Weather API

Weather API is a free service that offers information on forecasted, real-time, and historical weather, including marine weather.

This component enables users to load data from Weather API and bring it into Keboola Connection.


To use Weather API, you must first register to obtain an API token. Subscribe to the plan that fits your expected consumption.


  • Authentication
    • Api Token: Your Weather API token
  • Fetching Settings
    • Parameters From
      • Choose Configuration Parameters to define the location, request type, and date range directly in the configuration.
      • Choose Using an Input Table to set the fetching parameters dynamically from an input table. This approach is beneficial when determining the location based on data from an upstream pipeline.
    • Request Type
      • Choose Forecast to obtain a forecast (use Forecast days to specify how many days ahead you want data for).
      • Choose History to retrieve historical data. Use Historical Date to specify the period in the past.
    • Location Query (location_query): A query parameter for location. It can be latitude and longitude in decimal degrees, e.g., 48.8567,2.3508, a city name, e.g., Paris, or even an IP address like Learn more in the documentation.
    • Forecast Days (forecast_days): The number of forecast days required.
    • Historical Date (historical_date): The date from which to fetch historical data, either in YYYY-MM-DD format or a relative date like last week.
    • Continue On Failure (continue_on_failure): A Boolean value. If set to True, the process will continue dispite fetching errors, and failed responses will be saved in the failed_fetches.csv file. If set to False, the component run will terminate with an error as soon as one request fails.
  • Destination Settings (destination_settings)
    • Load Type (load_type): If full load is selected, the destination table will be overwritten with every run. If incremental load is used, data will be upserted into the existing destination table. Full load overwrites the destination table each time.


If fetch_parameter_from is set to input_table, you can use a single input table to set fetching parameters.

  • The table must have a ‘location’ column, or both latitude and longitude columns to specify the data fetching location.
  • If the request type is forecast, you can include a forecast_days column. It will define the forecast_days configuration value. If not provided, the forecast_days value will default to 10.
  • If the request type is history, a historical_date column must be added. It will define the historical_date configuration value.

Each row in the input table represents a single API request.


There are four output tables:

  • weather_astronomical.csv: Contains daily historical and future astronomical data.
  • weather_daily.csv: Contains daily forecasts and historical data.
  • weather_hourly.csv: Contains hourly forecasts and historical data.
  • failed_fetches.csv: If the ‘continue on failure’ parameter is set to true, this table will record any errors that occurr during data fetching.