Not that every photo is a first, but they have been observed for the first time at some point and not identified since. If one such observation was identified it would officially be an inat first for said species.
You miss my point. You were writing about the discrepancy between growth in numbers of observations and growth in numbers of species. Some species are always going to be difficult to identify – in all likelihood there will be lots of unidentifiable observations of those species before someone manages to post an identifiable one. This means that the numbers of species identified on iNat will always lag behind the number of species actually seen. It also contributes to growth in numbers of observations without a corresponding increase in number of species. In effect, for some taxa in some places iNat has already reached the maximum number of species that can realistically be expected to be represented from field photos (the default type of observation on iNat), even though the number of species that exist in the real world may be larger – because getting identifiable observations of the remaining species will require an additional effort that goes beyond field photos (photos of specimens, DNA barcoding, etc.)
I decided to look into this further. I will release the chart that separates individual orders, but for now I think this is sufficient.
In your post you seem to imply that we have already observed all of the ‘low hanging fruit so to speak’ and that fewer species will be observed. But the current data does not suggest this. For arthropod as a whole, there does not seem to be any asymptote developing, meaning we are still in the high growth stage. What looks like an asymptote at the end is just because we are still early in 2026 and thus a year’s worth of new species have not accumulated yet. But from 2008 to 2025 there seems to be no signs of slowing.
1 Like
No. I said that for some taxa in some regions this is the case.
There are also other factors at play in your graph, such as the growth and expansion of iNat (more people making observations in more places) that mean the growth curve does not show the same pattern as it would if the number of observations were consistent over time. I imagine if you were to control for the number of observations this graph would look very different – note that the change in the number of species over time in your chart appears roughly linear while the change in number of new observations per year on iNat is more exponential. I expect that the ratio of of new species to observations is likely broadly decreasing – it would be highly surprising if this were not the case, for all the reasons I gave above, regardless of how difficult or well studied the taxon in question is.
1 Like
Here is the accumulation of species for each arthropod order between 2008 and 2025. Nothing from 2026 or from before 2008 is included here.
1 Like
Don’t quite a lot of sweat bee species (and I assume other insect ID) require stuff like dissections, examining genitals under a microscope, and other details that almost no observer will be willing to do due to the high amount of effort and cost?
