Hi,
MetPy 0.8.0 has been released. This release has a wide collection of new
features as well as minor bug fixes, including several contributions from
our community.
Highlights
-------------
* MetPy now requires XArray. Using XArray, MetPy provides some custom
accessors under the `.metpy` attribute on `Dataset` and `DataArray`. This
allows simplified access to the underlying grid projection as a CartoPy
projection, as well as simplified access to data as an array with units
attached. (See the `XArray_Projections.py` example) MetPy calculation
functions will also now automatically convert an XArray `DataArray` with
units metadata into a proper array with units. Future releases of MetPy
will improve the integration with XArray.
* `GINIFile` now works directly as an XArray storage backend. See the
updated `GINI_Water_Vapor.py` example for details. Support for Unidata
NEXRAD composites in GINI format was also added.
* MetPy's colortables are now available from `metpy.plots.colortables`
rather than needing to do `metpy.plots.ctables.registry`.
* Fixed a few corner cases with isentropic interpolation, including
problems with columns of entirely missing data (frequently encountered with
NARR data)
* `lfc` and `el` now work with arbitrary parcel profiles
* Gradient and related functions now properly propagate masked values
* Text and time formatting for `add_timestamp()` are now configurable
* New calculations
- Absolute vorticity
- Ageostrophic wind
- Apparent temperature (combines wind chill and heat index)
- Brunt-Väisälä frequency and period
- Critical angle (severe weather parameter)
- Dewpoint (from specific humidity)
- Exner function
- Inertial advective wind
- Parse angle (converts strings like "ESE" to degree values)
- Potential temperature to temperature
- Potential vorticity
- Q Vector
- Saturation equivalent potential temperature
- Static stability
- Wet bulb temperature
* Other assorted bug fixes
* Reminder: MetPy plans to drop support for Python 2.7 in Fall 2019
API Changes
-----------------
* `lat_lon_grid_spacing` has been deprecated in favor of
`lat_lon_grid_deltas`. The latter gives properly signed values depending on
the ordering of points in the lat/lon grid.
* The deprecated upper air data reading support has been removed. Use the
upper air support in siphon (https://unidata.github.io/siphon) instead.
* The function `delete_masked_points` has been renamed to
`_delete_masked_points`, as it was not intended to be part of the supported
API.
* `GINIFile.to_dataset()`, which gives a netCDF4-python-like interface to
the information in the file, has been deprecated. Instead, now an instance
of `GINIFile` can be passed directly to `xarray.open_dataset()` to provide
a similar interface. `to_dataset()` (as well as the non-public `cdm`
module) will be removed in 0.11.
Contributors
------------
@ahuang11, @akrherz, @asselapathirana, @eliteuser26, @jthielen, @kishan3,
@Meteodan, @mwilson14, @shofer16450, @tjwixtrom, @WEP11, @jrleeman, and
@dopplershift all contributed commits to this release.
For full release notes see the GitHub release page.
https://github.com/Unidata/MetPy/releases/tag/v0.8.0
MetPy packages are available for Conda on the conda-forge channel
and for pip from the Python Package Index.
https://anaconda.org/conda-forge/MetPy
https://pypi.python.org/pypi/MetPy
Let us know if you run into any problems, either at MetPy issue tracker,
using support-python@xxxxxxxxxxxxxxxx, or on
this list.
You can also ask questions using the "MetPy" tag on Stack Overflow.
https://github.com/Unidata/MetPy/issues
https://stackoverflow.com/questions/tagged/python-siphon
Ryan
--
Ryan May, Ph.D.
Software Engineer
UCAR/Unidata
Boulder, CO
--
Ryan May, Ph.D.
Software Engineer
UCAR/Unidata
Boulder, CO
