meteora 0.14.0

Pythonic interface to access data from meteorological stations

Meteora

Pythonic interface to access observations from meteorological stations. Key features:

• easily stream meteorological observations from multiple providers, from global (e.g., GHCNh) and regional (e.g., MeteoSwiss) networks to citizen weather stations (e.g., Netatmo) into pandas data frames.
• user-friendly arguments to filter data by region, variables or date ranges.
• request and download caching with requests-cache and pooch to avoid re-downloading data and help bypassing API limits.
• postprocess the collected data, e.g.:
  • compute climate indices
  • detect heatwave periods
  • produce ground truth datasets to benchmark weather forecasts
  • perform quality-control on citizen weather station data.

Overview

Meteora provides a set of provider-specific clients to get observations from meteorological stations. For instance, it can be used to stream data from the Global Historical Climatology Network hourly (GHCNh) into a pandas data frame:

from meteora.clients import GHCNHourlyClient

region = "Canton de Vaud, Switzerland"
variables = ["temperature", "precipitation", "wind_speed"]
start = "12-11-2021"
end = "12-12-2021"

client = GHCNHourlyClient(region)
ts_df = client.get_ts_df(variables, start, end)
ts_df.head()

[########################################] | 100% Completed | 925.68 ms

station_id	time	temperature	precipitation	wind_speed
SZI0000LSMP	2021-12-11 00:20:00	2.0	NaN	4.6
	2021-12-11 00:50:00	2.0	NaN	5.1
	2021-12-11 01:20:00	2.0	NaN	4.6
	2021-12-11 01:50:00	2.0	NaN	3.6
	2021-12-11 02:20:00	2.0	NaN	4.6

We can also get the station locations using the stations_gdf property:

import contextily as cx

ax = client.stations_gdf.assign(
    T_mean=ts_df.groupby("station_id")["temperature"].mean()
).plot(
    "T_mean",
    cmap="winter",
    legend=True,
    legend_kwds={"label": "$\overline{T} \; [\circ C]$", "shrink": 0.5},
)
cx.add_basemap(ax, crs=client.stations_gdf.crs, attribution=False)

(C) OpenStreetMap contributors, Tiles style by Humanitarian OpenStreetMap Team hosted by OpenStreetMap France

See the user guide for more details about the features of Meteora as well as the list of supported providers.

Installation

The easiest way to install Meteora is with conda/mamba:

conda install -c conda-forge meteora

Alternatively, if geopandas dependencies are installed correctly, you can install Meteora using pip:

pip install meteora

See also

Meteora intends to provide a unified way to access data from meteorological stations from multiple providers. The following libraries provide access to data from a specific provider:

• martibosch/agrometeo-geopy
• martibosch/netatmo-geopy

Eventually these packages will be fully integrated into Meteora.

Acknowledgements

• The logging system is based on code from gboeing/osmnx.
• This package was created with the martibosch/cookiecutter-geopy-package project template.
• With the support of the École Polytechnique Fédérale de Lausanne (EPFL).