ISO Examples/Good Practices/Tips for Training Identifiers

I have set up a training with staff from two of my state’s natural resource agencies. The goal is to increase the number of identifiers who can weigh in on aquatic macroinvertebrate identifications.

For those who have trained or otherwise supported the development of identifiers regardless of taxon, I am interested in

  • Resources you have developed or use in training (slide decks, hand outs, websites etc.)
  • Good practices, especially when orienting folks to the identify side of things and setting up filters
  • What not to do

I’m also interested in hearing from the identifier community about what you wish you would have known when you first started out.

Looking forward to hearing the collective wisdom from this community.


One thing that is greatly under utilized is how much can be learned just from studying the range maps on iNat. For some taxa this will be more reliable than others and talking to the identifiers for the taxon in question will tell he state the particular one is in.

Be aware of who the major identifiers are in what ever the preferred taxon/region are, it can be embarrassing to “explain” species characters to an expert only to have them point something else that you weren’t aware of (I’ve both done this, and had many people do it to me). These people will also likely be able to point towards useful resources.

If you get an ID from someone not on iNat, please list who they are and their ID reasoning.


Tip: Just because it looks like a particular species, doesn’t mean that can’t be another species which looks identical. One has to be aware of which species can actually be distinguished given the limited characters visible.


Encourage them to not agree to each other’s identifications at first. If you have say 5 people all agree to each other’s identifications, it can be a PITA to override incorrect identifications.

When I’ve trained undergrads, I’ve had them go slow at first and I’ve gone through all their IDs and given them tips. May not be feasible with a large group, but you’ll pretty quickly see who has an eye for it and who could ahem spend their time with other things. Focus on common species that people can master as oppose to covering everything. If identifiers can ID as “Yes A” vs “Not A” it can make a huge difference even if they don’t know if “Not A” is B,C,D, etc.

My collaborator and I made visual dichotomous keys that were very specific to the region which people have found helpful. Obviously not aquatic macroinvertebrates, but a framework for making the guide. We then went through the guide over a series of four 1-hour Zoom classes.


Wow! ‘A Shutterbee Guide to the Bees of St. Louis’ is so beautiful and (if I may submit humbly) well-conceived. A real pleasure to navigate, even just casually.

1 Like

aw gee thanks. tho comments like these are gonna force me to dust off my notes about what I need to change for the next addition :dotted_line_face:
but in all seriousness, besides bumblebees it should work for most of Eastern North America


General advice for identifiers:

  1. Do your own research and come to your own conclusions. Don’t blindly agree to an ID just because someone else identified it that way. You should be able to defend your ID if someone asks why you chose it.

  2. Be accountable for your IDs! Review your messages and follow-up if someone disagrees or asks a question. It can be a very long process fixing community ID when people do not respond to messages, and too often there are too few experts to help get the numbers to research grade if there are more than one wrong IDs.

  3. Taxonomy is constantly being updated. iNat may use different names than your references.

  4. Don’t be afraid to withdraw or revise your IDs, it is a normal occurrence and we are constantly learning new information! Don’t be afraid to ask others for help or ask for clarification.

  5. Use the iNat computer vision suggestions only as a guide or starting place, and realize that CV is not always the best tool for identifying things. It is most accurate with common, distinctive species, and less so with obscure or rare species. Regional guides are the best references.

  6. Be aware some users, usually new users, will use the agree button as a “like” button. They will agree with your ID to get the observation to research grade without really having the knowledge of why you chose that ID. This is not necessarily a bad thing, but can make it harder to fix later if the ID turns out to be wrong. Just something to consider if you are questionable on an ID.


I’m not sure how useful this might be, but here’s a link to the journal posts I wrote for the first year I organized a identification blitz for New England plants:

The first few posts might be helpful, I hope. And thanks so much for organizing training for potential identifiers!


For aquatic invertebrates like stoneflies, I hope you will urge identifiers to hit the “annotate” tab and note life stage when they make an ID. That information can be very valuable.