No one has made an observation of a creosote bagworm (Thyridopteryx meadii) on iNaturalist since September 2022… at least as of February 6, 2023. Because there is a history of observations in December and January, their absence this year is surprising. I wonder if it is early evidence of a population crash of this moth? I just spent about twelve hours over a three day period searching creosote bushes over a large part of the Colorado and Mojave deserts without finding any, so I think it might be. https://www.inaturalist.org/observations?order_by=observed_on&place_id=any&subview=table&taxon_id=475010
I know there have been studies using iNat data to show general declines as well as declines linked to climate change ( See: https://forum.inaturalist.org/t/published-papers-that-use-inaturalist-data-wiki-3-2022-and-2023/34753/1 )
Do you know of other examples like this where observations for a taxon were being made fairly regularly and then stopped? I can imagine scenarios where someone who used to be interested in a particular species suddenly stops using iNat. Or a particular project is disbanded. But I don’t think that is what is going on in the case of the Creosote Bagworm.
Of course any advice on how to follow up on this preliminary investigation is appreciated.
I think using iNat data to assess abundance is really tricky/difficult, and 59 would be a low sample size for multiple years, even if the data collection were scientifically rigorous. So I would hesitate to draw any strong conclusions in this situation. The challenge in using iNat data for assessing population numbers (abundance) is that, to do this well, you need to account for observer behavior, including number of observers out looking, their ability to detect the focal organism/s, and their interest in uploading observations of the focal organism/s.
As an example, I’ve run projects on iNat that have students looking for anole lizards during late winter/early spring because that’s when it fits into the school program. At one point the phenology graph for some anoles in the US showed a massive spike in February. It wasn’t because anoles were more common then, but because there were a bunch of students in one county out looking for them and making observations. Likewise, we skipped that project during Covid. There were many fewer observations made during February than usual. It wasn’t a population crash, just a lack of observer activity. I’m not saying this is the case with the bagworms (I’ve no idea), just that iNat data is tough to use here. I think that there have been some papers that have used iNat data (along with other data) in large quantities and with some more sophisticated statistical techniques to look at broad scale patterns in abundance, but don’t know of specific ones off the top of my head.
The male moth has nearly transparent wings about 2 cm across and the female is wingless. Most observations are of the larval case which is typically constructed of bits of dull brown creosote leaves, is well camouflaged… http://mothphotographersgroup.msstate.edu/species.php?hodges=455 I wouldn’t have called it “rare,” but it’s not common either unless you come across one of the “infestations” reported in the literature, which I have not. In fact, I’ve never found more than one per bush. But there are quite a number of folks who routinely search creosote for galls (there is quite a diversity including undescribed species!), so those are the folks who also have been making observations of the bagworm.
Thanks. It makes sense not to draw strong conclusions. 59 is a small sample.
I have no formal experience doing even casual arthropod population surveys, but I imagine that it is at least “number seen/unit effort.” It was my own “seen/effort” experience (zero/12 hrs) during the past trip compared to my less intense indirect effort last year (5/~<6 hrs), that caused me to check the pattern of observations on iNat and see that none have been observed for the past four months. I anticipated the episodic observations that could be possible thanks to blitzes or classes, but there is no evidence that is going on here.
All psychids look pretty much the same way as imago. But you could get to a place where none live this year, they could all be 2 metres from where your “path” went or because of camouflage you could miss those 1-2 that were there, hiding behind the branches and leaves. I think you definitely could document population crash if you have a group or 1 user who observe the same group the same way over long period of time, but with such insects it’d be hard to draw conclusions from this species’ stats.
I’m using iNat to track the range extension of a taxon.
We have locally here a crash in Saturniidae (which may not be so local) though I don’t know if iNat can YET be employed to track this.
The greatest challenge is the immaturity of iNat; the records (observations) just aren’t that old, and going back even a few years, there weren’t that many observers / observations. In 20 years, iNat will have the data to track ranges and population crashes; then, fed into AI, we should have instantaneous analyses which right now I have to do by hand, looking at every record.