Identify best approach for incorporating protected lands data

[jcoliver]

A few questions/challenges to be addressed:

• Which protected lands categories are we interested in?
• How will we store protected lands data (may be too large for GitHub)?
• How do we decide future areas to consider (minimum number of species)?

[ezylstra]

Worked on this a bunch in several commits, but mostly this one: 9df52c7. Tough to get PA data for all of North America (and not just one country/state). Best option looked like the data set from CEC (which is based on the PAD database from USGS and similar datasets in Mexico, Canada). I used all the areas that were classified as PAs by the IUCN (categories Ia/Ib - VI), which correspond to GAP status codes 1 and 2. It's a massive dataset, that you can visualize by clicking on the "Map Viewer" button on the CEC website above.

Not sure about storing this dataset. I have the original shapefile zipped on our GoogleDrive folder. For now, there's a section at the top of the protected-areas.R script where you take that unzipped shapefile and process it a bit. I then saved the new shapefile on my computer since it's too big for GitHub. We'll need to figure out where to store this moving forward.

The rest of the protected-areas.R script, calculates the proportion of each insects' distribution (for each climate scenario and type of distribution [overlap with at least one host plant or areas that are climatically suitable for the insect, regardless of host plant suitabilities]) that falls in a protected area. This took a bunch of fussing as R kept crashing when I tried to crop spatial data and/or do extraction calculations with terra. I got things to work eventually with the exactextractr package, though it does take a long time to run (few hours for all species, distributions). The script produces a csv with summary statistics (output/summary-stats/protected-areas-allspp.csv), which is under version control so you should be able to look at it. You'll see that the proportions of areas protected are pretty low, but I think that's because of the fairly strict IUCN categories.

[jcoliver]

For storage, three options (listed in relative order of preference):

1. ReDATA. Creates DOI and API allows fairly easy retrieval of compressed files.
2. Zenodo. Creates DOI, maybe be relatively easy to use API.
3. Google Drive. Least amount of metadata & curation, but technically not an "archive".

@jcoliver will look at ReDATA submission as option.

[jcoliver]

@ezylstra For exploring how the proportion of range in protected areas changes, there is a start at src/summary/plot-protected-proportion.R.

[ezylstra]

• I edited the script that Jeff started to plot changes in protected area coverage over time in beccac4.
• I also added a new script that will calculate protected area coverage for swallowtail diversity "hotspots" (with options to define hotspots by varying the minimum number of species present) in 1d2e14e.
• Finally, I downloaded the shapefile with "other" protected areas that don't have an IUCN classification and generally fall into GAP status 3 or 4 if in the US. This shapefile is even bigger than the file with IUCN-classified areas, so I haven't done anything more with this dataset.

[jcoliver]

With categorization of areas into four groups (National, State, Local, and Private), the current version of the shapefile lives on Google Drive. ReDATA is probably the the way to go for longer term storage.