lair.soundings.get_soundings#

lair.soundings.get_soundings(station='SLC', start=None, end=None, sounding_dir=None, months=None, driver='xarray', download=False)[source]#

Get upper air soundings from the Wyoming archive.

Reads the soundings that are in the station’s folder. It downloads only when asked: see When get_soundings downloads.

Parameters:
  • station (str) – The 3-letter station identifier (e.g. 'SLC'). Names the subdirectory of $LAIR_SOUNDING_DIR and the file prefix.

  • start (datetime | str, optional) – The start date and time (inclusive). A time without a time zone follows the UTC rule; one with a time zone is converted to UTC.

  • end (datetime | str, optional) – The end date and time (inclusive).

  • sounding_dir (str) – The directory containing the station’s sounding files. Defaults to $LAIR_SOUNDING_DIR/<station>.

  • months (list[int], optional) – Only use soundings from these months (with download, only download those).

  • driver ({'xarray', 'pandas'}, optional) – Return interpolated profiles as an xarray Dataset (default), or the merged raw soundings as a pandas DataFrame. The profiles share one set of heights (Read many soundings says which).

  • download (bool, optional) – Download the soundings the folder lacks, from start to end, before reading them. The default, False, downloads nothing.

Returns:

xr.Dataset | pd.DataFrame – The sounding data.

Raises:

ValueError – If no sounding is found for the period (the message says to pass download=True when it is False); if download is True without start and end; or for an invalid driver.

See also

download_soundings

Download soundings without reading them.

Examples

Two made-up soundings in a folder, read as one dataset on the same heights (Read a sounding and compute its potential temperature reads real ones):

>>> import pathlib
>>> import tempfile
>>> import pandas as pd
>>> from lair.soundings import get_soundings
>>> folder = pathlib.Path(tempfile.mkdtemp())
>>> for hour, ground in [("2024011500", -8.0), ("2024011512", -5.0)]:
...     pd.DataFrame(
...         {
...             "height": [1289.0, 5100.0],
...             "temperature": [ground, -30.0],
...             "station": "SLC",
...             "time": pd.to_datetime(hour, format="%Y%m%d%H"),
...             "elevation": 1289.0,
...         }
...     ).to_csv(folder / f"SLC_{hour}.csv", index=False)
>>> ds = get_soundings(sounding_dir=str(folder))
>>> ds.temperature.sizes
Frozen({'time': 2, 'height': 372})
>>> ds.temperature.sel(height=1289.0).values
array([-8., -5.])