Hi Kevin,
This is an excellent question, and I'm happy to weigh in. I wrote the book on bird sounds (well, one of the books), and I use Merlin Sound ID all the time.
I used to do bird surveys for Bird Conservancy of the Rockies, but had to quit a number of years ago when I realized my high-pitched hearing was much worse than other surveyors'. During training, everybody else could hear the Blue-gray Gnatcatcher at 100 yards and I still couldn't hear it at 50. I feared my bad ears would skew the data.
But nobody has exactly the same abilities, either physical or mental. Differences between observers are an inevitable confounding factor in all big-data citizen science initiatives, including the Christmas Bird Count, eBird checklists, and Breeding Bird Survey routes. The idea is that these differences mostly wash out across the huge number of repeated surveys over time.
Merlin helps me be more accurate. It can hear things I can't. But it also makes lots of mistakes. In some situations it performs better than the average human, in some situations worse. It helps to know what it does well and what it does poorly.
I don't think we should be too afraid of assistive technology -- after all, eyeglasses and binoculars and scope are assistive technologies that greatly change detection rates of birds. But we shouldn't let technology operate unsupervised by human judgment. If Merlin detects a bird I haven't detected, I pay attention to context. It's got a great track record of detecting real Bushtits and Cedar Waxwings before I do. If it's detecting those species repeatedly during an observation, in decent habitat, I might put them on my list. I usually like to be able to see the shape of the call on the spectrogram, though, if I can't spot the birds.
If it's telling me there's an Indigo Bunting around, I'm going to be skeptical. Merlin can't tell Indigo from Lazuli in Colorado. Neither can humans. Neither can buntings. (That's why we have a lot of hybrids here -- the two species learn elements of each other's songs in the overlap zone.) With experience and study, you can learn when to place more trust in technology and when to place less. It's not a matter of using it or not using it, it's a matter of using it carefully.
Nathan Pieplow
Boulder, Colorado
On Fri, Feb 6, 2026 at 6:09 PM Kevin Schutz <kschutz@gmail.com> wrote:
Hello CoBirds Community,Today's post by Bill Kosar and subsequent responses prompted me to start the following discussion.First, some background. We all hear and read of the risks and appeals of not relying upon common technology in popular use today - Merlin, etc. In the past, I've been forthright in admitting some of my identification limitations when inquiring about participation with various surveys and describing how I use multiple tools to (personally) learn and narrow on any reported identifications I may make. As such, I have been declined for participation - no technology allowed, etc. That's fine as I know the intent is to provide (reasonably) accurate data for various scientific purposes. What I've found curious is reading subsequent posts from other participants for the same surveys indicating their use of technology. Argh!! - so some routes potentially went uncovered. We know of examples of published experts/authors of books covering their "big year" efforts and garnering numerous speaking engagements afterwards that use technology such as frequency shifting headphones to assist them with identification while birding.I admit confusion, especially in the context of mixed inputs pleading for more data reporting juxtaposed against pleas and warnings that at times feel more like one is receiving a "thou shalt not" style sermon. When is technology use appropriate when recording an identification? Does use of technology depend on specific surveys/records/databases? What constitutes "technology" and how is technology defined? Does technology include sound amplification headphones, recordings that can be compared post observation at a later time against vetted libraries, optics, photography, electronic or printed guide books? Even consultation to more experienced birders relies upon some form of technology (vocalizations, photographic confirmation, etc.). All of the examples listed above could be characterized as "technology" in the context of humankind, and in some cases would seem to be dismissed out-of-hand.Today, within eBird, when one submit a checklist, one is asked "Are you submitting a complete checklist of the birds you were able to identify?" I think eBIrd used to ask something along the lines of "... to the best of your ability", but I can't attest to that with certainty. I've always adopted the philosophy that I would record identifications to the best of my ability, which includes the use of various forms of technology to assist me with a confident identification.What are the current, best practices deemed acceptable today for bird identification? Should technology use be context specific (eBird database, bird surveys, other...)? While database corruption is and always will be a concern, are we artificially limiting community science resources over such concerns? Humans will always be fallible. When technology limitations are appropriate, how much cheating is likely occurring? Are we at a point where we are past being able to use an honor system, of relying upon one's best abilities? When some form of technology has been used to assist with an identification, is it incumbent to disclose all forms of technology used (optics, photography, recordings, various forms of guidebooks, applications such as Merlin, various AI applications)?I am sincerely interested in understanding the breadth of views present. I may find myself having to reconsider my philosophy of using "my best ability" as no longer being appropriate.I hope this post results in a respectful, thoughtful discussion.KSEl Paso County--
--
You received this message because you are subscribed to the Google Groups "Colorado Birds" group.
To post to this group, send email to cobirds@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/cobirds
* All posts should be signed with the poster's full name and city. Include bird species and location in the subject line when appropriate.
* Join Colorado Field Ornithologists https://cobirds.org/membership/
---
You received this message because you are subscribed to the Google Groups "Colorado Birds" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cobirds+unsubscribe@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cobirds/e7168f62-d4fa-4ae9-a246-23797720569cn%40googlegroups.com.
--
You received this message because you are subscribed to the Google Groups "Colorado Birds" group.
To post to this group, send email to cobirds@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/cobirds
* All posts should be signed with the poster's full name and city. Include bird species and location in the subject line when appropriate.
* Join Colorado Field Ornithologists https://cobirds.org/membership/
---
You received this message because you are subscribed to the Google Groups "Colorado Birds" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cobirds+unsubscribe@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cobirds/CAFhaDVJ-0pRZjJveTBKk_GaUurh8L0Ak_Npqak-7-wk%3DH7jrLQ%40mail.gmail.com.
No comments:
Post a Comment