Kangaroos - The 100 Days Project: Day 39 
[reference Day 68: ‘The Numbers Game’]
Photo: Andras Berkes-Brandl
aerial surveys are undertaken with the use of fixed-wing aircraft and, in Queensland and New South Wales, also with helicopters. The fixed-wing aircraft technique is known as strip transect sampling. Two trained observers, travelling at 185 kilometres per hour at about 76 metres (250 feet) above the ground, count the animals within a 100- or 200-metre wide area on the ground that is delineated by streamers or fibreglass rods attached to the wing struts on either side of the aircraft and trailing parallel to the fuselage. The observers count in 97-second units, with a seven-second break in between counts, giving a sampling unit of half or one square kilometre
– Stephen Jackson and Karl Vernes, Kangaroo (Sydney: Allen & Unwin, 2010), 176/77.
The ‘national’ kangaroo and cull quota figures aren’t really national. They exclude Victoria, Tasmania, the Northern Territory and the Australian Capital Territory. True, figures are quite low in Tasmania and the A.C.T. (see the posts on the Forester kangaroo, and the Canberra Roos), and lowish in Victoria, though culling still takes place there. The exclusion of the Northern Territory, with its significant population of Red kangaroos, Antelopine kangaroos and Wallaroo/Euros is another matter. Some have suggested that an addition of 25-30% to the overall kangaroo figures might be appropriate.
The ‘national’ figures as they are presented, moreover, are based upon submissions made by the participating states. Each state, presumably, does its own count – or arrives at its own estimate – and submits this to the Commonwealth Government. To arrive at a better understanding of the final figures one would need to know whether each state followed the same counting/estimating procedures, and, if not, what the variations in procedures were, whether they followed their own procedures in finalising and submitting estimates, or whether and to what extent they followed protocols given them by the Commonwealth Government, and whether, once the figures were submitted to it, the Commonwealth made any adjustments of its own.
One might think that counting/estimating procedures were relatively straightforward, but to do, annually, a full count of all kangaroos in each state would be a very expensive undertaking indeed. One can’t imagine that any of the participating states would commit themselves to it. The method instead seems to be to select certain regions within the state – relatively small regions, in the overall, typically called ‘Kangaroo Management Units’ (KMUs) – and to then expand/multiply/average-out the counts from those regions to the rest of the state. The manner in which this is done, and the factors taken into consideration as it is done, are crucial. Clearly, as I’ve said before, there would be no point in ‘counting’ the kangaroos in a region where there are none, and so regions where kangaroos are a countable presence will be used to determine the ‘population’ of regions where they are less present or not present at all. Are these areas where kangaroos are not present – locally extinct – taken into consideration? How? Clearly, not to take into consideration areas where kangaroos are less present or not present at all will lead to grossly inflated population figures.
Clearly, too, KMUs themselves will not all cover uniform terrain. A typical KMU may have areas of open grassland, lightly wooded areas, perhaps areas of heavier forest cover. You might see most of the kangaroos in an area of open grassland, but you’ll miss some in the wooded areas. Counters will therefore employ a ‘correction factor’ for such areas: you might multiply the kangaroos seen by two (a factor of 2.0), you might employ a factor of 5.5. How are these factors determined? You might determine that the seven roos you have seen mean that there were fourteen there; you might determine that there were really thirty-five; but perhaps there really were only seven. It’s rumoured that the Commonwealth Government adds in its own correction factors. What and why are these?
So too with the benchmark percentages the Commonwealth uses to arrive at sustainable culling/‘harvesting’ quota figures. Kangaroos face far too many adversities – the danger of collision with vehicles driven by humans, hunting by humans, whether or not for commercial purposes, drought, diseases, predation by raptors, foxes, dingoes – for there to be anything like a 100% survival rate from new-born into adulthood. What would be an acceptable survival rate to bear in mind as one anticipated the growth, or otherwise, of a population from year to year. The Commonwealth Department of Environment and Energy, in its concern to be operating on figures which are sustainable – i.e. will not put at risk, by overkilling, the survival of the species as a ‘harvestable’ concern – appears, in one of its columns, to operate on a figure of approximately 15%, though in another of its columns, where it estimates overall population growth rates, it manages to arrive at a much higher figure than this (36.3 and 34.9% respectively for the years 2011 and 2013). A glance at the Kangaroos at Risk website, where this question is surveyed reveals wide discrepancies, even amongst experts, as to what are reasonable presumptions of survival rates, and, most interestingly, some of the fundamentally impossible year-by-year population explosion rates put forth for some individual KMUs. To give you a taste, such as might lure you to the site:
Any reported population growth rates over 20% seem biologically unlikely, however the summary above provides “estimates” which represent serious anomalies in the data set, which are discussed nowhere. 100% population growth rates – higher than can be achieved by feral goats and feral pigs (discussed previously) – are biologically impossible. 270%, 209% and 313% growth rates are patently absurd, yet these figures are not even questioned, much less discarded. Would the researchers think that all the kangaroos had twins perhaps, including the males? Or are these data sets simply invalid? Why have these values never been examined?
There do seem, all this is only to say, to be some serious anomalies, both within the ‘system’ and in the field. Anecdotally – accounts by individual/private visitors to areas which the state or Commonwealth Government says are heavily populated by kangaroos, saying that they have found very few if any kangaroos at all – but also in terms of official records: a recent study of chiller box deposit records, for example (roos shot for the ‘meat’ industry must be taken to chiller boxes [dotted throughout each state] for subsequent collection and processing), suggests that the number of deposits to those boxes by kangaroo ‘harvesters’ has been dropping, and that shooters are having to travel further and further afield in order to find the kangaroos in the first place. It’s also been suggested that this drop in chiller box deposits reflects a decline in the number of roo hunters, i.e. that there is less and less appeal in the profession, but of course these things are more likely to be connected than not. But these things are local, and are at every point contradicted by accounts (usually from newspapers) of/from landholders that they are ‘overrun’ by roos.
Plenitude on paper, scarcity in the field: what’s it to be?
I wish it were a simple matter of finding someone from within the system who could explain that system clearly to me, but with, let’s not beat around the bush, the murder of millions at the heart of the issue, who on earth to trust?