I recently attempted to determine when most plant species in an area could be observed & recognized to guide my decision-making for when/where to visit during next year’s growing season. To accomplish this, I used the explore page to search the area and filtered by plant and every month of the year (separately). A few issues with this method immediately became apparent. (1) Highly motivated and skilled naturalists caused the data to skew from month-to-month based on their activity. (2) Observations seemed to favor rare and endemic species, so species richness among observations was not an accurate reflection of species richness in the area.
Is there a way to use iNaturalist to accomplish my goal? Are there websites that might better help me accomplish my goal. For example, is there a way to use an herbarium database for this purpose?
I have also been trying to use iNaturalist to determine biodiversity. Active Observers and rare focus are big issues. I recently dealt with some of those issue when creating a map based on county level species count (the code allows for any state but it is set to Colorado) see: https://github.com/JaredLincenberg/iNaturalist-Biodiversity but it doesn’t yet handle the time component.
There is a powerful option with pyinatualist. It is a python client to access the iNaturalist API. One of the examples tries to address your question of unique species in an area.
I normally check google maps for any promising patches of forest, then check Inat to see what lives there so I know what to expect.
Observations are definitely more biased towards rarer species though so I’d say it’s more an indicator of the total number of species rather than how common/rare they are.
The compare tool is perfect for this analysis. I used it in 2024 to plan when to visit Chile to see Eriosyce in flower. Take a look at this phenology plot and adjust it to meet your needs.
I don’t have much experience on flora, but with fauna it works quite well because the most common species are often the top observations. The same with diving observations.
It’s definitely not scientifically accurate or exhaustive, because there are a lot of common species that are clearly overlooked (for example, pigeons, moss and lichen in general and sponges when diving). But it’s a nice glance of what to expect.
My experience is in Europe and Southeast Asia mostly, so it might be quite different in other areas.
As you’ve said there’s often biases towards certain species / groups of plants. Although this is true the process you’re describing is basically what I’ve been doing for the past two years in Australia as a way for me to familiarise myself with local / regional species and how their distributions are shaped by geology, geography and climate. Though I’ve often investigated an area for interesting and restricted endemics, I also try my best to capture the full pallete of species present in an area (which does present the issue of countless hours of photo processing in lightroom though I find this rewarding in its own way). So at the very least I’ve found iNat as a great resource to point me in the direction of interesting locations for endemic plants then prompts further investigation from other resources such as local ID books / herbarium records / national parks documentation.
I don’t really have a specific workflow I follow necessarily but have found a lot of success in getting to grips with the range of diversity present across Eastern Australia. I’d mostly say don’t worry so much about optimizing a particular workflow, and if anything by you going out to areas for something specific you’ll be able to fill in the gaps of species that haven’t been logged in that area before.
I also have done the same thing, plus if there’s a species I’m interested in, I will check the seasonality chart, of course knowing that it’s affected by elevation etc.
The reality is that this type of data collection only does one thing - confirm presence of a single species, and even that can become questionable if it’s a species easily confused with others.
I wonder if anyone has done a project that includes a plot protocol? Or a protocol to observe, say, the 5 closest plants, link the observations, and state the distance to the original observation?
There are a lot of approaches to correcting “biodiversity” for observer/observations, the simplest is to divide observations by observers to get a relative count.
To focus on species of interest, you can filter for species that you haven’t seen (add to the URL: &unobserved_by_user_id=).
I frequently use INat to look for good birding spots when I’m on a trip- I filter by the general area I’m in, “birds”, and the same month or two out of the year (ex. every April), then look at big clusters. The problems you’ve mentioned are of course still present, but if I see a park that has a big cluster of different bird species on it, it’s probably worth checking out.
Figuring out the best time to go to an area is a bit trickier, but it seems to me that this filtering option ought to still work- sure, someone making a ton of observations in April and none in May (for reasons other than “there were a lot more birds in April due to migration” and the like) might skew things for that year, but the chance of them repeating that pattern every April for several years is relatively unlikely.
I travel full time so I use iNat for this constantly. But one of the most useful tips I have is to filter for which users area doing the most IDs and observations in an area, read their profiles, and pick a few to direct message. I’ve gotten amazing advice that way and even a personal tour of a local botanical garden from the director.
If you show interest in a place someone loves, they’re most likely going to help you out.
you can get more granular data – weekly, daily, hourly, etc.-- using either the compare tool that mrtnlowr mentions above, or you can get it directly from the API. below is a calendar heatmap visualization that i like to use to view that data from the API. normally what i would do for a case like yours is to look first at all observations in a given place or area to see if there are any spikes due to events like the City Nature Challenge. then, i can add a filter for, say, taxon to see the more specific case that i’m interested in. example:
i wonder if this sort of comparison would be useful to see applied to a calendar heatmap, as opposed to just a regular map? (if anyone’s interested in that sort of thing, let me know, and i might add that as an option in the calendar heatmap page.)
I have only dabbled a little in coding before, so it’ll take some time for me to learn how to use these resources. I’ve got them saved so I can come back to them this winter when I have more time. Thank you for sharing!
I’m looking at areas anywhere between 14 to 55 square miles (the size of the smallest WMA in my state and the size of the largest wildlife refuge in my state, respectively). For the smaller areas, I probably won’t have much luck with the flower attribute, but I could probably use that for the larger wildlife refuges.
I haven’t been doing this on an individual basis when visiting new WMA, but I think this is a good idea. I feel like knowing what species exist somewhere (and what they look like) acts as a sort of primer to recognize the visual cues that you need to spot them in a busy landscape.
Perhaps its a difference in culture or the size of the land. The southeast USA has a lot of great naturalists, many of whom I know. But, I get the impression that the population of this region is not very interested/educated in the natural sciences relative to other parts of the country.
Also, I get the sense that much of the land in Georgia, USA is privately owned and closely monitored for trespassers (to prevent poaching on private hunting land, squatting, etc). So, I am limited to looking within specific points on a map, which may not be the case elsewhere.
That’s a good point. You recommendation is similar to what I’ve been doing, but as I’m expanding my search I’m coming up against time and financial constraints. This is the main reason I’m trying to optimize my workflow.
But, some of the observations that have fascinated me the most are the ones I didn’t expect, so maybe there’s something to be said for taking a more casual approach to naturalism.
I like this idea. I think that observation count is often reflective of the goals of the observer. so, If I remove those whose observation count is significantly higher (those using inaturalist to count populations and such) and those whose observation count is significantly lower (who just wanna know that the cool bug is called), then I should be able to select for those who use the tool for similar purposes.
It may take me some time to learn how to use those resources, but I’ll let you know if one of those is a definitive solution… someday that isn’t a work night
I’m mainly searching flora because my iphone camera is a limiting factor in what observations I can make. If I ever get a proper camera, I would definitely switch to considering fauna with my annual searches. This does have me thinking that I may want to use this method in local parks since they’re visited more often than some far-flung WMA, so the large number of inaturalist users would hopefully smooth out some of the observation biases caused by super-users. Perhaps more accurately estimating peak biodiversity will allow me to see more species in a landscape managed for non-biodiversity reasons.
It’s funny, sometimes I get so caught up in the wealth of data that I forget that inaturalist users are real people I can contact. I always try to guide my friends around my favorite trails during peak bloom periods, so it makes sense that a motivated observer would want to do the same. Great idea.
A lot of phones cameras have got much better at macro, which is very accessible for insects while observing plants. So don’t forget invertebrates in general. A lot of them often like to stay still for the photos provided we move slowly and with no shadows.