Generate daily stations from meteo france data

README.md

Récupération des stations quotidiennes depuis les données Météo France

import json
import urllib.request
from glob import glob
import pandas as pd
import geopandas

dataset_id = '6569b51ae64326786e4e8e1a'
url = f'https://www.data.gouv.fr/api/1/datasets/{dataset_id}/'

with urllib.request.urlopen(url) as resp:
    json_content = json.load(resp)

urls = [resource.get('url') for resource in json_content.get('resources') if 'RR-T-Vent' in resource.get('url') and resource.get('type') != 'documentation']

mydict = {}
for url, dep in [[url, url.split('/')[-1].split('_')[1]] for url in urls]:
    if dep not in mydict:
        mydict[dep] = []
    mydict[dep].append(url)

for dep,values in mydict.items():
    frames = [pd.read_csv(url, compression='gzip', sep=';', quotechar='"') for url in values]
    df = pd.concat(frames)
    stations = df[['NUM_POSTE', 'NOM_USUEL', 'LAT', 'LON', 'ALTI', 'AAAAMMJJ']]
    stations['AAAAMMJJ'] = pd.to_datetime(stations['AAAAMMJJ'], format = '%Y%m%d')
    stations['MIN_DATE'] = stations.groupby(['NUM_POSTE'])['AAAAMMJJ'].transform('min')
    stations['MAX_DATE'] = stations.groupby(['NUM_POSTE'])['AAAAMMJJ'].transform('max')
    stations.drop(columns=['AAAAMMJJ'], inplace=True)
    stations.reset_index().drop_duplicates('NUM_POSTE').drop(columns=['index']).to_csv(f'stations-RR-T-Vent-dep-{dep}.csv', index=False)

files_stations_rr_t_vent = glob('stations-RR-T-Vent-dep-*.csv')
frames_stations_rr_t_vent = [pd.read_csv(input_file) for input_file in files_stations_rr_t_vent]
df_stations_rr_t_vent = pd.concat(frames_stations_rr_t_vent)
gdf_stations_rr_t_vent = geopandas.GeoDataFrame(
    df_stations_rr_t_vent, geometry=geopandas.points_from_xy(df_stations_rr_t_vent.LON, df_stations_rr_t_vent.LAT), crs="EPSG:4326"
)
gdf_stations_rr_t_vent.to_file('stations_rr_t_vent.geojson', driver='GeoJSON')

get-stations-daily-climatology-infos.py

import json
import urllib.request
from glob import glob
import pandas as pd
import geopandas

dataset_id = '6569b51ae64326786e4e8e1a'
url = f'https://www.data.gouv.fr/api/1/datasets/{dataset_id}/'

with urllib.request.urlopen(url) as resp:
    json_content = json.load(resp)

urls = [resource.get('url') for resource in json_content.get('resources') if 'RR-T-Vent' in resource.get('url') and resource.get('type') != 'documentation']

mydict = {}
for url, dep in [[url, url.split('/')[-1].split('_')[1]] for url in urls]:
    if dep not in mydict:
        mydict[dep] = []
    mydict[dep].append(url)

for dep,values in mydict.items():
    frames = [pd.read_csv(url, compression='gzip', sep=';', quotechar='"') for url in values]
    df = pd.concat(frames)
    stations = df[['NUM_POSTE', 'NOM_USUEL', 'LAT', 'LON', 'ALTI', 'AAAAMMJJ']]
    stations['AAAAMMJJ'] = pd.to_datetime(stations['AAAAMMJJ'], format = '%Y%m%d')
    stations['MIN_DATE'] = stations.groupby(['NUM_POSTE'])['AAAAMMJJ'].transform('min')
    stations['MAX_DATE'] = stations.groupby(['NUM_POSTE'])['AAAAMMJJ'].transform('max')
    stations.drop(columns=['AAAAMMJJ'], inplace=True)
    stations.reset_index().drop_duplicates('NUM_POSTE').drop(columns=['index']).to_csv(f'stations-RR-T-Vent-dep-{dep}.csv', index=False)

files_stations_rr_t_vent = glob('stations-RR-T-Vent-dep-*.csv')
frames_stations_rr_t_vent = [pd.read_csv(input_file) for input_file in files_stations_rr_t_vent]
df_stations_rr_t_vent = pd.concat(frames_stations_rr_t_vent)
gdf_stations_rr_t_vent = geopandas.GeoDataFrame(
    df_stations_rr_t_vent, geometry=geopandas.points_from_xy(df_stations_rr_t_vent.LON, df_stations_rr_t_vent.LAT), crs="EPSG:4326"
)
gdf_stations_rr_t_vent.to_file('stations_rr_t_vent.geojson', driver='GeoJSON')

stations_rr_t_vent.geojson