eBird's missing species

As of early June 2014, eBird’s database has validated records for an amazing 9902 species. The August 2013 eBird taxonomy (following version 6.8 of the Clements Checklist) recognizes 10,324 species of birds worldwide, so roughly 96% of the world’s bird species have at least one confirmed record entered into eBird. Below we investigate the 4% that have yet to be recorded in eBird—a snapshot of the most isolated, elusive, threatened, and under-studied birds in the world. If you are a world birder and spot any species on this list that you have seen, please get those records in eBird!  7 AUG UPDATE: 34 species have already been added, and there are more coming in! This amazing progress speaks volumes to the dedication of the eBird community. Even so, there’s always more work to be done, especially when it comes to mapping the ranges and understanding patterns of avian distribution! Expect more details later in the month when we publish an official update to the missing species article! Check out eBird on Facebook to see some of the additions.

For a project that began in 2002, and expanded globally in 2010, this extent of coverage of the world’s species is pretty amazing. On the other hand, why is the total not closer to 100 percent? Which species have eluded eBirders, and why? To answer these questions, Andrew Dreelin and Reid Rumelt, two undergraduates at Cornell University who help out with eBird and other Lab projects, pored through the eBird database to find the species that have yet to be entered into eBird. Their analysis and downloadable list give a compendium of some of the rarest and most poorly-known species in the world. Thanks to Andrew and Reid for this superb analysis.

Note that this article and analysis addresses only species; records of subspecies groups, hybrids (e.g., Palm x Yellow-rumped Warbler), and other taxa are not included here (but would be interesting to look into as well).

Explanation and Analysis
Our list of eBird’s missing species includes 422 birds and can be downloaded in PDF, CSV, and Excel formats. Each entry is highlighted according to the species’ IUCN (International Union for the Conservation of Nature) Red List status and includes a brief description of its distribution, most of which come directly from the Clements Checklist range descriptions. Links to additional sources of information are provided at the end of the document. We also include information from Birdlife International. It should be noted that the species list was generated in early June 2014, so it is possible that some species could have been entered into eBird in the meantime.

Additionally, all of the species on our list that are extant in the wild have been assigned a “sighting probability.” Species which conceivably could be observed in their native range and habitat are designated as “possible,” while the “highly unlikely” designation is given to species that are known from only a few specimens, have gone more than 40 years without a confirmed report, or are listed as probably extinct (CR-PE) by BirdLife International. While all of the species on our list are under-studied, geographically isolated, or uncommon, those species that warrant a “highly unlikely” designation are truly the rarest and least understood species of birds in the world.

The primary goal of this article is to summarize conservation status (IUCN status), patterns of species distribution, and taxonomic problem areas that we noticed while compiling our list. We hope our analyses will help members of the eBird community learn more about the species on our list, not just by showcasing rare and interesting birds, but also by demonstrating why gathering distribution and abundance data for these species is highly desirable from a conservation standpoint. Science is built on data, and these species are in desperate need of it.

IUCN Breakdown
Of the 422 species on our list, 76 are listed by the IUCN as extinct. An additional eight extinct species are entered as Not Evaluated because Clements gives them full species recognition, while the IUCN lists them as subspecies, bringing the total to 84 extinct species. Many were never common in recorded history; the Stephen’s Island Wren, for instance, is represented by little more than a few collection skins. On the other end of the spectrum are species, such as the Atitlan Grebe, which were intensively studied prior to their demise. Two species, Alagoas Curassow and Socorro Dove, are extinct in the wild, although captive breeding programs are ongoing for both species and there is some hope for eventual reintroduction.

The majority of the remaining species are listed as threatened in some way by the IUCN. We chose to split the 58 Critically Endangered species on our list into two groups based on the ease with which eBirders might encounter them on checklists. BirdLife, the IUCN’s taxonomic authority, recognizes 7 of our Critically Endangered species as Possibly Extinct (CR-PE), including the Jamaican Pauraque. Proof of their continued existence has not been forthcoming despite concerted efforts to find remaining individuals, making it unlikely that they have escaped extinction. An additional 51 species, including the Cherry-throated Tanager and Blue-winged Racquet-tail, are listed under the standard IUCN Critically Endangered listing. While these species are slightly further from extinction than their CR-PE counterparts, and may have been observed by scientists in the past decade or so, their continued existence is nonetheless extremely tenuous. Approximately equal numbers of species on our list are Endangered (41), Vulnerable (45), and Near Threatened (50).

We were surprised to find that 69 of the species on our list are designated as Least Concern. However, the fact that eBird lacks data for these relatively stable species suggests that not all Least Concern species are easy to track down. For instance, while the Bamboo Warbler is listed as Least Concern based on its large (albeit highly fragmented) range, this species is rare and local throughout its range, and any intrepid eBirder hoping to see one would need to plan their trip carefully.

Interestingly, 43 extant species on our list have not yet been evaluated by the IUCN. Many of these species are the results of recent lumps or splits, or are recognized as species by some authorities but not by others. Other birds have not yet been evaluated because they are new to science, including fifteen species described from the Amazon in August 2013 [Team eBird note: some of these were not ratified by the South American Classification Committee, so will be demoted to subspecies groups in the August 2014 taxonomic update]. Meanwhile, 30 species on the list are listed as Data Deficient, meaning that data on their long-term population trends simply isn’t available for the IUCN to use when evaluating their statuses. Some of these species may be evaluated as more information about them becomes available, but for now they remain shrouded in mystery.

The Mindoro Imperial-Pigeon is one of many extremely rare and localized species that could still be found by a lucky or adventurous eBirder; it is barely hanging on in mountaintops on Mondoro (Philippines).

Problem Areas
A critical part of conservation ecology is identifying areas that are in great need of study and protection. To identify global “dark spots,” or areas with high concentrations of understudied and threatened species, we analyzed the distributions of the species on our list that are extant in the wild. The areas depicted in the tables below are those most in need of coverage. The Malay Archipelago has a far greater number of missing species than any other political region, despite possessing a smaller total area than Africa, South America, or East Asia. This density of missing species is also evident when distributions are matched to biogeographic regions, since the area between Wallace’s and Lydekker’s lines (the Moluccan Islands, Sulawesi, and the Lesser Sundas), the Greater Sundas, the Philippines, and New Guinea are all part of the Malay Archipelago. Although the Malay Archipelago hosts a dazzling array of bird species such as the Sulu Hornbill, Kofiau Paradise-Kingfisher, and the incredible Birds-of-Paradise, it remains significantly less covered by eBirders because of its remote location and fragmented geography. We hope that these areas will garner more coverage in the future as eBird expands its global community. If you have visited some of these remote and bird-rich areas, please enter those lists!

Political regions

1. Malay Archipelago – 151 species 
2. Africa – 76 species
3. South America – 45 species
4. East Asia – 26 species
5. Pacific Islands – 16 species 

Biogeographic regions

1. Wallacea – 55 species
2. Philippines and Greater Sundas – 42 species
3. Central Africa – 40 species
4. New Guinea and nearby islands – 38 species
5. Amazonia – 28 species

Problem families
We also ranked families based on the number of species in that family on our list. This analysis paints a striking picture of rarity and threat status across the avian landscape. All of these groups have large numbers of geographically-limited or island endemic species, many of which are highly threatened (Sulu Bleeding-heart and Blue-fronted Lorikeet are good examples). Rallidae (Rails) ranks much lower when extinct species are removed from the tally, which hints at the high extinction rate in this family, especially for flightless island endemics (e.g., Laysan Rail). Similarly, Fringillidae (Finches) doesn’t rank at all when extinct species are removed, providing a haunting reminder of the impact that avian malaria has had on the Hawaiian Finches (Ou). The high rank of Strigidae (True Owls), meanwhile, is likely associated with the nocturnal habits and secretive nature of the group. Finally, although Zosteropidae (White-eyes) ranks highly in terms of species, many of these are listed as Least Concern. This family is particularly diverse in Southeast Asia and the Pacific Islands, highlighting the areas with high levels of endemism but low levels of eBird coverage.

Families with the most missing species (including extinct species)

1. Columbidae (Pigeons and Doves) – 25 species
2. Psittacidae (Parrots) – 25 species
3. Rallidae (Rails) – 23 species
4. Fringillidae (Finches) – 19 species
5. Zosteropidae (White-eyes and Yuhinas) – 19 species

Why haven’t we found them?
To begin with, practically all of the birds on this list occur in highly isolated areas that are challenging or treacherous for birders to visit. On top of that, many of the species are rare, secretive, or both. Some, such as the Cayenne and Nechisar Nightjars, are known with certainty from just a single specimen. However, the secretive nature of some these species are not the only impediments delaying their entry into eBird’s database. Tragically, political conflict prevents a significant portion of these species from being seen or studied. Most notably, ongoing conflicts in Somalia and the Democratic Republic of the Congo make finding endemics like the Somali Thrush and Congo Bay-Owl incredibly dangerous, and instability in both regions shows few signs of abating in the foreseeable future.

The Ou is one of many Hawaiian finches (Hawaiian Honeycreepers) that went extinct well before eBird was invented, so acquiring records is going to depend on the entry of historical notes. This story is all-too-common with island endemics.

Where do we go from here?
Given the elusive nature of many of the species on this list, as well as the potential dangers of traveling to some of their native ranges, actively pursuing these species could be exceedingly challenging, not to mention dangerous. However, there are far easier ways eBirders can help. If you have any paper checklists which you think might contain records for one or more of these species, or if you know someone who might, now is the perfect time to get those sightings entered. eBird continues to expand its global partnerships (e.g., Eremaea eBird) and this influx of records has already added new species to eBird. Please share eBird with your friends at home and abroad; the larger the eBird community, the better the coverage for all birds, from the very rare to the very common!

On the whole, this list serves as an indicator of where ornithological study, as well as avian conservation, is most needed. The more data we can gather about birds that are highly threatened or have limited ranges, the more effective future conservation programs can be.

There are many ways that you can help the species on this list. BirdLife International runs a number of conservation efforts dedicated to preserving avian biodiversity, such as the Preventing Extinctions and Invasive Species Programmes. Similarly, supporting the Cornell Lab of Ornithology helps further conservation science methods and techniques, while donating to eBird allows us to continue monitoring global patterns of bird abundance and distribution. Contributing to these organizations and others like them supports birds both locally and abroad.

Download eBird’s Missing Species as spreadsheet or PDF

Thanks again to Cornell students Andrew Dreelin and Reid Rumelt for this analysis and contribution. Andrew and Reid both wanted to thank Jeff Gerbracht–our lead application programmer at eBird– for his help and support in keeping the list updated during the time it took to write this article. On particular, he helped to reassess the records when data from Eremaea Birds was ported into eBird, which subtracted a number of species from the “missing species” list.”

Andrew and Reid provided these short bios. If you run into them in the field, be sure to say hi!

Andrew Dreelin is a rising sophomore in the College of Arts and Sciences at Cornell University. He fell in love with birds, birding, and Ornithology during high school, and since then he has committed himself studying ecology and evolutionary biology, with the ultimate goal of becoming a conservation biologist. He is a passionate eBirder and helps with eBird and other projects at the Cornell Lab of Ornithology. Andrew wrote the Jamaican Pauraque account for Neotropical birds.

Reid Rumelt is a rising Junior in Cornell University’s College of Agriculture and Life Sciences. He is passionate about applying his interests in computer science and analytical biology to pressing questions in ecology and conservation. His interests have led to his participation in a number of citizen science initiatives, including eMammal and the Encyclopedia of Life project at the Smithsonian Institution, as well as eBird and Merlin at the Cornell Lab of Ornithology.

The Jamaican Pauraque was last recorded in 1860, so probably is lost for good. Be sure to read the Neotropical Birds account by Andrew Dreelin for the full account what little is known of this single island endemic.

References: 

BirdLife International (2014) IUCN Red List for birds. Downloaded from http://www.birdlife.org on 03/02/2014

Clements, J. F., T. S. Schulenberg, M. J. Iliff, B.L. Sullivan, C. L. Wood, and D. Roberson. 2013. The eBird/Clements checklist of birds of the world: Version 6.8. Download it here.

eBird Basic Dataset. Version: EBD_relNov-2013. Cornell Lab of Ornithology, Ithaca, New York. November 2013.

IUCN 2013. IUCN Red List of Threatened Species. Version 2013.2. www.iucnredlist.org. Downloaded on 03 March 2014.