I’ve done quite a bit of plant survey work. The method to use depends on your goal. If you want to list all (or most of) the species, or if you want to find all the listed rare species, then wandering around all the habitats at all the appropriate seasons is the way to go, especially if you have the experience that will allow you to recognize some of the subtle habitat differences.
This doesn’t sound “scientific,” at least to people who don’t do this kind of field work. They prefer transets or sampling on a grid or randomly place plots. Well, if you’re trying to measure relative densities or estimate populations or look for certain kinds of associations, that’s the way to go. If you’re trying to list biodiversity, that’s stupid because you’ll miss lots of records unless your sampling is incredibly dense and you’ll waste a lot of time recording the common species.
In our work, the field botanists would OK our plans and then some body who works mainly with papers any more and vaguely remembers some ecology class from college would object. We developed some term for our system (intuitively controlled surveys of appropriate habitat types, I think – maybe I’ll remember to ask my colleagues). Keep doing what you’re doing.
Is iNaturalist a reliable enough source of identifications for a scientific paper? Well, if you just accepted all names anybody put on your photos, I’d yell, “Whoa! Stop!” But I’ll bet that’s not what you did. I’ll bet you paid attention to whether the names made sense. You used iNaturalist to confirm your ID’s, and you probably checked names that seemed unlikely or just unfamiliar. And/or you paid attention to who the identifiers are. Although many of us are more enthusiastic than knowledgeable, and all of us are ignorant of some taxonomic groups*, among the identifiers are regional or even world experts in various groups. The same experts we’d consult “in real life,” if we could.
[this was supposed to be an asterisk.] I keep learning that I’m more ignorant of various organisms than I thought I was. Though I am very good at some taxa.
If you’re getting this “you can’t use iNaturalist” from a reviewer on a paper you’re trying to get published, correspond with the editor of the journal. Explain why the reviewer is wrong. Editors know that happens sometimes.
My extreme novice point of view… Even “professionals” can skew their data to whatever suits their goal. That’s why we get “salt is good”, “salt is bad” etc. I honestly see nothing wrong with a study/survey saying this is my data findings based on iNaturalist observations.
There are a huge number of survey methods. You have everything from the extremely detailed and even more time consuming point-transect method to rapid assessments that simply take walking a designated track and quickly making an estimate of the range of %cover of a given species or number of species (0%, 1-5%, 6-20%, 21-50% etc…). Both are valuable in their own way. While P-T is going to give you very detailed and precise information, the survey covers <10% of a given area whereas rapid assessment methods are less detailed, but allow for surveying up to 50% (or potentially even beyond) of a given area.
For me and my application, the latter is a far more useful when it comes to rangeland management. We are talking about management at a landscape scale rather than at an individual scale so the fact we found 10 purple three awn plants in a field of western wheat grass is pretty useless knowledge as opposed to “we have an area that is >50% cheatgrass.”
Also take into account the amount of time and manpower (AKA resources) that something like a point-transect survey requires, things I quite frankly, don’t have at my disposal in any given field season (and I actually have a lot of seasonals at my disposal compared to many other refuge complexes).
In reality, a rapid assessment is basically just “walking around.” Generally, we TRY to maintain a specific “track” and try to maintain those tracks over time to assess changes in cover (i.e. %cover of invasive weed species) but in reality, it is FAR less “rigorous” than much more detailed and intricate surveys. But, it is still considered valuable and usable scientific data, at least by the standards of USFWS field techs (such as myself) and Inventory and Monitoring program who designed the survey. In fact, we are using the data compiled through this method to assess the condition of xeric tallgrass prairie at one of the refuges in our complex.
iNaturalist does lack scientific rigor. I think that’s OK.
I think the best description of iNat lists is that they include the commonly observed (or “glamorous”) species in an area plus any less glamorous or cryptic species that have caught someone’s special interest. In an African university, someone is adding an observation for each insect in the university’s museum collection.
The failing is not in iNaturalist itself but in the way people use and reference iNat data. In your examples, iNat lists are used as an apparently definitive report on those locations as opposed to citizen science observations. In the days of yore, lists of species in an area were compiled by dredging thru scientific papers, reference manuals, and field notes to see what species were recorded there. There might be a field survey to check some of the species and scientist would examine and compare specimen collections, iNat doesn’t currently have that historical depth of data.
It doesn’t have the same rigor of identification either. Some people know their species well and do lots of research, while others just agree based on what their granny called that plant or whether a flower looks the same to them although they don’t know to count petals or look for hairs. Many species can’t be definitively identified to species from casual photos alone. As an example, to differentiate between sea slugs of similar appearance, sometimes you have to count the number of denticles (teeth) on the radula (tongue), which I hope other scuba divers don’t do just for fun, 'cause it will ruin lots of microscopes AND sea slugs . But people ID them in iNat based on outward appearance anyway and I’m OK with that. A scientist who is working with species having overlapping ranges pf appearance should know that and be able to account for it in their use of those observations; peer reviewers should know how to call them out if they don’t.
FYI, when I ID things that I’m not really sure about, I ID them to the next taxa “upstream” because a lot of new users tend to just agree with the first ID they get, without even checking the pictures of that species. (I don’t like the Agree button on Observation lists for that reason.) Agreeing with a genus or family is less troubling to me. I sometimes encounter RG observations that are clearly (to me) wrong. I flag them, then add the closest level I can and explain in comments. I’ve learned when checking photos for an ID that I should view only photos from RG observations, to avoid anything only one person thinks is the species.
But scientists are recording their IDs in iNaturalist. I’ve seen at least one with drawings of body parts in addition to photos. That will improve the quality of data going forward, until the next big taxonomic revision…
I’m ecstatic to have the iNat community to help me identify things I see and to be able to share my joy of nature (and identification) with others. I tend to add odd facts to the comments of interesting species. Nobody has complained to me yet.
. But people ID them in iNat based on outward appearance anyway and I’m OK with that. A scientist who is working with species having overlapping ranges pf appearance should know that and be able to account for it in their use of those observations; peer reviewers should know how to call them out if they don’t.