I recently went on a pretty large hike, and ended up with 434 observations from that day. I tried to observe everything unique/interesting that I could, but it had me thinking, would it be at all beneficial to observe EVERY organism I see? Every plant, insect, lichen, just everything I can photograph reasonably where they’re identifiable?
I figured I would ask here, because this forum is much more knowledgeable about what data is useful, and also have a very big chunk of identifiers to boot. I just don’t know if posting 100+ observations of the same plant species in the same area, or the same galls on that plant species would contribute anything other than making identifiers hate me
I personally avoid uploading duplicates of the same species when they’re within a kilometer or so of eachother (at least during the same day). For most data uses, in my opinion, super dense data, having multiple obsevations of the same species within eyesight is not the most useful. In fact, for some of my projects I have downloaded data and then used a GIS filter to remove these dense data so that organisms of the same species can be only 10 km apart for example.
You probably won’t get very far on your hikes if you post literally every single organism you see :D If you’re sure something’s the same organism and aren’t interested in documenting every individual either for yourself or for a purpose (a project, etc.), it may be easier to skip those. For example, there may not be much point to observing every individual of the 3 dominant tree species that make up the forest.
But there are some organisms (e.g., ferns) that I’ve observed “repeatedly” on a hike and only figured out later that they were actually multiple species!
If you tried to record “everything” it wouldn’t be much of a hike
Record what interests you. if you are a birder record birds. If you participate in a pollinator project (and you should) record pollinators. Do you love plants? Record them and ignore the birds and the bees.
The short of it is that you can record as many observations as you have the time and interest to do. Don’t worry about identifiers, the algorithm, or scientists.
Now the hard question: should I delete most of my observations (one sp. point every 0.001° or ~70m, also if on a same day) to save dataset users some filter/decimation steps? what is the consensus of “acceptable density” to iNat experts?
(published biogeographic grids in my area tend to be in the 100m-1km range, with 10km for continent-scale only…)
There are a number of different possible uses that researchers might find for iNat data, and I don’t think we can know for sure whether any particular observation will be useful or not.
Someone who is using iNat to study variation or phenology might find a densely grouped set of observations from the same place and time to be really helpful.
And for things like, say, pollinator associations, the relationships mapped by the observations are probably going to be more important than repeat records of the same organism at a particular place.
I personally don’t generally find it meaningful to observe the same species multiple times at the same location, unless there is something notable about the different specimens that I want to record. Also, I would find it exhausting to try to properly photograph everything on a particular outing, so I mostly observe what happens to capture my interest.
I agree with the interests thing; I usually go for birds if I’m walking along roads, take pictures of flowers here and there, and photograph all the insects I see. There isn’t a definitive answer to this; do whatever floats your boat!
It’s your choice. Any survey is potentially useful, even if it is not a “formal grid survey”. You can think of it as a mini “City Nature Challenge”. “USA National Phenology Network” volunteers visit the same sites repeatedly to document phenophases. Observe to your hearts content.
It’s ultimately your choice how you’d like to upload them. If you’re doing it for yourself to survey an area and learn about it I’d say go for it. But if you are only doing it in hope that scientists will find your data useful, I would thin them out. I don’t think there’s a consensus. The numbers I used of 1 km between observations of the same species is based on my work with species distribution modeling.
I’ll just add that firstly - for many of the invertebrates that i like, they’re small and dull - so often overlooked. For those anything new is welcome. There’s already tons of photos on the system of the big, bright and hence frequently observed creatures.
Then secondly, please consider that in many cases with more obscure creatures then multiple photos can be really helpful to show different angles and features. So, please consider the trade off in your time where you could be making one observation really well (justified with many photos - some hopefully showing the key diagnostic feature/s) versus many single photos which may all lack any critical diagnostics
Thanks. I do it primarily to fill gaps in seldom surveyed areas, for municipality-level biodiversity inventories. I suppose that sampling and exporting 1 point every km could do the job (~0.0166° instead of 0.001: that’s a 94% reduction in obs count at best, starting with all those stuck at ‘Needs id’!), as most parts of municipalities are larger than that.
(Of course the tricky part is now to decide on how to place the future 1-km grid of points to survey, to maximize the likelihood of sampling all taxa of interest for conservation…)
edit: as an added benefit it will remove me/keep me out of these pesky “leaderboards” :D
edit²: finally “thinned things out” - removing at least 5000 out of 10.000 superfluous observations, keeping only 1 record per species per municipality (~25km² on average). Less redundancy, and hopefully less work for identifiers.
I wouldn’t feel obligated to make a lot of observations of the same organism. If you enjoy it, that’s great, but do keep in mind that add lots of them to iNat does create more work for other users - it’s not without an effect on the community. I wouldn’t let that be the only factor in your decision, but it’s something to keep in mind.
One thing I like to do is, if I see a species multiple times along a trail, I’ll only post observations of that organism I see further along the trail. A lot of us (myself definitely included) will only make an observation of a species the first time we see it on the trail, which biases the data toward trailheads. So I like to fill in the gaps that open up deeper along the trail.
When doing this hiking alone I try to observe the first time I see something then semi-randomly repeating, especially if I get to a potentially different habitat type. If it’s on, say, a mountain slope I might also try to capture the last elevation I saw it if I still have the energy on the way back. In addition to better ranges, this also helps if I have mistakenly believed that I had seen a plant before but had not; for example, if a species concept is later split.
The disadvantage to taking a lot of observations is that I have a semi-absurd amount of observations on my phone that I have not actually uploaded, probably more than the total of what I have actually posted to date… I’ll get around to it someday
Hiking with other people forces me to ration stopping much more, and also decreases the relative density at the trailhead as compared to whenever they happen to stop to take a break
Unfortunately, that too often also means forever unidentified. “Round-backed Millipedes” is one taxon on iNaturalist, but encompasses how many species in the real world?
This matters to the original question because the degree of benefit is affected by the granularity of the identification. We don’t learn anything new from learning that there are “round-backed millipedes” in an area, but we may learn something new from learning which ones are there.
Think about what people will want to see in 50 years, and why they want to see it.
Both of these are unknown. Search tools that have not yet been invented can pull data out of landscape photos.
150 years ago, people took photos of houses. These are invaluable to building restoration. Unfortunately, people tended to take photos of what was extraordinary, and leave out the ordinary. For example, expensive trophy houses were well documented, working class housing not so much.
Documenting only the trophy finds of the natural world ignores building a data set of what is here now. It would have been interesting if many people had documented forest succession as the Chestnut trees died off in New England. A picture of one Chestnut tree near a trail head would not show what the forest looked like. Knowing that one example of a plant existing in a town doesn’t seem like much of a data set.
I see iNaturalist as a chance to pass on information otherwise unavailable in the future. Instead of pre-judging what the future will want to see, I try to include as much as possible. That includes an observation, its environment, groups of observations in one place, and multiple examples of one species in an area.
No, it depends. I don’t think it’s correct to generalize that only species-level observations tell us anything meaningful. More specific is of course going to be more useful than more general, but this doesn’t mean that an observation at a higher level is automatically useless.
Some examples:
I helped out with a local citizen science project on monitoring watercourses that included sampling the macroinvertebrates; the material they provided focused mostly on identifying to family level (because of the difficulty of identifying many aquatic larvae, and also to allow participation by laypeople without prior experience or training). But within the framework of the project, this was considered sufficient, because the families of organisms present and their relative frequency still provide information about the health of the ecosystem, the presence of pollutants, etc.
Sometimes even studies carried out by scientists with access to labs have to use taxa higher than species. I was reading an article on the use of “ballooning” as a dispersal method by spiders. With many spiders, only adults can be reliably ID’d even with a specimen in hand because the relevant features haven’t yet fully developed in juvenile specimens. The results table in the study therefore included separate entries for the juvenile spiders captured (usually at genus/family level) and for the adults (genus or species). This doesn’t make the results useless or invalid – on the contrary: had they omitted the juvenile specimens because they couldn’t be ID’d to species, this would have made their results essentially meaningless, because ballooning is a technique that is most frequently used by smaller and juvenile spiders.
The presence of aphid mummies on my balcony plants tells me something meaningful (that the natural predators are doing their job to help keep the pests in check), even if I will probably never get a species ID for any of them unless I manage to capture one of the tiny wasps that emerges and examine it under a microscope.
I’d argue that these cases tell us more than a species-level observation of some of the more ubiquitous organisms – one that comes to mind around here is Pyrrhocoris apterus. Its presence tells us virtually nothing, because it can be found in nearly all local environments, most of the year. And yet even here, the observations may provide other types of data: whether they are congregating in large groups for the winter or adventuring on their own, when they mate, what they are eating (or who is eating them), whether they are nymphs or adults, etc.
PS: If the observer is unhappy with their observations of round-backed millipedes staying at Juliformia, what is to stop them from learning what characteristics are required for ID, finding a key for the local area, and taking photographs and/or collecting specimens to allow for a more specific ID? Just because there are few people who are IDing the taxon on iNat at present doesn’t mean that it has to stay this way forever.
I do something similar, but sometimes if I’m on a hike and I see something I feel like having a map for along that trail for whatever reason I’ll take observations pretty frequently.
Edit: oh lol hey you’ve identified a lot of those, thanks
