daniel5's Journal

Thanks to platforms like iNaturalist, even in these bleak times for science, making notable contributions to our knowledge, particularly for understudied taxa, is almost stupidly easy. Before the Internet, taxon specialists had to make extensive travels and correspondence to compile a general distribution map, and their efforts were nothing short of heroic. Now, the number of local correspondents is virtually limitless, and the amount of detailed data they give is truly unprecedented.

Case in point: In just the last week of browsing audio observations of crickets and other insects from the U.S., I managed to find first documented state records for three species and three states. While none of these new records is all that surprising, they demonstrate the incredible impact of people simply going outside, observing what’s around them, and submitting to a website. When iNaturalist started allowing direct upload of audio observations from its apps, it truly opened the floodgates. I can attest that many, many audio observations come in every day, particularly during mating seasons. Research papers have covered iNaturalist’s incredible effect far better than I can, but I thought I’d just add my two cents with the following examples from recordings of singing crickets (note that many of these are not yet research-grade):

• The False Jumping Bush Cricket (Hapithus luteolira) was previously known only from peninsular Florida and the Atlantic Coastal Plain north to Virginia. Based on song, habitat and season, it now has new state records for Alabama (#1, #2, #3, #4) and from Mississippi. The very closely-related and much more widespread Jumping Bush Cricket (Hapithus saltator) has been known to occur in both states for decades, and audio observations on iNaturalist back this up. The two species cannot be distinguished morphologically, so it is very possible that some photo-only observations of H. saltator on iNaturalist could refer to H. luteolira instead, and vice versa.
• The Jamaican Field Cricket (Gryllus assimilis) is native to Texas, introduced in Florida, and also occurs widely outside the United States. There are now two audio observations from Louisiana (#1, #2]. I presume it has spread there naturally from Texas.
• The Pink-spotted Trig (Anaxipha rosamacula), first described only in 2014, occurs in much of the Southeast from Georgia to Florida and Texas. It now has two observations from Alabama (#1, #2. Additionally, @dan_johnson may have been the first to find both this species and the Florida False Katydid (Amblycorypha floridana) in Texas.

Finally two bonuses in which I played a small part:

• This photo was the first observation of the Florida True Katydid (Lea floridensis) from the South Carolina Sea Islands. I independently discovered this species during a trip there in 2024 and again in 2025. My recordings on Xeno-Canto and my two iNaturalist observations (#1, #2), were the first state records documented with audio. It now appears that the first observation from the South Carolina mainland has finally been added.
• Lastly, not a new state record, but I may (may) have gotten the first iNaturalist observation of the Banded Scaly Cricket (Cycloptilum albocercum).

All this to say that there has really never been anything quite like iNaturalist when it comes to collecting biodiversity data, especially for taxa that don’t get a lot of attention. The reach and resolution we get with this continually growing treasure trove is almost indescribable. So, a gigantic shout-out to iNaturalist itself, and perhaps most especially to all those humans curious enough about the natural world to record what they hear (even crickets!) And help science in the process. Keep it up, everyone!

Following is a little article I just posted on Xeno-Canto dealing with the different call notes of various populations of a Caribbean thrush species. I did not use any recordings from iNaturalist for this piece, mostly due to the paucity of audio here for the Red-legged Thrush, but I thought it was worth posting in my journal so it can be accessible to people who don't know of Xeno-Canto. The article itself is more or less a sequel to Peter Boesman's analysis of Red-legged Thrush songs. My part is nothing special, but I hope it can be of use or interest to someone. The text is copied below, but you have direct access to play the recordings from your browser by clicking on the Xeno-Canto link. In the future I may edit this so that the web players are embedded here, too.

---

This is a preliminary and amateur study of the call types (as opposed to songs) of the Red-legged Thrush (Turdus plumbeus) as currently constituted by most major checklists. Specifically, this article intends to provide a list of calls by island or island group using recordings from Xeno-Canto (XC) and the Macaulay Library (ML). The goal is to summarize what presently seems to be known about the repertoire of each and to catalog differences where they may exist. As this complex has recently been under taxonomic review, I hope that this article will provide a starting point from which to assess differences in calls as is already happening with songs (e.g., Boesman (2016)). No attempt is made to determine the function of any given call type, or whether any call types may be analogous. Nor, for the time being, are any differences scored using criteria such as those found in Tobias et al. (2010), but others are encouraged to quantify this should they find anything of note in these results.

Background and Literature Review

As currently understood by most checklists, Red-legged Thrush is composed of six extant subspecies:

• T. p. plumbeus of the Bahamas,
• T. p. rubripes of Cuba except the east,
• T. p. schistaceus of eastern Cuba,
• T. p. coryi of Cayman Brack (Cayman Islands),
• T. p. ardosiaceus of Hispaniola and Puerto Rico, and
• T. p. albiventris of Dominica.

Although unnoticed at the time, I had heard and recorded this species on a trip to the Dominican Republic in April 2022. I had misidentified its soft wuk-wuk calls (see below) as conversational notes of the abundant Hispaniolan Woodpecker (Melanerpes (Chryserpes) striatus), whose conspicuous laugh calls can be reminiscent of a whinnying Turdus, e.g. T. migratorius. This was corrected after comparing recordings in the Macaulay Library. However, if I had heard any thrush songs at all, they were very brief and easily overlooked, which factored into the misidentification. For this reason, and because others have analyzed the song in detail (e.g., Boesman (2016)), the focus of this article will be on calls only, of which there may not yet be a published analysis broken down by subspecies or geographic area..

In several 2022 proposals regarding taxonomy of Turdus thrushes to the North American Classification Committee (NACC) of the American Ornithological Society (AOS), committee member Van Remsen noted that “voice is the key indicator of species status in Turdus and relatives”, citing the discovery of T. sanchezorum by voice alone as evidence. As for T. plumbeus, I think there may be some overlap in the various groups, as also noted by Boesman). Since Turdus are oscines, there is also the issue of learned songs and dialects to consider. All this is to say I have not quite determined exactly what in Turdus songs may signal a specific difference and what is mere dialectal variation, or as in the case of T. migratorius at least, individual variation. However, as calls are presumably mostly innate, they should be reliable indicators of difference no matter what the songs, if it can be determined which calls are analogous. The latter is beyond the scope of the present article.

Notes

• Though a passionate birder, I am not a professional scientist and have limited background in technical sound analysis or taxonomy. I encourage feedback and technical corrections in the comments and will reflect these in future edits.
• As my blindness makes reading and interpreting sonograms impossible, phonetic renderings and other descriptions will be used here. However, anyone who wishes to analyze sonograms and add their findings would be most welcome and appreciated.
• I would like to mention Simon Colenutt’s recordings ML561208711, ML561630671 and ML561632431, the first recordings of song from albiventris on Dominica, to my knowledge (David Hollie’s recordings of calls from Dominica are listed below). One recording seems similar to the whistled song of ardosiaceus rather than the harsher songs from Cuba, but is perhaps higher and shriller on average than ardosiaceus. However, another from the same general area seems harsher and more akin to Cuban examples, and a third seems to be in the middle. Clearly, more study is warranted. Additionally, in May 2024, Nicole Martin uploaded what may be the first recording of the song of coryi to ML.

Methodology

There are currently 306 recordings of T. plumbeus between XC and ML, including duplicates. In some ML recordings, some clearly call-type vocalizations are listed as songs or not labeled at all, so every recording, with the exception of some longer than about ten minutes, was analyzed for the sake of completeness. Each recording found to contain a call is roughly classified in Appendix 1.

For the purposes of this article, song of T. Plumbeus is defined as any vocalization consisting of whistled or squeaky phrases, possibly with some immediate repetitions but normally with spaced repetitions, with regular pauses, even those with the sound type listed as ‘call’ in collections. Whether or not T. plumbeus uses song in call-like circumstances, such as alarm, is another topic beyond this article’s scope. However, out of an abundance of caution, recordings labeled as calls but consisting of a single, repeated, clearly song-like phrase with no different phrases for the duration have been included below and marked with an asterisk (*).

Results

In an effort to isolate vocal data from current subspecies boundaries, recordings are listed in Appendix 1 by island or group. While this is imperfect as there are two subspecies on Cuba, this article is concerned mostly with differences between the three major subspecies groups (i.e., plumbeus, rubripes / schistaceus / coryi and ardosiaceus / albiventris). Differences within groups are noted separately. Recordings from Cuba are listed in the appendix with the province where they were taken, as well as any subspecies identification given by the recordist, so as to more easily compare them with traditional subspecies boundaries.

Following is a list of call types sorted by island or group, with representative XC recordings where possible. Appendix 1 has the full list with ML recordings as well:

Bahamas

• Rising or level weep, from slow, low and squeaky to high, fast and trilled: XC277422
• See-there *: XC331868
• Shrieking weeeeeep: XC50147
• Low wok-wok, clearer and lower-pitched than on Hispaniola: ML611564854

Cuba

• Rapid, squeaky wup-wup-wup or riwp-riwp-riwp: XC834456 (Pinar del Río), XC659373 (Holguín)
• High piercing eet: ML167512401 (La Habana)
• Whining, scolding waaa: XC399145 (Artemisa)
• Higher wurrip-wurrip, trilled but not sharp: ML395902911 (Pinar del Río)
• Scratchy tsiewp *: ML547120351 (Pinar del Río)
• See-there *: XC548271 (La Habana)
• Quavering, nasal dzeyeyeyeep: ML615443403 (Ciego de Ávila)
• Knocking dup-dup-dup: ML614239529

Dominica

• Whiny, scolding vriwp-vriwp, much slower than on Cuba, higher than on Hispaniola: ML602434051
• Nasal, squeaky tick-tick: ML608413069

Florida, United States (vagrant)

• Rising weep, nasal and whiny: ML176031491

Hispaniola

• Low, nasal wuk-wuk-wuk, often in series, may grade into long higher squawks, but not harsh: XC800407
• Higher wurrip-wurrip, trilled but not sharp: ML145705
• Trilled weer-weer in series, faster and shorter than in Puerto Rico: ML163158491
• Whinnying chew-chew-chew at various pitches: XC800407, ML145627
• Squeaky would-it?: XC800407
• Burry cheer * (cf. XC103837): XC103838
• Whining, scolding “waaa”: ML145704)
• Emphatic chip (could be Northern Mockingbird or Palmchat): ML600420261

Puerto Rico

• Harsh, sharp rup-rup or rup-wurrip often in series: XC173659, XC32054, XC173413
• Whistled or squeaky pweer-pweer-pweer in series: XC32057
• Squeaky, emphatic quiwp: XC173668
• Low, nasal wuk-wuk-wuk, often in series, may grade into long higher squawks, but not harsh: ML566783761
• Whistled mew *: ML546245111
• Squeaky would-it?: ML530498561
• Low, rapid wup-wup-wup: ML615760926
• Quavering, nasal dzeyeyeyeep: ML566783761
• Rich, burry cheer *: ML594348141

Discussion

As is to be expected, many of the calls between islands or subspecies are broadly similar, but unlike the repertoire of T. migratorius, for example, there are clear differences between groups. The riwp-riwp of Cuban birds seem absent, as such, from the repertoire of ardosiaceus, and the variable weep-weep of Bahamian birds may or may not be analogous to the wuk-wuk or rup-wurrip of ardosiaceus. Also, on average, the riwp-riwp in Cuba seems to be more rapid and less pure-toned than the weep-weep in the Bahamas. Conversely, the slow, low wuk-wuk or rup-wurrip of ardosiaceus is not heard from birds to the north and west, though it seems likely that the aforementioned calls from the Bahamas and Cuba are analogous, as they are the most common call for their respective group. The see-there call (or song phrase?) of the Bahamas and Cuba also does not seem to exist in ardosiaceus, although the use of analogous phrases by that subspecies cannot be ruled out. Two calls so far only heard from ardosiaceus are the squeaky would-it?, not commonly given but almost identical between the islands, and the whinny, whose analogues in other groups are unknown.

Also intriguing is the variation within ardosiaceus. Specifically, the burry rup-wurrip call so common on Puerto Rico may be replaced by the nasal wuk-wuk grading into a squawking waak-waak on Hispaniola, though this is based on a very small sample size. One recording from Puerto Rico, ML566783761, has birds which appear to give a fast wuk-wuk-waak-waak leading up to rup-wurrip in the midst of many other calls. I do not know whether this is indeed the same call given by Hispaniolan birds or if it only sounds that way superficially, or if Puerto Rican birds commonly use it in situations where Hispaniolan birds do also. In most recordings, where a Hispaniolan thrush would give wuk-wuk or waak-waak for minutes on end, as I myself heard there, the Puerto Rican birds give rup-rup or rup-wurrip with the same delivery pattern. We have no equivalent call recordings from albiventris to compare, and those calls we do have sound rather different from any ardosiaceus vocalizations. I attribute this mostly to small sample size on Dominica, but a real vocal divergence would be an interesting find. In the case that vocal and other evidence should lead to the eventual recognition of an endemic Puerto Rican species or subspecies, the name Turdus ardosiaceus portoricensis Bryant, 1867, should be available (Lepage, 2024).

Limitations and Recommendations for Future Research

With the exception of one recording I made myself, this analysis was conducted entirely away from the Caribbean using data gathered there by a number of diligent residents, visitors, birders, and scientists. The majority of islands, and call types on each island, have very few recordings, so contextual information is lacking, and we have no way of knowing how often a particular call may be given and which situations may elicit different calls. Consequently, I am limited by the data available to me, and I had no ability to conduct experiments with living specimens. Future studies should perform full spectrographic analyses, attempt to determine analogous calls between groups and score these with criteria such as those in Tobias et al. (2010). Playback experiments should test the responsiveness of birds on one island to calls from another, e.g., playing the Puerto Rican rup-wurrip to Cuban birds. Tests should also be done between Hispaniolan and Puerto Rican ardosiaceus regarding the wuk-wuk and rup-wurrip calls and might also investigate whether both populations can produce either call. Lastly, we need many more recordings from all islands, but particularly Cayman Brack and Dominica, to fully document the repertoires there.

Acknowledgements

I would first like to express profound gratitude to the many local birders, ornithologists, conservationists and others in the Caribbean who are fighting to preserve their islands, communities, and precious natural resources, including endemic birds, which are uniquely affected by the worst of the climate crisis and other anthropogenic hazards. Those who appreciate and love the Caribbean, its people and its birds from afar should do everything in our power to support them and to encourage eco-friendly policy changes at home. Additionally, this small study would not have been possible without the many recordists who have spent their time with these birds and uploaded their data to Xeno-Canto and the Macaulay Library, and without the hosting services for these sites provided by Naturalis Biodiversity Centre and the Cornell Lab of Ornithology, respectively.

References

Boesman, P. (2016). Notes on the vocalizations of Red-legged Thrush (Turdus plumbeus). HBW Alive Ornithological Note 310. In: Handbook of the Birds of the World Alive. Lynx Edicions, Barcelona. https://doi.org/10.2173/bow-on.100310
Larsen, N. (2024). Red-legged Thrush (Turdus plumbeus), version 2.0. In Birds of the World (S. M. Billerman, Editor). Cornell Lab of Ornithology, Ithaca, NY, USA. https://doi.org/10.2173/bow.relthr1.02

Lepage, D. (2024). Turdus ardosiaceus ardosiaceus Vieillot, LJP 1822. Avibase. https://avibase.bsc-eoc.org/species.jsp?lang=EN&avibaseid=D8710D4CBA7024BE&sec=synonyms

Remsen, V. (2022). Proposal 2022-A-3a: Treat Turdus plumbeus (Red-legged Thrush) as two species. / Proposal 2022-A-3b:  Treat Turdus plumbeus (Red-legged Thrush) as three species. In 2022 Proposals. North American Classification Committee (NACC) of American Ornithological Society (AOS)https://americanornithology.org/wp-content/uploads/2022/04/2022-A-final.pdf.

Tobias, J.A., Seddon, N., Spottiswoode, C.N., Pilgrim, J.D., Fishpool, L.D.C. & Collar, N.J.
(2010). Quantitative criteria for species delimitation. Ibis 152(4): 724–746.

Appendix 1

Following is the full List of Recordings used for this survey, Sorted by Island and Call Type:

Bahamas

• Rising or level weep, from slow, low and squeaky to high, fast and trilled: ML14108, ML115837, ML48557751, ML173619821, XC277422
• See-there *: XC331868 / ML33097180, ML614669324
• Shrieking weeeeeep: XC50147
• Low wok-wok, clearer and lower-pitched than on Hispaniola: ML611564854

Cuba

• Rapid, squeaky wup-wup-wup or riwp-riwp-riwp: XC834456 (Pinar del Río), XC659373 (Holguín), ML624817505 (Cienfuegos), ML325038931 (rubripes; Matanzas), ML274114771 (Florida vagrant), ML274114751 (Florida vagrant), ML273108891 (Florida vagrant), ML270780611 (Florida vagrant), ML203701321 (Cienfuegos), ML614793508 (Ciego de Ávila), ML182695161 (La Habana), ML162222051 (La Habana), ML164218401 (La Habana), ML167512401 (Matanzas), ML179882551 (La Habana), ML426850351 (Mayabeque, ML611953337 (Holguín), ML611953503 (Holguín); ML613135476 (Las Tunas), ML614793508 (Ciego de Ávila), ML616441709 (Artemisa), ML617746046 (Villa Clara), ML619035939 (Holguín), ML619880271 (La Habana), ML619880272 (La Habana)) High piercing eet: ML167512401 (La Habana)
• Whining, scolding waaa: XC399145 (Artemisa), ML610231573 (Cienfuegos), ML613159304 (Las Tunas), ML615592961 (Ciego de Ávila), ML617007451 (Matanzas), ML621800477 (Artemisa), ML619627832 (La Habana), ML622613897 (Ciego de Ávila), ML226824781 (La Habana)
• Higher wurrip-wurrip, trilled but not sharp: ML395902911 (Pinar del Río)
• Scratchy tsiewp *: ML547120351 (Pinar del Río), ML413581291 (Artemisa)
• See-there *: ML395902921 (Pinar del Río), XC548271 (La Habana), ML226824781 (La Habana), ML229055231 (La Habana), ML614239529 (Ciego de Ávila)
• Quavering, nasal dzeyeyeyeep: ML615443403 (Ciego de Ávila)
• Knocking dup-dup-dup: ML614239529 (Ciego de Ávila)

Dominica

• Whiny, scolding vriwp-vriwp, much slower than on Cuba, higher than on Hispaniola: ML602434051
• Nasal, squeaky tick-tick: ML608413069

Florida, United States (vagrant)

• Rising weep, nasal and whiny: ML176031491, ML166365001

Hispaniola

• Low, nasal wuk-wuk-wuk, often in series, may grade into long higher squawks, but not harsh: XC800407, ML35342, ML145705, ML180738, ML527582601
• Higher wurrip-wurrip, trilled but not sharp: ML145705, ML163144881, ML163145301, ML163158491
• Trilled weer-weer in series, faster and shorter than in Puerto Rico: ML163158491
• Whinnying chew-chew-chew at various pitches: XC800407, ML145627, ML180738
• Squeaky would-it?: XC800407
• Burry cheer * (cf. XC103837): XC103838, ML145704
• Whining, scolding “waaa”: ML145704)
• Emphatic chip (could be Northern Mockingbird or Palmchat): ML600420261

Puerto Rico

• Harsh, sharp rup-rup or rup-wurrip often in series: XC173659, XC32054, XC173413, ML129735, ML53130, ML52727471, ML52796611, ML465086301, ML471781401, ML530498561, ML566783761, ML614448564, ML615205396, ML616070868, ML624087461
• Whistled or squeaky pweer-pweer-pweer in series: XC32057, ML14103, ML83564251, ML140209, ML188070, ML188091, ML566783761, ML614448564, ML615820806, ML615881062
• Squeaky, emphatic quiwp: XC173668, ML471781671, ML560210231, ML621341024
• Low, nasal wuk-wuk-wuk, often in series, may grade into long higher squawks, but not harsh: ML566783761
• Whistled mew *: ML546245111
• Squeaky would-it?: ML530498561
• Low, rapid wup-wup-wup: ML615760926
• Quavering, nasal dzeyeyeyeep: ML566783761
• Rich, burry cheer *: ML594348141

It may be getting later in the summer, but the Common True Katydids (Pterophylla camellifolia) are still singing! There is a veritable flood of new recordings in iNaturalist just over the last few seasons, and more being added every day. I have begun analyzing these to figure out what Inat data can tell us about the distribution of katydid song types (this page contains sound files for all the different types as well as maps with their approximate distribution). However, at the time those maps were created, INaturalist wasn’t a thing!

Thanks to you all, we can now get a picture down to the county or town level of what katydids sound like, and there are already surprises! The katydids sounding consistently different from Fort Worth to Dallas? Song type transition zones around St. Louis and Kansas City? Mosaic patterns in multiple states? The Southeastern song type being heard as far north as Maryland or possibly New Jersey? A song type known from Iowa and Missouri being recorded for Inaturalist in Arkansas and possibly Oklahoma? All this and more!

So once again, go and record your local katydids! If you think you’ve found something interesting, no matter what it is, feel free to tag @daniel5. Also, if you’re interested in contributing to my amateur effort at sorting through recordings, please get in touch!