Largest genera with every species covered by CV?

General
21

CV has definitely misID’d some of my Larus observations

1 Like
22

CV is not going to be perfect for genera with frequently observed hybrids (like Larus) because those aren’t in the CV. Larus glaucescens × occidentalis (Olympic Gull), for instance, has ~15,000 observations on the west coast of N. America, but will never be suggested by the CV (afaik).

Whether or not the CV would overall be more or less accurate if hybrids were included in the CV might depend on the genus in question. I feel like I read somewhere on the forum that hybrids were excluded from the CV because it was oversuggesting less common bird hybrids (some Anas hybrid I think?).

1 Like
23

The observation I had in mind does not seem to be a case where the organism is a hybrid https://www.inaturalist.org/observations/339083299

24

If the “expected_nearby=true” trick still works, there are 316 species of Euphorbia included in the model: https://www.inaturalist.org/observations?expected_nearby=true&taxon_id=51822&view=species

2 Likes
25

@quyksilver I think I just had hybrids on the mind, but you’re right, Larus seems like a tricky genus in general for the CV, hybrids or no.

To get the ball rolling on insects, Leucorrhinia is a dragonfly genus with all 14 species in the CV.

2 Likes
26

I’m not sure it’s reliable. I’ve found that it reduces the number of observations for many species in the model. Especially for the common pigeon (Columbia livia). I think this might be because it omits the vast majority of observations whose observation taxon is a subspecies/variety/form/infrahybrid. All except for 21.

1 Like
27

Well, if you only care about the first 1000 and last 1000 alphabetically, there’s https://www.inaturalist.org/geo_model.

You can see all Acacia and Yucca species in the model. But most genera are too far middle in the alphabet.

You can see the 1000 with the smallest threshold with https://www.inaturalist.org/geo_model.json?order_by=threshold.

1 Like
28

For plants native to North America, the genus Lomatium is probably one of the toughest because it is speciose and many species are uncommon. I can’t speak to other U.S. states, but in Washington I count that 28 of 37 species are in the CV model. A few of the remaining ones are very rare, but the CV should get up to 33 of 37 or so this year.

1 Like