Number of observations seems to not be accurate upon search for any organism between pages (PC)

Problem is present in-browser on a desktop PC running Windows 10, currently in firefox browser.

Description of problem When searching for an organism on explore, the number of observations is lower than the number of observations supposedly existing under the listing for the species. In the case of the Black Caterpillar Hunter (Calosoma sayi), total observations is listed as 640 on the species page image yet upon searching with grid, this number is magically 739

, and then it’s 739 on the entire map , but if you search with a bounding box over the area it’s observed, it’s somehow 791 image . This can be replicated with many listings. Tribe Syrphini states 21,338 observed: https://www.inaturalist.org/taxa/121991-Syrphini upon grid search this is 208,779 observed. image Upon bounding box it’s 208,046. image

Step 1: Choose listing for an organism

Step 2: Look at observed on it’s listing page

Step 3: Look at observed when searched on grid, then map

Step 4: Look at observed with a bounding box over the entire area the organism is present on the map (easiest by zooming all the way out and then selecting a bounding box).

If this is not a bug I would like an explanation for why it does happen because it does not make any sense. Is this a discrepancy of research grade vs casual? If so why isn’t the default for research grade chosen first? Why does bounding box break the numbers?

1 Like

on the taxon page, you probably have the filter in the upper-right corner set to the United States. so it’s probably providing you counts only for the US. if you clear that filter, then you should see a count for everything.

on the Explore page, even if i zoom to the lowest zoom level available, i don’t see the whole world. parts of the extreme north or extreme south will be cut off. so if you limit observations by a bounding box, you’ll cut off observations that are at the extreme north or south, and also you’ll potentially filter out observations that aren’t mappable.

1 Like

Why isn’t the filter the same for every page? And no, the listing’s I’ve input don’t have ranges into the extreme north and south. There isn’t any issue there.

the pages are different. if you click the View All button under the count on the Taxon screen, it should take you to the Explore screen, with the same location applied as the filter.

for example, observations with private locations: https://www.inaturalist.org/observations?place_id=any&taxon_id=244519&geoprivacy=private.

1 Like

And there is no way to set a standard search-by-default for your account?

in your account settings, you can set a default place to search, and that will get applied to the Explore page. is that what you’re looking for?

No, as in a way to enable only one way of displaying numbers of animals (all or one) not pick and choose. It should be set up by default to display all observations, or just research grade. Not both.

i still don’t understand what you’re talking about here. which page(s) are you talking about? what does numbers of animals mean? (is that number of observations or number of species or something else?) what does “both” mean?

As in, why does it not allow the setting of a default “set” (research grade/private location on or off/both research grade and "unverified) as a default parameter when displaying the numbers of a species? It being different on the page for an organism vs bounding boxes vs basic search does not make any sense.

i still don’t understand what you’re really asking for, but i’ll offer a couple of thoughts before leaving this discussion:

  1. if you want to change the way the taxon screen can be filtered, you might want to contribute some thoughts to this feature request (https://forum.inaturalist.org/t/add-filter-panel-to-taxon-pages/4087)
  2. some observations simply cannot be represented properly on a map. for example, if an observation’s geoprivacy is set to private, its coordinates will not be available to the general public. this means such an observation will be excluded if you do any sort of location filtering, such as filtering within a bounding box.

Not going to explain further on my main point, but secondarily I’m saying that having Private location observations shown at all by default is useless in many cases, especially when you’re (generally) trying to see where these animals are. Obscured location is even fine for this. Is there a way to by default disable private listings from even being included in the numbers?

It looks like this is working as designed; if you have a request to change the functionality, check through existing #feature-requests and then post it as a new topic if it hasn’t already been requested. Thanks!