Extract historical climate data from the Norwegian Meteorological Institute (MET)

Author: Willeke A’Campo

Description: This notebook extracts historical climate data from the Norwegian Meteorological Institute (MET) using the FROST API. The data is stored in a pandas dataframe and saved as a csv file. The data is used to detect the hottest summer on record in Bærum, Bodø, Kristiansand, and Oslo.

Data source: Documentation Frost MET

Result Summary:

The table shows hottest summer on record in Bærum, Bodø, Kristiansand, and Oslo.

City	Station	SourceID	Coordinates	Hottest Summer during Landsat-8 observation period	Temperature	Hottest Summer on record	Temperature
Bærum	Oslo (Blindern)	SN18700	10.723072, 59.940625	2018	18.8	1997	18.8
Bodø	Bodø (Lufthavn)	SN82310	14.375128, 67.282669	2023	14.1	2023	14.1
Kristiansand	Kristiansand (Lufthavn)	SN39040	7.995733, 58.14615	2018	17.5	2018	17.5
Oslo	Oslo (Blindern)	SN18700	10.723072, 59.940625	2018	18.8	1997	18.8

import requests
import pandas as pd
import yaml
import os
from pathlib import Path

project_path= Path.cwd().parents[0]
credentials_file = os.path.join(project_path, 'conf/local/credentials.yml')

# read client id from conf/local/credentials.yml
with open(credentials_file, 'r') as f:
    credentials = yaml.safe_load(f)
    client_id = credentials['frost_api']['client_id']
    client_key = credentials['frost_api']['client_key']

Stations Request

# Define stations endpoint
endpoint_stations = 'https://frost.met.no/locations/v0.jsonld'
# Issue an HTTP GET request
r = requests.get(endpoint_stations, auth=(client_id,''))
# Extract JSON data
json_stations = r.json()

# search for station with name Oslo, Kristiansand, Bodø and Bæerum
for station in json_stations['data']:
    # print station name and coordinates
    if station['name'] == 'Oslo':
        print(station['name'], station['geometry']['coordinates'])
    if station['name'] == 'Kristiansand':
        print(station['name'], station['geometry']['coordinates'])
    if station['name'] == 'Bodø':
        print(station['name'], station['geometry']['coordinates'])
    if station['name'] == 'Blindern':
        print(station['name'], station['geometry']['coordinates'])

Blindern [10.723072, 59.940625]
Bodø [14.375128, 67.282669]
Kristiansand [7.995733, 58.14615]
Oslo [10.746092, 59.912728]

Observations Request

# Define endpoint and parameters
endpoint_observations = 'https://frost.met.no/observations/v0.jsonld'

parameters = {
    'sources': 'SN18700,SN39040,SN82310', # oslo/bærum, kristiansand, bodø
    'elements': 'mean(air_temperature P3M)', # best_estimate_mean(air_temperature P3M)
    'referencetime': '1900-01-01/2023-10-01',
}
r = requests.get(endpoint_observations, parameters, auth=(client_id,''))
json = r.json()

# Check if the request worked, print out any errors
if r.status_code == 200:
    data = json['data']
    print('Data retrieved from frost.met.no!')
else:
    print('Error! Returned status code %s' % r.status_code)
    print('Message: %s' % json['error']['message'])
    print('Reason: %s' % json['error']['reason'])

Data retrieved from frost.met.no!

# Return a Dataframe with all of the observations in a table format
df = pd.DataFrame()
for i in range(len(data)):
    row = pd.DataFrame(data[i]['observations'])
    row['referenceTime'] = data[i]['referenceTime']
    row['sourceId'] = data[i]['sourceId']
    df = pd.concat([df, row], ignore_index=True)

df = df.reset_index()

# These columns will be kept
columns = ['sourceId','referenceTime','elementId','value','unit','timeOffset']
df2 = df[columns].copy()
# Convert the time value to Python standard datetime format
df2['referenceTime'] = pd.to_datetime(df2['referenceTime'])

# only show the data for the summer months
df_summer = df2[(df2['referenceTime'].dt.month >= 6) & (df2['referenceTime'].dt.month <= 8)]

Oslo and Bærums Hottest Summer on Record

The hottes summer on record for Bærum and Oslo during Landsat 8 observation period (2013 - present) is 2018.

# separate df for oslo
df_oslo = df_summer[df_summer['sourceId'] == 'SN18700:0'].copy()
df_oslo = df_oslo.reset_index()
# sort by warmest summer
df_oslo.sort_values(by=['value'], inplace=True, ascending=False)
display(df_oslo.head(3))

	index	sourceId	referenceTime	elementId	value	unit	timeOffset
60	241	SN18700:0	1997-06-01 00:00:00+00:00	mean(air_temperature P3M)	18.8	degC	PT0H
81	325	SN18700:0	2018-06-01 00:00:00+00:00	mean(air_temperature P3M)	18.8	degC	PT0H
10	41	SN18700:0	1947-06-01 00:00:00+00:00	mean(air_temperature P3M)	18.7	degC	PT0H

Kristiansand Hottest summer on record

The hottest summer on record for Kristiansand during Landsat 8 observation period (2013 - present) is 2018.

# separate df for kristiansand
df_kristiansand = df_summer[df_summer['sourceId'] == 'SN39040:0'].copy()
df_kristiansand = df_kristiansand.reset_index()
# sort by warmest summer
df_kristiansand.sort_values(by=['value'], inplace=True, ascending=False)
display(df_kristiansand.head(3))

	index	sourceId	referenceTime	elementId	value	unit	timeOffset
77	655	SN39040:0	2018-06-01 00:00:00+00:00	mean(air_temperature P3M)	17.5	degC	PT0H
56	572	SN39040:0	1997-06-01 00:00:00+00:00	mean(air_temperature P3M)	17.4	degC	PT0H
6	372	SN39040:0	1947-06-01 00:00:00+00:00	mean(air_temperature P3M)	17.2	degC	PT0H

Bodø Hottest summer on record

The hottest summer on record for Bodø during Landsat 8 observation period (2013 - present) is 2023.

# separate df for bodø
df_bodø = df_summer[df_summer['sourceId'] == 'SN82310:0'].copy()
df_bodø = df_bodø.reset_index()
# sort by warmest summer
df_bodø.sort_values(by=['value'], inplace=True, ascending=False)
display(df_bodø.head(3))

	index	sourceId	referenceTime	elementId	value	unit	timeOffset
5	697	SN82310:0	2023-06-01 00:00:00+00:00	mean(air_temperature P3M)	14.1	degC	PT0H
3	689	SN82310:0	2020-06-01 00:00:00+00:00	mean(air_temperature P3M)	13.5	degC	PT0H
2	685	SN82310:0	2019-06-01 00:00:00+00:00	mean(air_temperature P3M)	13.2	degC	PT0H