frostrose 0.1.1

A Python library for accessing the Norwegian Meteorological Institute frost.met.no API

Frostrose

frostrose is a Python library for easy access to the Norwegian Meteorological Institute's frost.met.no API. It simplifies the process of fetching weather data from various endpoints.

Installation

pip install frostrose

Usage

Setting up the Client

You can use the FrostClient either by setting an environment variable FROST_CLIENT_ID or by providing the client ID directly during instantiation.

Using an Environment Variable

Set the environment variable FROST_CLIENT_ID with your API key:

export FROST_CLIENT_ID='your_api_key_here'

Then, in your Python code:

from frostrose import FrostClient

client = FrostClient()

Providing Client ID on Instantiation

Alternatively, you can provide the client ID directly:

from frostrose import FrostClient

client = FrostClient('your_api_key_here')

Working with Sources (Weather Stations)

Weather stations are referred to as "sources" in the frost.met.no API. frostrose provides methods to find sources by name, near a specific longitude and latitude, or near another source.

Finding Sources by Name

from frostrose import Sources

sources = Sources(client)
sources.fetch_all_sources()
tynset_sources = sources.get_sources_with_name_icontains('tynset')

Finding Sources Near a Specific Location

nearby_sources = sources.get_sources_within_radius_of_lonlat(lon=5.1963, lat=59.2555, radius_km=50)

Finding Sources Near Another Source

nearby_sources_of_id = sources.get_sources_within_radius_of_source_id('SN47230', radius_km=50)

Finding Weather Elements

To find available weather elements:

from frostrose import Elements

elements = Elements(client)
elements_data = elements.get_elements()

Getting Observation Time Series

To get an observation time series for a specific weather element:

from frostrose import Observations

observations = Observations(client)
temperature = observations.get_data_for_source_element_for_period('SN9580', 'air_temperature', date_start='2023-01-01')

Converting to Pandas DataFrame

To convert the observation time series into a pandas DataFrame:

import pandas as pd

flat_data = temperature.to_flat_dict_list()
df = pd.DataFrame.from_records(flat_data)

License

Frostrose is licensed under the MIT License. See the LICENSE file for more details.