import metar or synop data fo metpy

[fipoucat]

I am facing an error for metar not in the registry, which means associated with package:

ValueError: File 'metar_20230114_1200.txt' is not in the registry

1)I need help on how to load other imported metar and synops to be handled by the station_plot.py script.

2. I am looking for the script to compute and display the nice GDI map shown in the related issue section.

THank you

[DWesl]

What code are you trying that produces the above ValueError? Is it lifted exactly from an example? Did you tweak it to plot a different date?

Which is "the nice GDI map"? What "related issue section" do you mean? I would guess "the nice GDI map" is one of the ones from the example gallery, perhaps this one, but none of them mention a GDI.

[fipoucat]

Thank you @dwes, I am trying the example station_plot.py which works as it is. Now when I download other metar datasets from the server https://thredds.ucar.edu/thredds/catalog/noaaport/text/metar/catalog.html I am getting this error message. I tried 3 different dates.

For the GDI it relates to the following issue.: THe example related to GDI is : #3121

[DWesl]

The catalog you linked only displays files for the last month or so: does 2025-01-17T12Z work for you?

There's tests calculating GDI here:

MetPy/tests/calc/test_thermo.py

Lines 2527 to 2566 in 1a00500

	def test_gdi_xarray(index_xarray_data_expanded):
	"""Test the GDI calculation with a grid of xarray data."""
	pressure = index_xarray_data_expanded.isobaric
	temperature = index_xarray_data_expanded.temperature
	dewpoint = index_xarray_data_expanded.dewpoint
	mixing_ratio = mixing_ratio_from_relative_humidity(
	pressure, temperature, relative_humidity_from_dewpoint(temperature, dewpoint))
	result = galvez_davison_index(
	pressure,
	temperature,
	mixing_ratio,
	pressure[0]
	)
	assert_array_almost_equal(
	result,
	np.array([[[189.5890429, 157.4307982, 129.9739099],
	[106.6763526, 87.0637477, 70.7202505]]])
	)
	def test_gdi_arrays(index_xarray_data_expanded):
	"""Test GDI on 3-D Quantity arrays with an array of surface pressure."""
	ds = index_xarray_data_expanded.isel(time=0).squeeze()
	pressure = ds.isobaric.metpy.unit_array[:, None, None]
	temperature = ds.temperature.metpy.unit_array
	dewpoint = ds.dewpoint.metpy.unit_array
	mixing_ratio = mixing_ratio_from_relative_humidity(
	pressure, temperature, relative_humidity_from_dewpoint(temperature, dewpoint))
	surface_pressure = units.Quantity(
	np.broadcast_to(pressure.m, temperature.shape), pressure.units)[0]
	result = galvez_davison_index(pressure, temperature, mixing_ratio, surface_pressure)
	assert_array_almost_equal(
	result,
	np.array([[189.5890429, 157.4307982, 129.9739099],
	[106.6763526, 87.0637477, 70.7202505]])
	)

with data defined here:

MetPy/tests/calc/test_thermo.py

Lines 2365 to 2406 in 1a00500

	@pytest.fixture()
	def index_xarray_data_expanded():
	"""Create expanded data for testing that index calculations work with xarray data.
	Specifically for Galvez Davison Index calculation, which requires 950hPa pressure
	"""
	pressure = xr.DataArray(
	[950., 850., 700., 500.], dims=('isobaric',), attrs={'units': 'hPa'}
	)
	temp = xr.DataArray([[[[306., 305., 304.], [303., 302., 301.]],
	[[296., 295., 294.], [293., 292., 291.]],
	[[286., 285., 284.], [283., 282., 281.]],
	[[276., 275., 274.], [273., 272., 271.]]]] * units.K,
	dims=('time', 'isobaric', 'y', 'x'))
	profile = xr.DataArray([[[[299., 298., 297.], [296., 295., 294.]],
	[[289., 288., 287.], [286., 285., 284.]],
	[[279., 278., 277.], [276., 275., 274.]],
	[[269., 268., 267.], [266., 265., 264.]]]] * units.K,
	dims=('time', 'isobaric', 'y', 'x'))
	dewp = xr.DataArray([[[[304., 303., 302.], [301., 300., 299.]],
	[[294., 293., 292.], [291., 290., 289.]],
	[[284., 283., 282.], [281., 280., 279.]],
	[[274., 273., 272.], [271., 270., 269.]]]] * units.K,
	dims=('time', 'isobaric', 'y', 'x'))
	dirw = xr.DataArray([[[[135., 135., 135.], [135., 135., 135.]],
	[[180., 180., 180.], [180., 180., 180.]],
	[[225., 225., 225.], [225., 225., 225.]],
	[[270., 270., 270.], [270., 270., 270.]]]] * units.degree,
	dims=('time', 'isobaric', 'y', 'x'))
	speed = xr.DataArray([[[[15., 15., 15.], [15., 15., 15.]],
	[[20., 20., 20.], [20., 20., 20.]],
	[[25., 25., 25.], [25., 25., 25.]],
	[[50., 50., 50.], [50., 50., 50.]]]] * units.knots,
	dims=('time', 'isobaric', 'y', 'x'))
	return xr.Dataset({'temperature': temp, 'profile': profile, 'dewpoint': dewp,
	'wind_direction': dirw, 'wind_speed': speed},
	coords={'isobaric': pressure, 'time': ['2023-01-01T00:00Z']})

Loading the data required for the calculation from a file (downloaded from, say, NCEP for the US models or Copernicus for ERA5 data) and feeding it to the function would produce the GDI to plot.
The plots shown there look like a contourf plot with a contour interval of 5 from -100 to 100 and a RdBu_r colormap on a Plate-Caree/equirectangular projection with state borders and ocean and lake coastlines. Possibly a levels argument of np.arange(0, 101, 5) with a colormap of Reds would also work.

[DWesl]

Here's the implementation of metpy.cbook.get_test_data:

MetPy/src/metpy/cbook.py

Lines 14 to 40 in 1a00500

	pooch_kwargs = {'path': pooch.os_cache('metpy'), 'version': 'v' + __version__,
	'base_url': 'https://github.com/Unidata/MetPy/raw/{version}/staticdata/'}
	# Check if we have the data available directly from a git checkout, either from the
	# TEST_DATA_DIR variable, or looking relative to the path of this module's file. Use this
	# to override Pooch's path and disable downloading from GitHub.
	dev_data_path = os.environ.get('TEST_DATA_DIR', Path(__file__).parents[2] / 'staticdata')
	if Path(dev_data_path).exists():
	pooch_kwargs['path'] = dev_data_path
	pooch_kwargs['version'] = None
	pooch_kwargs['base_url'] = pooch_kwargs['base_url'].format(version='main')
	if pooch.__version__ >= 'v1.6.0':
	pooch_kwargs['allow_updates'] = False
	POOCH = pooch.create(version_dev='main', **pooch_kwargs)
	POOCH.load_registry(Path(__file__).parent / 'static-data-manifest.txt')
	def get_test_data(fname, as_file_obj=True, mode='rb'):
	"""Access a file from MetPy's collection of test data."""
	path = POOCH.fetch(fname)
	# If we want a file object, open it, trying to guess whether this should be binary mode
	# or not
	if as_file_obj:
	return open(path, mode) # noqa: SIM115
	return path

as the name implies, it's only designed for the MetPy test data, not arbitrary dates.

I don't remember if Siphon will download metar data for you, but you could download the files manually with wget and pass the path to parse_metar_file, or download with response = requests.get(url); fid = io.StringIO(response.text), and pass fid to parse_metar_file. Alternately, you could re-work the example above to use the THREDDS catalog you posted earlier instead of MetPy's test data repository and use that.