The hopstpot view for users is really great:
It would be nice if we could use the hotspot view for species (or levels above) as well.
When I try replacing portions of the url with species ID #s instead it defaults to the blocky individual observation view rather than a heatmap view.
I’ve figured out how to do it for a project:
But when I try to use a species name or number it either gives an error (using the name), or defaults to the cell-block view if using the taxon ID:
Is there a way to get the heatmap view for taxon?
Or is this something that needs to be in a feature request instead?
Not that I know of, but if you know about a county and state heat map like ebird for Inat please tell me.
You can look at any place iNat has via the heatmap, but it’s looking at all observations, not ones of specific species.
“place_id=57241” is the identifier for Mainland Southeast Asia, replace that with whatever the place ID is for your area of interest.
the choice to display the “grid” style map rather than the “heatmap” style seems to be on purpose.
you can always create your own heatmap (see: https://forum.inaturalist.org/t/in-pursuit-of-mappiness-part-1/21864), or you’re welcome to use my map: https://jumear.github.io/stirfry/iNat_map?taxon_id=47532&view=heatmap&defaultzoom=3¢erlat=36.424¢erlng=62.104.
do you mean a choropleth? if so, you could look at https://forum.inaturalist.org/t/some-choropleth-maps-using-inat-data/21776. that thread contains links to an R script and also an Observable notebook that can help you generate state-level choropleths of species count. if you wanted to count observations instead, you could easily adapt either to get data from the API via /v1/observations instead of /v1/observations/species_counts. the necessary changes to generate county-level choropleths are a little more complicated, but if you’re interested, i can explain how to adapt either of these things.
It doesn’t seem to work so well at the county level. Maybe it’s intended for larger landscape levels? Also, it would be great if there were a way to adjust the assignment of colors to data values – in some places/scales there are so many observations that the entire map is red. If I zoom out so that the heat map is displayed for my county, the entire county is pretty much red. If I zoom in to see detail, it leaves the heat map view and instead shows individual observations. And at finer scales, the 10km box for obscuring locations really messes things up. All this is visible here:
it’s relatively easy to create your own map using the heatmap tiles (as noted above), and if you do so, you’ll be able to make the heatmap as transparent as you like. for example:
you could also hue-shift the colors of the heatmap or do any other manipulation you like.
i looked at this code again, and it looks like you can add a
color=heatmap parameter to force a heatmap style. for example:
Very cool. Thanks for figuring that out.
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.