MetPy

1.7.1 · active · verified Tue Apr 14

MetPy is a collection of tools in Python for reading, visualizing and performing calculations with weather data. It provides meteorological calculations, data input/output, and specialized plotting capabilities like Skew-T diagrams and station plots. Following semantic versioning, MetPy 1.x releases are backward compatible, with breaking changes reserved for major version increments. The current version is 1.7.1 and it requires Python >= 3.10.

Warnings

breaking The MetPy 1.0 release introduced significant API changes, particularly in `metpy.calc` functions. Many argument names were changed for consistency (e.g., `rh` to `relative_humidity`, `dewpt` to `dewpoint`). Functions like `geostrophic_wind()` and `ageostrophic_wind()` now expect `latitude` instead of the `coriolis_parameter f`. Code using keyword arguments with old names will break.
Fix: Consult the MetPy 1.0 API Changes and Upgrade Guide in the official documentation to update argument names. Pass latitude instead of the Coriolis parameter where required.
breaking Prior to MetPy 1.0, many `metpy.calc` functions consistently returned `pint.Quantity` objects even when provided `xarray.DataArray` inputs. From MetPy 1.0 onwards, these functions will now properly return an `xarray.DataArray` when given `xarray.DataArray` input, preserving xarray's coordinate information.
Fix: Adjust code expectations for return types from `metpy.calc` functions when using xarray inputs. Leverage xarray's capabilities for coordinate-aware analysis.
gotcha MetPy heavily relies on unit-aware quantities provided by the `pint` library. Most MetPy functions explicitly require inputs to have units attached (e.g., `25 * units.degC`). Passing raw numerical arrays without units will often result in a `UnitsError` or incorrect calculations.
Fix: Always attach appropriate units from `metpy.units` (e.g., `from metpy.units import units`) to your data when calling MetPy functions.
gotcha Import errors related to `pint` (MetPy's unit library) can occur if the `typing_extensions` package in your environment is outdated. This conflict prevents proper loading of MetPy's dependencies.
Fix: Upgrade `typing_extensions` to its latest version: `pip install --upgrade typing_extensions`.

Install

pip install metpy PyPI
conda install -c conda-forge metpy Conda

Imports

units
```
from metpy.units import units
```
Required for unit-aware calculations.
mpcalc
```
import metpy.calc as mpcalc
```
Common alias for meteorological calculations.
io
```
from metpy import io
```
For reading meteorological data formats like METARs or NEXRAD.
plots
```
from metpy import plots
```
For specialized meteorological plots like Skew-T diagrams.

Quickstart

This quickstart demonstrates how to define quantities with units using `metpy.units`, perform a common meteorological calculation (dewpoint) using `metpy.calc`, and convert units. MetPy's strong unit handling ensures correct calculations.

from metpy.units import units
import metpy.calc as mpcalc
import numpy as np

# Define temperature and relative humidity with units
temperature = 25 * units.degC
relative_humidity = 75 * units.percent

# Calculate dewpoint
dewpoint = mpcalc.dewpoint_from_relative_humidity(temperature, relative_humidity)

print(f"Temperature: {temperature:.1f}")
print(f"Relative Humidity: {relative_humidity:.1f}")
print(f"Dewpoint: {dewpoint:.1f}")

# Example of converting units
temp_f = temperature.to('degF')
print(f"Temperature in Fahrenheit: {temp_f:.1f}")

view raw JSON →