Happy New Year!
A note of tremendous appreciation for everyone submitting data to this project! I'm am thrilled to find the project at close to 10k observations by the end of its first year! WOW!
In 2024, we'll be working on the following projects under the PIP umbrella:
1) Synthesizing crowd-sourced interaction data to develop a plant selection tool for the Northeast.
2) Installing experimental gardens in VT, NH (in collaboration with the Native Plant Trust) and in CT and NY (in collaboration with Plan it Wild)
3) Continuing our field study comparing methods for quantifying plant-pollinator networks in the Upper Valley, VT.
Every iNaturalist observation you annotate with interaction data helps these projects! Here's to another year of plants, pollinators and scientific discovery!
Thank you!
Desiree
Comentarios
Happy New Year to you too! I really like this project. And I like your goals too. A great addition to the NWF's list of plants that host caterpillars. (I trust you would be focusing on native plants of course!)
Proposal: Besides "Interaction->Visited flower of" there are a lot more fields that annotate insect/plant interactions. I am trying to come up with a script to efficiently convert those to your favorite field, but I am just a beginning coder. Is there someone more skilled who can make a tool? One way it could work would be to run through each page (filtered by whatever criteria) that has a given observation field, show it to you, you decide if it's one you want, and then it creates an "Interaction" field with the same taxon. (That works for fields that have taxon links as values not text.)
Hi @lmtaylor, thanks so much! I'm glad you like the project (and NWF's plant finder too, which is based on some data from myself, Doug Tallamy and Kimberley Shropshire).
You're totally right about other fields being used. I've got a list of about 15 (it's even worse for caterpillars on host plants!). For pollinators on flowers, the "interaction->visited flower of" field seems to be used much more often and consistently, which is why I used that one.
The short answer is that these iNat observations in this project are only one part of the dataset we're using. The full dataset download will include Globi and GBIF data as well. Globi (https://www.globalbioticinteractions.org/data) already does some screening to convert iNat data from a whole bunch of interaction fields into their database. So if Globi includes it, then it's a 'good' observation, so to speak, for our purposes. The list of interactions they include is here: https://github.com/globalbioticinteractions/inaturalist/blob/main/interaction_types_mapping.csv
The important thing is that the field is a 'taxonomically-referenced' interaction field, i.e., that it has a iNat searched plant taxonomy. Some of the fields are just text fields, which is not ideal for curation and I don't believe are included in Globi's curation. Another important thing is that the field indicates what the interaction is. For example, "Associated species" is a field, which can be a plant-pollinator interaction, but not always. It can especially be an issue with Lepidoptera where it could be a host plant or a flower visit.
We can search these kinds of fields and add the 'interaction->visited flower of" field to curate it. Is there a way to automate that process in iNat? Not that I know of. I've seen other folks ask these kind of 'batch annotation' questions, but I haven't seen any solutions. If you see any, let me know!
In any case, this means finding observations without the indicated interaction field is really useful for our purposes! I've been 'hacking' the iNat search to find information in the notes and other fields to add them to the project. For example, this link searches any observation with 'visited' in the notes that is not already in the project: https://www.inaturalist.org/observations?has%5B%5D=geo¬_in_project=pollinator-interactions-on-plants-pip-of-the-ne-us&place_id=48,47&q=visited&search_on=description&subview=table&taxon_ids=47157,47201,47822,47208,3&verifiable=any
You can also find plenty of observations where there is just a picture of an observation but the plant is identifiable. Tons of observations can be added this way!
Thanks for bringing this up!
Desiree
I've gone through all NH observations with the Interaction field, and I'm now working through NY counties one by one (the vast majority of those 7k observations are in NYC so I'm focusing on upstate to fill in the gaps on the map). I'd be interested to know where the experimental gardens in NH will be!
Hi Desiree, Don't know where to ask this question so I'm commenting here. I have tried adding a couple observations of mine to the project, but I'm getting the message: 'Didn't pass rules: must be in taxon Lepidoptera OR must be in taxon Hymenoptera OR must be in taxon Diptera OR must be in taxon Coleoptera OR must be in taxon Icteridae OR must be in taxon Parulidae OR must be in taxon Trochilidae.' The observations are clearly in Vermont such as this one: https://www.inaturalist.org/observations/7037157. Most of my others are accepted with no problem.
Hi @susanelliott! Thanks for your contributions! That particular insect is a mantidfly which is in the order Neuroptera which are not included in this project. They are related to other insects like lacewings. Some mantidflies are wasp-mimics, and are predators. So they are visiting a flower but likely not for the nectar or pollen.
If you have any other observations you're having trouble adding, feel free to send me a dm!
Ah ha! Thanks so much for the quick response.
Thanks for that note. I have been submitting observations to the project when the photo depicts an insect on a flower, even when I suspect that the insect isn't interested in it AS a flower (for example, a lady beetle walking across it while searching for prey) -- is that appropriate? I thought it was best to be inclusive and let you guys sort out which ones don't meet your criteria.
This project is limited to major groups that include pollinators, including Lepidoptera (moths and butterflies), Diptera (flies), Hymenoptera (Bees and wasps) and Coleoptera (beetles) as well as a few families of birds that are flower visitors. When I'm going through observations, I tend to focus on taxa that are well known and abundant pollinators (bees, butterflies, flower flies). If you tag taxa with the interaction field that are not pollinators (e.g., ladybeetles and other predators), we can easily filter them out. But for some groups (wasps, flies) it's not always clear. For example, they can visit for floral resources and be a predator. So I agree with you that it is a good idea to be inclusive. But in the interest of efficiency, you could also exclude some exclusively predatory groups using the "&without_taxon_id=" function.
----And just to add on to what @dlnarango said about predation not being all or nothing, I spent a little time looking up the diet of some lady beetle species and found that, according to one entomological source, "Unlike most lady beetles, plant pollen may constitute up to 50% of the diet" of Coleomegilla maculata. Sure enough, iNat has lots of pictures of C. maculata neck-deep in the pollen-bearing parts of flowers. I guess I'm adding those to the project. Some other coccinellids supposedly eat a bit of pollen or nectar too ... I will take a look and see if there are any good photos which seem to show them feeding.
Check out this Propylea quatuordecimpunctata for example--pretty interested in that flower, isn't it? Just at a quick glance, there seem to be more photos of P. quatuordecimpunctata on flowers than just about any other species of coccinellid.
Link(which I haven't read yet): Giorgi, J.A., et al. (2009) The evolution of food preferences in Coccinellidae. Biological Control 51 :215–231
And here, if I'm not mistaken, is C. maculata chowing down on pollen (and maybe eating the entire center out of the flower)
ecology is complex! :) Thanks for sharing these cool photos!
It turns out some Neuroptera can also eat pollen too (Chrysopidae and probably others). I'm going to keep the project with the orders I originally included, but you can always tag flower visitors with the field and just not add to the project. The field will still be accessed thorugh iNat API and be sorted into GLOBI, gbif etc.
Añade un comentario