AttributeError: vertical attribute is not available. #3160

tik619 · 2023-08-20T12:39:18Z

tik619
Aug 20, 2023

I am trying to calculate potential vorticity at isentropic level, I am successful in calculating the potential vorticity at pressure level, but as I am trying to interpolate it to isentropic level I am getting the error as :

Input In [12] in <cell line: 2>
    isent_data = mpcalc.isentropic_interpolation_as_dataset(
  File ~/miniconda3/lib/python3.8/site-packages/metpy/calc/thermo.py:2767 in isentropic_interpolation_as_dataset
    all_args[0].metpy.vertical,
  File ~/miniconda3/lib/python3.8/site-packages/metpy/xarray.py:486 in vertical
    return self._axis('vertical')
  File ~/miniconda3/lib/python3.8/site-packages/metpy/xarray.py:418 in _axis
    raise AttributeError(axis + ' attribute is not available.')
AttributeError: vertical attribute is not available.

I am also giving the code which I am using for the calculation:

import cartopy.crs as ccrs
import cartopy.feature as cfeature
import matplotlib.pyplot as plt
import metpy.calc as mpcalc
from metpy.units import units
import numpy as np
import xarray as xr
#############################
data1 = xr.open_dataset("/mnt/h/MPAS/HICS/VAR/ERA/AMPHAN/AMPHAN.uzonal.nc").metpy.parse_cf()
data2 = xr.open_dataset("/mnt/h/MPAS/HICS/VAR/ERA/AMPHAN/AMPHAN.umeridional.nc").metpy.parse_cf()
data3 = xr.open_dataset("/mnt/h/MPAS/HICS/VAR/ERA/AMPHAN/AMPHAN.temperature.nc").metpy.parse_cf()
data1 = data1.metpy.parse_cf('u',
                                  coordinates={'T': 'time', 'Z': 'level',
                                               'Y': 'lat', 'X': 'lon'})
data2 = data2.metpy.parse_cf('v',
                                  coordinates={'T': 'time', 'Z': 'level',
                                               'Y': 'lat', 'X': 'lon'})
data3 = data3.metpy.parse_cf('atmp',
                                  coordinates={'T': 'time', 'Z': 'level',
                                               'Y': 'lat', 'X': 'lon'})
############################################
lats = data1.lat.values
lons = data1.lon.values
#########################################
pres = data1['level'].values[:] * units('Pa')
##########################################
tmpk_var = data3.isel(time=10).squeeze()
tmpk = mpcalc.smooth_n_point(tmpk_var, 9, 2)
thta = mpcalc.potential_temperature(pres[:, None, None], tmpk)
############################################
uwnd_var = data1.isel(time=10).squeeze()
vwnd_var = data2.isel(time=10).squeeze()
uwnd = mpcalc.smooth_n_point(uwnd_var, 9, 2)
vwnd = mpcalc.smooth_n_point(vwnd_var, 9, 2)
######Compute dx and dy spacing for use in vorticity calculation################
dx, dy = mpcalc.lat_lon_grid_deltas(lons, lats)
###############Comput the PV on all isobaric surfaces##################
pv = mpcalc.potential_vorticity_baroclinic(thta, pres[:, None, None], uwnd, vwnd,
                                           dx[None, :, :], dy[None, :, :],
                                           lats[None, :, None] * units('degrees'))
isentlevs = [320.] * units.kelvin
isent_data = mpcalc.isentropic_interpolation_as_dataset(
    isentlevs,
    tmpk,
    pv,
    )

Answered by dcamron

Sep 12, 2023

Hi! The core of the issue is that your vertical coordinate (here lev and level in some of your examples) is not compliant with CF conventions. MetPy tries to understand relevant metadata and even match common names, but we your coordinates are lacking any metadata and we don't currently check for "lev" or "level" (@jthielen should we?).

You were on the right track by specifying parse_cf(coordinates = {'Z': 'level'}) in your earlier example, but the MetPy internal coordinate identifier is 'vertical', not 'Z' (an alias we could consider adding.) You are also missing unit information. The following should work with your data with those two fixes:

import xarray as xr
import metpy.calc as mpcalc

View full answer

dopplershift · 2023-08-20T21:50:18Z

dopplershift
Aug 20, 2023
Maintainer

That error indicates that MetPy's xarray machinery isn't able to discern what the vertical coordinate/dimension is from the available metadata on the input data. I don't have access to the data, so I'm not sure what that problem is exactly. Could you share the output of print(data1.level)?

A work-around is to skip the xarray-aware isentropical inteprolation and just use isentropic_interpolation directly.

0 replies

tik619 · 2023-08-21T03:40:15Z

tik619
Aug 21, 2023
Author

Thanks for the solution. Let me try, and I see. Also, I have posted below the output of `print(data1.level)`: ``` <xarray.DataArray 'level' (level: 6)> array([200, 250, 500, 700, 850, 925]) Coordinates: * level (level) int64 200 250 500 700 850 925 metpy_crs object Projection: latitude_longitude Attributes: _metpy_axis: Z ```

…

On Mon, Aug 21, 2023 at 3:20 AM Ryan May ***@***.***> wrote: That error indicates that MetPy's xarray machinery isn't able to discern what the vertical coordinate/dimension is from the available metadata on the input data. I don't have access to the data, so I'm not sure what that problem is exactly. Could you share the output of print(data1.level)? A work-around is to skip the xarray-aware isentropical inteprolation and just use isentropic_interpolation <https://unidata.github.io/MetPy/latest/api/generated/metpy.calc.isentropic_interpolation.html> directly. — Reply to this email directly, view it on GitHub <#3160 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ASYYLPK65S3VVJMBXLKR6VLXWKBCLANCNFSM6AAAAAA3XIFD5Q> . You are receiving this because you authored the thread.Message ID: ***@***.***>

1 reply

dopplershift Aug 21, 2023
Maintainer

What about print(tmpk)?

tik619 · 2023-08-22T01:46:15Z

tik619
Aug 22, 2023
Author

tmpk is 4 dimensional having dimensions time x level x lat x lon (49,6,300,350)

…

On Tue, 22 Aug, 2023, 5:11 am Ryan May, ***@***.***> wrote: What about print(tmpk)? — Reply to this email directly, view it on GitHub <#3160 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ASYYLPPTYG5V4ZSMVFUSHJTXWPW3TANCNFSM6AAAAAA3XIFD5Q> . You are receiving this because you authored the thread.Message ID: ***@***.***>

1 reply

dopplershift Aug 22, 2023
Maintainer

yes, but what I'm trying to figure out is:

Is it a DataArray?
What are its coordinates?

Alternatively, can you share the data somewhere?

tik619 · 2023-08-22T16:42:41Z

tik619
Aug 22, 2023
Author

Data. Please go to this link for the data.

2 replies

dopplershift Aug 22, 2023
Maintainer

Those data seem different than what's expected by the code (different variable and coordinate names). Can you post different data or share code for that data that's failing in the same way?

tik619 Aug 28, 2023
Author

Sorry for the late reply. Actually, I tried to do the interpolation with some other data, but I got the same error. Below I am posting the code that I used to do the interpolation along with the link to the data.

import numpy as np
import xarray as xr
from metpy.cbook import get_test_data
import metpy.calc as mpcalc;from metpy.units import units
########################################################
d1 = xr.open_dataset('/run/media/maslab-20/DATA/MPAS/ERA/AMPHAN.air_temp.nc')
d2 = xr.open_dataset('/run/media/maslab-20/DATA/MPAS/AMPHAN.pot_vort.nc')
pv_list = []
for t in range(len(d1.time)):
    d3 = d1['air_temp'].isel(time=t)
    d4 = d2['pv'].isel(time=t)
    isentlevs = [320.] * units.kelvin
    isent_data = mpcalc.isentropic_interpolation_as_dataset(
        isentlevs,
        d3,
        d4,
        ) 
    pv_list.append(isent_data)
# Convert the list to a numpy array
var_data = xr.concat(pv_list, dim='time')
var_data1 = var_data.squeeze(dim='isentropic_level')
data = var_data1.pv.values
dr = xr.DataArray(data, coords=[('time', d1.time.values), ('lat', d1.latitude.values), ('lon', d1.longitude.values), ])
dr.name = 'pv'
dr.to_netcdf('/mnt/h/ERA5_DATA/MULTI_LEVEL/pv.isen.nc', encoding={'time': {
    'units': 'days since 1800-01-01 00:00:00',
    'dtype': 'float'
    }})

And the link to data: data

dcamron · 2023-09-12T22:37:00Z

dcamron
Sep 12, 2023
Maintainer

Hi! The core of the issue is that your vertical coordinate (here lev and level in some of your examples) is not compliant with CF conventions. MetPy tries to understand relevant metadata and even match common names, but we your coordinates are lacking any metadata and we don't currently check for "lev" or "level" (@jthielen should we?).

You were on the right track by specifying parse_cf(coordinates = {'Z': 'level'}) in your earlier example, but the MetPy internal coordinate identifier is 'vertical', not 'Z' (an alias we could consider adding.) You are also missing unit information. The following should work with your data with those two fixes:

import xarray as xr
import metpy.calc as mpcalc
from metpy.units import units

air_temp = xr.open_dataarray("AMPHAN.air_temp.nc")
potential_vorticity = xr.open_dataarray("AMPHAN.pot_vort.nc")

air_temp.metpy.assign_coordinates({"vertical": "lev"})
air_temp["lev"].attrs["units"] = "hPa"

potential_vorticity.metpy.assign_coordinates({"vertical": "lev"})
potential_vorticity["lev"].attrs["units"] = "hPa"

isentlevs = [320.] * units.kelvin

mpcalc.isentropic_interpolation_as_dataset(isentlevs, d1, d2)

1 reply

dcamron Sep 13, 2023
Maintainer

In fact, you can get by with only assigning units here, and we'll assume that the 1-d coordinate with units of [pressure] is your vertical coordinate:

import xarray as xr
import metpy.calc as mpcalc
from metpy.units import units

air_temp = xr.open_dataarray("AMPHAN.air_temp.nc")
potential_vorticity = xr.open_dataarray("AMPHAN.pot_vort.nc")

air_temp["lev"].attrs["units"] = "hPa"
potential_vorticity["lev"].attrs["units"] = "hPa"

isentlevs = [320.] * units.kelvin

mpcalc.isentropic_interpolation_as_dataset(isentlevs, d1, d2)

tik619 · 2023-09-14T06:47:41Z

tik619
Sep 14, 2023
Author

Yes, the solution worked. Thanks *Debashis Paul*

0 replies

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

AttributeError: vertical attribute is not available. #3160

{{title}}

{{editor}}'s edit

{{editor}}'s edit

Replies: 6 comments 5 replies

{{title}}

{{title}}

{{editor}}'s edit

{{editor}}'s edit

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{editor}}'s edit

{{editor}}'s edit

{{title}}

{{title}}

{{title}}

{{editor}}'s edit

{{editor}}'s edit

Select a reply

AttributeError: vertical attribute is not available. #3160

tik619 Aug 20, 2023

Replies: 6 comments · 5 replies

dopplershift Aug 20, 2023 Maintainer

tik619 Aug 21, 2023 Author

dopplershift Aug 21, 2023 Maintainer

tik619 Aug 22, 2023 Author

dopplershift Aug 22, 2023 Maintainer

tik619 Aug 22, 2023 Author

dopplershift Aug 22, 2023 Maintainer

tik619 Aug 28, 2023 Author

dcamron Sep 12, 2023 Maintainer

dcamron Sep 13, 2023 Maintainer

tik619 Sep 14, 2023 Author

tik619
Aug 20, 2023

Replies: 6 comments 5 replies

dopplershift
Aug 20, 2023
Maintainer

tik619
Aug 21, 2023
Author

dopplershift Aug 21, 2023
Maintainer

tik619
Aug 22, 2023
Author

dopplershift Aug 22, 2023
Maintainer

tik619
Aug 22, 2023
Author

dopplershift Aug 22, 2023
Maintainer

tik619 Aug 28, 2023
Author

dcamron
Sep 12, 2023
Maintainer

dcamron Sep 13, 2023
Maintainer

tik619
Sep 14, 2023
Author