Hello,
I am doing some development to analyze the iNaturalist observation data. I believe the Notes and the Observation fields would be very valuable for my analysis. For example, “Visited Flower Of”, “Host of” etc., would help me to see relationships among organisms. Is it possible to download these fields using the iNaturalist API for the whole global dataset?
Thanks,
Misha
the API isn’t really designed to get more than a few tens of thousands of records. so if by “whole global dataset” you mean all observations with observation fields, i don’t think there’s a good way to get that much data.
if you’re looking for interactions associated with specific taxa, that might make it more feasible (if there are few enough observations) to get that data via API or other means.
some data from iNaturalist does end up in globalbioticinteractions.org, and you can download the data they have over there. that might be your best bet if you’re trying to analyze interactions.
@pisum Thanks. Sorry, I asked the wrong question. What I am really looking for is to download this data as a csv/gz file similar to the other files (observers, taxa, etc) that are available on S3 (https://registry.opendata.aws/inaturalist-open-data/). Is there some way to download this useful meta data from the cloud?
the AWS open data set metadata is limited and does not include observation fields.
as noted, probably the best source for you is GLOBI, but if you don’t like that, you can also look at the file that iNat sends to GLOBI. (i didn’t read all the details about that, but there’s some conversation about that here: https://github.com/globalbioticinteractions/globalbioticinteractions/issues/427 .)
the GBIF DWCA file doesn’t contain obsservation fields as far as i know, but maybe there’s a related file that does. there’s some discussion in the above link that may touch on this, but i’m not sure.
@pisum Thanks so much! This is very useful information. I’m going to take a look at the GLOBI dataset and the file(s) that iNat sends to GLOBI.
If there are specific observation fields you are interested in, you can query them via the URL (e.g. https://www.inaturalist.org/observations?verifiable=any&place_id=any&field:Visited%20flower) and then click on “Filter” and “Download” to download a csv of the data. That might be the easiest option.
If you wanted to download the observation fields for a specific list of observations, you can probably adapt the code at https://github.com/cafundis/inatdatagrabber to do what you need. That code is currently tailored to only download the observation fields of interest to the Fungal Diversity Survey, but you could easily change it to download other observation fields or all the observation fields. As pisum mentioned, this option is only appropriate for relatively small sets of data (>10,000 observations).