---
id: earthkit-meteo
version: "1.1.0"
license: Apache-2.0
license_treatment: permissive
maintenance: active
---

# earthkit-meteo — A Python library for meteorological computations

License: permissive · Maintenance: active · Downloads: 76.5K/mo

## What it is and what it does

earthkit-meteo is a meteorological computation library from ECMWF that wraps standard atmospheric physics calculations to work with multiple array backends. It lets you compute thermodynamic properties like potential temperature from temperature and pressure data, accepting NumPy arrays, Torch tensors, CuPy arrays, xarray objects, or fieldlist formats as input. The library is part of the broader earthkit ecosystem and is classified as Graduated and Production/Stable.

You use it when you need to perform standard meteorological calculations on weather or atmospheric data without writing the physics formulas yourself. It abstracts away the array-backend differences so the same code works whether your data lives in NumPy, on a GPU via Torch, or in other formats. The package has low install friction, depends only on deprecation, earthkit-utils, and numpy at runtime, and is actively maintained.

Use it for:

- Compute potential temperature from model or observational data for atmospheric analysis.
- Process weather datasets in multiple array formats without rewriting calculation logic.
- Integrate meteorological computations into data pipelines using NumPy or Torch tensors.
- Perform thermodynamic calculations as part of ECMWF earthkit-based workflows.
- Accelerate atmospheric physics on GPUs by passing CuPy or Torch arrays to the library.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Computes meteorological quantities (potential temperature, thermodynamic properties) from atmospheric data using NumPy, Torch, CuPy arrays, xarray, or fieldlist formats.

Yes, if you work with atmospheric or weather data and need standard meteorological calculations. The package is actively maintained, has low install friction, carries a permissive license, and supports modern Python versions. It is worth installing for meteorological workflows, especially if you already use NumPy or Torch and want to avoid reimplementing atmospheric physics.

## Install

pip install earthkit-meteo
uv add earthkit-meteo
poetry add earthkit-meteo

## Installing earthkit-meteo

Before you install: Low install friction; pure Python wheel with three lightweight runtime dependencies (deprecation, earthkit-utils, numpy). Actively maintained with a recent release 9 days ago.

License in practice: Apache-2.0 permissive license allows commercial and private use with minimal restrictions; you must retain license notices and copyright attribution.

Quickstart:

pip install earthkit-meteo

from earthkit.meteo import thermo
import numpy as np

t = np.array([264.12, 261.45])  # Kelvins
p = np.array([850, 850]) * 100.0  # Pascals
theta = thermo.potential_temperature(t, p)

Requires Python 3.10 or later.

Verify before relying:

- Whether Torch and CuPy support is optional or requires separate installation.
- Full scope of meteorological functions beyond potential temperature.
- Performance characteristics when working with large datasets or GPU tensors.

## Package facts

- License: Apache-2.0 (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 76.5K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

meteorological calculations python, atmospheric thermodynamics library, potential temperature computation, weather data array processing, meteorology numpy torch, ECMWF weather calculations, atmospheric physics computations, meteorology, atmospheric-physics, array-agnostic

[View on SkillFed](https://skillfed.io/packages/earthkit-meteo) · [View on PyPI](https://pypi.org/project/earthkit-meteo/)