I’d like to demonstrate changes in habitat in a specific park over time. For example, if there were 55 species observed one year and 65 observed the next, then perhaps we could say that the park’s habitat is improving in biodiversity. Whether it’s a “valid” claim is an important question. But I can’t even figure out how to gather the data without identifying multiple time periods, extracting the data, and manipulating in Excel. And that seems too cumbersome.
This line reminded me of something that Barbara said:
3 Likes
I don’t think that there’s a way to do this within iNat itself - iNat doesn’t really have tools for data analysis, though it does have some data visualization options (none of which meet your needs). Downloading the data and manipulating in your program of choice will be your best avenue I think.
iNat’s option’s for downloading data are:
https://help.inaturalist.org/en/support/solutions/articles/151000170342-how-can-i-download-data-from-inaturalist-
is it too cumbersome because you don’t know how to do it? or is it too cumbersome because it’s not worth your time to do it?
if the latter:
- how long do you think it will take you to get this information?
- what sort of process do you envision where you wouldn’t have to identify your (multiple) time periods at some point in your process?
Sorry that this isn’t the answer you’re looking for, but as someone who works in this field (tracking plant community behavior over time), I would emphatically encourage against using iNaturalist data in your effort to show any sort of change in biodiversity and/or community health.
Looking at a rate of change in species richness (diversity), composition, density, etc. over time require a lot of controlled variables to definitively prove. And sure, once you’ve proven there has been a change, interpreting the change is a whole other ballpark.
The data one can mine from iNat is much too broad and imprecise for this purpose.
That being said, I pull data from iNat frequently and it’s honestly much easier than it seems! I would experiment with generating spreadsheets and sorting the columns you find useful. Any part of a spreadsheet can be turned into a CSV, just highlight which part you want and export.
4 Likes
It is too cumbersome.
This statistic (# of species over time) should be easy to pull out of a database like this. We can discuss the validity of interpreting such a statistic. But as a data person who has used other databases, it seems like a slam dunk.
I’ll pull the data out, dump it to Excel, and then manipulate. It’ll be one step better than copying the data by hand.
very few systems allow you to perform just any query directly from their underlying database. so you work with what you have available.
you have the option of either:
- exporting a set of observations and then aggregating to species / time period, or
- getting a count of species for a set of time periods (or something like that).
for the latter, there are many manual and automated ways to do this, but since you like using Excel, you could do some variant of the process described here: https://forum.inaturalist.org/t/using-excel-api/22378#step-6-optional-count-of-identifications-for-any-given-date-range-17.
it sounds like you just need to familiarize yourself with how to get data from the system. it’s really not that hard once you learn how it works.
1 Like
Since iNat data is generally accessed via GBIF, it is essentially the same level of work to access it as any other type of biodiversity data commonly used. It can be pulled via “manual” (web-based interface) download from iNat or GBIF, API calls however someone wants, Python package, R package, etc. If using iNat data is
then using pretty much any biodiversity data is.