Thursday, February 22, 2007

DNA Reveals Possible Bird List Changes

Genetic barcoding of North American bird species suggests that fifteen current species may need to be split into two or more species, while forty-two current species may actually represent seventeen species. The barcoding of North American birds was done in support of a project to create a database of the DNA of all living creatures. So far, 643 out of 690 North American breeding species have been analyzed.

The project analyzed a small strand of mitochondrial DNA from each species sampled and compared the results against current taxonomy lists. It found that the vast majority (94%) of recognized bird species in North America correspond closely with distinct genetic clusters. Four percent are closely-related species represented by a single distinct cluster. Two percent of current species show two distinct clusters, suggesting the need for a split. (An academic paper describing the genetic barcoding project is available for free. The article describing the avian research is published in Molecular Ecology Notes, Comprehensive DNA barcode coverage of North American birds. If you cannot access the bird paper via the Blackwell link, try this pdf link. The project's website is here.)


In birding terminology, dividing a species into two or more is called a split and combining two or more species into one is called a lump. Species with a genetic variation over 2.5% were recommended for splits. The differences ranged from 3.1% in Northern Fulmar and Western Screech Owl to 7.9% in Marsh Wren. Potential splits are listed below. The genetic variation cited by the article corresponds to recognized subspecies with distinct geographic ranges. The Sibley Guide illustrates recognizable differences among different populations for most of these species. However, the differences are usually subtle, so field identification of new species will remain a challenge unless their range boundaries are clearly demarcated.

1Northern fulmarFulmaris glacialis
2Solitary sandpiperTringa solitaria
3Western screech owlMegascops kennicottii
4Warbling vireoVireo gilvus
5Mexican jayAphelocoma ultramarina
6Western scrub-jayAphelocoma californica
7Common ravenCorvus corax
8Mountain chickadeePoecile gambeli
9BushtitPsaltriparus minimus
10Winter wrenTroglodytes troglodytes
11Marsh wrenCistothorus palustris
12Bewick's wrenThyromanes bewickii
13Hermit thrushCatharus guttatus
14Curve-billed thrasherToxostoma curvirostre
15Eastern meadowlark
Sturnella magna


The species that share close genetic characteristics are shown in the table below. Some of these are easily separable in the field, such as Black Duck vs. Mallard, the two teals, or the two eiders in breeding plumage. Others present real identification challenges, even though some individual species may be easily identifiable. Many birders may be relieved to see eight species of gulls lumped together, except for those who already have gone to the trouble of finding and identifying all of those species.

1Snow gooseChen caerulescens
Ross's gooseChen rossii
2American Black duckAnas rubripes
MallardAnas platyrhynchos
Mottled duckAnas fulvigula
3Blue-winged tealAnas discors
Cinnamon tealAnas cyanoptera
4King eiderSomateria spectabilis
Common eiderScomateria mollissima
5Sharp-tailed grouseTympanuchus phasianellus
Greater prairie-chickenTympanuchus cupido
Lesser prairie-chickenTympanuchus pallidicinctus
6Western grebeAechmophorus occidentalis
Clark's grebeAechmophorus clarkii
7Laughing gullLarus atricilla
Franklin's gullLarus pipixcan
8California gullLarus californicus
Herring gullLarus argentatus
Thayer's gullLarus thayeri
Iceland gullLarus glaucoides
Lesser black-backed gullLarus fuscus
Western gullLarus occidentalis
Glaucous-winged gullLarus glaucescens
Glaucous gullLarus hyperboreus
9Red-naped sapsuckerSphyrapicus nuchalis
Red-breasted sapsuckerSphyrapicus ruber
10Black-billed magpiePica hudsonia
Yellow-billed magpiePica nuttalli
11American crowCorvus brachyrhynchos
Northwestern crowCorvus caurinus
12Townsend's warblerDendroica townsendi
Hermit warblerDendroica occidentalis
13Golden-crowned sparrowZonotrichia leucophrys
White-crowned sparrowZonotrichia atricapilla
14Dark-eyed juncoJunco hyemalis
Yellow-eyed juncoJunco phaeonotus
15Snow buntingPlectrophenax nivalis
McKay's buntingPlectrophenax hyperboreus
16Great-tailed grackleQuiscalis mexicanus
Boat-tailed grackleQuiscalis major
17Common redpollCarduelis flammea
Hoary redpollCarduelis hornemanni

(For the full tables and analysis, see the article, Comprehensive DNA barcode coverage of North American birds. The tables shown here reproduce only a portion of the information in the article.)

So how does this affect birders? Probably not much, at least initially. Headlines for this story excitedly proclaim the discovery of potential new species. In fact, the North American bird list may see a net reduction if all of the proposed changes are accepted by the appropriate committees. Coverage has also suggested that the results, extrapolated to the whole world, may imply an additional 1,000 species to add to the 10,000 already recognized. Again, extrapolating results suggests that some species groups will also be condensed, so it is too soon to predict reaching the 11,000 mark.

This study is was designed as a test of the barcoding methodology, as the authors mention several times. As such, the DNA samples are too small to be used for distinguishing species based on genetics alone. The authors suggest that further DNA analysis will be necessary before any of the recommended splits or lumps are implemented. Even then, genetic analysis is only one factor among many in distinguishing separate species. Interbreeding, behavior, range, and physical appearance are among the other factors considered. There also seems to be some skepticism about the results in the wider birding and ornithological communities, so records committees are likely to move very slowly in acting on the recommendations.

Several other blogs have covered this story, including Search and Serendipity, Hawk Owl's Nest, and Nemesis Bird. There has been extensive discussion of the issue on the ID-Frontiers list. Snail's Tales has an unrelated post on defining species.