After I’ve made an observation in the field, I often archive the specimen as an herbarium mount and deposit it in an appropriate herbarium. It would be nice to be able to cite and link the herbarium record in the iNat record. An earlier topic ( https://forum.inaturalist.org/t/linking-inat-records-w-museum-specimens-collected/8586 ) suggested that one way to do this would be to add a tag to the iNat observation, but I’m not sure how this would work. Can the tag be a link to an herbarium specimen database? Is there a way to add tags to existing iNat records with a batch process? I’m hoping someone has figured this out.
When we voucher a specimen, I put the collector and collector number in the notes section. It is usually some months between taking the voucher and entering it into the herbarium database. Sometimes, I add the accession number to the iNat observation when I enter info into the db.
I have seen other people just put the accession number in the notes section.
Putting the herbarium’s accession number into the notes section of the iNaturalist observation has the advantage of making it very visible to users, but as far as I know, there’s no way to add this information other than editing each individual notes section. When we’re talking about hundreds or thousands of entries, we’d need something more efficient. But I’m glad you’re doing that.
I’m not sure what your labeling, pressing, reorienting, confirming, and organizing workflows look like, but in mine adding a number usually my collection number (so I can have a stable inaturalist observation number to put on my label) represents very little time commitment compared to the rest of the process. Collecting and taking notes on location, habitat, associated species takes me a couple minutes at my fastest (4 minutes per specimen on average on my fastest collecting trip usually significantly more), putting them in the press takes a fair amount of time (sometimes more than 10 minutes per specimen if they are unwieldy or I have to make decisions about what material to include vs exclude) I always try to restack my press a day later to reorient specimens and ensure that the right parts are on top of other parts (another couple minutes), confirming IDs and making sure key features are usable (another few minutes), entering label data takes significant time per specimen to ensure correct spelling and check authors (5+ minutes per label) not to mention formating and printing, and even bringing them to the museum takes hours (another minute or two per specimen)!
That puts my grand total time per specimen at something like 20 minutes of my time per specimen. That is the unfortunate reality of collecting (and in my opinion why listing and photographing has an important but different place than vouchering) even fast collectors (which I do not consider myself) take minutes per specimen to take it from the field to accession. Adding in the backlog can seem daunting, but an extra minute per specimen seems to me in line with another of many steps in my voucher workflow. Collecting good specimens seems to not be a very efficient process, at least for me.
It’s a labor of love, for sure. But here I’m not talking about writing a number on a specimen or just entering a number into a database. Picture that we have an iNaturalist observation that is also a physical specimen, and that physical specimen is eventually databased with its museum or herbarium number and ALSO the iNat number. Now the database has information linking both of them, but the iNat observation doesn’t because it was created first. I’m looking for a way to attach the museum number to the iNat observation so that iNat users will be able to use it to get to the museum specimen. You might be right that I could do this in a minute per specimen with a lot of searching and clicking. I just hope there’s a better way.
Can you provide a little more information about the type of process that you are looking for? It sounds like there are at least two questions you are asking about (where to add the info on iNat, how to add it), but I’m not sure what exactly you are asking for.
Adding an herbarium specimen number or similar info in the notes/description, a comment, or an observation field (what I would recommend) are probably the main options available to you. You can see available observation fields for similar functionality here: https://www.inaturalist.org/observation_fields?utf8=✓&q=specimen&commit=Search or whatever search for fields you want to try. Other users might be able to recommend specific fields.
It also sounds like you want some kind of programmatic way to add this info that isn’t just entering the data individually yourself. To get a useful answer for this, you’ll probably need to identify a specific place you want to add the info (description, observation field) and then say how that info is currently stored. You’ll need a digital source (spreadsheet, database) to do any automated addition. I’m sure this type of work can generally be done via the API and presumably via existing R or Python packages/libraries, so if you are familiar with using any of those (or have a preferred one) it might help to list that as well.
Janet, do you have a downloaded spreadsheet of the herbarium specimens with their accession numbers? I would think some smart person (@pisum, @alan_rockefeller) might be able to tell you how to do a bulk edit of your iNat observations to add the individual accession number to each specimen.
I hope, @lynnharper!
You’re correct that I would be interested in advice on WHERE to add a museum/herbarium number to an iNat observation and also HOW to do it in bulk fashion. Let’s assume I can get a spreadsheet that has both the iNat observation number and the museum number, and also that I don’t have facility with R or Python. Thanks for helping me clarify.
I have been using a related approach, although it mainly solves the link in the opposite direction — from the physical herbarium specimen back to the iNaturalist observation.
The idea is based on Heberling & Isaac (2018), iNaturalist as a tool to expand the research value of museum specimens. They proposed including the iNaturalist observation URL as a QR code directly on the herbarium label, so someone examining the physical voucher can immediately access the corresponding field photographs and other information that is not preserved well on a pressed specimen. Heberling & Isaac 2018
For my collections I have implemented this as a LaTeX workflow. My collecting spreadsheet contains the iNaturalist observation URL/ID together with the usual specimen metadata, and LaTeX generates the herbarium labels in batches. If an iNaturalist link is present, a QR code pointing to that observation is automatically placed on the label, with the observation ID printed below it as a human-readable reference.
This works very well for linking the voucher to the iNaturalist record, particularly because field images of habit, flower colour, habitat, etc. remain directly accessible. It does not by itself solve the reverse problem you describe — adding the herbarium accession/database record back to iNaturalist once that identifier is assigned later. For that, a scalable way of updating observations in bulk would indeed be very useful.
@jwidness wrote something back in the day that handles something similar: https://forum.inaturalist.org/t/how-to-bulk-batch-add-its-sequence-data-to-an-observation-field/31382. it would only take a couple of minor edits to make it handle your case. it’s Python in a Colab environment, which means that you would need a Google account or your own Python environment to run the notebook/script after modifications.
the notebook is well documented and it’s not a super complicated process. so i think even someone not super familiar Python would be able to run it successfully, with some minimal additional guidance on the necessary modifications. (if you’re interested in trying, i can explain what would need to be changed.)
…
if i were to rewrite jwidness’s notebook to make it a little more generalizable, i would probably either:
- ask the user to define the observation field ids they’re interested in adding to in the header row of the CSV to be imported, or else
- ask the user to define the observation field ids they’re interested in adding to a variable
either approach above would make the notebook a little more flexible and allow for multiple observation fields to be added to any given observation.
i’ve also been adding my own Python notebooks to my own JupyterLite environment which allows folks to run things entirely in a browser without Google account or knowledge of how to set up a Python environment. if there’s anyone that doesn’t want to use a Google account or have their own Python environment, but still is willing to run Python, i could add a notebook to my JupyerLite instance.
…
it’s worth noting that iNat’s Batch Edit screen does allow you to add observation fields (but not edit them). you would still have to define the herbarium values one observation at a time, but you can do it batches of 50 or so on one page. so that could be slightly faster than doing it via the Identify screen, and it would be much faster than doing it one observation page at a time.
…
but if you want to fully automate that process for a big set, that still requires some sort of scripting, unless someone builds a webpage to help automate the process. creating such a page would be a nice project for someone interested in that sort of development. i’m not going to do it (since i think scripting offers a little more flexibility in this case), but it wouldn’t be super difficult for someone with the necessary technical background (and/or a coding AI) to do.
Thanks, @timboeh – yes we already do this, but your directions will help others do it too. Our herbarium labels and herbarium database give the URL for the iNaturalist observation. Your QR code is a step further – nice! I am pleased to be finding out how many museum people are interested in linking to iNaturalist. I hope this becomes a standard practice.
Thank you, @pisum. I’ll look into that and may get back to you when we are farther along in the current herbarium project.
I have a personal collection of mostly insects. Each iNat observation of something in my collection get a “My Collection” tag and then the observation field “My personal collection #” with the number from my catalog. The catalog then has an iNat field that has the link. The number is also on the specimen label allowing all three to be easily associated.
Often, GBIF will automatically link related records, assuming your herbarium specimen is getting ingested by GBIF and your inat post become RG. https://www.gbif.org/occurrence/2242954936
What I do personally is have my collection number in the description of my iNat post. I figure that if anybody comes around trying to back associate from iNat to a specimen, that they will figure it out if they’re motivated, or at least comment or message me for more details about the specimen.
That’s awesome. Someone had told me that before but I hadn’t looked for it; thanks for the example! (Unfortunately most of my linked observations are mosses or liverworts with microscopic characters, which are very slow to reach RG status as there are very few identifiers, but others may have better luck.)
This is an amazing tool by @alan_rockefeller
I use it for all my insect vouchers as well.
The folks at Carnegie (CM) have a system that pulls specific iNat observation fields (used in their project) from a data export and adds them to an herbarium label and prints a QR code linking to the iNat observation. I haven’t used that system myself, but it seems to work very well for them. Might be worth reaching out to them to see how it’s set up if you’re interested- I know Mason and Bonnie (Heberling & Isaac respectively, mentioned above via their 2018 paper), both of whom are still active on iNat, generate all (or most) of their labels this way.
Thanks, this is good to know. We have a similar system but it’s clunky and doesn’t include the QR. (This thread is actually about going the other way, back to iNat, but I’m finding these suggestions pretty awesome.)
Oops, that’s what I get for responding before my morning caffeine kicked in. Sorry about that!