dmi-open-data-pandas 0.1

A package to extract data from DMI Open Data into pandas dataframes

DMI Open Data API

Simple Python interface to the The Danish Meteorological Institute's (DMI) v2 Open Data API.

Forked from this repo to add support for more data sources as well as pandas output.

Currently the following APIs are supported:

• climateDataAPI
• forecastedr
• metObsAPI

See API status here

Requirements

• Python 3.9+
• API Keys for the different Open Data APIs: see https://dmiapi.govcloud.dk/#!/apis

Installation

$ pip install dmi-open-data-pandas

Examples

Forecast API

from dmi_open_data import DMIOpenDataClient

# initialize client with API key stored in env variable
client = DMIOpenDataClient(env_var_name="DMI_FORECAST_EDR_API_KEY")

# get list of models available
models = client.get_models(as_df=True)

# get info about a specific model
model_info = client.get_models("harmonie_dini_sf")

# list parameters available for specific model
params = client.list_model_parameters("harmonie_dini_sf", as_df=True)

# get a forecast for wind speed and temperature
fc = client.get_forecast(
    latitude=55.767247, 
    longitude=12.464263,
    parameter=["temperature-0m", "wind-speed-10m"]
)

Observations and Climate Data API

from dmi_open_data import DMIOpenDataClient, Parameter, ClimateDataParameter
from datetime import datetime
import pandas as pd

# initialize client with API key stored in env variable
client = DMIOpenDataClient(env_var_name="DMI_API_KEY")

# get all stations
stations = client.get_stations()

# get station closest to a lat-lon
closest_station = client.get_closest_station(
    latitude=55.767247, 
    longitude=12.464263,
)

# Get available parameters
parameters = client.list_observation_parameters()

# Get temperature observations from DMI station in given time period
observations = client.get_observations(
    parameter=Parameter.TempDry,
    station_id=closest_station['properties']['stationId'],
    from_time=datetime(2021, 7, 20),
    to_time=datetime(2021, 7, 24),
    limit=1000
)

# you can also use the simple parameter string, and ask to return a pandas dataframe
observations = client.get_observations(
    parameter="temp_dry",
    station_id=closest_station['properties']['stationId'],
    from_time=datetime(2021, 7, 20),
    to_time=datetime(2021, 7, 24),
    as_df=True
)

# initialize climate data client
climate_data_client = DMIOpenDataClient(env_var_name='DMI_CLIMATE_DATA_API_KEY')

# get climate data, also using pandas Timestamps
climate_data = climate_data_client.get_climate_data(
    parameter=ClimateDataParameter.MeanTemp,
    station_id=closest_station['properties']['stationId'],
    from_time=pd.Timestamp("2023-01-01"),
    to_time=pd.Timestamp("2023-01-03"),
    time_resolution='day',
    as_df=True
)

Tests

Run tests (after installing pytest)

$ python -m pytest

Contributing

Please send an email to [e-mail]. We welcome contributions!