On Tue, May 28, 2013 at 10:31:45AM -0400, Bert Harris wrote:
> Andrew Taylor raises an interesting point, but a separate analysis we
> did on one prominent Kangaroo Island endemic, South Australian Glossy
> Black Cockatoo, suggests that some subspecies could be in trouble even
> if other subspecies (of the same species) are found further north. We
> modeled the range of the cockatoo's obligate food plant Allocasuarina
> verticillata based on the plant's range across South Australia and
> found that its prefered climate will contract substantially on Kangaroo
> Island, which will likely threaten the cockatoo.
Actually the modelling in your paper has exactly the same flaw. I'm no
botantist but when I read A. verticillata grows from coastal shale near
Sydney to rocky hills out towards Cobar, and papers say it "tolerates a
broad range of environmental conditions" and its been grown overseas in
places like Israel & Egypt, I wonder why a 1-2C warming of Kangaroo's
Island temperate climate would have the impact you describe.
The reason would seem to be your modelling excluded A. verticillata's
distribution outside SA. I looked at the citation you give to support
this choice, and it definitely does not support your decision saying:
"More generally, studies aimed at modeling species distributions with
climatic suitability models should consider global ranges, or at least
ranges across large biogeographic units (such as the Palaearctic),
otherwise they can lead to unfounded conclusions"
And this is also why i think Garnett et al.'s headline claim that the
"climate space" of 106 bird taxa will be gone by 2085 is similarly
It is possible climatic tolerance may not be uniform within
A. verticillata, so using the distribution of the entire species might not
accurately reflect the Kangaroo island populations climatic tolerance.
But given the large effect this choice has on the modelling I'd want to
model several distributional choices and look elsewhere for information
as to the plausibility of the choices.
Even if you had good reason to believe the SA population has a different
climatic tolerance, modelling an envelope derived from the entire species
would provide insight to the feasibility of introducing A. verticillata
from other populations as a conservation measure for the Kangaroo Island
I don't know what the future holds for the tiny population of Kangaroo
Island Glossy-blacks, but I don't think your modelling yields useful
information about how many Allocasuarina verticillata they'll have to
feed on in upcoming decades.