Get all HUC4 for CONUS as GeoDataFrame

[rsignell-usgs]

Struggling to figure out how I can get a geopandas dataframe of all the huc4 for CONUS.

I see:

from pygeohydro import WBD

wbd = WBD("huc4")
hudson = wbd.byids("huc4", ["0202", "0203"])

Would I need to get the geometry of CONUS and request .bygeom ?

[cheginit]

Yes, the best way is to pass the bounds and use .bygeom. For example, for CONUS you can do:

wbd = WBD("huc4")
conus_bound = (-124.848974, 24.396308, -66.885444, 49.384479)
huc4 = wbd.bygeom(conus_bound)

[cheginit]

FYI, you can get US geometries using pygeohydro:

import pygeohydro as gh

conus = gh.helpers.get_us_states("contiguous")

You also can pass any US state code: TX, CA, ...

[cheginit]

I have written a little function for my personal use that gets the full HUC geometries. Since it's useful at least for one person, I just added it to PygeoHydro! You can reinstall from the repo to use it. It's called huc_wb_full. You can find the function here

[rsignell-usgs]

@cheginit , thanks so much!

Although the response back sounded like failure:

from pygeohydro import WBD
import pygeohydro as gh
conus = gh.helpers.get_us_states("contiguous")
bbox = tuple(conus.total_bounds)
wbd = WBD("huc2")
dhuc2 = wbd.bygeom(bbox)

Found 1 failed requests. Retrying ...
Total of 18 out of 18 requested features are not available in the dataset. Returning the successfully retrieved features. The failed object IDs have been saved in the file cache/failed_ids_7c0f214ff8bd460fb6dc8142fd010047.txt. The service returned the following error message for the failed requests:
 Error performing query operation
CPU times: user 11 s, sys: 541 ms, total: 11.6 s
Wall time: 1min 46s

It would seem like it worked:

[cheginit]

Right, the warning message was wrong. I just pushed a fix to pygeoogc. Thanks!

[rsignell-usgs]

@cheginit , this is great! I used it in this notebook!
https://nbviewer.org/gist/85e37c7ef05c4a51ccec62612b22272c

[cheginit]

Cool! That's a useful notebook, thanks for sharing.

BTW, you missed bbox = tuple(conus.total_bounds) in the fourth cell.

[rsignell-usgs]

Grrr, thanks! Fixed in https://nbviewer.org/gist/32bc2c9ee50dd26335de2e54663b6531