On Sat, Jun 20, 2009 at 08:25:30AM +1000, L&L Knight wrote:
> ScienceDaily (June 19, 2009) ? Climate change is altering North American
> winter bird communities in ways that models currently favored by
> ecologists fail to predict.
Attribution in studies like this should be treated with great caution.
For example, both temperature and Crested Pigeon abundance in inner Sydney
increased over the same time period (1975-2000). This association is
insufficient to attribute the growth of the Crested Pigeon population to
global warming because there are plausible alternate explanations, e.g
colonization facilitated by past vegetation modification. Secular trends
present a particular problem as they can be found in many variables,
Sydney house prices increased 1975-2000 too, but probably didn't affect
Crested Pigeon numbers.
The study is based on Christmas Bird Count data largely from the NE US.
Having participated in a couple of NE US CBCs in 1980s its seems very
plausible that climate change, particular snow cover reduction, could
have be important to birds, but there are other plausible factors: e.g
changes in supplementary feeding or environmental pollutants such as DDT.
The authors exclude some species (Turkey, House Finch, Bald Eagle,
Hooded Merganser, Cooper's Hawk, Evening Grosbeak) because they have
shown such large population changes that they assume factors other that
climate change must be the cause. This suggests similar factors may have
smaller effects on other species which aren't excluded in their analysis.
The inconsistency of their observations with ecological expectations for
warming climates can be taken as evidence that the expectations are wrong
but conversely it could be taken as evidence that their attribution of
the observations to warming climate are incorrect.