6 ms·

Just to challenge your scepticism a bit > Fourth, roughly 30% of those estimated animals are birds, of which other experts say 99% would survive because, ya kn

by Cogito 7y ago

Just to challenge your scepticism a bit > Fourth, roughly 30% of those estimated animals are birds, of which other experts say 99% would survive because, ya know, they can fly away. Of the other 70%, many others would have just walked or slithered or run and gotten away from the fire. The fires we're talking about here aren't ones you can easily get away from if you're caught by them. If they're near a firebreak of some kind (large river, cliff, etc) the animals may be able to escape, but otherwise these are very fast moving and very hot fires. It's reasonable to think that for animals caught up in the fires, at least for ground dwelling animals, more were severely impacted than were not. I'm not sure about birds. Where did you see the 99% number? The fires can get hot enough to kill you from 100m away, cause firestorms, and are massive. A lot of birds would have needed to fly a very long way to be safe, and I've no reason to think that they either could or did.

jcranmer 7y ago
The intellectually honest thing to do here would be to either try to find some precedent study of wildfire mortality for wildlife to get a coarse estimate on what happens, or to admit that you don't know. Instead, we have an "affected" number that is technically not presented as "killed", but we-kinda-sorta-hope-you-interpret-it-as-such. I'm personally frustrated that no one has been able to dredge up an actual study on wildfire mortality patterns.
- Cogito 7y ago
  I couldn't find the publication myself (though I now see others have posted it in other threads [0]) so I emailed Professor Dickson to get his input. In the meantime I looked for some studies, and found two that are interesting and include some background on the subject. The Responses of Mammals to La Niña (El Niño Southern Oscillation)–Associated Rainfall, Predation, and Wildfire in Central Australia [1] > The responses of vertebrates to wildfires are complex, reflecting the varying food and shelter requirements of different species, their exposure to predation, and other factors. Vertebrates that shelter above the ground and are not particularly mobile are often killed directly by fire (Chew et al. 1959; Newsome et al. 1975; Silveira et al. 1999; Simons 1991). Conversely, species capable of extensive movements such as large predators and birds frequently escape the flames and may increase in abundance after wildfire when food resources become available (Loyn 1997; Newsome et al. 1975).,Burrowing herbivores and omnivores are likely to survive fires and, depending on predation, may increase in abundance after postfire flushes of vegetation (Newsome et al. 1975). Postwild-fire decreases in the abundance of some species have been attributed to a lack of food resources and increased exposure to predation (Groves and Steenhof 1988; Lawrence 1966; Newsome et al. 1975, 1983). Immediate and short‐term responses of bird and mammal assemblages to a subalpine wildfire in the Snowy Mountains, Australia [2] > Over 35 days in January–February 2003, wildfires burnt across much of the subalpine/alpine landscape of south‐eastern Australia, including about 70% of the land above 1500 m in the Snowy Mountains. At the time of the fire, studies of two subalpine faunal assemblages were being undertaken. The opportunity to resurvey the assemblages was taken in order to examine the immediate impact of fire in an environment where it is uncommon but predicted to occur increasingly with global warming. A study area in the Whites River Valley, where the number of bird species was counted monthly from 1996 to 2003, and weekly in late winter–spring from 2000 to 2003, was burnt in one fire. As well as the expected decrease in the number of individual birds, the fire resulted in an immediate decrease in the number of bird species, unlike in previously studied montane forest, with only the regularly wintering species plus the olive whistler and the ground‐feeding flame robin remaining. During the post‐winter avian immigration, few regular spring migrants appeared on burnt sites despite their nearby presence on the unburnt side of the valley. Five of six small mammal trapping grids were burnt. As with fires at lower altitudes, there was an immediate reduction in mammal numbers on burnt grids following the fire, but in addition, one species, Mastacomys fuscus, declined further in the ensuing 2 months both on burnt and unburnt sites. Numbers of Antechinus swainsonii and Rattus fuscipes stabilized until autumn/winter when there was a further decline due to the unavailability of subnivean space to allow winter foraging, allied with a concentration of fox predation on areas still carrying populations of small mammals. [0] https://www.researchgate.net/publication/318029981_Impacts_of_Landclearing_the_impacts_of_the_approved_clearing_of_native_vegetation_on_Australian_wildlife_in_New_South_Wales https://www.researchgate.net/publication/318029981_Impacts_o... [1] https://academic.oup.com/jmammal/article/86/4/689/876336 https://academic.oup.com/jmammal/article/86/4/689/876336 [2] https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1442-9993.2006.01629.x https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1442-9993....
- nl 7y ago
  A fairly old book[1] has bird mortality rates between 40% and 54% for similar fires. [1] https://books.google.com.au/books?hl=en&lr=&id=4f8anIoS33MC&oi=fnd&pg=PA351&dq=Wildlife+mortality+in+bushfires+in+Australia&ots=kGisLYv0km&sig=81d1pW9C_IVGRyZ8fqJfmHCRHaw#v=onepage&q=Wildlife%20mortality%20in%20bushfires%20in%20Australia&f=false https://books.google.com.au/books?hl=en&lr=&id=4f8anIoS33MC&... (page 358)
- nl 7y ago
  Small mammal populations in a eucalypt forest affected by fire and drought [2009][1] tracks 5 species in a particular area from 1970 until 2005. Only the abstract is available (unfortunately), but by my reading of that it shows that a 1973 bushfire wiped out all 5 species in the area ("Following the 1972 fire, numbers fell to the lowest level recorded during the study and each population subsequently disappeared from the plot between the 1973 and 1974 winter censuses"). In further reading (I read a lot of abstracts trying to find one that had numbers!) it appears this pattern of some surviving the immediate fire but 18 months later having disappeared dur to predators and lack of habitat is fairly common. [1] https://www.publish.csiro.au/WR/WR08086 https://www.publish.csiro.au/WR/WR08086
  - wutbrodo 7y ago
    Thank you, this is _exactly_ the kind of analysis I'm interested in! (but unfortunately, far too nuanced to be comprehended or covered by media with any sort of reach). > it appears this pattern of some surviving the immediate fire but 18 months later having disappeared dur to predators and lack of habitat is fairly common. I don't think your reading of this is quite correct (or perhaps the implication that I'm incorrectly reading into it). The species didn't quite "disappear", as all of the ones claimed to have done so after the 1972 fire were able to be tracked during the 1980 fire: > Following the 1972 fire, numbers fell to the lowest level recorded during the study and each population subsequently disappeared from the plot between the 1973 and 1974 winter censuses. The less intense 1980 fire did not lead to extirpation, but numbers of A. agilis, A. swainsonii and R. fuscipes declined as drought conditions persisted through 1983. This suggests to me that they were repopulated according to the carrying capacity of the habitat (which itself recovered), either from areas outside of the study area or from the small populations that remained (the abstract is careful to repeatedly say "disappeared from the plot", a qualifier that you dropped despite it being significant, as sampling has error). These dynamics are what need to be studied, or at least extrapolated from. The study you shared is based on an 0.02 square mile study area (ಠ_ಠ), of a single ecosystem type; It's possible that the vast scale of the current wildfires mean that recovery from external populations is much less likely, but the claim under discussion here is very far off either way. (For more information about these dynamics, the "Population refuges and immigration" section of this paper is interesting). At any rate, the centrally-trumpeted claims that these animals have already died based on habitat destruction is plainly false. I guess I should note that I don't think pushing back against these estimates suggests that this isn't a tragedy, or that it rebuts calls to action against climate change or mitigation or whatever. I just don't think it serves anyone to misrepresent the facts, or in this case, for such a sloppily-calculated, off-the-cuff figure to be spread so widely as gospel. This is exactly the kind of thing that gives anti-science skeptics traction. OTOH, maybe I'm typical-minding too hard here: most people aren't interested in an accurate view of reality in any case, and maybe being dishonest and plying the public with bullshit figures is how things need to be done. Maybe the harm from giving stupid denialists fodder is outweighed by the benefit of inciting stupid non-denialists to care. ¯\_(ツ)_/¯
wutbrodo 7y ago
I read the GP comment with this caveats in mind, and I still think it's accurate. The figure used assumes that 100% of animals in areas affected by fires were killed. Even a quote like: > It's reasonable to think that for animals caught up in the fires, at least for ground dwelling animals, more were severely impacted than were not. taken literally, mean that the estimate is off by a factor of 2. I understand that you didn't mean this literally, but it's illustrative of the structure of the claims and skepticism here. On top of that, even the core methodology of the study doesn't make any sense. Per the university's methodology notes[1], they copied the method used by a previous study on land clearing, ie, permanent destruction of habitat. It's a somewhat reasonable approach in that context: cleared and developed land is lost to wildlife in the equilibrium state, and it only relies on some fairly mild assumptions about a constant carrying capacity per species in order to use this methodology to calculate losses. There's no indication that destruction of the habitat leads to immediate death of the animals, and no attempt to account for the proportion of animals that may have successfully fled. In the case of these wildfires, the burned land isn't being permanently and explicitly kept out of the ecosystem. It's easy to imagine species density increasing in the remaining areas in the short term, and I don't think anyone expects the habitats not to grow back onto the destroyed land. Unless the claim is that the burned habitats are lost forever, the same logic obviously isn't applicable. The spread of this "fake news" is down to an irresponsible claim from the university and an incompetent/irresponsible game of Telephone from journalists. That being said, this isn't anything new for journalism. I'm in the habit of checking the sources of dramatic claims, as reported even by "prestigious" outlets like the New York Times, and the hit rate is laughably low if you follow the trail of sources and have even a modicum of scientific literacy. Again, this is true of _every_ topic they report on, not just specific, politically-loaded areas, which suggests it's not just a question of dishonesty and bias: Journalism's incentives are fucked right now, epistemology in the modern era is _really_ hard, and the hiring bar for journalism doesn't select all that stringently for intelligence. I don't know what the solution to this is... [1] https://sydney.edu.au/news-opinion/news/2020/01/03/a-statement-about-the-480-million-animals-killed-in-nsw-bushfire.html https://sydney.edu.au/news-opinion/news/2020/01/03/a-stateme...
chongli 7y ago
Birds have very sensitive respiratory systems (which is why canaries were traditionally taken into mines to detect toxic gases). Given the prevalence of fires in these Australian bush habitats, it’s not unreasonable to expect birds in those areas to be extremely intolerant of smoke, allowing them to fly away and escape long before the fire gets close enough for them to feel the heat.
- Cogito 7y ago
  That seems plausible, but it doesn't seem to be what's happening. The author of the 480 million affected animals estimate has co-authored an article which talks about the impact on birds (as well as many other animals). The relevant quote from the article: One might think birds and other fast-moving animals can easily escape fires. But smoke and strong winds can badly disorient them, and mass bird deaths in severe bushfires are common. We saw this[0] in the current fire crisis, when dead birds including rainbow lorikeets and yellow-tailed black-cockatoos washed up on the beach at Mallacoota in Victoria. The Age article referenced has more graphic detail: Following the fires that have torn through the region, bird carcasses – perhaps in their thousands – have washed up on Mallacoota's once-pristine beaches. Mr Semmens, 87, has counted 100 dead birds from 25 species on a half-kilometre stretch of Tip Beach, just outside Mallacoota, half-buried in blackened piles of leaves. “It’s a sorry sight,” he said. And he has only viewed a small section of the beach following the fires – he estimates that "thousands" more birds would have died in the area. “The birds who avoided the firestorm are struggling to survive about the town, where residents are putting out water, seed and fruit, hoping it will save some of them.” I've posted the first article at https://news.ycombinator.com/item?id=21990825 https://news.ycombinator.com/item?id=21990825 [0] https://www.theage.com.au/national/victoria/it-s-a-sorry-sight-dead-birds-wash-up-on-mallacoota-s-beaches-20200107-p53plm.html https://www.theage.com.au/national/victoria/it-s-a-sorry-sig...

by Cogito 7y ago

jcranmer 7y ago

The intellectually honest thing to do here would be to either try to find some precedent study of wildfire mortality for wildlife to get a coarse estimate on what happens, or to admit that you don't know. Instead, we have an "affected" number that is technically not presented as "killed", but we-kinda-sorta-hope-you-interpret-it-as-such. I'm personally frustrated that no one has been able to dredge up an actual study on wildfire mortality patterns.

Cogito 7y ago
I couldn't find the publication myself (though I now see others have posted it in other threads [0]) so I emailed Professor Dickson to get his input. In the meantime I looked for some studies, and found two that are interesting and include some background on the subject. The Responses of Mammals to La Niña (El Niño Southern Oscillation)–Associated Rainfall, Predation, and Wildfire in Central Australia [1] > The responses of vertebrates to wildfires are complex, reflecting the varying food and shelter requirements of different species, their exposure to predation, and other factors. Vertebrates that shelter above the ground and are not particularly mobile are often killed directly by fire (Chew et al. 1959; Newsome et al. 1975; Silveira et al. 1999; Simons 1991). Conversely, species capable of extensive movements such as large predators and birds frequently escape the flames and may increase in abundance after wildfire when food resources become available (Loyn 1997; Newsome et al. 1975).,Burrowing herbivores and omnivores are likely to survive fires and, depending on predation, may increase in abundance after postfire flushes of vegetation (Newsome et al. 1975). Postwild-fire decreases in the abundance of some species have been attributed to a lack of food resources and increased exposure to predation (Groves and Steenhof 1988; Lawrence 1966; Newsome et al. 1975, 1983). Immediate and short‐term responses of bird and mammal assemblages to a subalpine wildfire in the Snowy Mountains, Australia [2] > Over 35 days in January–February 2003, wildfires burnt across much of the subalpine/alpine landscape of south‐eastern Australia, including about 70% of the land above 1500 m in the Snowy Mountains. At the time of the fire, studies of two subalpine faunal assemblages were being undertaken. The opportunity to resurvey the assemblages was taken in order to examine the immediate impact of fire in an environment where it is uncommon but predicted to occur increasingly with global warming. A study area in the Whites River Valley, where the number of bird species was counted monthly from 1996 to 2003, and weekly in late winter–spring from 2000 to 2003, was burnt in one fire. As well as the expected decrease in the number of individual birds, the fire resulted in an immediate decrease in the number of bird species, unlike in previously studied montane forest, with only the regularly wintering species plus the olive whistler and the ground‐feeding flame robin remaining. During the post‐winter avian immigration, few regular spring migrants appeared on burnt sites despite their nearby presence on the unburnt side of the valley. Five of six small mammal trapping grids were burnt. As with fires at lower altitudes, there was an immediate reduction in mammal numbers on burnt grids following the fire, but in addition, one species, Mastacomys fuscus, declined further in the ensuing 2 months both on burnt and unburnt sites. Numbers of Antechinus swainsonii and Rattus fuscipes stabilized until autumn/winter when there was a further decline due to the unavailability of subnivean space to allow winter foraging, allied with a concentration of fox predation on areas still carrying populations of small mammals. [0] https://www.researchgate.net/publication/318029981_Impacts_of_Landclearing_the_impacts_of_the_approved_clearing_of_native_vegetation_on_Australian_wildlife_in_New_South_Wales https://www.researchgate.net/publication/318029981_Impacts_o... [1] https://academic.oup.com/jmammal/article/86/4/689/876336 https://academic.oup.com/jmammal/article/86/4/689/876336 [2] https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1442-9993.2006.01629.x https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1442-9993....
nl 7y ago
A fairly old book[1] has bird mortality rates between 40% and 54% for similar fires. [1] https://books.google.com.au/books?hl=en&lr=&id=4f8anIoS33MC&oi=fnd&pg=PA351&dq=Wildlife+mortality+in+bushfires+in+Australia&ots=kGisLYv0km&sig=81d1pW9C_IVGRyZ8fqJfmHCRHaw#v=onepage&q=Wildlife%20mortality%20in%20bushfires%20in%20Australia&f=false https://books.google.com.au/books?hl=en&lr=&id=4f8anIoS33MC&... (page 358)
nl 7y ago
Small mammal populations in a eucalypt forest affected by fire and drought [2009][1] tracks 5 species in a particular area from 1970 until 2005. Only the abstract is available (unfortunately), but by my reading of that it shows that a 1973 bushfire wiped out all 5 species in the area ("Following the 1972 fire, numbers fell to the lowest level recorded during the study and each population subsequently disappeared from the plot between the 1973 and 1974 winter censuses"). In further reading (I read a lot of abstracts trying to find one that had numbers!) it appears this pattern of some surviving the immediate fire but 18 months later having disappeared dur to predators and lack of habitat is fairly common. [1] https://www.publish.csiro.au/WR/WR08086 https://www.publish.csiro.au/WR/WR08086
- wutbrodo 7y ago
  Thank you, this is _exactly_ the kind of analysis I'm interested in! (but unfortunately, far too nuanced to be comprehended or covered by media with any sort of reach). > it appears this pattern of some surviving the immediate fire but 18 months later having disappeared dur to predators and lack of habitat is fairly common. I don't think your reading of this is quite correct (or perhaps the implication that I'm incorrectly reading into it). The species didn't quite "disappear", as all of the ones claimed to have done so after the 1972 fire were able to be tracked during the 1980 fire: > Following the 1972 fire, numbers fell to the lowest level recorded during the study and each population subsequently disappeared from the plot between the 1973 and 1974 winter censuses. The less intense 1980 fire did not lead to extirpation, but numbers of A. agilis, A. swainsonii and R. fuscipes declined as drought conditions persisted through 1983. This suggests to me that they were repopulated according to the carrying capacity of the habitat (which itself recovered), either from areas outside of the study area or from the small populations that remained (the abstract is careful to repeatedly say "disappeared from the plot", a qualifier that you dropped despite it being significant, as sampling has error). These dynamics are what need to be studied, or at least extrapolated from. The study you shared is based on an 0.02 square mile study area (ಠ_ಠ), of a single ecosystem type; It's possible that the vast scale of the current wildfires mean that recovery from external populations is much less likely, but the claim under discussion here is very far off either way. (For more information about these dynamics, the "Population refuges and immigration" section of this paper is interesting). At any rate, the centrally-trumpeted claims that these animals have already died based on habitat destruction is plainly false. I guess I should note that I don't think pushing back against these estimates suggests that this isn't a tragedy, or that it rebuts calls to action against climate change or mitigation or whatever. I just don't think it serves anyone to misrepresent the facts, or in this case, for such a sloppily-calculated, off-the-cuff figure to be spread so widely as gospel. This is exactly the kind of thing that gives anti-science skeptics traction. OTOH, maybe I'm typical-minding too hard here: most people aren't interested in an accurate view of reality in any case, and maybe being dishonest and plying the public with bullshit figures is how things need to be done. Maybe the harm from giving stupid denialists fodder is outweighed by the benefit of inciting stupid non-denialists to care. ¯\_(ツ)_/¯

wutbrodo 7y ago

I read the GP comment with this caveats in mind, and I still think it's accurate. The figure used assumes that 100% of animals in areas affected by fires were killed. Even a quote like: > It's reasonable to think that for animals caught up in the fires, at least for ground dwelling animals, more were severely impacted than were not. taken literally, mean that the estimate is off by a factor of 2. I understand that you didn't mean this literally, but it's illustrative of the structure of the claims and skepticism here. On top of that, even the core methodology of the study doesn't make any sense. Per the university's methodology notes[1], they copied the method used by a previous study on land clearing, ie, permanent destruction of habitat. It's a somewhat reasonable approach in that context: cleared and developed land is lost to wildlife in the equilibrium state, and it only relies on some fairly mild assumptions about a constant carrying capacity per species in order to use this methodology to calculate losses. There's no indication that destruction of the habitat leads to immediate death of the animals, and no attempt to account for the proportion of animals that may have successfully fled. In the case of these wildfires, the burned land isn't being permanently and explicitly kept out of the ecosystem. It's easy to imagine species density increasing in the remaining areas in the short term, and I don't think anyone expects the habitats not to grow back onto the destroyed land. Unless the claim is that the burned habitats are lost forever, the same logic obviously isn't applicable. The spread of this "fake news" is down to an irresponsible claim from the university and an incompetent/irresponsible game of Telephone from journalists. That being said, this isn't anything new for journalism. I'm in the habit of checking the sources of dramatic claims, as reported even by "prestigious" outlets like the New York Times, and the hit rate is laughably low if you follow the trail of sources and have even a modicum of scientific literacy. Again, this is true of _every_ topic they report on, not just specific, politically-loaded areas, which suggests it's not just a question of dishonesty and bias: Journalism's incentives are fucked right now, epistemology in the modern era is _really_ hard, and the hiring bar for journalism doesn't select all that stringently for intelligence. I don't know what the solution to this is... [1] https://sydney.edu.au/news-opinion/news/2020/01/03/a-statement-about-the-480-million-animals-killed-in-nsw-bushfire.html https://sydney.edu.au/news-opinion/news/2020/01/03/a-stateme...

chongli 7y ago

Birds have very sensitive respiratory systems (which is why canaries were traditionally taken into mines to detect toxic gases). Given the prevalence of fires in these Australian bush habitats, it’s not unreasonable to expect birds in those areas to be extremely intolerant of smoke, allowing them to fly away and escape long before the fire gets close enough for them to feel the heat.

Cogito 7y ago
That seems plausible, but it doesn't seem to be what's happening. The author of the 480 million affected animals estimate has co-authored an article which talks about the impact on birds (as well as many other animals). The relevant quote from the article: One might think birds and other fast-moving animals can easily escape fires. But smoke and strong winds can badly disorient them, and mass bird deaths in severe bushfires are common. We saw this[0] in the current fire crisis, when dead birds including rainbow lorikeets and yellow-tailed black-cockatoos washed up on the beach at Mallacoota in Victoria. The Age article referenced has more graphic detail: Following the fires that have torn through the region, bird carcasses – perhaps in their thousands – have washed up on Mallacoota's once-pristine beaches. Mr Semmens, 87, has counted 100 dead birds from 25 species on a half-kilometre stretch of Tip Beach, just outside Mallacoota, half-buried in blackened piles of leaves. “It’s a sorry sight,” he said. And he has only viewed a small section of the beach following the fires – he estimates that "thousands" more birds would have died in the area. “The birds who avoided the firestorm are struggling to survive about the town, where residents are putting out water, seed and fruit, hoping it will save some of them.” I've posted the first article at https://news.ycombinator.com/item?id=21990825 https://news.ycombinator.com/item?id=21990825 [0] https://www.theage.com.au/national/victoria/it-s-a-sorry-sight-dead-birds-wash-up-on-mallacoota-s-beaches-20200107-p53plm.html https://www.theage.com.au/national/victoria/it-s-a-sorry-sig...