A word about MDLepsOdes as we end the 2021 season

[Rick Borchelt]

Thanks everyone for continuing to make MDLepsOdes a valued complement to the field observations we make on butterflies, moths, and odes.  I've learned a tremendous lot of things from all of you this year, and that's informed a lot of my thinking about the nature and future of this listserv.

You may remember I asked for feedback on the utility of MDLepsOdes earlier in the year, and appreciate the useful and candid comments I got back.  Given the responses, I plan to keep MDLepsOdes going for at least 2022.  We have 113 members currently, and continue to grow.  I hope you'll all continue to post here!

I also toyed with the idea of name change so that folks from other municipalities in the region felt welcome to post.  Because we have pretty good name recognition, I've decided to keep the name as it is but add a welcome in the "About" section:

"We won’t be too picky about geographic limits of MDLepsOdes as long as the location is within a day trip’s distance for most Marylanders. That includes most of VA, much of PA, DE, DC, eastern WV and southern NJ.  If you have an observation you believe would be of interest or value in helping our list members better understand local butterfly fauna, you're welcome to post here -- worse that could happen is I might take it down, but that's an option I use exceedingly sparingly."

A lot of you commented during the discussions about the future of MDLepsOdes that you were not on Facebook and never would be, or that you didn't or wouldn't use iNaturalist. I understand that -- both are heavily focused on pictures, and folks who don't have a photo of their sighting I think often feel reluctant or shamed into not posting.  That isn't the case here.  MDLepsOdes serves (in my mind) as one of the few such photo-neutral platforms remaining -- most other butterfly listservs have quietly pulled up stakes or gone silent.  

I think there is still an active and loyal following here at MDLepsOdes. And, for that matter, for the weekly butterfly field forecast I do for LepLog and post here.  I'm still on the fence about the LepLog forecasts; it's a pretty huge undertaking.  But I also see LepLog as finding regional observations and reporting them for those of you who aren't on multiple platforms, and I think it's particularly useful to those of us in the MDLepsOdes "catch basin."  So watch that space as we get to April (the official start of the weekly forecasts).  

Even though the season is almost at an end (we're still seeing some sulphurs about, and the occasional checkered-skipper or variegated frit), there are plenty of other things we can and I hope will discuss here:  nomenclature changes, field guide or gear reviews, interesting butterfly research, winter observations, plans for field trips in 2022, etc.  Please share these during the winter hiatus.

A very happy winter holiday to you all; I look forward to continuing to see all my colleagues and friends here on MDLepsOdes for some time to come!

[Harry Pavulaan]

Rick:

 

Great job keeping MDLepsOdes going, as well as the LepLog blog.  These are a valued source of regional information and a great “meeting place” for our region’s naturalists.  I’ll cc all my future Leesburg, VA. posts to MDLepsOdes (I have another sighting today!)

 

There will be a lot of interesting tidbits coming up, which I will post to the group.  One concerns our region’s population of Euphydryas phaeton (Baltimore).  Our region’s population was once described as a subspecies (schausi) but was completely ignored and forgotten in the literature.  I plan to highlight and “ressurect” this subspecies in a research paper, plus it will be renamed due to a technicality of the International Code Of Zoological Nomenclature.  The presence of a regional subspecies is very important to know because there have been discussions of attempts at introducing nominate phaeton stock from regions up north, which, while it might create a local colony, it is not advisable due to differing genetic stock from our local populations.

 

Going long here.

 

Harry Pavulaan

Leesburg, VA.

 

Sent from Mail for Windows

--
You received this message because you are subscribed to the Google Groups "MDLepsOdes" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mdlepsodes+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/mdlepsodes/51f54e0d-771c-42e1-9f50-17c8184e490bn%40googlegroups.com.

 

[Rick Borchelt]

The Euphydryas news is very exciting!b be is schausi as it was described the only obligate turtlehead species (at least in early instars)?

On Dec 4, 2021, at 4:28 PM, Harry Pavulaan <harr...@hotmail.com> wrote:



To view this discussion on the web visit https://groups.google.com/d/msgid/mdlepsodes/BL0PR05MB5474EFAA12DF9C913B5EE082D86B9%40BL0PR05MB5474.namprd05.prod.outlook.com.

[Harry Pavulaan]

Rick:

 

I’ve mined the literature for information on Baltimore (Euphydryas phaeton) hostplants.  Chelone glabra (White Turtlehead) was the primary host from Virginia to Nova Scotia until the New England populations switched over to Plantago lanceolata (Plantain) as their primary host.  That’s within the nominate phaeton range.  It will be interesting to see if this switchover remains within the nominate population (southern New England to New Jersey then west to the Great Lakes).  That would be interesting as it would indicate some early form of evolution within the Baltimore (hostplant changes within a butterfly population segment over the millennia can lead to speciation).  If it remains within the nominate population, that could make it easier to determine if schausi and borealis have a more defined range with nominate phaeton, but I believe these are all part of a huge north-south cline.

 

So far, our local subspecies schausi and Canadian Maritimes subspecies borealis remain primarily Turtlehead feeders and have not switched to Plantain. 

 

I know, some people are probably wondering where these subspecies (schausi and borealis) popped up from.  My paper discusses the description and naming of these – all valid subspecies names per the I.C.Z.N. code.  Upcoming…

 

I often wonder how far the switch to Plantain will spread.  In southern New England, Plantain is a common plant and often fills entire fields, thus the explosion of Baltimores in some localities.  I have not seen such profusion of Plantain in northern Virginia.

 

Harry Pavulaan

To view this discussion on the web visit https://groups.google.com/d/msgid/mdlepsodes/DCC71BEB-B55C-46A6-8EAD-5B536CCC151A%40gmail.com.

 

[Rick Borchelt]

Interesting that turtleheads used to be placed in the figwort family, Scrophulariaceae. Well after the New England checkerspots switched to plantain as a primary host plant, botanists reclassified turtlehead drom figworts into Plantaginaceae — the plantain family. Many of the other hosts for later instars that have been reported for turtlehead specialist checkerspots are also Plantaginaceae. Euphydryas apparently makes a passable plant taxonomist!

To view this discussion on the web visit https://groups.google.com/d/msgid/mdlepsodes/BL0PR05MB5474E6F2A85EDFAA25E8489AD86B9%40BL0PR05MB5474.namprd05.prod.outlook.com.

[mmon...@aol.com]

Thank you, Rick, for all you do to make us a cohesive and communication community.  I've made friends (like yourself) through this group and traveled from PA to MD to see butterflies and dragonflies and flowers I haven't seen anywhere else. I'm looking forward to our collective adventures in 2022! Thanks again!

Monica

Allegheny Co, PA

-----Original Message-----
From: Rick Borchelt <rbor...@gmail.com>
To: MDLepsOdes <mdlep...@googlegroups.com>
Sent: Sat, Dec 4, 2021 1:20 pm
Subject: [MDLepsOdes] A word about MDLepsOdes as we end the 2021 season

[lakek...@gmail.com]

Rick:

I appreciate this group and your efforts in keeping it going. A regional lep and ode group fills a niche that the national sites cannot achieve.  I do have one issue and that is the seeming prejudice against iNaturalist that I read expressed in some posts. I use both this site and iNaturalist.  Some of my photos I submit to INat are good and some are not, but I have never felt ashamed to post there. Nor do I believe that the iNat reviewers are less capable than on other sites like BugGuide for example.

Jim Wilkinson

Columbia, MD

[Rick Borchelt]

I think the crowd sourcing function on iNat for difficult IDs is a showstopper for me most of the time; remember I go through all MD, VA, and PA iNat records weekly in the summer. And both BugGuide and iNat are biased toward photos, which in my mind skews the whole enterprise toward showy butterfly megafauna. Much prefer the field lists we share on MDLepsOdes.  Nothing against iNat proper but — as a recent academic paper on community observation platforms points out — it’s not a good place for making or informing decisions about biodiversity. I too rely on and post to iNat for educational purposes but am keenly aware of its process and geographic shortcomings. 

On Dec 11, 2021, at 4:44 PM, lakek...@gmail.com <lakek...@gmail.com> wrote:

Rick:

--
You received this message because you are subscribed to a topic in the Google Groups "MDLepsOdes" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/mdlepsodes/732T66hj0qQ/unsubscribe.
To unsubscribe from this group and all its topics, send an email to mdlepsodes+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/mdlepsodes/af0f1635-9d0c-4919-aedb-71bb571087e4n%40googlegroups.com.

[Harry Pavulaan]

Rick:

 

And - the “community consensus” I.D. method is not always correct.  I have seen many records where a number of observers agreed on a misidentification, and it gets moved to “research grade”.  I turn out to be the lone voice in disagreeing, and the status of the misidentified species never gets updated because I was outvoted or the reviewers don’t seem to notice corrections.  Problem is, many of the people “voting” on an I.D. are rank amateurs who have trouble identifying any species in the problem groups or even some common species (I know this from multiple mis-ID’s from the same people).  Also, some are using NABA terminology, which is vastly outdated and biased to species-level names which have long been changed via published research studies.  I disagree with Jim on the capabilities of the iNat reviewers vs. BugGuide, but that could be because we’re looking at different groups of lepidoptera.  In my experience, some iNat reviewers are repeatedly agreeing with misidentifications.  BugGuide seems to get more serious peer review from people specialized in certain groups and I find fewer misidentifications.

 

Then, when I disagree with some, I get a lecture why I’m wrong.  Let me see, a rank amateur telling me, with 61 years of collecting/observing/rearing/researching experience that I’m wrong.  If I’m not sure of an ID, I’ll state so, or suggest something like “Tiger Swallowtail group”.

 

My biggest beef is when people obscure locations for even common species.  One person recently obscured a number of common moths.  I asked why.  The person responded that they did not want to identify people’s properties!  Seriously?  I suggested at least listing the nearest city or town to help naturalists working on state distributions.  Another rank amateur then agreed with the first.  What, busloads of collectors are going to someone’s front porch weeks later to catch that long-gone common Geometer?

 

Other problems, iNat submissions are basically clustered around population centers (backyard naturalists) and there are considerably fewer posts from out in the countryside.  Give the false impression that butterflies are more common in the immediate D.C. area than in surrounding rural regions – which most of us agree on the opposite.  Also, reporting often is biased to Monarchs and Black Swallowtails.  How many thousands of Monarch images are necessary for the Washington D.C. metro area?

 

I do use iNat but I vet each image separately.  I gave up correcting Skippers in Virginia – a horror show.  Takes too long, and the corrections are never updated in the search results.  Corrections just sit in the comments area.  And searching through several thousand Monarch posts in Virginia, just to find a new county or state host record, is, sadly a waste of my time.  Who actually has use for all those Monarch posts?  I know, some people post their images because they want an I.D.  Just look at the images and make up your own mind.  I suggest to people to “check your field guide”, which I’m certain many do not have.  Makes me wonder how reliable Ebutterfly is!

 

Anecdote:  Just saw a post in the NYLeps group where someone observed the Common Blue (European introduction that just made its way into northern N.Y.), but they could not add it to their life list for 2021 because it was not on the NABA Checklist!  Seriously?  It’s sad that some people are so beholden to the NABA Checklist like this.  Turns out that Mike Reese, the Sightings coordinator got emergency approval to add Common Blue to the North American list, so now the person who posted can add it to their life list.  Wow. 

 

Harry Pavulaan

 

Sent from Mail for Windows

 

From: Rick Borchelt
Sent: Saturday, December 11, 2021 5:40 PM
To: mdlep...@googlegroups.com

--

You received this message because you are subscribed to the Google Groups "MDLepsOdes" group.

To unsubscribe from this group and stop receiving emails from it, send an email to mdlepsodes+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/mdlepsodes/48CC6A09-1B4B-44F5-8B4F-8CD52DDE0C51%40gmail.com.

 

[Jim Moore]

I think everyone is raising some interesting points on iNat and other reporting methods.  Some miscellaneous comments:

– My experience at iNat is pretty similar to the other Jim's.  I mostly submit Maryland sightings of a broad range of organisms, and the IDs on my submissions are generally accurate and careful as far as they go. I only recall one instance where a teenager inexplicably disagreed with an expert.  I also note that all the Maryland sightings at iNat are automatically considered for inclusion in the Maryland Biodiversity Project, which vets the IDs more rigorously than "research grade."

– iNat is collecting loads of data on thousands of species and in thousands of areas around the globe that was simply never collected before.  And all the photos and ID discussions are retained, so anyone who needs to evaluate the IDs more carefully can do so.  So, for all its shortcomings, I think it is still a great resource.

– iNat's Computer Vision, while obviously not perfect, can be a real timesaver in identifying candidate species for many organisms,  e.g. with moths.  It can also prevent misidentifications by suggesting alternate species.

– EBird/Ebutterfly complete lists of species seen provide, in theory, more valuable data than iNat since they also identify what species are not seen.  But they also rely to a greater degree than iNat on the expertise of the observer to correctly identify what was seen because not everything is photographed.  Reports posted on this group provide great information, but, at least to my knowledge, no one is collecting these lists into a database, which makes them less valuable.  I think all these methods have advantages and drawbacks.

Jim Moore
Rockville

To view this discussion on the web visit https://groups.google.com/d/msgid/mdlepsodes/BL0PR05MB5474A97D2E47427720BA971ED8739%40BL0PR05MB5474.namprd05.prod.outlook.com.

[Rick Borchelt]

I *think* -- and Bill or Jim can correct me -- MBP automatically ingests only iNat at Research Grade or Needs ID (which presumably they would then ID or vet).  Even for a popular family like Geometridae, this is half or fewer of submissions even after six months.  Many of these are quite valid, but nobody has stepped up to ID them.  I personally have more than half of my observations (from various taxa) sitting on iNat with no second ID (even though I'm 99% sure about most of them).  Maybe they will be identified in the future and be accessible; right now there's a lot of churn and only the most popular taxa are regularly making Research Grade.  

--

Rick Borchelt
College Park, MD
preferred personal email:  rborchelt |AT| gmail |DOT| com

http://leplog.wordpress.com

[Rick Borchelt]

What I should have said was that MBP considers Research Grade and ID Needed for MBP; they do of course perform some vetting of these but I leave it to Jim and Bill to tell us more about the level of review they give iNat records.  Here's what MBP says:

[Jim Brighton]

Hello everyone,

Research Grade on iNaturalist has absolutely nothing to do with the way iNat records are ingested into the MBP database. We have accumulated a list of people that we deem as experts. If these experts make an identification, whether their determination is in agreement with the original poster, or not, the record will then be put in a queue for final verification. A final verification is done by MBP staff, and only after this final verification will the record be ingested into the MBP database.

I think one important point that hasn't been made in this discussion, is that iNaturalist and MBP have totally different missions. iNaturalist does not want to be the global repository of all nature records. Their mission has more of a social bent. They have made a platform for the amateur and professional to interact in a way that was never before possible. In my opinion, this is what makes iNaturalist revolutionary. I have never had a list of peruvian orchid experts at my fingertips. Never before have I had the opportunity for an expert to so easily look at my data, and confirm and correct that data. 

Mistakes are made across the board. I have accidentally gotten common names mixed up and called a common Maryland organism something from Uganda. But I don't think this is really important. Most of the time, someone is quick to point out the mistake, and most of the time in a nice and non-condescending way. I think that iNaturalist is a platform that connects people, so our overall knowledge of the natural world can be increased. When this connection helps the scientific community, which it often does, then iNaturalist become doubly awesome. 

Jim Brighton

Cambridge, MD

jimbri...@gmail.com

www.marylandbiodiversity.com

To view this discussion on the web visit https://groups.google.com/d/msgid/mdlepsodes/CAN0Du2tzYmCjwho0Gi9Pvwm8SdFtqvQta%2Bd6DJb-7cgz73LOsw%40mail.gmail.com.

[Rick Borchelt]

Thanks, Jim.  So, out of curiosity, what % of records submitted to iNat for MD in 2021 do you expect to pull into iNat?

To view this discussion on the web visit https://groups.google.com/d/msgid/mdlepsodes/CAPiwJXT7GyCcnMPUS5KZOSJT5TRMxrkrot5ETwcHJgf8AwfXyg%40mail.gmail.com.

[Jim Brighton]

Hi Rick,

There is a large percentage of records that are entered into iNaturalist that can't be determined. Many photos are of domesticated/cultivated material. Throw out all of the out of focus records, and then take out all the photos of plants/animals that don't highlight the prerequisite identification characteristics (think symphyotrichums/lichens/fungi). Think about all the wonky skipper photos that are only identifiable to genus because the photograph does not show the proper markings to require 100% identification. Of the photos I look at, I can only identify with 100% certainty somewhere between 40 - 60 percent of those records. Think about all the people that take photos of Daddy Longlegs. There are only a handful of people that can reliably identify that order. Those Daddy Longlegs records will just sit there until one of those experts comes around and identifies the records. 

To get back to your original question. I think of all the records submitted with a Maryland location, MBP will ingest maybe 30 - 40% of those records. BIll would probably be able to answer this question with better certainty. Of course, MBP has to be conservative. A species that one person might consider easy to identify, we might think differently. 

Jim Brighton

Cambridge, MD

jimbri...@gmial.com

www.marylandbiodiversity.com

To view this discussion on the web visit https://groups.google.com/d/msgid/mdlepsodes/CAN0Du2uKHCobPBgiEFQcBrB%2BgMdRegbjUFFK-Yk%2Bcqz7Cpv2XA%40mail.gmail.com.

[Peter DeGennaro]

Like others, I was also initially skeptical of iNaturalist given the ID errors and the ability of anyone to make IDs. I only started paying attention to it when I moved to Arizona in 2020 since it was very helpful to find butterfly locations. The sheer amount of significant records posted to the site, often by people who didn't know the ID of the species they photographed, alone proves the site's worth. There are plenty of record early and late dates, county records, and even range extensions that have been recorded. Only a small fraction of these records would get posted to another site if iNat didn't exist.

Also, the "Research Grade" designation is a bit of a misnomer. It's simply a way of moving the observation out of the "needs identification" queue. The iNat founders and curators repeatedly mention that this designation should not be relied upon when using the site for one's research.

Peter DeGennaro

To view this discussion on the web visit https://groups.google.com/d/msgid/mdlepsodes/CAPiwJXT7GyCcnMPUS5KZOSJT5TRMxrkrot5ETwcHJgf8AwfXyg%40mail.gmail.com.

[Rick Borchelt]

I don’t think anyone (least of all I) think iNat is worthless.  I do think it is relied on for definitive ID far beyond its capabilities.  And as the literature suggests, is a capture mostly of biodiversity in urban and suburban locations.  This is geographically variable, of course, and a strong field lep community might overcome this to some degree if they regularly post to iNat.  

Sent from my iPad

On Dec 14, 2021, at 3:08 PM, Peter DeGennaro <dege...@gmail.com> wrote:



To view this discussion on the web visit https://groups.google.com/d/msgid/mdlepsodes/CAENXc_3cdMakqcfQ30fdwO5iwwodw6ezTCr9n3fjm1VcjVDrYg%40mail.gmail.com.

[Sharon F]

Here is a screenshot of iNat records from the Osa Peninsula of Costa Rica, which, if you haven't been there, is not the suburbs. Conservationists working in Latin America understand the potential of iNat, maybe more so than some in the U.S. In many areas, people don't have access to books or specialists, and iNat is their best source of information about the local flora and fauna. 
Sharon Forsyth

[Rick Borchelt]

This is the corresponding iNat screen shot just for butterflies.  Two things emerge.  80% of the sightings are in or around national parks or refuges (same as in the US); the remainder are in highly-trafficked coastal resort regions (the Costa Rica equivalent of suburbs).  Looks like about 2/3 of the peninsula lacks records entirely. I don't believe one would get a good picture of overall Costa Rican butterfly diversity from this capture.

 I don't dismiss the potential value of iNat as an educational tool, but based on its error rate domestically for butterflies I would be somewhat skeptical. 

 

To view this discussion on the web visit https://groups.google.com/d/msgid/mdlepsodes/501a5281-c22b-4902-8820-da9b156f8fe5n%40googlegroups.com.

[Rick Borchelt]

Costa Rica doesn't look that much different from, say, Wyoming -- where the bulk of iNat butterfly observations are clustered around parks and urban centers (such as they are in Wyoming).  Luckily we know a lot more about butterfly biodiversity in Wyoming because of diligent academic and "hobbyist" butterfly observers and (especially) collectors that are not reflected in iNat.  Plus the butterfly biota of the western states is much trickier to ID than in MD (hosts of sibling, cryptic, and local endemics), even in hand let along from photos on iNat.  Much like Costa Rica ...

[Jeff Cagle]

MDLepsOdes is a fabulous way to connect and discuss.  It could not be replaced by either Facebook (too much valuable text) nor by iNat (limited networking ability).   Thanks for the effort you put into it, Rick and all you posters.

[Jeff Cagle]

The limitation of iNat is reminiscent of the limitation of p-values: often misused because often misunderstood.   

A "Research Grade" should be understood as something much less than museum grade.  The set of Research Grade observations of a taxon will be a noisy estimate of the individuals of that taxon which came into contact with humans ... and nothing more.  It's not a survey, it's not well-curated, it's based only on visual morphology, and it's definitely not a guarantee.  Hence the terminology: "Suggest an ID."  

The value of a noisy signal is that the genuine signal can be extracted from it.  If I want to find Holly Azures or Northern Crescents, I can pull up putative observations of those and winnow them to genuine candidates.

So the mis-IDs don't trouble me overly.  They introduce a bit of noise, most of which is found by expert reviewers and neutralized.  The actual error rate is honestly fairly small for observations that have been around a month or more.

(I occasionally look at the species list and search for outliers -- an "Echo Azure" sticks out like a sore thumb.)  

For tough IDs (say, C ladon vs C neglecta in April), there will be a higher error rate.  That just means we interpret with more salt.

Over against the noise is the educational value.  iNat allows casual butterfliers ("lepers") to increase their knowledge quickly in a way that field guides or burrowing through literature cannot.  A lot of the education happens as a contested ID is talked through.

The bottom line is that iNat is a great place to find a moderately noisy signal and to contribute to the education of the Lep community.

On Tuesday, December 14, 2021 at 3:24:59 PM UTC-5 rbor...@gmail.com wrote:

[Bert Harris]

Hi all,

I just wanted to add that iNaturalist can be a valuable tool for research as well as for education, as long as the "grain of salt" method that Jeff Cagle described is applied. For example, I just worked with a student to model the distributions of Piedmont grassland plants and we used iNat for a good portion of our locality data. I checked the identification of every record we used and we were able to harvest a few thousand data points. There is no other comparable database in the world in terms of accessibility, and most iNat records specify their spatial accuracy, which is key for this kind of analysis. Sure there are lots of herbarium specimens out there, but many of them are not digitized and it's much harder to check the identity of each one. I know this is not a lep or ode example but I thought it was relevant.

Looking forward to spring and the insect season!

Bert

To view this discussion on the web visit https://groups.google.com/d/msgid/mdlepsodes/0f009242-3bfd-498c-93a3-26c56b95b0bcn%40googlegroups.com.

--

Bert Harris, Ph.D.
Executive Director
The Clifton Institute
6712 Blantyre Rd.|Warrenton, VA 20187
www.cliftoninstitute.org | Facebook

Click here to support our education, restoration, and research activities!

Submit observations to our iNaturalist project