Skip to content

Coverages

Weather forecasts from Météo-France APIs are based on key parameters that vary from indicator to indicator, which makes them complex to use.

Understanding coverages is a must to have a comprehensive usage of Météo-France forecasting models like AROME, AROME INSTANTANE, ARPEGE or PIAF.

Coverage_id

Introduction

A coverage_id looks like that:

WIND_SPEED__SPECIFIC_HEIGHT_LEVEL_ABOVE_GROUND___2024-01-16T09.00.00Z

It contains several information in a single string:

  • WIND_SPEED: Indicates that the data pertains to wind speed.
  • SPECIFIC_HEIGHT_LEVEL_ABOVE_GROUND: Specifies that the measurement is taken at a particular height above the ground.
  • 2024-01-16T09.00.00Z: Represents the date and time of the measurement, in ISO 8601 format (January 16, 2024, at 09:00 UTC).

if you want to display all valid coverage_ids, you can use get_capabilities:

1
2
3
4
5
from meteole import AromeForecast

arome = AromeForecast(application_id=application_id)

arome.get_capabilities()

Time-series coverages

Some coverages can contain an additional suffix named interval:

TEMPERATURE__SPECIFIC_HEIGHT_LEVEL_ABOVE_GROUND___2024-01-16T09.00.00Z_PT1H

PT1H Specifies the interval, meaning the data is provided at 1-hour intervals.

When no interval is specified, it means coverage returns a single datapoint instead of a timeseries.

Others parameters

Forecast_horizons

The time of day to which the prediction corresponds must be specified. For example, for a run of 12:00, in 1 hour's time, we have the weather indicator prediction of 13:00. The get_coverage method takes as (optional) parameter the list of desired forecast hours (in dt.timedelta format), named forecast_horizons.

To get the list of available forecast_horizons, use the function get_coverage_description as described in the example below.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
from meteole import arome

arome_client = arome.AromeForecast(application_id=APPLICATION_ID)

# Create coverage_id
capabilities = arome_client.get_capabilities()
indicator = 'V_COMPONENT_OF_WIND_GUST__SPECIFIC_HEIGHT_LEVEL_ABOVE_GROUND'
coverage_id = capabilities[capabilities['indicator'] == indicator]['id'].iloc[0]

# get the description of the coverage
coverage_axis = arome.get_coverage_description(coverage_id)

# retrieve the available forecast_horizons
coverage_axis['forecast_horizons']

Heights and Pressures

Atmospheric parameters can be measured at various heights and pressure levels, providing comprehensive data for weather analysis and forecasting. In consequence, some coverages must be queried with a height or pressure parameter.

To get the list of available height or pressure parameters, use the function get_coverage_description like forecast_horizons.

1
2
coverage_axis['heights']
coverage_axis['pressures']

Geographical Coverage

The geographical coverage of forecasts can be customized using the lat and long parameters. By default, Meteole retrieves data for the entire metropolitan France.