Not an unbiased dataset

На мой взгляд иНат показывает не реальное распространение видов, а наиболее наблюдаемые виды. То есть он показывает с какими видами и где чаще всего контактирует человек.

4 Likes

You can adjust for it, as this classic paper illustrates. Although methods may have advanced since 1996, the observation bias for this species is still extreme. Clearly, more iNaturalist observers are needed by whiskey distilleries.

McNoleg, O., 1996. The integration of GIS, remote sensing, expert systems and adaptive co-kriging for environmental habitat modeling of the Highland Haggis using object-oriented, fuzzy-logic and neural network techniques. Comput. Geosci. 22, 585–588.

; )

2 Likes

I have a topic about the problem in the other direction:

https://forum.inaturalist.org/t/abnormality-bias/10626

People also are biased in the other direction, because when people go to a novel location, they are more likely to want to capture everything that they see, because it is a rarity to them. A person going to a pristine beach is going to be much more interested in what they see than a person who is exploring an empty lot near their home, with its “typical” weeds and insects.

7 Likes

For the Southern Bioblitz Cape Town was asked to target our nature reserves.
But even aiming at that request Wolfgat Reserve is still zero obs, for the second time (not safe, not many iNatters nearby …)

https://www.inaturalist.org/projects/great-southern-bioblitz-2020-cape-town/journal/41620-monday-night-all-done

3 Likes

It’s interesting to know what species people notice and pay attention to. One could argue that’s useful information as well.

6 Likes

Good point. Very small life forms will be ignored by people who don’t have specialised equipment to photograph them. I guess that there are also very few observations of skittish critters by those who only use their phones as their wide angle lenses mean you would have to get closer than the critter will allow.

3 Likes

I would like to better understand the automated cleaning section of that article that talks about flagging “coordinates in the sea for terrestrial taxa” as errors. It’s simply not always the case, as demonstrated in this observation: https://www.inaturalist.org/observations/33219090

2 Likes

Agreed. Personal experience has also revealed that specialized equipment can also limit what is recorded. Because I was interested in making observations of pollinators this summer, I had my macro lens on my camera. Even if I had my telephoto lens with me, it was extra effort to change lenses and I tried to limit how much I expose the inner workings of the camera to the elements. I tended not to record larger fauna this summer unless that was all I was doing.

7 Likes

Yup. I make a decision to either put my macro lens on my camera and look for bugs, or my telephoto zoom and look for birds. (Many herps will work either way.) If I see something interesting that doesn’t match the current lens I will generally swap the lens, but I use most of my attention trying to see things that do match the current lens.

5 Likes

2 cameras help avoiding that. :D

5 Likes

I have two bodies but one is older and is not as flexible in settings as I would hope. One thing is that the older only has about 4 fixed ISO settings - and best I can tell, for the amount of time I have put in, I have to go into settings to chage it - good place to store my lens and take the odd quick shot but thst is it. Plan to get two identical bodies, the I’ll be set.

The real point is, as far as bias is concerned, if you only have a hammer, you are looking for nails. If you you don’t have hiking boots, you are staying on the trails. If you can’t get there from here, you probably are not going there.

5 Likes

That’s a pity, I’d left the old one for macro where you’d need stable iso, though you’re the one who knows how it works right now. I remember times when I had to change lenses, it resulted in all shapes of strings of dust on the matrix.

3 Likes

Not in the localities where I provide identifications. I have lost count of the times I brought up an observation tagged “Flowering plants,” and was unable to improve it because it was just a leafy bough of some tropical tree or shrub that was not in flower. It is amazing how similar many tropical leaves can look.

I confess to that bias myself: Except for less than a half dozen California observations, I just don’t upload observations within North America. I am much more fascinated by the exotic places I have been blessed to visit.

But on the other hand, it could be argued that my personal bias helps to counteract the overall bias that the developed world is so much more thoroughly documented. I’ve been a member of the International Aroid Society for 21 years, and in every issue of their journal Aroideana there are new-species descriptions, whether from Colombia, Ecuador, or Southeast Asia. Whereas the number of North America’s Araceae species has stayed pretty constant over the same period.

3 Likes

Although it may actually not be that biased since there are so many people on iNat that almost everything is covered by at least one person, but more likely many people.

2 Likes

Well, in fact not, considering both geography and taxonomy, 50 mln is only a glance on Earth biodiversity.

3 Likes

These biases are not new. Herbarium specimens are mostly from easily accessible sites. Maps of botanist Morton Peck’s collections from more or less 80 years ago trace major Oregon roads (though they are not confined to them). There’s still an area of central Oregon with very few records – bad roads, probably overgrazed vegetation.

3 Likes

A lot has been said here about “invasive species bias” but I believe the abundance of observations of invasive species is sometimes fairly realistic to just how much many overpower life in the area, especially ones that invade urban landscapes. Where I’m at, there are no natural geckos for tens of miles. Invasive Mediterranean geckos are introduced at some point, and they absolutely blanket anywhere they can inhabit. The closest lizards to their niche naturally in this location are fence lizards, and further out, green anoles. In an area you may find less than ten of either of those lizards there will be hundreds of these geckos. And they are observed on Inat accordingly in scale.

8 Likes

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.