDITIONAL COUNTS AND RECORDS OF FLOCK
COMPOSITION OF CARNABY'S COCKATOO
(CALYPTORHYNCHUS LATIROSTRIS) AT
TWO OVERNIGHT ROOSTING SITES IN
METROPOLITAN PERTH
By P.F. BERRY
Honorary Associate, Western Australian Museum
([email protected])
and
M. OWEN
([email protected])
ABSTRACT
Daily counts of Carnaby's Cockatoo from April 2006 to
August 2009 at an overnight roosting site centred at
Hollywood Private Hospital in the western suburbs, were
made over a period of 41 consecutive months. Data from
the first 13 months were published in 2008. Data from the
28 subsequent months confirm a clear seasonal trend in
abundance. Highest numbers were recorded from February
to June followed by a rapid decline in July with no birds
roosting in September and October. Thereafter there was a
progressive build-up in numbers throughout the summer
to a peak in February, with maximum numbers of >400
birds.
Counts made at the only other known roosting site in the
western suburbs, centred at nearby Perry Lakes, indicate
that from February a progressively increasing proportion of
the local cockatoo population roosts at the Perry Lakes site
in autumn and winter rather than at the Hollywood site.
The total numbers of birds remaining in the western
suburbs from February to July is thus higher than
previously reported based only on counts at the Hollywood
roosting site. The mean numbers at both roosting sites
combined show that a population of around 250-300 birds
is present for at least four months of the year (March to
June) and over around 150 birds for at least 7 months (Jan.
to July). However, between March and June maxima ranged
between about 400 and 500 birds.
The additional data confirm earlier findings that the flocks
27
of Carnaby’s Cockatoo comprised a mean of approxi¬
mately 60% pairs and 40% triplets (pair and fledged
young that was still being fed) over the year. However, the
data for 2007/08 and 2008/09 show a progressive
monthly increase in the proportion of pairs and corre¬
lated decrease in proportion of triplets from December to
July/ August. This may reflect an exodus of parents to
breed, leaving behind juveniles which then form non-
breeding pairs.
INTRODUCTION in its non-breeding range in
metropolitan Perth, is the first
Carnaby’s Cockatoo
attempt to quantify its
(Calyptorhynchus latirostris) is
abundance on an ongoing basis,
endemic to the south-west
with daily counts now having
region of Western Australia. It is
been made over 41 consecutive
considered Endangered under
months at a habitual overnight
IUCN criteria (Burbidge 2004)
roosting site in Hollywood
and is currently listed as a
(Nedlands). The objectives are to
threatened species under State
provide a quantitative record of
and Commonwealth legislation,
its status and to contribute to
due to an apparent rapid decline
knowledge about its resource
in its abundance and distri¬
requirements and behaviour,
bution associated with land
particularly in relation to
clearance in the wheatbelt and
natural and man-made changes
the Swan Coastal Plain. It breeds
in the urban environment of
largely in the wheatbelt and
Perth’s western suburbs.
moves to the Swan Coastal Plain
Daily counts of Carnaby’s
to feed in the non-breeding
Cockatoo made from April 2006
season (Davies 1966; Saunders
to April 2007 at an overnight
1977, 1980, 1990; Saunders and
roosting site centred at Holly¬
Ingram 1995, 1998). Clearing of
wood Hospital in the western
native vegetation on the Swan
suburbs, as well as concurrent
Coastal plain, particularly of
records of flock composition,
Banksia/Tuart Woodland used
have recently been published
by Carbaby’s Cockatoo for feed¬
(Berry 2008). Additional daily
ing, has been extensive in recent
counts from the Hollywood site
years and continues at an
made between May 2007 and
accelerating rate. Government of
August 2009 are reported here, as
Western Australia (2000) esti¬
well as counts made on between
mated that 28% of the original
4 and 15 days a month from a
vegetation of the Perth Metro¬
second roosting site centred at
politan Region remained in 2000.
Perry Lakes that were started in
This would have declined since
February 2008. The Hollywood
then.
and Perry Lakes roosting sites are
This study of Carnaby’s Cockatoo, 2.7 km apart and are the only
28
known roosting sites in the counted at the Perry Lakes site
western suburbs of Perth. The with a corresponding decline at
total number of birds roosting at the Hollywood site. This indicates
both sites is thus thought to that a pro-gressively increasing
approximate the total number of proportion of the local cockatoo
Carnaby’s Cockatoo in the population roosts at the Perry
western suburbs on any day. Lakes site in autumn and winter
(Table 1, Figures 3 and 4) and that
the population remains in the
METHODS
western suburbs in greater
The methods used to count numbers for longer than had
Carnaby's Cockatoo are as been thought, based on Holly¬
described previously (Berry 2008). wood site data alone. The mean
numbers counted at both
roosting sites combined indicate
RESULTS
that a population of around 250-
The clear trend of mean and 300 birds is present in the
maximum seasonal abundance western suburbs for at least four
recorded in 2006 and 2006/07 months of the year (March to
was repeated at the Hollywood June) and over around 150 birds
roost site in 2007/08 and 2008/ for at least 7 months (Jan. to July)
09 (Table 1, Figures 1 and 2). After However, between March and
a rapid progressive increase June maxima ranged between
starting in November, mean about 400 and 500 birds. (Table
monthly numbers peaked in 1, Figures 5 and 6).
March or April (222, 291, 197, & As in 2006/07 the composition of
262) followed by a rapid decline flocks in 2007/08 and 2008/09
with lowest numbers of birds consistently comprised an annual
recorded from August to mean of approximately 60% pairs
October (Figure 1). Maximum and 40% triplets (pair and fledged
numbers of birds were recorded young that was still being fed)
at the Hollywood site in January/ (see Table 2). A variable low
February (331), March (676), April proportion of single birds
(479) and May (450), (Figure 2). In recorded is thought to represent
August 2009 there was an individuals that are temporarily
apparent influx of birds to the separated from their mates or
western suburbs that roosted at parents. However, on a monthly
the Hollywood site. basis an increase in the percentage
From February 2008, additional of pairs and correlated reduction
counts were made on between 4 in percentage of triplets was
and 15 days a month by Margaret evident, particularly over the
Owen at the roosting site near months when a good sample size
Perry Lakes. Monthly mean and was achieved. This trend is
maximum counts over both years particularly clear in 2008/09 (see
show an increase in numbers Table 3 and Figure 7).
y Lakes <s3x O O —’—!' CaOMdN CM rrO~-L CO10M OCCCMOON 0in Osn NOOON CCMO X0 NdCMO kets) 200(5) 299(9)
r c
9) and Per "3 r0CCl aind 0rCnM0 t't--4 CnMOo0 N0O ICCCnMO- CCrrnMO- 140.90 149.1 csT XCM O07N- iOnN sites in bra 269(12) 235(8)
ust 200 0G3 n0rOn xv0Od in O0CO0N mCcOMl'. mumn X 0rrnn NCCMMO NOvOOn 0innO C<ICnMMn 0rmCnM MCmM y Lakes 284(7) 510(13)
g r
u er
2006-A s>CO• Ou0CMi CCMM xo 0aCMd0 O<O>-tNNn- rCrmCnnMM CiCcrnnMM7 0CicrnmnM0 N'CwMO' vnOO Xrn SCCMM' pcOCMNi OrXC<Mn d and P 405(8) 486(15)
o
unts of Carnaby’s Cockatoos at overnight roosting sites at Hollywood (Aprilugust 2009). (Numbers in brackets = days counted). -Xzc<2POct Jan *nuc40O>0a43tCm-o2->2m OCCoMM1111111 i x0NOOCOrOCOamOi0rd0nnnMMddON O0NxOm\CiCrOO102.1 dnnMMdNO} nOOCrOOO—CC—OCrmr119.1 nOMnnMMMNNn« N_’T3xTTXZ*uJw’Cc>cO0CO’>£•auuQOO£OcCccC0OOcOUacc£i—QCO—-SJ33o2OoO:>rjS>~<ii i JiOCnnM1111111 ' rNvCNrC0rOCCCCrcOcCCrcXmCO4(31) M=OnnnMMMMMMMoOO0cood"r OCCiC§riCrCCON0F00rrCrn•nnnMnnnnMM"MONO1o^'1f^*t_' - N0CICOJCrcCCnmICiC0m(cid:9632)OOmOMT0(26) —nnMnnOMMMMMnOONNf--"- - NOOrmCiinnnMON11111 ' 0N<CO0dCir00007.3 nnMM00ON0 -NZJ3JJTTJii•'CO0s0>OOOCc00o0£50C•7OUu0ck0££707OOO1—S-OO5t>32*2323.33J)S3S-.jJJ1).• ii arNCO0OnOnO10ON1111d l COrCOc2COOOrCO-rTtnMMMnM~-n- sG0’07S223i OC0rOclCOcCMM01--11l1n ii - NirCmOj0O»rOmC0ri40.3 0(2) nZnnMnnMOn)'' •02S i ers of coincident days each month on which there were counts at the Hollywo _____ (3) 392(7) 394(8)331 0(3) 0(2) 34(2) 69(4) 254(6) 368(4) 473(8) 506(11)
oA c c 73 73 b
ost c008- <73v 2XCO 27032 X^3 _C+mJ + Num
Table 1. Ro(February 2 —£XT7O£O3: xCO0OM c000CM- 00OOCM0 O0O000N01 I•r"00>r>OT"*u nOOrMO CcO00Co-l Cx000M OO000CMN0 cs>030-232- 0000CM0 OOa00Cdl' -—>•CC0J2O22•5 XCO00M OXO00CNl T•X>OOO—3-* X0OCOl COOC00MOn XT*OO£53 Maximum (2007/08 2008/09
30
Figure 1. Mean numbers of Carnaby's Cockatoo at the Hollywood roost site over
41 consecutive months.
Sept Oct Nov Dec Jan Feb Mar Apr May June July Aug
MONTHS
Figure 2. Maximum numbers of Carnaby’s Cockatoo at the Hollywood roost site
over 41 consecutive months.
31
300
Figure 3. Mean numbers of Carnaby’s Cockatoo at overnight roosting sites at
Hollywood (September 2007-August 2009) and Perry Lakes (February 2008-
August 2009).
Sept Oct Nov Dec Jan Feb Mar Apr May June July Aug
MONTHS
Figure 4. Maximum numbers of Carnaby’s Cockatoo at overnight roosting sites
at Hollywood (September 2007-August 2009) and Perry Lakes (February 2008-
August 2009).
32
400
Figure 5. Mean numbers of Carnaby’s Cockatoo at Hollywood and Hollywood
plus Perry Lakes combined in 2007/08 and 2008/09.
MONTHS
Figure 6. Maximum numbers of Carnaby’s Cockatoo at Hollywood and at
Hollywood plus Perry Lakes combined in 2007/08 and 2008/09.
33
90
Dec-05 Jan-06 Apr May-06 Jun-06 Aug-06
Months
Figure 7. Monthly proportion of pairs and triplets (parents and juvenile)
discriminated within flocks at the Hollywood roost site in 2007/08 and 2008/09.
Solid points are months with fewer than 50 pairs/triplets recorded i.e.
comparatively small samples (see Table 3).
Table 2. Mean annual flock composition of Carnaby’s Cockatoo at the
Hollywood Roost site
YEAR mean % mean % mean % with n (singles, pairs
single pairs juvenile or triplets
birds (triplets) counted)
May 06 - Apr 07 4 58 38 3859
May 07- Apr 08 3 60 37 1930
May 08 - Apr 09 2 60 38 1774
May 09 - Aug 09 3 69 28 810
That Carnaby’s Cockatoo un¬ and secondary feathers were
dergo possibly a complete moult collected between January and
while on the Swan Coastal Plain May beneath the roosting trees at
was confirmed photographically the Hollywood site for DNA
and is demonstrated by the ease analysis.
with which 64 rectrices, primary
34
e
h
t
t
a
g
ostin <530 <530 <530 <530 _rX Rn<n r<^)n Ocgno'
o
r
s
k
oc r>: m O — oo T} ^ oo
fl O r\0i nK -^ t^idn oin S uX OvO' OOJ' ^(N
n
i
h
t
i
w
Q\ (N O' R °9 *0 i- in 9 >n co e in P CN <N
ed u<-x \o£) roroi VO <n co cn ^ 2 ?n Jr' vVOO om CoN
t
a
n
i
m
discri ^O'— ' RNiTL O(N) vO* oo o. O Srt ' ^^ nmt-"* uCNi ^O'
)
e
nil
uve <a <a or1 - ^Tint- m>oo0 Oo<n' <a ^n ohm ^^^ Ovo' <a o ^in n; 5ID
j
d
n
a
ents cino ^tO ^^Tt- V2O^ mRin Om^ ^C~N CN ci^nd n\O£)' ONO-'
r
a
p
(
s
t
ple oo i' croi l>9n * X*n) JoL o « ^O' n
i ^ in <n 22 04 in ^ Pl ^ in «n fo
r
t
d
n
airs a2009. v^^O "minT ^on 'oOo ^(cid:9632);j CiTnTt"n ^Rt vc^Oo vx°O icinnn C(cid:9632)ttN oo^o
pg
ds,Au
r-
ngle biy 2006 oRi JoO vmO o^ dI - [P$ [2 ^ o^ xO dtOo T[f ^
siMa
ofe
s sit TO
berost jaC—j>
um ro Po
able 3. Nollywood X£O</>X '»a3-- --1"Ccu2L Xaa£^-3r) 42 «CL 2*a-j -5^9 XTcox)j i£C2L R£oj JT<c0Lx) a q
TH c£ ^ Z
35
DISCUSSION the Hollywood roost. This large
flock was seen feeding most
Up until February the mean
evenings on the ground in
numbers of Carnaby's Cockatoo
Karrakatta cemetery on Norfolk
counted at the Hollywood
Island Pine (Araucaria
roosting site are thought to
heterophylla) seeds which were
approximate the size of the
present in great abundance (2009
population present in the west¬
appears to have been a mast year
ern suburbs. Thereafter, when
for seeding, R. Dixon, pers. com.).
they start to roost at Perry Lakes
They also regularly fed on
as well, the mean counts at the
two roost sites combined better Tipuana (Tipuana tipua) and
Banksia praemorsa seeds in the
indicate the total size of the
cemetery.
population present. Maximum
counts are probably swelled by The most plausible explanation
transient flocks, but the pro¬ for the progressive monthly
longed period (March to June) increase recorded in the
when between about 400-500 proportion of pairs is that it is
birds are present means that food associated with departure of
resources are required to sustain parents to breeding sites, leaving
populations of this magnitude. behind juveniles which then
The consistency of the numbers form non-breeding pairs. In
of Carnaby’s Cockatoos recorded 2007/08 and 2008/9 there is a
in the western suburbs sub¬ sudden decrease in mean and
population also indicates that maximum numbers roosting in
there is a correlation between the western suburbs in April and
numbers of birds present and May respectively marking the
food resources available. Pre¬ beginning of an exodus (Figures
sumably a balance is readied 5 and 6). This correlates with an
between food availability and increase in the rate at which the
numbers of birds supported. proportion of pairs starts to
Why otherwise are flocks of increase (although this is in April
thousands never recorded in the in both years, see Figure 7). More
western suburbs as they are work needs to be done on the age
regularly at the Gnangara pine structure of birds within pairs
plantations (Johnstone and and triplets to confirm this.
Kirkby 2008)?
The decline in Carnaby's
In August 2009 the increase in Cockatoo has been attributed
the number of birds recorded largely to loss of feeding habitat
roosting at the Hollywood site is available to the breeding popu¬
interpreted as an influx to the lation in the wheatbelt (Saunders
western suburbs. Why they 1977, 1980, 1990; Saunders and
roosted at Hollywood and not Ingram 1995, 1998a, 1998b).
Perry Lakes is unknown, but the However, the most rapid and
most likely explanation is that extensive clearing of the wheat-
their food resource was closer to belt occurred in the 1930’s and
36
