Stream: xarray

Topic: ✔ landunit to lat/lon

view this post on Zulip Adrianna Foster (Nov 15 2023 at 20:37):

Hi,

I am trying to analyze some data in /glade/campaign/cgd/tss/people/oleson/CLM_LAND_ONLY_RELEASE/CLM5/clm50_r270_1deg_GSWP3V1_iso_newpopd_hist/lnd/proc/tseries/month_1, and the history variables are indexed by landunit. I'm not sure how to work with this data using typical xarray plotting. For example, a command that @Will Wieder uses:

cf = ax.pcolormesh(clm_sub.lon, clm_sub.lat,
                   clm_sub.H2OSNO.isel(time=slice(0,120)).max(dim='time'),
                   transform=ccrs.PlateCarree());

gives this error:

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
/glade/derecho/scratch/afoster/tmp/ipykernel_31860/3617445073.py in <module>
     20 ax.add_patch(poly)
     21
---> 22 cf = ax.pcolormesh(clm_sub.lon, clm_sub.lat,
     23                    clm_sub.H2OSNO.isel(time=slice(0,120)).max(dim='time'),
     24                    transform=ccrs.PlateCarree());

/glade/work/afoster/miniconda3/envs/analysis/lib/python3.8/site-packages/cartopy/mpl/geoaxes.py in wrapper(self, *args, **kwargs)
    316
    317         kwargs['transform'] = transform
--> 318         return func(self, *args, **kwargs)
    319     return wrapper
    320

/glade/work/afoster/miniconda3/envs/analysis/lib/python3.8/site-packages/cartopy/mpl/geoaxes.py in pcolormesh(self, *args, **kwargs)
   1794         # Add in an argument checker to handle Matplotlib's potential
   1795         # interpolation when coordinate wraps are involved
-> 1796         args, kwargs = self._wrap_args(*args, **kwargs)
   1797         result = matplotlib.axes.Axes.pcolormesh(self, *args, **kwargs)
   1798         # Wrap the quadrilaterals if necessary

/glade/work/afoster/miniconda3/envs/analysis/lib/python3.8/site-packages/cartopy/mpl/geoaxes.py in _wrap_args(self, *args, **kwargs)
   1823         X = np.asanyarray(args[0])
   1824         Y = np.asanyarray(args[1])
-> 1825         nrows, ncols = np.asanyarray(args[2]).shape
   1826         Nx = X.shape[-1]
   1827         Ny = Y.shape[0]

ValueError: not enough values to unpack (expected 2, got 1)

I'm guessing because the function expects the data to be a 2-d array. But for this it's a 1D array indexed by landunit (n=98467)

view this post on Zulip Notification Bot (Nov 15 2023 at 21:30):

Adrianna Foster has marked this topic as resolved.

view this post on Zulip Katie Dagon (Nov 16 2023 at 21:15):

Hi @Adrianna Foster it looks like you marked this as resolved, but if others are wondering about how to analyze 1D output from CESM there are a few resources I can point to. We have an ESDS blog post on using CAM-SE output which is indexed by ncol instead of lat/lon. There is another blog post on analyzing CLM output indexed by pft. I'm not sure we have an example for data indexed by landunit but I would imagine the remapping would be similar to these other cases. If you do find that the solution or analysis is different, this would be a useful contribution!

view this post on Zulip Adrianna Foster (Nov 27 2023 at 21:56):

Thank you! I ended up using a different dataset that was indexed by lat/lon instead of landunit but this blog post was helpful!

Last updated: May 16 2025 at 17:14 UTC