The Shrike
Philosopher
As I first noticed in this story on the BBC, some researchers in the US state of Tennessee have just released a startling paper in Current Biology on the alleged behavior of songbirds to pre-emptively leave the path of oncoming tornadoes.
Background: Golden-winged Warbler is a species of conservation priority in the US, so there has been a great deal of attention paid to this species' habitat needs, reproductive success, and migratory connectivity. The latter refers to identifying specific linkages between breeding and wintering territories that are often several thousands of km apart.
Tracking 9g-birds over such distances is tricky business, so many researchers attach small geolocators to the birds. These are much lighter and smaller than satellite or radio transmitters. The devices record daylength; researchers match the data to geographic locations to infer the location of the subject. One big drawback is that the device actually has to be physically recovered by catching the bird again to download the data.
In the linked article, the researchers determined that the warblers had just returned to their breeding territories from wintering areas in Colombia this past April. Shortly after arriving, however, they abandoned their territories, flew hundreds of km back south again (to the Gulf Coast), and then returned north to their territories a few days later. This behavior coincided with a massive outbreak of tornadoes in the central US and, incredibly, the birds left before conditions deteriorated in the study area. The authors concluded that the birds detected tornadoes 1–2 days before the storm made its way to the study area, and that they detected the tornadoes via infrasound.
1) IF the behavior actually happened (rather than some malfunction or misreading of the geolocator data), this would truly be an extraordinary finding. It would signal far greater plasticity in migratory behavior than generally accepted, as well as indicate a heretofore undescribed mechanism (weather) affecting territoriality.
2) Given the massive spread of tornadic storms and the relatively tiny path width of a tornado, I don't see it likely that tornado risk could have been a selective pressure sufficient to induce such behavior. Hail damage seems actually far more destructive and widespread with such storms, and much more likely to kill a lot of birds.
3) Infrasound, schminfrasound: The authors have not ruled out other forms of detection to my satisfaction, e.g., polarized light, changes in atmosphereic pressure, odors, etc.
Your thoughts?
Background: Golden-winged Warbler is a species of conservation priority in the US, so there has been a great deal of attention paid to this species' habitat needs, reproductive success, and migratory connectivity. The latter refers to identifying specific linkages between breeding and wintering territories that are often several thousands of km apart.
Tracking 9g-birds over such distances is tricky business, so many researchers attach small geolocators to the birds. These are much lighter and smaller than satellite or radio transmitters. The devices record daylength; researchers match the data to geographic locations to infer the location of the subject. One big drawback is that the device actually has to be physically recovered by catching the bird again to download the data.
In the linked article, the researchers determined that the warblers had just returned to their breeding territories from wintering areas in Colombia this past April. Shortly after arriving, however, they abandoned their territories, flew hundreds of km back south again (to the Gulf Coast), and then returned north to their territories a few days later. This behavior coincided with a massive outbreak of tornadoes in the central US and, incredibly, the birds left before conditions deteriorated in the study area. The authors concluded that the birds detected tornadoes 1–2 days before the storm made its way to the study area, and that they detected the tornadoes via infrasound.
1) IF the behavior actually happened (rather than some malfunction or misreading of the geolocator data), this would truly be an extraordinary finding. It would signal far greater plasticity in migratory behavior than generally accepted, as well as indicate a heretofore undescribed mechanism (weather) affecting territoriality.
2) Given the massive spread of tornadic storms and the relatively tiny path width of a tornado, I don't see it likely that tornado risk could have been a selective pressure sufficient to induce such behavior. Hail damage seems actually far more destructive and widespread with such storms, and much more likely to kill a lot of birds.
3) Infrasound, schminfrasound: The authors have not ruled out other forms of detection to my satisfaction, e.g., polarized light, changes in atmosphereic pressure, odors, etc.
Your thoughts?
