• Welcome to BirdForum, the internet's largest birding community with thousands of members from all over the world. The forums are dedicated to wild birds, birding, binoculars and equipment and all that goes with it.

    Please register for an account to take part in the discussions in the forum, post your pictures in the gallery and more.
ZEISS DTI thermal imaging cameras. For more discoveries at night, and during the day.

Time calibration and Linnean ranks in birds (1 Viewer)

Mysticete

Well-known member
United States
I thought about making this part of making this a part of an existing thread, and it has certainly come up several times. But I think its a distinct enough idea to deserve its own thread, even if a whopping four people care about this stuff here (lolz?)

I am a taxonomy nerd. Before I was even a birder I was making lists of critters. One of my current projects is to create my own checklist, the subspecies groups project an outgrowth of that, as are my posts. I am also a paleontologist by training. Those two things make me incredibly interested in the concept of using time-calibration to create an intrinsically less random and arbitrary system of classification. At present, other than marking the existence of a clade, whether something is a family or a order or a subfamily doesn't actually mean anything. But what if it could? What if you COULD fix the age of families to a set time period, and directly correlate them to events in geologic history? A large number of zoom meetings this week and....well...boredom during them, has made me explore this concept in more detail.

This is not an original idea. Avise a long-time ago argued for this to be considered for all life, and the Avitaxonomicon (a great reference site!) previously has made a big push. My big complaint with the latter is that the dates the author chose are arbitrary. But what if you could pick major periods of climate and geologic change, which might be expected to influence diversification rates. I already apply this to orders in my own taxonomy, using the Paleocene-Eocene Thermal Maximum as the cut-off age for orders of birds, as it is well known to be a major period of diversification for mammals and birds (and works with very little modification of the existing taxonomy). Could other events be just as useful, and how might it change ranks?

Caveats of course: Many bird groups lack any sort of time-calibrated phylogeny, and methods, choice of sampled taxon, and calibration points can also mess things up.

At any rate, to start off I decided to test the most conservative cut off for Families: For a Family to exist it had to have been present in the Eocene or earlier, or close enough. A hard cut off would automatically demote ANY FAMILY that was younger, while a soft cut off would allow younger families but elevate anything present in the Eocene to family level if not already recognized. The Eocene was a time of warm tropical temperatures that started in the PETM (the warmest period in Earth History since the extinction of the dinosaurs, with some gradual cooling towards the end. It was a major period of diversification for life, with tropical rainforests above the Arctic circle and vasts circumpolar forests that allowed free interchange of critters. Personally this would not be my preferred cut off point since I would prefer one that marked a more significant global change from my cut-off point for Orders.
.
In the next post I will show my results for the Eocene
 
The Eocene ended 33 Mya. If we use this as a hard cut-off date, the the following changes would need to be made. If a order is not listed, then nothing changes. Also the baseline for comparison is IOC

Tinamiformes (Net +1)
Tinamidae would need to be split into two families, which correspond to the subfamilies Nothuridae and Tinamidae. Possibly into 4 families, but I took a conservative interpretation

Anseriformes (Net +1)
This would require Dendrocygnidae (Whistling-ducks) to be elevated to family status, a move already advocated for by Sun et al 2017

Galliformes (Net +1)
Rollulidae (Hill Partridges, a largely SE Asian group) elevated from Subfamily to Family status

Cuculiformes (Net +1)
Anis and Ground Cuckoos (Crotophaginae) elevated to family status. I have heard this is something SACC might consider taking up in the future anyway

Gruiformes (Net -1)
Dates here are still a bit in flux. My interpretation of some studies would suggest elevating Himantornithinae to family status, a group that includes the Nkulengu Rail and a few Indonesian oddballs. At the same time, Limpkin would be demoted to a subfamily of Cranes and Finfoots and Flufftails would be combined into a single family

Charadriiformes (Net +0)
Additions and subtractions cancel each other out here. On one hand, Magellanic Plover would be lumped into Chionidae with Sheathbills, something Howard and Moore already do. Ibisbills and Stilts/Avocets would also be lumped with Oystercatchers into a expanded Haematopidae. On the other hand, Golden-Plovers (Pluvialiidae), Curlews (Numeniidae) and Godwits (Limosidae) would all be elevated to family level. Pluvialiidae has been suggested before and Taxonomy in flux recently brought up the idea for the Curlews and Godwits.

Eurypgyiformes (Net -1)
Kagu and Sunbitterns would be lumped into the same family

Procellariformes (Net - 1)

Hydrobatidae would (maybe) get lumped into Procellaridae

Suliformes (Net -1)
Anhingas would be lumped with Cormorants into Phalacrocoracidae

Accipitriformes (Net +1)
Elanidae would be split from Accipitridae, something already advocated by Starikov et al 2020

Bucerotiformes (Net -2)

Wood-Hoopoes would be lumped with regular Hoopoes, and Ground Hornbills and Tree Hornbills would once more be lumped.

Coraciiformes (Net +3)
Bearded Bee-eaters would be split from regular Bee-eaters, while all three Kingfisher subfamilies would be elevated to subfamily status (although some are kind of borderline)

Piciformes (Net -5)
All barbets and toucans would be lumped into a single family, while honeyguides would be lumped with woodpeckers.

Falconiformes (Net +1)
Herpetotheridae would be split from Falconidae, a move already suggested by others

Psittaciiformes (Net -1)
Psittaculidae would be relumped with Psittacidae

So far nothing here is especially radical, and most groups have no change at all. But then we come to Passeriformes. Effectly, you would go from 144 families recognized by IOC to only 28. To give you a sense of the effect and how much demoting would occur for different families, Corvidae goes from a family to a TRIBE, and you have some truly massive families including Tyrannidae (which also includes cotingas, tityras, and manakins), Sylviidae which includes chickadees, swallows, bulbuls plus all old world warblers, Muscicapidae which includes all wrens, creepers, nuthatches, waxwings, thrushes and starlings, and Passeridae, which includes all sunbirds, waxbills, pipits, and nine-primaried oscines.
Passeriformes (Net - 116)
 
The Eocene ended 33 Mya. If we use this as a hard cut-off date, the the following changes would need to be made. If a order is not listed, then nothing changes. Also the baseline for comparison is IOC

Tinamiformes (Net +1)
Tinamidae would need to be split into two families, which correspond to the subfamilies Nothuridae and Tinamidae. Possibly into 4 families, but I took a conservative interpretation

Anseriformes (Net +1)
This would require Dendrocygnidae (Whistling-ducks) to be elevated to family status, a move already advocated for by Sun et al 2017

Galliformes (Net +1)
Rollulidae (Hill Partridges, a largely SE Asian group) elevated from Subfamily to Family status

Cuculiformes (Net +1)
Anis and Ground Cuckoos (Crotophaginae) elevated to family status. I have heard this is something SACC might consider taking up in the future anyway

Gruiformes (Net -1)
Dates here are still a bit in flux. My interpretation of some studies would suggest elevating Himantornithinae to family status, a group that includes the Nkulengu Rail and a few Indonesian oddballs. At the same time, Limpkin would be demoted to a subfamily of Cranes and Finfoots and Flufftails would be combined into a single family

Charadriiformes (Net +0)
Additions and subtractions cancel each other out here. On one hand, Magellanic Plover would be lumped into Chionidae with Sheathbills, something Howard and Moore already do. Ibisbills and Stilts/Avocets would also be lumped with Oystercatchers into a expanded Haematopidae. On the other hand, Golden-Plovers (Pluvialiidae), Curlews (Numeniidae) and Godwits (Limosidae) would all be elevated to family level. Pluvialiidae has been suggested before and Taxonomy in flux recently brought up the idea for the Curlews and Godwits.

Eurypgyiformes (Net -1)
Kagu and Sunbitterns would be lumped into the same family

Procellariformes (Net - 1)
Hydrobatidae would (maybe) get lumped into Procellaridae

Suliformes (Net -1)
Anhingas would be lumped with Cormorants into Phalacrocoracidae

Accipitriformes (Net +1)
Elanidae would be split from Accipitridae, something already advocated by Starikov et al 2020

Bucerotiformes (Net -2)
Wood-Hoopoes would be lumped with regular Hoopoes, and Ground Hornbills and Tree Hornbills would once more be lumped.

Coraciiformes (Net +3)
Bearded Bee-eaters would be split from regular Bee-eaters, while all three Kingfisher subfamilies would be elevated to subfamily status (although some are kind of borderline)

Piciformes (Net -5)
All barbets and toucans would be lumped into a single family, while honeyguides would be lumped with woodpeckers.

Falconiformes (Net +1)
Herpetotheridae would be split from Falconidae, a move already suggested by others

Psittaciiformes (Net -1)
Psittaculidae would be relumped with Psittacidae

So far nothing here is especially radical, and most groups have no change at all. But then we come to Passeriformes. Effectly, you would go from 144 families recognized by IOC to only 28. To give you a sense of the effect and how much demoting would occur for different families, Corvidae goes from a family to a TRIBE, and you have some truly massive families including Tyrannidae (which also includes cotingas, tityras, and manakins), Sylviidae which includes chickadees, swallows, bulbuls plus all old world warblers, Muscicapidae which includes all wrens, creepers, nuthatches, waxwings, thrushes and starlings, and Passeridae, which includes all sunbirds, waxbills, pipits, and nine-primaried oscines.
Passeriformes (Net - 116)
I want to send you something and then you can tell me what you think about it
 
It's an interesting thought exercise, which certainly uncovered some splits and lumps that really should be given more credence. Especially when we get better at doing time calibrations. Currently I'm very sceptical of their precision!
 
Great post(s)! I think it would be very welcome and neat if the linnean taxonomy rank gave some more additional information on the clade. As for cut-off dates, I wonder if one could find a (later) moment in Earth history which would fit better for Passeriformes families, so we don't get 1000 species strong families as a result. I might have a look in "The Largest Avian Radiation" book later.
I think different cut-offs could be used for passerines and non-passerines, in the same way we would presumably use different cut-offs for other organisms than birds.

PS. I'm sure more than four people care about this stuff here, just many of them - like me - don't have enough knowledge to contribute, so mostly stick to just reading ;)
 
Last edited:
Add me to those interested; as a keen family lister I am most interested in the evidence for the Nkulengu Rail and relatives.
I'm also intrigued by that. From the little I've seen, Nkulengu Rail and the three Gymnocrex species are sister to the Fulica/Gallinula/Porzana clade. Splitting them off would mean making much more drastic changes to the wider Rallidae as well.
 
Tinamiformes (Net +1)
Tinamidae would need to be split into two families, which correspond to the subfamilies Nothuridae and Tinamidae. Possibly into 4 families, but I took a conservative interpretation
Eudromiinae has priority
Anseriformes (Net +1)
This would require Dendrocygnidae (Whistling-ducks) to be elevated to family status, a move already advocated for by Sun et al 2017
This is the option I chosen but I finally put it in Anatidae
Galliformes (Net +1)
Rollulidae (Hill Partridges, a largely SE Asian group) elevated from Subfamily to Family status
Why not
Charadriiformes (Net +0)
Additions and subtractions cancel each other out here. On one hand, Magellanic Plover would be lumped into Chionidae with Sheathbills, something Howard and Moore already do. Ibisbills and Stilts/Avocets would also be lumped with Oystercatchers into a expanded Haematopidae.
Recurvirostridae has priority. I put all in one family: Recurvirostridae
Accipitriformes (Net +1)
Elanidae would be split from Accipitridae, something already advocated by Starikov et al 2020
I did it
Falconiformes (Net +1)
Herpetotheridae would be split from Falconidae, a move already suggested by others
The same
So far nothing here is especially radical, and most groups have no change at all. But then we come to Passeriformes. Effectly, you would go from 144 families recognized by IOC to only 28. To give you a sense of the effect and how much demoting would occur for different families, Corvidae goes from a family to a TRIBE, and you have some truly massive families including Tyrannidae (which also includes cotingas, tityras, and manakins), Sylviidae which includes chickadees, swallows, bulbuls plus all old world warblers, Muscicapidae which includes all wrens, creepers, nuthatches, waxwings, thrushes and starlings, and Passeridae, which includes all sunbirds, waxbills, pipits, and nine-primaried oscines.
Passeriformes (Net - 116)
Worse than suggestion of Sibley and Ahlquist
 
I really love this as a mind game, as it really helps understanding some things. On the other hand, I really don't like some of the outcomes, with the changes to Passeriformes being the worst and having little to no real-life value. The more I think about it, the better I think the current system already is, even if perhaps not based on objective markers.
This does not however mean, that I don't fully appreciate learning from the information presented here.
What other cut-off dates could be chosen and what effect would that have?
 
I think Mysticete would be the first to say that this is purely a thought exercise. There are some merits to it, particularly in uncovering some non-passerine groups that could be elevated to the family level.

It breaks down though with Passeriformes, I'm guessing mostly because it doesn't take into account things like generation time. Passerines are evolving at a faster rate than, say, canes, because cranes live many times longer than Great Tits. This highlights why taxonomists in general never take one measure when decided whether to split or lump taxa.
 
I would be happy with lumping some of the oscines. Enlarged Corvidae and "one-Tyrannidae" were in my very first Sibley & Monroe-based list. I agree that generation times should play a role, but currently the oscines are ridiculously oversplit.
I do not see much merit in lumping Kagu and Sunbittern. Flufftail/finfoots is an odd family, but it mirrors rails & coots.
Anhingas with cormorants makes sense.
 
I'm also intrigued by that. From the little I've seen, Nkulengu Rail and the three Gymnocrex species are sister to the Fulica/Gallinula/Porzana clade. Splitting them off would mean making much more drastic changes to the wider Rallidae as well.
Garcia and Matzke 2021, which I was using for Rails, has them as the earliest diverging lineage of Rallidae. They are with coots in Kirchman et al 2021, although here they do have some pretty long branch lengths. Clearly more work needs to be done...Garcia and Matzke seem to show the rest of Rallidae as being a younger radiation
 
I think Mysticete would be the first to say that this is purely a thought exercise. There are some merits to it, particularly in uncovering some non-passerine groups that could be elevated to the family level.

It breaks down though with Passeriformes, I'm guessing mostly because it doesn't take into account things like generation time. Passerines are evolving at a faster rate than, say, canes, because cranes live many times longer than Great Tits. This highlights why taxonomists in general never take one measure when decided whether to split or lump taxa.
Oh yes...I am not going to batter down the doors of AOS/SACC/IOC and demand changes. I am just looking to see what ways things could be changed so that a rank meant something more concrete then "This is nested in that" I also picked the most conservative age cut off possible for this list, in part to see how it would effect different groups (Barely really changes things in Non-passeriformes, but slaughters Passeriformes to the point of not being at all informative. I have some other dates I am testing, I just started here.
 
The Eocene ended 33 Mya. If we use this as a hard cut-off date, the the following changes would need to be made. If a order is not listed, then nothing changes. Also the baseline for comparison is IOC

Tinamiformes (Net +1)
Tinamidae would need to be split into two families, which correspond to the subfamilies Nothuridae and Tinamidae. Possibly into 4 families, but I took a conservative interpretation

Anseriformes (Net +1)
This would require Dendrocygnidae (Whistling-ducks) to be elevated to family status, a move already advocated for by Sun et al 2017

Galliformes (Net +1)
Rollulidae (Hill Partridges, a largely SE Asian group) elevated from Subfamily to Family status

Cuculiformes (Net +1)
Anis and Ground Cuckoos (Crotophaginae) elevated to family status. I have heard this is something SACC might consider taking up in the future anyway

Gruiformes (Net -1)
Dates here are still a bit in flux. My interpretation of some studies would suggest elevating Himantornithinae to family status, a group that includes the Nkulengu Rail and a few Indonesian oddballs. At the same time, Limpkin would be demoted to a subfamily of Cranes and Finfoots and Flufftails would be combined into a single family

Charadriiformes (Net +0)
Additions and subtractions cancel each other out here. On one hand, Magellanic Plover would be lumped into Chionidae with Sheathbills, something Howard and Moore already do. Ibisbills and Stilts/Avocets would also be lumped with Oystercatchers into a expanded Haematopidae. On the other hand, Golden-Plovers (Pluvialiidae), Curlews (Numeniidae) and Godwits (Limosidae) would all be elevated to family level. Pluvialiidae has been suggested before and Taxonomy in flux recently brought up the idea for the Curlews and Godwits.

Eurypgyiformes (Net -1)
Kagu and Sunbitterns would be lumped into the same family

Procellariformes (Net - 1)
Hydrobatidae would (maybe) get lumped into Procellaridae

Suliformes (Net -1)
Anhingas would be lumped with Cormorants into Phalacrocoracidae

Accipitriformes (Net +1)
Elanidae would be split from Accipitridae, something already advocated by Starikov et al 2020

Bucerotiformes (Net -2)
Wood-Hoopoes would be lumped with regular Hoopoes, and Ground Hornbills and Tree Hornbills would once more be lumped.

Coraciiformes (Net +3)
Bearded Bee-eaters would be split from regular Bee-eaters, while all three Kingfisher subfamilies would be elevated to subfamily status (although some are kind of borderline)

Piciformes (Net -5)
All barbets and toucans would be lumped into a single family, while honeyguides would be lumped with woodpeckers.

Falconiformes (Net +1)
Herpetotheridae would be split from Falconidae, a move already suggested by others

Psittaciiformes (Net -1)
Psittaculidae would be relumped with Psittacidae

So far nothing here is especially radical, and most groups have no change at all. But then we come to Passeriformes. Effectly, you would go from 144 families recognized by IOC to only 28. To give you a sense of the effect and how much demoting would occur for different families, Corvidae goes from a family to a TRIBE, and you have some truly massive families including Tyrannidae (which also includes cotingas, tityras, and manakins), Sylviidae which includes chickadees, swallows, bulbuls plus all old world warblers, Muscicapidae which includes all wrens, creepers, nuthatches, waxwings, thrushes and starlings, and Passeridae, which includes all sunbirds, waxbills, pipits, and nine-primaried oscines.
Passeriformes (Net - 116)
I think you should publish on these proposals. Sounds very sensible to me. We have too many families in birds, compared to other groups such as Insects. Passerine families in particular seem to be converging to a 1980s concept of genus-level diversification. There is an inclination on the part of molecular scientists to publish new families instead of lumping existing ones or using subfamily/tribe categories. Many of these should probably best be Tribes.
 
So like I said, there is two ways we can look at the Eocene cut off date: a hard cut off point that would sink anything younger into a larger clade, or a "soft" cut off point that would simply recognize Eocene aged lineages without necessitating lower level lumps. If we take the soft approach, the following should be recognized as "new" families:

Eudromiidae (Arid land Tinamous)
Dendrocygnidae (Whistling Ducks)
Rollulidae (Hill Partridges)
Crotophagidae (Anis and New World Ground Cuckoos)
Himantornithidae (Primitive Rails)
Pluvialiidae (Golden-Plovers)
Numeniidae (Curlews)
Limosidae (Godwits)
Elanidae (True Kites)
Nyctyornithidae (Bearded Bee-eaters)
Cerylidae (Water Kingfishers)
Halcyonidae (Tree Kingfishers)
Herpetotheridae (Forest and Laughing Falcons)

Of these, the only ones that I would say are really novel are Rollulidae, Himantornithidae, and Nyctyornithidae. And Himantornithidae is still a little uncertain since it moves around between studies. The others have been argued in recent papers (Elanidae, Dendrocygnidae, Herpetotheridae), Suggested or implemented by online checklists (Pluvialiidae, Numeniidae, Limosidae) or were split by Sibley-Ahlquist.
 
So like I said, there is two ways we can look at the Eocene cut off date: a hard cut off point that would sink anything younger into a larger clade, or a "soft" cut off point that would simply recognize Eocene aged lineages without necessitating lower level lumps. If we take the soft approach, the following should be recognized as "new" families:

Eudromiidae (Arid land Tinamous)
Dendrocygnidae (Whistling Ducks)
Rollulidae (Hill Partridges)
Crotophagidae (Anis and New World Ground Cuckoos)
Himantornithidae (Primitive Rails)
Pluvialiidae (Golden-Plovers)
Numeniidae (Curlews)
Limosidae (Godwits)
Elanidae (True Kites)
Nyctyornithidae (Bearded Bee-eaters)
Cerylidae (Water Kingfishers)
Halcyonidae (Tree Kingfishers)
Herpetotheridae (Forest and Laughing Falcons)

Of these, the only ones that I would say are really novel are Rollulidae, Himantornithidae, and Nyctyornithidae. And Himantornithidae is still a little uncertain since it moves around between studies. The others have been argued in recent papers (Elanidae, Dendrocygnidae, Herpetotheridae), Suggested or implemented by online checklists (Pluvialiidae, Numeniidae, Limosidae) or were split by Sibley-Ahlquist.
I've come across "Tundra Plovers" somewhere as a nice alternative to "Golden-Plovers"
Can't remember where🤔
 
I've come across "Tundra Plovers" somewhere as a nice alternative to "Golden-Plovers"
Can't remember where🤔
European GP is breeding in the UK and Denmark it seems, not exactly places I would consider Tundra ...
 

Users who are viewing this thread

Back
Top