Top Observers On The Site

Some folks, like me, prefer to identify, rather than observe. I have a daily walk, and take photos whenever there are things I can find. But I prefer to identify, and don’t take trips to look for life forms. Individual preference.

An addendum: There is a fellow on iNat with zero observations, but over 340,000 identifications. Again, individual preference.


I find that taking photos for iNat via the iNat app on my cellphone (iPhone X) makes creating large numbers of daily observations possible.

I do try to put at least a rough ID on the obs while I am still out, or while I am traveling back home from where I went.

Then I press the upload button and let it all upload automatically.

Then later that day I go through the obs and try to fine-tune the IDs as best as I can, letting the Computer Vision help me in the cases where I know nothing, or next to nothing.

I don’t have a car, but some people live out in the suburbs where they have a lot of access to interesting different biomes, and they drive a lot, and have time each day to go out somewhere, well, they often do sort of mini-bioblitz each place they go and rack up a lot of obs that way.

As for me, often during summertime, in the morning I am able to make 10 or 12 observations right here on my city block, where there is a garden outside my building, tree pits on my block, and a small garden in a traffic island on the avenue, all of which I take care of. And then, I am often able to add a whole bunch more observations if I am lucky enough to go out on a nature outing somewhere during the day, usually via taxi, like to a park that is a mile or a few miles away. And errands or doctor’s appointments sometimes take me to other blocks in other areas, where I can add a few more observations.

I also have a bird feeder on my window which often allows me to capture a few images of cool birds.


I usually take photos with a camera that I can then download to my phone and upload via the iNat app, which probably isn’t the most efficient way, but it’s the only way I’m really that familiar and comfortable with. For me, actually finding and photographing a very large number of organisms and different species isn’t difficult, but uploading them takes a really long time and is really exhausting mentally. For the 2021 City Nature Challenge, I ended up making over 1,700 observations over 4 days, and was able to post them all over the next 6 days. I had been dealing with a lot of other stressful things (not-nature-related stuff) as well, and I had been getting really burn-out. Something about posting that many observations over such a short period of time (compared to what’s usual for me) took a toll on my ability to even muster up the willpower to post many more observations. As a result of that, I haven’t posted many iNat observations since early May, even though I’ve been out observing at least as much as usual and have probably several thousand photos to post to iNat. It’s just something that I couldn’t keep up anymore. I’m sure I’ll get them all posted eventually, but I have no idea when, but right now posting so many observations so intensively would not be good for my mental health. Maybe I just need to find a better way of uploading that many observations that isn’t as time consuming.


What criteria did you use to assess the quality? Do you know what is required to reliably identify all the subjects you looked at, and how commonly recorded they are within a given region?

The iNat App makes recording certain subjects very easy (some might say, almost too easy). No expertise is required: all you need to do is take a picture with your phone, then “identify” it with the CV and tap a button to upload it. This will usually produce a perfectly valid observation that can often achieve RG status almost immediately.

But what will be learnt from such a process? I suppose repeating it thousands of times might make the user a better recorder - but will it necessarily also make them a better naturalist?


What can we learn from this? Spots of distribution of those species in date and time, we will learn we have another sould interested in what lives around us, there’s nothing inferior in being a good recorder, and it qualifies under naturalist term pretty well. I don’t think we need anything more, but we can also see from it where users are/were active and where we need more study, both in urband and “wild” settings, both changing rapidly and in need of attention right now. Quality for iNat is identifiable photo, not just thousands of blurry ones where you can’t id anything (which everyone will make from time to time under challendging circumstances, but they shouldn’t be all like that).


I just meant photographic quality in this context…( the top observers at present are not active in taxa or geographies I’m particularly familiar with ).

It’s certainly visible in a few of the top observers that (rightly or wrongly) they are sort of gaming the system by making multiple observations of the same species in close proximity as you mentioned. But there is at least one high observer at present ( @reiner) who seems to have high quality photography, and doesn’t seem to be doing this. I have also seen @alexis_orion on the leaderboard previously, a user who has consistently high quality obs of mesofauna. ( hope you folks don’t mind me tagging you in! - I’m curious to know about how you optimise your workflows if you are open to sharing )

I do think the gamification aspect of having a leaderboard within iNat should promote quality over quantity though, as I’ve said. I have attempted the 1000 species in 1 sq km challenge for two years running in different locations - the great thing about this objective is that it forces you to step outside of taxa you are familiar with and to increase quality and detail in observations in order to attain it. I believe there is potential benefit both as a naturalist and a recorder to push yourself to attain a goal like this. But I accept that’s not for everyone and of course there are many other ways to develop your skills as a naturalist too :)


I’d be curious to hear from “power observers” what their hardware is. One of my biggest limitations in observation speed is my mediocre phone. There is delay between each step of the process: unlocking phone, new observation, from photo, try to focus image, still focusing, try capture, select retry, repeat, select okay, review observation page, (sometimes manually trigger gps location), (sometimes attempt ID), complete observation. Processing delay or data lag all contribute to the length of each observation. I’d be curious if a nice phone clicks thru those steps quickly and would enable fast observations.


rather than upload each photo/observation as you take it, just photograph everything using the normal phone camera, and then upload everything when you get home; this will dramatically cut the time taken


The site is called “iNaturalist”, not “iRecord”. The primary focus is on the observer, not the data. What third-parties may learn from the data is only of secondary importance. Becoming a more efficient recorder won’t necessarily make you a better naturalist. The two aspects are orthogonal to each other. It makes no sense to claim that one might be “inferior” to the other, so I don’t understand why you are suggesting that.

There’s much more to iNaturalist than merely accumulating data. In the UK, I use a site called iRecord for that specific purpose. Its primary focus really is on the data, so I know that, in the long run, it will make much better use of my observations. I mainly use iNaturalist as a learning tool. In that role, it has certainly helped make me a better naturalist, and I hope that by sharing what I’ve learned, it has helped some other users do the same.

1 Like

I am by no means a power observer, but I have many of the same problems with my DSLR. Even when the camera is out, zoom, focus, exposure, etc. all create a delay in which time the organism can vanish. Getting the camera out of the bag just makes it all worse, something you would think I should have learned by now.

I don’t think photographic quality has much relevance at all. It is certainly nice if an observation includes some artistic and/or technically accomplished photos. But if you can’t reliably identify the subject from them, the observation will obviously lose a lot of value. As an identifier, I don’t much care if the photo is blurry, badly exposed or poorly framed, so long as i can see all the relevant characters.

In the context of scientific nature recording, the informational quality is what matters most. This is where the two aspects of recorder and naturalist intersect. If you don’t know how to identify things, or what can be identified, that will significantly reduce your chances of taking good quality photos (i.e. ones with high informational value).

Creating records in that kind of way is entirely valid, and can produce lots of perfectly good RG observations. My point was simply that it doesn’t necessarily require any kind of super-human effort or unusally efficient workflow to generate lots of observations. This may sometimes result in a somewhat lower signal to noise ratio (which is why I asked how you assessed quality), but that isn’t necessarily a problem in itself. Some people just enjoy the process of recording for its own sake, and don’t mind too much if their efforts don’t always produce RG observations.


Because there’s nothing orthogonal in that, our top observers are not just recoders, you really can’t be just a recorder if you spend your time with nature, it’s impossible to not learn things. Focus is on observer, that’s why the question of how someone becomes “recorder” is illegitimate, everyone observes the way they like. It’s unfortunate that you think that focus on data means better use of it, but it’s your deal and not other observers, if you use it as learning tool it’s great, but others can do their own thing, including photography of each plant they meet, what’s your problem with that?

Why all the hostility? Please read my posts more carefully before jumping to unwarranted conclusions.

You call people recorders, that they don’t do iNatting the supposed way, making RG observations easy as that, but somehow still we have a tiny proportion of people who actually have 100+k records, or 500 a day, so maybe it’s not that easy after all?

1 Like

I don’t understand what you’re arguing against. Surely you must agree that people who know more about about nature make better nature recorders?

iNaturalist makes it easy for people with no experience whatsoever to make useful observations. And that is a good thing, because it allows them to engage with nature in a positive way without any prior knowledge. But photographing a butterfly and using the CV to find out what species it is won’t make you an expert identifier, no matter how many times you do it. At some point, you need to start reading books and learning from other people if you want to deepen your knowledge. Recording data is just one part of what we do. There’s much more to being a naturalist than that.


But they’re observers, not iders. If the tool is here so you can skip routine that not everyone likes, why not use it? You can appreciate nature and not know how to separate two similar species. We talk about top contributors, not new users.

As a newish user here (introduced via a local Bioblitz) I have learned much about what it takes to document an organism. As long-time user of eBird user (online only until Merlin was released) this is a very different experience. Birders are only required to record observations as a “witness” but without photo or audio. eBird includes a self-report “experience/knowledge level” when signing up.

What isn’t clear to newbies outside their expertise, like me, is what it takes to get a firm ID for which organism. It may be that DNA is necessary so that might be useful to know when posting. Some sort of red flag might pop up for the most difficult. But for plants, a flower is often helpful but flower plus leaves plus stem and maybe root may be necessary. That takes time and experience to know how deep to go.


I think a couple of keys for getting large numbers of observations are these: 1) Use a dedicated camera along with geotagging software on your smartphone. This is much faster than using the iNaturalist smartphone app. 2) Be an expert on the taxa you are observing. For example, if you are well versed in botany, it is super easy to find large numbers of species at just about any roadside or other site and you’ll be able to determine quickly which species are there. 3) Be aware of all different taxa and make observations of things that you are not targeting. For example, if you are targeting plants, and you see interesting insects, spiders, fungus, animal tracks, etc, make observations of them as well.


Use a camera with GPS and whilst out - if you see something you know then photograph it; but if you see something you do not know - then also photograph it - a very good way of learning new species.


Because it often doesn’t work, and when it does, you can’t learn much from it.

That seems self-contradictory. If you can’t recognise the differences between things, what is there to appreciate? One of the greatest pleasures of engaging with nature is developing a deeper appreciation of its amazing diversity. It seems to me that you are at risk of dumbing things down to the point where it’s becoming almost condescending. In my experience, people genuinely like learning about nature, and it really bothers them if they don’t know how to identify what they’ve found.

Everyone has to start somewhere. What was the journey that took them from zero observations to becoming “top” contributors? Did it really differ significantly from everyone else’s journey, and if so, how? That is what we are talking about.