Corrupt indexing of observer statistics at Southern Africa scale

Bug Reports
1

Platform and Browser

Platform (Android, iOS, Website): Website

Browser: Chrome

Description of the issue
There appears to be a problem with how observer data is indexed or aggregated for the Southern Africa place filter.

This issue relates to the user simonescrase.

When viewing this user’s mammal observations directly, the following is observed:

Total of 98 observations

Covering 36 mammal species

All observations are Research Grade

Most observations have an accuracy value, with only a few lacking accuracy

Link to the user’s mammal observations in Southern Africa:
https://www.inaturalist.org/observations?iconic_taxa=Mammalia&place_id=113055&user_id=simonescrase

Unexpected behaviour
When using Explore → Observers for Southern Africa and filtering by Mammalia, this user does not appear in the observers list, even when scrolling down to the level where 98 observations should place them.

The same applies to the species count view for observers.

Southern Africa observers view (Mammalia):
https://www.inaturalist.org/observations?place_id=113055&view=observers&iconic_taxa=Mammalia

This is unexpected, as 98 observations clearly falls within the top 500 observers, which iNaturalist uses for these rankings.

Expected behaviour
The user simonescrase should appear in the Southern Africa observers list and species count rankings for Mammalia, consistent with their observation totals.

Additional evidence for inconsistency
The same user does appear correctly at other geographic scales using the same filters:

KwaZulu-Natal: Ranked #71
https://www.inaturalist.org/observations?place_id=13313&subview=map&view=observers&iconic_taxa=Mammalia

South Africa: Ranked #353
https://www.inaturalist.org/observations?place_id=6986&subview=map&view=observers&iconic_taxa=Mammalia

Summary
Observer statistics for Mammalia appear to be incorrectly indexed or excluded at the Southern Africa place level, while functioning correctly at both smaller (province) and national scales.

Thank you

2 Likes
2

I’m not sure if this applies here, but when iNat lists the top 500 observers, that list is based on number of observations. Then if you look at who’s seen the most species, that is just generated by resorting that same list based on overall number. So, if you’re not making the top 500 list based on overall observation numbers, you won’t show up on the most species observed list, either, even though some people who posted more observations than you may have seen fewer species.

1 Like
3

This is within the top 500.
It should be around 460 on observations, and around 360 on species.

2 Likes
4

I suspect this isn’t a bug but rather just a side effect of the way we make these counts, which are estimates. So the top 500 is based on estimated counts and it’s probable that simonescrase’s estimated count didn’t make the cut. The listing will fluctuate, especially around the bottom. Since his species count is much higher in those smaller area searches, it’s less likely to be left off the list.

2 Likes