Proposal (981) to South American Classification Committee

Treat Campephilus splendens as a species separate from Campephilus haematogaster

Background: Although two subspecies have long been recognized based on plumage, Ridgely and Greenfield (2001) were the first to highlight potential differences in vocalizations/drumming that might suggest Campephilus haematogaster splendens (Panama, Pacific slope of Colombia and northwest Ecuador) be treated as a separate species from nominate (eastern Andes, southeast Colombia to northwestern Bolivia; eBird). As noted by Ridgely and Greenfield (2001) and del Hoyo et al. (2022), splendens differs in plumage from nominate in having less black on the upper breast, but slightly more black barring on the underparts. Del Hoyo et al. (2022) also noted that splendens has three large buff spots vs. two in nominate on the wing (visible only when wing is spread), and the buff sub-moustachial (cheek stripe) extends onto the upper neck in male splendens (see above refs for depiction of these characters).

Donegan et al. (2015) presented the most in-depth analysis of vocalizations and drumming for these two taxa. They reported that the single-noted, harsh calls were higher in maximum frequency in splendens than in nominate. Limited material on other calls (“interaction” and “rattle”) made comparisons difficult and inconclusive. Donegan et al.’s (2015) extensive review of drumming demonstrated that splendens typically does a double rap, whereas the drum in nominate consists of multiple raps, usually 5-6.

No genetic data have been published for comparison of these two taxa.

New information: Because new material is constantly being added to on-line audio resources, I listened to every audio recording of Campephilus haematogaster on xeno-canto and the Macaulay Library (accessed on 1-2 August 2023). Note that there are at least two recordists who have deposited the same recordings on both sites, so the actual sample size is smaller than one may initially think (see specific comment below). Here are my results:

Even eight years after Donegan et al.’s (2015) assessment, there still isn’t enough vocal material of various calls. Furthermore, not surprisingly given how difficult it is to audio record most Campephilus, the majority of recordings are of poor quality. It does appear based on very little material that the “rattle” call of splendens may be lower in frequency than in nominate, but again, more material is needed, especially of splendens.

However, there are enough drumming recordings across the distribution of both taxa to make definite conclusions. As Donegan et al. (2015) documented, there clearly is a striking dichotomy in drumming between splendens and nominate. Nominate almost invariably gives a multiple noted rap (e.g., XC683956, XC683954), whereas splendens gives the classic double-rap (XC455497, XC275336) that is shared with other Campephilus. There is a single exception (XC220123 & ML286793; clearly these are the same recording deposited on both sites - ugh!), where at Cana, Darién, Panama, there is a multi-rap drum attributed to haematogaster. However, I suspect that this is a mis-identification given that this is anomaly for splendens and the fact that Campephilus melanogaster is known from this locality (Robbins et al. 1985, eBird) and gives the multi-noted rap, that is very similar to nominate haematogaster.

Recommendation: In my opinion, this is not a straightforward decision, especially given that there are no genetic data available. It is debatable on whether the differences in drumming and the rather minor plumage differences are enough to consider these separate species. I am not aware of any hybridization among Campephilus, nor am I aware of the above audio and plumage characters being important for mate selection in that clade. We do know that rather dramatic differences in plumage among Celeus woodpecker species do not preclude hybridization, e.g., between Celeus lugubris and C. elegans jumana (Benz and Robbins 2011). Whether that is relevant to the Campephilus clade is uncertain. If we use Celeus as a reference point, we know that there is considerable plumage variation among subspecies within a species, e.g., elegans and flavus. Moreover, I quote our paper (Benz and Robbins 2011) regarding vocalizations within Celeus: “Song structure and note composition vary considerably within clades and are largely incongruous with the molecular framework….”. Thus, one might argue that plumage differences and vocalizations in some New World picids may have limited utility for defining species limits. We could delay recognizing splendens as a species until we have genetic data, however, if the majority of the committee believes we should go ahead and split them, I will support that.

References:

Benz, B.W. and M.B. Robbins. 2011. Molecular phylogenetics, vocalizations and species limits in Celeus woodpeckers (Aves: Picidae). Molecular Phylogenetics and Evolution 56:29-44.

del Hoyo, J., H. Winkler, D. A. Christie, and N. Collar (2022). Crimson-bellied Woodpecker (Campephilus haematogaster), version 1.1. In Birds of the World (N. D. Sly, Editor). Cornell Lab of Ornithology, Ithaca, NY, USA. https://doi.org/10.2173/bow.crbwoo1.01.1

Donegan, T., A. Quevedo, J. C. Verhelst, O. Cortes-Herrera, T. Ellery, and P. Salaman (2015). Revision of the status of bird species occurring or reported in Colombia 2015, with discussion of BirdLife International's new taxonomy. Conservación Colombiana 23:3-48.

Ridgely, R.S. and P.J. Greenfield. 2001. The birds of Ecuador. Field Guide. Vol. II. Comstock Publ. Associates, New York.

Robbins, M.B., T.A. Parker, and S.E. Allen. 1985. The avifauna of Cerro Pirre, Darien, Eastern Panama. Pp. 198-232. In P.A. Buckley, M.S. Foster, E.S. Morton, R.S. Ridgely, and F.G. Buckley (eds.), Neotropical Ornithology. Ornithol. Monogr. No. 36. https://doi.org/10.2307/40168283

Mark Robbins, August 2023

Comments from Lane: “NO for now. I find the evidence Mark provides to be convincing regarding the differences in drums--the drums of Campephilus are probably usually the first line of conspecific recognition in regions where multiple species occur, so the difference between the two forms may be a significant one here. However, I find it interesting that both taxa overlap with C. pollens, but don't necessarily divvy up "drum-space" with that species, which has me scratching my head, and suddenly weakens my last comment significantly. In addition, I am not that impressed with the difference in vocalizations (body size could well be responsible for higher vocalizations in splendens) nor plumage, and would be happy to consider these within the realm of subspecific variation. Additional investigation would likely help change my mind on this case.”

Comments from Areta: “NO. Some drummings of haematogaster have several notes, but this could well be (as mentioned by Andrew Spencer in his XC recordings) a matter of excitement (e.g., after playback). Then, there are also several recordings of haematogaster (notably, those said to be unsolicited) which have three notes (the last one being, typically but not always, fainter than the two preceding ones; check this one https://xeno-canto.org/262898). Finally, most of the available recordings assigned to splendens consist of a double-knock (the default drumming in Campephilus). Check also this one of haematogaster, in which at least one of the drummings has so much emphasis on the first two notes that this is all that would be audible if recorded from a distance. I think that the case has some merit, given the plumage distinctions, biogeography, and likely drumming differences. But I also think that it falls short of being clearly convincing. I vote NO to the split, in the hope that a rigorous study will provide more support for the 2-species treatment.”

Comments from Zimmer: “NO. As others have voiced, the combination of plumage differences, drumming distinctions, and biogeography are all suggestive that two species may be involved, and of these, I also feel that potential differences in the drums of the two taxa are the most significant isolating mechanisms. However, I don’t think the evidence is conclusive yet. “Crimson-bellied” Woodpeckers are exceptionally sneaky, quiet, and notoriously difficult to audio record in my experience (from Panama and Peru), and that is reflected in the small sample sizes of quality recordings of either taxon in the various archived audio collections. Pending a more intensively sampled analysis of the various vocalizations and drums, I would prefer to maintain the two taxa as one species.”

Comments from Robbins: “NO. As I detailed in the proposal, this is not a straightforward decision, but for now, primarily because I would like to see nuclear DNA data, I vote NO.”

Comments from Remsen: “NO. Just too many unknowns, as the proposal and others’ comments have outlined above. This is a case in which a more detailed analysis of voice and drumming is needed, with an N that surpasses anecdote level. The evidence is trending towards species-level differences, but I don’t think it is sufficient to alter current taxonomy under a scientific framework.

Comments from Claramunt: “YES. Tentative. I’m persuaded by the coincidence of both drumming differences and plumage differences (multiple) that seem fully diagnostic. I admit that the evidence is not very strong because we are missing a detailed analysis of character variation. But this seems the case of two taxa lumped together just because they look similar on a cursory examination, but there is actually zero evidence of conspecificity.”

Comments from Stiles: “NO. Although several lines of evidence point to splitting these two, sample sizes are very small, the differences in plumage could go either way, and genetic information is lacking – especially from the nuclear genome. Also, I would have appreciated seeing some data on measurements and their analysis. In sum, NO for now: more data needed.”

Comments from Niels Krabbe (voting for Del-Rio): “NO. I agree with other no-voters that genetic and more vocal material is needed to support a split.”