Thanks for the replies and thoughts. I was motivated to think about it after noticing that the Narrabundah regular flocks overflying at dusk are absent now
most of the time and I’ve not noticed any roosting at Callum Brae (though they could still be doing so), and therefore I wonder about the extent to which the different observations in different areas of Canberra relate to: a) different birds, or b) same birds
moving around seen at different times. Ryu’s and Philip’s geographically separated birds
might still represent a single collection that gets around a lot. Or partially so, a disparate group that occasionally gets together for fiesta overhead my place.
The Telopea Park pack are certainly not there all the time, nor the Nbundah/Callum Brae birds, and I assume neither are the ANU roosters nor the Tuggeranong+
Fadden group(s). It is possible that one larger flock shifts multiple roost locations every now and again, and completes a regular circuit around significant distances most days, being seen in all the mentioned locations. Conceivably, they have moved
roost from Callum Brae + ?Telopea? to Fadden + the ANU in more recent times. As Philip mentions, it might be possible to deduce some such movements from the database, though somehow I think the available info would be too sparse to draw useful conclusions.
Unless we can do that, how do we know if there are “… surely simultaneous flocks around Tuggeranong, Lanyon, Woden, Narrabundah, etc.”?
Then I started wondering what I even meant by ‘the same flock’. Is a flock the group that roosts together, or perhaps birds from multiple roosts who gather
together to go marauding in the daytime?
Either way I’m not convinced about the answer to my central question which really boils down to how many corellas there are in Canberra. I’ve seen ~240 for
a while over Nbundah. Is that it? … 240 of them, seen in different locations regularly (either all together or split into subgroups), or are there several times this number keeping to several different roosting and marauding areas?
It’s the same question that was posed by the Gang-gang survey (and by John Leonard I think before that) about Gang-gangs … and the same difficulty in measuring
absolute numbers of very mobile birds, even when they are geographically restricted and highly conspicuous.
As an amateur researcher possessed of more curiosity and armchair-dreaming than field effort, I can think of three ways of answering the question:
1. If you could identify all roost sites you would have the answer, but I don’t know how you would ever demonstrate completeness. There could always be one
or many roost sites of which you were unaware, and it’s not so easy to find these sites, especially sites used by small numbers of birds. (As an aside, I once observed two SC cockatoos fly into a quite sparse eucalypt nearby at dusk. Watching without a break
I walked right around this tree but never found either of the birds again who were no doubt still there somewhere, settling down for the night). I also decided that census by roost sites was exceedingly difficult after Michael Lenz’s very interesting presentations
on his study of roosting Currawongs.
2. You could arrange to have a large number of observers at one time concentrating on the one species, and use times of their observations to prevent double
counting. This requires an enormous effort since it only works if you cover all the territory at one time and I don’t think is actually feasible other than for a very restricted area.
3. All in all, I can’t see a better way than the technique used for some blood counts and in other spheres including bird-banding – tag a known number of Corellas
or Gang-gangs and then measure the proportions of tagged birds found subsequently throughout the study area. e.g. if you tag 20 birds and find that subsequent observations over the study area settle consistently at 5% of birds observed with tags, you can
infer with reasonable accuracy that the total population is is 20/5% or 400. The greater the number of observations over time, the more accurate the measurement. The main limitations would be: a) knowing how many of your tagged birds still survive over the
years and remain in the study area, so you’d need to keep track of individual labels as well to try to answer that question; and b) it only really works for geographically restricted, or at least sedentary populations.
This (above) is my application for a moderate PhD grant, and I’m ready to start any time
J. Seriously, I think it’s an interesting question. With all due respect to many years of survey
effort, we still don’t really have a clue how many corellas or gang-gangs there are in our area (or any other species for that matter, in absolute terms), though I look forward to reading the GG Survey report due soon that might throw some light on the matter.
There would most certainly be numerous flocks of corellas plus others.
From my obs, the largest congregations seem to be in Tuggeranong, Inner north, and the lot near Callum Brae. I am led to believe Corellas are quite few in Weston Creek and
present in mush smaller groups in Gungahlin and Belconnen.
There is a roost flock in Fadden with the count standing at 190 corellas near the end of January. It has been interesting to observe Sulphur-crested Cockatoo numbers roosting
at the site drop to about 25% of the original number whereas corella numbers have at least doubled, although there seems to be seasonal fluctuations.
