InsectNet forum

Having seen an astonishing number of Giant Swallowtails (Papilio cresphontes) in the most northern portion of NY, and hearing of the same on the Canadian side, I did a look on iNat for observations in 2024 and it was interesting: https://www.inaturalist.org/observation ... n_id=85024

The majority of US observations are southern US states, and in Canada along the north shore of Lakes Ontario and St. Lawrence River. In fact, for August 2024, Canada probably accounts for 50% of the total North American observations. Second is Florida.

Now, apparently this is not new; I ran the same search for 2020 and the results are largely the same, albeit without the extremely high quantity in Canada of 2024. But still, both maps/ plots for 2020 and 2024 show that most observations are in Florida, and Canada, with Canada exceeding Florida. Interesting. Why?

We have more people posting on iNat. In Florida, this species is dirt common that no one cares probably. Think about how often anyone posts about Pieris rapae. What iNat captures is not true population density, but a function of population x frequency of human observation x popularity of iNat.

Don’t get me wrong. Cresphontes is indeed common here. But we aren’t going to beat Florida!

eurytides wrote: Fri Aug 23, 2024 1:55 pm Don’t get me wrong. Cresphontes is indeed common here. But we aren’t going to beat Florida!

I will note, that when in northern New York in late July I saw more cresphontes in one day than I've ever seen in one day in Florida.

eurytides wrote: Fri Aug 23, 2024 1:54 pm We have more people posting on iNat. In Florida, this species is dirt common that no one cares probably. Think about how often anyone posts about Pieris rapae. What iNat captures is not true population density, but a function of population x frequency of human observation x popularity of iNat.

I have noticed a similar pattern with iNaturalist here in Oregon, the amount and type of reporting reflects more on the location of observers
than the actual species. Just from looking at iNat, I would believe that Papilios are very rare in Oregon, which is not accurate, as well as
the ability to get a good photo, some animals are simply easier to photograph.

kevinkk wrote: Fri Aug 23, 2024 4:46 pm
I have noticed a similar pattern with iNaturalist here in Oregon, the amount and type of reporting reflects more on the location of observers than the actual species.

The same statement was recently published by Dr. Shapiro in LepSoc concerning records of butterfly species from the SW 100 years ago. Simply, more records occurred from near roadways.

And, not surprisingly, something like 50% of all butterfly observations on iNat are monarchs.

Clearly, one has to take iNat observations with a grain of salt and recognize the factors in "data" that may infer what is not true.

That said, in most cases when taken as an aggregate the range and frequency inferred in iNat maps is pretty close to the reality. Except in places where there are no humans, or no access or interest in iNat, the (clearly deficient) data is pretty good. Looking at iNat's Papilio cresphontes, for example, there are fewer records on the mid-Atlantic seaboard states when clearly there are significant population centers.

Similarly, flight periods are prone to false representation. The iNat "flight period" of White Tailed deer would indicate that the population doubles in early July, then tails off again. In reality the population jumps in April and May, and falls off in January & February. The false inference is simply due to seasonality and when humans are outdoors.

Mathematically speaking, modeling realistic species ranges and frequencies based on known faults in iNat shouldn't be that hard; however I've not heard of it being done.

About two years ago I published a 4-page species account on the occurrence of the 'Giant swallowtail' in Louisiana based upon 872 naturally wild adults. This species appears to have four naturally occurring annual broods in Louisiana (see phenogram). The four annual brood peak at naturally occurring approximate 67-day intervals in Louisiana.

Freely accessible Link: https://www.academia.edu/123162281/Hera ... _Louisiana

Chuck wrote: Fri Aug 23, 2024 3:00 pm

eurytides wrote: Fri Aug 23, 2024 1:55 pm Don’t get me wrong. Cresphontes is indeed common here. But we aren’t going to beat Florida!

I will note, that when in northern New York in late July I saw more cresphontes in one day than I've ever seen in one day in Florida.

I am sure there are going to be flukes here and there, and there may be days and locations where a place in upstate NY will beat some place in FL. But as a whole, butterfly generations are going to happen faster in the south than the north because of the climate, right? In late May when we start seeing only a few cresphontes after diapause, I’d imagine cresphontes is already widespread in FL. We can’t beat the head start you guys have down south. Winter sucks…

Interesting article on sampling bias.

Insect collecting bias in Arizona with a preliminary checklist of the beetles from the Sand Tank Mountains

https://bdj.pensoft.net/article/101960/ ... 2/8268698/

When I look at inaturalist data to try to get a more true view of how common things are in a large area, I divide by the number of plant records in each month to normalize.