Re: [washbutterflies] Washington Post article on decline in butterfly numbers

23 views
Skip to first unread message

Rick Borchelt

unread,
Mar 6, 2025, 3:38:14 PM3/6/25
to Rob Simmons, <mdlepsodes@googlegroups.com>, Thomas Stock, Washington Area Butterfly Club
I think this paper needs to be taken with a sizable dose of salt.
I agree with the overall conclusions, that many butterfly species are less numerous than they once were, although some are increasing.  But the methodology deserves looking at.

Take the picture example in the WaPo story: Giant Swallowtail.  In many parts of its range (northeast and midwest, and here in the mid-Atlantic), it is clearly in decline -- in our case, because we don't actually have any native citrus species and the ornamental ones (that cresphontes will eat) we grow are not hardy.  It's primary host here, prickly-ash, is vanishingly rare and a favorite deer snack. Yet Giant Swallowtail is a serious pest to citrus in Florida.  The decline numbers result because the data are overwhelmingly drawn from established NABA counts, and counts in the midwest and northeast that used to see this species don't see it as much.  Is the species "in trouble"? The conclusion one would draw from the paper is yes.  The reality for the species is less clear.

In an embargoed news conference yesterday, one of the authors was asked a question about Florida White, Appias drusilla.  The reporter asked what it meant that this species had a 100 percent decline, according to the paper.  The researcher (who I am fairly confident has never seen Appias in the field or knows anything about its biology), confidently replied it was extirpated in the United States. 

And yet ... I see it every time I go butterflying in Florida.  In its habitat -- hardwood hammocks and evergreen tropical forest -- it can be a very common butterfly.  It flies in the gloom of these habitats, and can even be observed nectaring in the dusk and at night.  The iNat map for this species shows a ton of recent observations in southern Florida and the LRGV (a whopping 150 observations in 2024-25 alone, astounding for an extirpated species even if many of them came from the National Butterfly Center and environs).  I did let the reporter (not WaPo) that Florida White is indeed alive and well in the US tropics/subtropics.

When I wrote to the researcher afterwards and questioned the characterization of Appias as extirpated, I was referred to the author who had crunched the numbers on this species--a data scientist, not a biologist or lepidopterist.  He based his analysis exclusively on data from 20 NABA counts, 13 of which were in FL (and I am willing to bet none of them were in hammock or evergreen forest habitat).  Some of the others were in the southwest or lower Great Plains, where Appias demonstrates a marked tendency toward vagrancy from its Mexican/borderland habitat, especially as a consequence of tropical weather events. Knowing that about Appias would have led me to exclude those counts as not representative of the health of Florida White populations in the U.S. That NABA counts -- which typically survey field/garden/open nectar habitats, let's be honest, unless we're looking specifically at/for a particular species -- were not seeing as many of this very distinctive pierid is, IMHO, not time to cue the pearl-clutching.  And certainly not sufficient to declare an extirpation event. 

I know that NABA and the fine insect data scientists who work with NABA numbers are at great pains to make the data as credible and robust as they can.  Nevertheless, I am skeptical of just how meaningful the conclusions are with the underpinnings in the vagaries of NABA counts.  

Makes for good reading perhaps, but not yet a foundation for good conservation action or conservation policy. 

Screenshot 2025-03-06 at 3.31.17 PM.png

Screenshot 2025-03-05 at 3.03.23 PM.png

On Thu, Mar 6, 2025 at 2:47 PM 'Rob Simmons' via Washington Area Butterfly Club <washbut...@googlegroups.com> wrote:

--
You received this message because you are subscribed to the Google Groups "Washington Area Butterfly Club" group.
To unsubscribe from this group and stop receiving emails from it, send an email to washbutterfli...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/washbutterflies/448927176.547805.1741290419844%40mail.yahoo.com.


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

http://leplog.wordpress.com

Frank Boyle

unread,
Mar 6, 2025, 5:03:36 PM3/6/25
to mdlep...@googlegroups.com, Rob Simmons, Thomas Stock, Washington Area Butterfly Club
I would caution in individual species cases, although in general I agree.

The Bath County NABA count was hardly of “gardens and field/open habitats”.  We counted methodically for almost 30 years in many mountain, forest, meadow and riparian habitats.  Clearly my Bath County, VA shows a marked decline in several mountain species, not least of which is the Diana Fritillary.  I can and have used the count data to change USFS policies on federal National Forest lands - habitats that have been degraded because of human development and over harvesting of timber, specifically.

So just be careful generalizing. I know you (sometimes) don’t like NABA counts, but some of us worked very hard for many  years on them.

Frank

Fran

"They who dream by day are cognizant of many things which escape those who dream only by night."
-E.A. Poe


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 visit https://groups.google.com/d/msgid/mdlepsodes/CAN0Du2v4B8ZoZ7jY2S8tPvzb8exwEfiFWPawDqUBg80XzZBt2Q%40mail.gmail.com.

Rick Borchelt

unread,
Mar 6, 2025, 5:12:05 PM3/6/25
to Frank Boyle, mdlep...@googlegroups.com, Rob Simmons, Thomas Stock, Butterfly Club Washington Area
The GRSF count too is broad ranging. The Florida counts are not. 

On Mar 6, 2025, at 5:03 pm, Frank Boyle <raven...@gmail.com> wrote:


I would caution in individual species cases, although in general I agree.

The Bath County NABA count was hardly of “gardens and field/open habitats”.  We counted methodically for almost 30 years in many mountain, forest, meadow and riparian habitats.  Clearly my Bath County, VA shows a marked decline in several mountain species, not least of which is the Diana Fritillary.  I can and have used the count data to change USFS policies on federal National Forest lands - habitats that have been degraded because of human development and over harvesting of timber, specifically.

So just be careful generalizing. I know you (sometimes) don’t like NABA counts, but some of us worked very hard for many  years on them.

Frank

Fran

"They who dream by day are cognizant of many things which escape those who dream only by night."
-E.A. Poe

On Thu, Mar 6, 2025 at 3:38 PM Rick Borchelt <rbor...@gmail.com> wrote:
I think this paper needs to be taken with a sizable dose of salt.
I agree with the overall conclusions, that many butterfly species are less numerous than they once were, although some are increasing.  But the methodology deserves looking at.

Take the picture example in the WaPo story: Giant Swallowtail.  In many parts of its range (northeast and midwest, and here in the mid-Atlantic), it is clearly in decline -- in our case, because we don't actually have any native citrus species and the ornamental ones (that cresphontes will eat) we grow are not hardy.  It's primary host here, prickly-ash, is vanishingly rare and a favorite deer snack. Yet Giant Swallowtail is a serious pest to citrus in Florida.  The decline numbers result because the data are overwhelmingly drawn from established NABA counts, and counts in the midwest and northeast that used to see this species don't see it as much.  Is the species "in trouble"? The conclusion one would draw from the paper is yes.  The reality for the species is less clear.

In an embargoed news conference yesterday, one of the authors was asked a question about Florida White, Appias drusilla.  The reporter asked what it meant that this species had a 100 percent decline, according to the paper.  The researcher (who I am fairly confident has never seen Appias in the field or knows anything about its biology), confidently replied it was extirpated in the United States. 

And yet ... I see it every time I go butterflying in Florida.  In its habitat -- hardwood hammocks and evergreen tropical forest -- it can be a very common butterfly.  It flies in the gloom of these habitats, and can even be observed nectaring in the dusk and at night.  The iNat map for this species shows a ton of recent observations in southern Florida and the LRGV (a whopping 150 observations in 2024-25 alone, astounding for an extirpated species even if many of them came from the National Butterfly Center and environs).  I did let the reporter (not WaPo) that Florida White is indeed alive and well in the US tropics/subtropics.

When I wrote to the researcher afterwards and questioned the characterization of Appias as extirpated, I was referred to the author who had crunched the numbers on this species--a data scientist, not a biologist or lepidopterist.  He based his analysis exclusively on data from 20 NABA counts, 13 of which were in FL (and I am willing to bet none of them were in hammock or evergreen forest habitat).  Some of the others were in the southwest or lower Great Plains, where Appias demonstrates a marked tendency toward vagrancy from its Mexican/borderland habitat, especially as a consequence of tropical weather events. Knowing that about Appias would have led me to exclude those counts as not representative of the health of Florida White populations in the U.S. That NABA counts -- which typically survey field/garden/open nectar habitats, let's be honest, unless we're looking specifically at/for a particular species -- were not seeing as many of this very distinctive pierid is, IMHO, not time to cue the pearl-clutching.  And certainly not sufficient to declare an extirpation event. 

I know that NABA and the fine insect data scientists who work with NABA numbers are at great pains to make the data as credible and robust as they can.  Nevertheless, I am skeptical of just how meaningful the conclusions are with the underpinnings in the vagaries of NABA counts.  

Makes for good reading perhaps, but not yet a foundation for good conservation action or conservation policy. 

<Screenshot 2025-03-06 at 3.31.17 PM.png>


<Screenshot 2025-03-05 at 3.03.23 PM.png>


On Thu, Mar 6, 2025 at 2:47 PM 'Rob Simmons' via Washington Area Butterfly Club <washbut...@googlegroups.com> wrote:

--
You received this message because you are subscribed to the Google Groups "Washington Area Butterfly Club" group.
To unsubscribe from this group and stop receiving emails from it, send an email to washbutterfli...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/washbutterflies/448927176.547805.1741290419844%40mail.yahoo.com.


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

http://leplog.wordpress.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 visit https://groups.google.com/d/msgid/mdlepsodes/CAN0Du2v4B8ZoZ7jY2S8tPvzb8exwEfiFWPawDqUBg80XzZBt2Q%40mail.gmail.com.

--
You received this message because you are subscribed to the Google Groups "Washington Area Butterfly Club" group.
To unsubscribe from this group and stop receiving emails from it, send an email to washbutterfli...@googlegroups.com.

Thomas Stock

unread,
Mar 7, 2025, 7:29:57 AM3/7/25
to Rick Borchelt, Frank Boyle, MdLepsOdes, Rob Simmons, Butterfly Club Washington Area
Frank, your comment got me to thinking about the Montgomery County count. When still active, it sadly charted the decline and extirpation of Baltimore Checkerspot from areas in the County where it once thrived, especially the marshy meadow downstream from the Beaver dam on Newcut Road - which is now a housing development. This area also had a huge number of Appalachian Browns. I think my high count there was over 60 one day. In habitats vulnerable to destruction, counts can remind us of what we have lost. Baltimore also disappeared from Little Bennett but the issue there was, I believe, successional changes in vegetation and hydrology. So I would be reluctant to disregard count data despite the inherent weakness of being "snapshots." 

Tom Stock

Rick Borchelt

unread,
Mar 7, 2025, 10:45:56 AM3/7/25
to Thomas Stock, Frank Boyle, MdLepsOdes, Rob Simmons, Butterfly Club Washington Area
I agree with you and Frank about the high value of *individual* NABA counts, if done systematically, methodically, and over multiple seasons/years.  From these individual counts it is often possible to tease out the signal of declines *at that site* or in similar habitat nearby. 

NABA has done a terrific job -- the work of Leslie Ries for example -- in normalizing the data from these disparate counts.  But for many species declines or increases observed in NABA counts *in aggregate* are IMHO not necessarily good proxies for the health of individual species.  The Florida White example was a good one. 

This is because some counts -- like our own GRSF counts -- are specifically designed to monitor the populations of species of conservation concern, in our case Olympia Marble, Cobweb Skipper, Silvery Blue, and Northern Metalmark.  The sectors are set up especially to give a window into the status of their populations, even as we collect data on many other species.  I can tell you to an individual how well these populations are doing in GRSF.

Plus, most counts are single-season summer counts; we're unusual in having done both spring and summer census work.

But our GRSF work would not have told you much about Hoary Edge, which flies in between count cycles. I think we might have seen it once on the summer count in its second flight, but then never again, and it would have been one of those 100% decline species. The GRSF spring data would also have suggested a rise in the population of Northern Azure -- but only because (thanks to Harry Pavulaan) we now are able to distinguish the azure species that fly in GRSF, a rise owing to observers, not butterflies.  It would not reflect that Giant Swallowtail is increasing in GRSF and nearby areas. And it would say nothing about the dramatic decline in Brown Elfin statewide, since we see it there regularly on the spring count but other populations in the state are winking out right and left. If it's a late spring, our spring GRSF count might miss Cobweb Skipper entirely -- doesn't mean it didn't fly that year.  It's just that our field count is religiously held around the end of the second week of April, no matter what the meteorological conditions have been. 

Access to higher quality field guides and online resources has also likely impacted field count data through citizen/community science, but you'd have to balance that against the generally declining numbers of people who are interested in doing field counts and the difficulty of recruiting knowledgeable participants. Many NABA counts flourish for a few years and go away, partly because of the rules and regulations surrounding the count management and process and lack of public access to the data. Or because they were started by a charismatic person who quit in frustration at the difficulty of arranging count logistics with few takers.

All I'm saying is that, interesting as the NABA count data are *in aggregate,* I consider them illustrative rather than definitive -- which is not how they are being portrayed in the media.  

There are other, higher-quality data (from Art Shapiro's decades-long transects in the Bay Area) in the Science paper that probably *are* more definitive for their areas, and in places where there is a lot of consistent effort across several sites in a region, probably give pretty spot-on insights about population health in that region. Our count work in GRSF likely falls somewhere in between a more typical NABA count and a methodologically rigorous transect effort. 

The last point I would make is that many butterfly counts, as with Christmas Bird Counts, are often competitive enterprises as much as scientific ones.  A premium is put on how many species one can find and count, and lots of tweaks are made to capture observations of otherwise rare or uncommon species rather than happily counting the numbers of Cabbage Whites.  
Reply all
Reply to author
Forward
0 new messages