Viewing observations with large accuracy values

[DT_Almquist]

I found a record in GBIF that originated in iNaturalist but was not showing up on the map when filtered to a particular speceis.
https://www.inaturalist.org/observations?subview=map&taxon_id=230339
Based upon this
https://forum.inaturalist.org/t/observations-lack-pin-in-map-view/7815
it appears that that is due to this observation
https://www.inaturalist.org/observations/122752441
having a large accuracy value. Is there any way within iNaturalist to have records like this show up in the map or is there maybe some sneaky workaround to find them? Also, what is the accuracy value cutoff? I realize that this may seem esoteric but if this record is anywhere near the centroid, it is a very valuable one as there are no other records within more than 50km.

[SQFP]

I still have (some) observations with accuracy>30km showing on the map as tiny squares, eg. Observations · iNaturalist

[cthawley]

I don’t think there’s anyway to have them show up in iNat itself (but am not 100% sure). You can download the data and use mapping software to see it, or GBIF may show them, not sure.

Info on the cutoff is here:
https://forum.inaturalist.org/t/research-grade-observation-not-showing-on-map/13608/5

[david99]

The bigger the area of uncertainty, the less likely that the observation was even seen anywhere within the circle. It’s not unusual for me to find observations with a 30-50km radius of uncertainty, and then when asked the observer lists a location that’s outside of even that large circle.

[pisum]

@DT_Almquist

what is the accuracy value cutoff?

• https://github.com/inaturalist/inaturalist/blob/main/app/models/observation.rb#L3068-L3080
• https://github.com/inaturalist/inaturalist/blob/main/app/models/observation.rb#L2518-L2531

@pisum

iNat won’t map anything that exceeds the diagonal of a bounding box of 0.2 deg lat and 0.2 deg long, which is roughly 30km, depending where you are in the world.

[DT_Almquist]

I understand that but if I don’t have a way to find the record then I can’t even ask the observer.

[DT_Almquist]

Thanks and interesting and corresponds with taxon geoprivacy spp getting a default ca. 29km accuracy value.

[DT_Almquist]

Seems odd to not be able to see iNaturalist records within iNaturalist. I sortof get why, but still weird that I can only easily see some in GBIF and there are records that don’t even get migrated to GBIF.

[pisum]

@DT_Almquist

corresponds with taxon geoprivacy spp getting a default ca. 29km accuracy value

yup. it’s the same calculation, since the obscuration box is also 0.2 lat by 0.2 long. conceptually, that accuracy value represents the maximum radius of a circle that encompasses the true location if the obscured location is in one corner of the box and the true location is in the opposite corner.

[thebeachcomber]

@DT_Almquist

Seems odd to not be able to see iNaturalist records within iNaturalist.

you can still see them if you use Identify

@tiwane

In Identify, mappable is set to any by default, but can be changed in the URL:

• &mappable=true to find obs with public lat/lon and location accuracy below threshold
• &mappable=false to find obs with either no public lat/lon or large location accuracy (or failing quality checks)
• &mappable=false&geo=true to find obs with large location accuracy (see previous section for details on geo)

[DT_Almquist]

@tiwane

&mappable=false&geo=true

I’m sorry; can you explain? I tried this and was still not able to see that record.
https://www.inaturalist.org/observations/identify?taxon_id=230339&mappable=false

[thebeachcomber]

@tiwane

&mappable=false&geo=true

https://www.inaturalist.org/observations/identify?quality_grade=needs_id%2Cresearch&place_id=21&taxon_id=211704&mappable=false&geo=true

[DT_Almquist]

Is there a way to find all records for a species regardless of accuracy? That link only brings up that one record whereas I’d like to be able to search for all records for a species and see them all on a map.

[pisum]

@DT_Almquist

I’d like to be able to search for all records for a species and see them all on a map.

if your set contains mappable=false records, they will never show up in the standard iNat mapping interface. the only reliable to map mappable=false records is to get their coordinates and map those as individual points (or to do your own other custom visualization using those individual points as a starting point).