Data for gsee via wetterdienst package

Readme.md

This is how we can use this function:

get_data_from_wetterdienst(dataset="temperature_air", year=2021, coords=(49.0, 8))

--> Loading temperature_air data from German Weather Service station: Bergzabern, Bad (Rheinland-Pfalz), distance: 11.9 km.
0        3.2
        ... 
8758    10.5
Name: value, Length: 8760, dtype: float64

get_data_from_wetterdienst(dataset="solar", year=2021, coords=sample_coords)

--> Loading solar data from German Weather Service station: Mannheim (Baden-Württemberg), distance: 83.8 km.
0       0.0
       ... 
8759    0.0
Name: radiation_global, Length: 8760, dtype: float64

data_from_wetterdienst.py

from wetterdienst.provider.dwd.observation import DwdObservationRequest
from datetime import datetime, timedelta
from typing import Tuple
import pandas as pd

def get_data_from_wetterdienst(
    dataset: str,
    year: int,
    coords: Tuple,
    verbose: bool = True,
) -> pd.Series:
    """Get solar global radiation or air temperature via wetterdienst package.

    `dataset` can be 'solar' or 'temperature_air'
    """

    _delta_to_UTC = 1  # 1 for Germany for UTC+1
    _si_units = dict(temperature_air=False, solar=False)
    _tidy = dict(temperature_air=True, solar=False)

    stations = DwdObservationRequest(
        parameter=dataset,
        resolution="hourly",
        start_date=datetime(year, 1, 1, 0, 0) + timedelta(hours=-_delta_to_UTC),
        end_date=datetime(year, 12, 31, 23, 0) + timedelta(hours=-_delta_to_UTC),
        si_units=_si_units[dataset],
        tidy=_tidy[dataset],
    ).filter_by_rank(*coords, 1)

    if verbose:
        provider_info = stations.provider.value[1]
        s = stations.df.loc[0, :]
        info = f"--> Loading {dataset} data from {provider_info} station: {s['name']} ({s['state']}), distance: {s['distance']:.1f} km."
        print(info)

    df = next(stations.values.query()).df

    if dataset == "solar":
        return df.radiation_global
    elif dataset == "temperature_air":
        return df[df.parameter == "temperature_air_mean_200"].value
    else:
        raise RuntimeError("`dataset` must be either 'solar' or 'temperature_air'!")