Rank filters sometimes show wrong results

Bug Reports
1

On the /observations page, when hrank and lrank are the same rank, it sometimes shows unexpected results, more observations than just those at that rank. This seems to be the case for:

The other ranks seem okay, e.g. order and subgenus.

I know I can change the URL, such as to rank=section to get the results I want, but hrank and lrank are the only publicly available/displayed options in the filters, and maybe this issue could be affecting something other than search results…

Some of these ranks aren’t even available on /identify…though the wrong results sometimes applies there too:
https://www.inaturalist.org/observations/identify?hrank=genushybrid&lrank=genushybrid vs
https://www.inaturalist.org/observations/identify?rank=genushybrid

3 Likes
2

Thanks @bouteloua, I’ll put this on the report for next week.

3

This is an old bug, from when I first joined iNat over a year ago.

I regularly hit my head on this, as for some projects I have to report on the level of identifications - i.e. to family, family to genus, species level and lower. If I include the subgeneric or subspecific ranks the results are garbage.: However, they are predictable - see below:

so - default for all obserrvations of the protea family:
https://www.inaturalist.org/observations?place_id=113055&subview=grid&taxon_id=64517&verifiable=any&view=species.
15,530 OBSERVATIONS 390 SPECIES 233 IDENTIFIERS 477 OBSERVERS
(I have chosen species view, as it allows me to identify the filtered set most easily)

so rank = genus: (this appears to be OK)
https://www.inaturalist.org/observations?hrank=genus&lrank=genus&place_id=113055&subview=grid&taxon_id=64517&verifiable=any&view=species
325 OBSERVATIONS 15 SPECIES 53 IDENTIFIERS 130 OBSERVERS

so rank = section (= total set of main filter)
https://www.inaturalist.org/observations?hrank=section&lrank=section&place_id=113055&subview=grid&taxon_id=64517&verifiable=any&view=species
15,530 OBSERVATIONS 390 SPECIES 233 IDENTIFIERS 477 OBSERVERS

so rank = subspecies (yields: total set of all subspecies, varieties and forms)
https://www.inaturalist.org/observations?hrank=subspecies&lrank=subspecies&place_id=113055&subview=grid&taxon_id=64517&verifiable=any&view=species
1,346 OBSERVATIONS 18 SPECIES 96 IDENTIFIERS 201 OBSERVERS

so rank = varieties (yields: total set of all subspecies, varieties and forms)
https://www.inaturalist.org/observations?hrank=variety&lrank=variety&place_id=113055&subview=grid&taxon_id=64517&verifiable=any&view=species
1,346 OBSERVATIONS 18 SPECIES 96 IDENTIFIERS 201 OBSERVERS

so rank range = subspecies to forms (yields: total set of all subspecies, varieties and forms)
https://www.inaturalist.org/observations?hrank=subspecies&lrank=form&place_id=113055&subview=grid&taxon_id=64517&verifiable=any&view=species
1,346 OBSERVATIONS 18 SPECIES 96 IDENTIFIERS 201 OBSERVERS

BUT rank range hybrid to infrahybrid yields species rank total
https://www.inaturalist.org/observations?hrank=hybrid&lrank=infrahybrid&place_id=113055&subview=grid&taxon_id=64517&verifiable=any&view=species
15,007 OBSERVATIONS 390 SPECIES 223 IDENTIFIERS 447 OBSERVERS
This is the same as the species rank total:
https://www.inaturalist.org/observations?hrank=species&place_id=113055&subview=grid&taxon_id=64517&verifiable=any&view=species
15,007 OBSERVATIONS 390 SPECIES 223 IDENTIFIERS 447 OBSERVERS
((hybrids only = 49 OBSERVATIONS 21 SPECIES 17 IDENTIFIERS 25 OBSERVERS))

So:

  • levels between genus and species do not work at all (i.e the filter fails and returns all data) < but subgenus might work - yields nothing which might be correct for Proteaceae? >
  • levels below species (*exc. hybrids) yield the total for all subspecific taxa, irrespective of which rank is requested.
  • levels including hybrids yield the species total
1 Like
4

This should be fixed, or at least much better now. Let us know how it’s looking.

1 Like
5

Infraclass, subterclass, section, and subsection seem okay now, but genushybrid, hybrid, subspecies, variety, form, and infrahybrid are still giving me the wrong results (check links in first post).

2 Likes
6

Anything below species rank doesn’t show up when searching for plants.

Also; https://www.inaturalist.org/observations?hrank=subspecies&lrank=subspecies&place_id=129265&subview=grid&taxon_id=51815&view=species

Shouldn’t this only show subspecies?

1 Like
7

@tonyrebelo recently reported this bug as well. https://forum.inaturalist.org/t/cannot-filter-hybrids/19828/

9

subspecies seemed to work (at least partly).

Do we have any idea on when this and the Explore page are to be redesigned?

10

this problem still exists unfortunately; I’ve been trying to filter to hybrid-hybrid and cannot get it to work, it just treats it as if I was using species-species

1 Like
11

As still present and continuing now, nearly a year later. Hybrids need to work with their priority level and filter properly.

12

This bug continues! @tiwane can you get a github issue made? It’s also worth noting that this bug also affects the identify modal along with explore.

Filters do not work for

Appears to work for subspecies however: https://www.inaturalist.org/observations?hrank=subspecies&lrank=subspecies&place_id=any&subview=map