Observation Field for Flowers and Pollinators

I photograph insects on flowers - it’s how I document pollinators and their host plants. Thanks to @andy71, I learned about a way to officially show this relationship in iNat!

The Observation Field “Nectar / Pollen delivering plant” is a great way to document the flower as well as the pollinator. I expect this dataset will be very useful to ecologists, so let’s spread the word!

There are already more than 80,000 observations with both insect taxa and flower taxa labeled:
https://www.inaturalist.org/observations?place_id=any&taxon_id=47158&field:Nectar%20%2F%20Pollen%20delivering%20plant

And the best part? You can search for pollinators based on the plant they were observed on. Here are all observations with Cirsium vulgare as a nectar plant:
https://www.inaturalist.org/observations?place_id=any&field:Nectar%20%2F%20Pollen%20delivering%20plant=52989

I’d like to be able to do the reverse search: find all the plant species a given taxa of pollinator was observed on. Is there a way to do this with the search URLs?

5 Likes

not every observation is properly licensed, but those that are will end up in GBIF. if those observations also have certain observation fields that indicate interactions, they will also get exported to GloBI. over at GloBI you can get all that data either through their web interface or by downloading their dataset. that’s probably the easiest way to search for interactions between taxa.

it’s harder to parse out all the unique values from the observation fields from a given set of observations. it’s doable, but to automate the process, you’d probably have to use the API in some way.

3 Likes

You can more or less do this by appending test=interactions to the URL of the pollinator taxon (see topic).

For example, for Bombus pascuorum, you would use this link: https://www.inaturalist.org/taxa/55637-Bombus-pascuorum?test=interactions and click on the Interactions tab.

I tend to use a different Observation Field, “Interaction->Visited flower of:” but they are all shown in the tab using this method.

5 Likes

Thank you. GloBI and the Interactions tab with GloBI data are interesting, but I would prefer to be able to search within iNaturalist.

URL codes are great because they can be combined with other search features on the Observation page: I can see the total number of observations, number of species, I can search within a specific geographic area, or analyze my own observations.

GloBI interaction data has the advantage of including alternate Observation Fields like “Interaction->Visited flower of:”, and I suppose with enough work I could use it. But there is a lot to be said for ease of use!

So: is it just not possible to reverse Observation Field searches? It is possible to search all organisms with Observation Field = X, but not possible to return a list of all the organisms in the Observation Field for a given taxa?

The Interactions tab is searching within iNatrualist (not GloBI), but if you don’t like the experimental Interactions tab or GloBI, you can export the details of a set of observations, along with particular observation fields, using the standard iNaturalist CSV export, or you can use the iNat API to do some variation of the same kind of thing.

i think that’s the best you can do without having direct access to a copy of the iNat database.

2 Likes

Please vote for this request and comment there to keep this topic on the agenda!

Another way to do this is through one of the Observation Fields that does the reverse, such as “Interaction: Flower visited by”. Here are all the observations with that field, where the flower visitor is the honeybee Apis mellifera:

https://www.inaturalist.org/observations?verifiable=any&place_id=any&field:Interaction->Flower%20visited%20by=47219

If you change the number 47219 in the above URL for that of another insect species, you will see the plants visited by that species. However, this field is less used than the field identifying the plant in observations of insects, so you won’t see all the information that iNaturalist really has to offer. For example, 4,612 observations of honeybees have the field “Interaction: visited flower of”, but only 224 observations of plants have the field “Interaction: flower visited by” with honeybee as the visitor [in fact, only 184, as 40 of these were observations of honeybees to which the field was erroneously added].

This is why we need a better way to record interactions on iNaturalist, so that an interaction can be recorded between two observations, and remain easily searchable from either taxon.

3 Likes

I didn’t realize I could easily export and download a CSV. That makes it relatively easy to search for this Observation field, download the data, and then summarize it either by insect on plant or plant visited by insect.

Thanks everyone!

2 Likes

GloBI does this automatically.

My main interest is insect-plant interactions, especially plant host relationships. I’m the lead user of the iNaturalist observation field Interaction->Visited flower of.

I rarely use the inverse relationship, Interaction->Flower visited by. Most of my observations are from my garden, or in New York City. I know most of the plants I’m observing, and many are in gardens, or otherwise planted, e.g.: street trees. So I rarely need identifications of the plants, and they would all be Casual, anyway, and unlikely to get identified by anyone else. Another issue is that you can only use these observation fields once. Typically, I’m observing multiple species on a plant, and the fields don’t support this many to one relationship. I would have to duplicate the plant observation just to record each interaction, even if they’re all visiting flowers on the same individual plant.

But GloBI understands that the relationship Visited flower of is reciprocal with Flower visited by. For example, I observed Eristalis arbustorum visiting Allium flowers in my garden. GloBI imports this insect-plant association in both directions.

However, I find searching on GloBI difficult, at best. And I have no idea how to download from GloBI. I much prefer searching, and downloading, interaction observations from iNaturalist.

1 Like

I was looking for a post about observation fields but probably this topic fits as well. There is no video about observation fields already, it there? Found only these https://www.inaturalist.org/pages/video+tutorials
https://github.com/globalbioticinteractions/globalbioticinteractions/issues/822