I’m currently working on a project using iNaturalist data, including both obscured and unobscured observations. Because the pseudo-coordinates of obscured observations are misleadingly precise, I was hoping to aggregate obscured observations to the level of the 0.2º x 0.2º obscuration cells and create a choropleth representing observation density by species. Is there a shapefile representing these cells (particularly for California) or a way to query this data?
Thanks!
1 Like
How are the obscured coordinates misleadingly precise? Their precision shows correctly in GBIF for me.
On the broader question, I don’t know of any existing shapefile for the obscuration cells. You could just create a raster file with 0.2 degree cells though and then sum all the occurrences in each cell - that might do more or less what you’re looking for.
1 Like
you can create your own grid. it should be relatively straightforward to accomplish in a GIS tool or programmatically.
not sure what you’re trying to accomplish exactly. sure, it’s possible to query such data, but there may or may not be an efficient way to do this, depending on what you’re trying to do.
1 Like
By that, I don’t mean there is any missing or incorrect information. My main point is that visually, displaying the private coordinates as the actually location of each observation presents a misleading degree of precision, since the true location of an obscured observation is equally likely to be anywhere within the cell. This is can be seen even in GBIF, as for Calochortus umbellatus. Although areas with high observation density (in this case, the west side of the bay) will show the limits of the obscuration cell, in areas with lower observation density (east bay), it is harder to determine visually whether these are true locations or obscured ones unless you know the taxon or observation geoprivacy. What I’d like to map out is similar to the GBIF display but with the obscuration cells as the spatial scale of aggregation instead of the hexagons.
I can produce the 0.2º x 0.2º grid using GIS software and do an overlay, which shouldn’t be an issue. As one final thing, I wanted to check if there are any offsets or adjustments to the iNat grid that I’m not aware of. I assume that the iNat grid is perfectly aligned with the coordinate system (cell boundaries fall precisely on every 0.2º interval globally), which seems consistent with some observations I’ve looked at, but not sure if there’s anything I’m missing.
1 Like
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.