meteoschweiz 0.1.0

Python package for accessing and processing MeteoSwiss weather data

MeteoSwiss

A comprehensive Python library for accessing, managing, and analyzing Swiss weather data from MeteoSwiss (Federal Office of Meteorology and Climatology). This toolkit provides high-level interfaces for working with weather stations, parameters, historic data, and forecasts.

---

Table of Contents

• Installation
• Usage
  • 1. Metadata
  • 2. Historic Weather Data
  • 3. Local Forecast Data
• API Reference
• Developer Guide
• Bugs / Issues
• Roadmap
• License

---

Installation

pip install meteoschweiz

Requirements: Python 3.9+ · pandas ≥ 1.5.0 · requests ≥ 2.28.0

---

Usage

1. Metadata

1.1 Stations

from meteoswiss.metadata.stations import SwissWeatherStations

stations = SwissWeatherStations()
stations_df = stations.load()  # returns all available weather stations in Switzerland

# Get a specific station by abbreviation
gre = stations.get_by_abbr('GRE')
print(f"{gre.point_name}: {gre.elevation:.0f}m")

# Find stations near a coordinate (lat, lon, radius in km)
nearby = stations.find_nearby(lat=47.0, lon=7.4, radius_km=30)

1.2 Parameter Metadata

MetaParametersLoader loads parameter definitions from any MeteoSwiss CSV source URL.

from meteoswiss.metadata.parameters import MetaParametersLoader

loader = MetaParametersLoader()

# Register sources
loader.add_source(
    name="Forecast Parameters",
    url="https://data.geo.admin.ch/ch.meteoschweiz.ogd-local-forecasting/ogd-local-forecasting_meta_parameters.csv",
    description="MeteoSwiss forecast parameters",
    encoding="latin-1",
    delimiter=";",
    key_column="parameter_shortname",
)

loader.add_source(
    name="Historic Parameters",
    url="https://data.geo.admin.ch/ch.meteoschweiz.ogd-smn/ogd-smn_meta_parameters.csv",
    description="MeteoSwiss historic/SMN parameters",
    encoding="latin-1",
    delimiter=";",
    key_column="parameter_shortname",
)

# Load all sources
loader.load_all()
print(f"Loaded {len(loader.data)} sources")

# Get all parameters for a source as a DataFrame
df_params = loader.get_all_params(source="Forecast Parameters")
print(f"{len(df_params)} parameters, columns: {list(df_params.columns)[:5]}")

# Search by keyword
temp_params = loader.search("temperature", source="Forecast Parameters")
print(f"Found {len(temp_params)} temperature-related parameters")

# Get a specific parameter by key
param = loader.get("tre200h0", source="Historic Parameters")
if param:
    print(f"Source: {param.source}, keys: {list(param.data.keys())}")

# Print summary
print(loader.summary())

---

2. Historic Weather Data

from meteoswiss.historic.historic_handler import HistoricWeatherHandler, MeteoSwissClient

client = MeteoSwissClient()
historic_handler = HistoricWeatherHandler(
    stations_handler=stations,   # from section 1.1
    params_loader=loader,        # from section 1.2
    meteoswiss_client=client,
    language="en",               # "en" or "de"
)

# Query by station abbreviation
result = historic_handler.get_historic_by_station_id(
    station_id="GRE",
    start_date="2023-06-01",
    end_date="2023-06-30",
    aggregation="daily",
    parameters=["tre200d0", "tre200dn", "tre200dx", "rre150d0"],
    rename_columns=True,
    include_units=True,
)

# result is a HistoricQueryResult dataclass:
# .station_id, .station_name, .canton, .latitude, .longitude,
# .altitude, .aggregation, .parameters, .start_date, .end_date, .data (DataFrame)
print(result.data.head())

# Convenience methods
temp = historic_handler.get_temperature_history("GRE", "2023-01-01", "2023-12-31")
precip = historic_handler.get_precipitation_history("GRE", "2023-01-01", "2023-12-31")

# Export
historic_handler.export_to_csv(result, "grenchen_june_2023.csv")

---

3. Local Forecast Data

from meteoswiss.forecasts.forecast_handler import LocalForecastHandler

forecast_handler = LocalForecastHandler(
    stations_handler=stations,   # from section 1.1
    params_loader=loader,        # from section 1.2
)

# By MeteoSwiss point ID
result = forecast_handler.get_forecast_for_point_id(
    point_id=774,
    parameters=["tre200h0", "rre150h0"],
)
print(result.summary())

# By station name
result = forecast_handler.get_forecast_for_station_name("Grenchen")

# By coordinates (uses nearest station)
result = forecast_handler.get_forecast_for_coordinates(lat=47.18, lon=7.41)

# Export
forecast_handler.export_to_csv(result, "forecast_grenchen.csv")

---

API Reference

SwissWeatherStations

Method	Description
load()	Load all stations from MeteoSwiss
get_by_abbr(abbr)	Get station by abbreviation (e.g. 'GRE')
get_by_name(name)	Search stations by name
get_by_id(id)	Get station by point ID
find_nearby(lat, lon, radius_km)	Find stations within a radius
find_nearest(lat, lon, n)	Find the n nearest stations
filter_by_bbox(lat_min, lat_max, lon_min, lon_max)	Filter by bounding box
filter_by_elevation(min, max)	Filter by elevation range
get_highest_stations(n)	Get the n highest stations
get_statistics()	Summary statistics for the station network

MetaParametersLoader

Method	Description
add_source(name, url, ...)	Register a CSV source
load_source(name)	Load data from a named source
load_all()	Load all registered sources
get_all_params(source)	Return all parameters as a DataFrame
search(keyword, source)	Search by keyword
filter(conditions)	Filter by field conditions
get(key, source)	Get a single parameter by key
export_to_csv(source, filepath)	Export to CSV
export_to_json(source, filepath)	Export to JSON
summary()	Print loader summary
list_sources()	List registered source names
clear_cache(source)	Clear cached data

HistoricWeatherHandler

Method	Description
get_historic_by_station_id(station_id, ...)	Query by SMN station ID
get_historic_by_name(station_name, ...)	Query by station name
get_historic_by_coords(lat, lon, ...)	Query nearest station
get_temperature_history(station, start, end)	Temperature convenience method
get_precipitation_history(station, start, end)	Precipitation convenience method
list_available_parameters()	List all available parameters
export_to_csv(result, filepath)	Export to CSV
export_to_json(result, filepath)	Export to JSON

LocalForecastHandler

Method	Description
get_forecast_for_point_id(point_id, ...)	Get forecast by MeteoSwiss point ID
get_forecast_for_station_name(name, ...)	Get forecast by station name
get_forecast_for_coordinates(lat, lon, ...)	Get forecast for nearest station
get_all_parameters()	List available forecast parameters
export_to_csv(result, filepath)	Export to CSV
export_to_json(result, filepath)	Export to JSON

---

Developer Guide

Full details are in CONTRIBUTING.md.

Quick setup

# 1. Fork the repository on GitHub, then clone your fork
git clone https://github.com/<your-username>/MeteoSwiss.git
cd MeteoSwiss

# 2. Create a virtual environment and install dev dependencies
python -m venv .venv
source .venv/bin/activate        # Linux/macOS
# .venv\Scripts\activate         # Windows

pip install -e ".[dev]"

Branch workflow

# Always branch from master
git checkout -b feature/your-feature-name

Branch prefix	Purpose
feature/	New functionality
fix/	Bug fixes
docs/	Documentation only
refactor/	Code quality, no behavior change

Run tests

pytest tests/ -v                              # all tests
pytest tests/ -m "not integration"            # skip network tests
pytest tests/ --cov=src/meteoswiss           # with coverage
tox                                           # all Python versions (3.8–3.13)

Lint and format

ruff check src/ tests/           # check for issues
ruff check --fix src/ tests/     # auto-fix where possible
ruff format src/ tests/          # format code

Branch protection on master

The master branch is protected:

• All CI checks (lint, tests on Python 3.8–3.13, type-check) must pass.
• At least 1 approving review is required before merging.
• Direct pushes are disabled — all changes go through Pull Requests.

Submitting a Pull Request

1. Push your branch: git push origin feature/your-feature-name
2. Open a PR against master and fill in the PR template.
3. Address review feedback; a maintainer will merge once approved and CI passes.

---

Bugs / Issues

Open an issue on GitHub: New Bug Report

Please include:

• Python version and OS
• meteoschweiz version (pip show meteoschweiz)
• A minimal code snippet that reproduces the problem

---

Roadmap

1. Better visualizations (station geo coordinates, districts, cantons)
2. Data preprocessing pipelines
3. Individual weather feature handlers

---

License

MIT — Copyright (c) 2026 Sai Jithendra Gangireddy