Rules of the information cost calculation

Basic concepts

The tariff for ECMWF Products is evaluated on the basis of an 'information unit' introduced by ECOMET, the EPU (ECOMET Product Unit).

The tariff is based on the cost of a global field at the resolution of the Centre's current resolution of the Atmospheric Model high resolution 10-day forecast (HRES), independent of its representation (spherical harmonics, regular and reduced Gaussian grid, regular latitude-longitude grid).

A unit price per EPU is applied to all Products in the Catalogue.

Unit Price (P)

A unit price of P = 0.25 EUR per EPU is applied to all Products in the Catalogue.

Please note that from 1 July 2024, the unit price of P = 0.15 EUR per EPU will be applied to all Products in the Catalogue.

Volume discount

The following discount scheme, based on the current unit price P = 0.25 EUR, applies.

Number of chargeable EPUs per year [N]	Discount Price [EUR]
0 - 2,000	N x P
2,001 - 20,000	[2,000 + 0.6 (N - 2,000)] x P
20,001 - 200,000	[12,800 + 0.4 (N - 20,000)] x P
200,001 and more	[84,800 + 0.2 (N - 200,000)] x P

Priced Items

An item is defined as one or more fields that are priced as one entity.

Below are the definitions of item based on the different Product Sets defined in the Catalogue:

Product Set		Definition of item for HRES & WAM HRES
I-i	Atmospheric fields	A single field
I-ii	Time series	The cost of a time-series is set to the cost of 8 parameters from the dataset, at full resolution, at all time steps in the time-series over the minimum area
II-i	WAM HRES (Global)	A single field. The 2D-wave spectra (all frequencies and directions) corresponds to 10 items.

Product Set		Definition of item for ENS
III-i	Atmospheric fields	All members for each field including perturbed and control forecasts and the equivalent deterministic product (same resolution).
III-ii	Clusters	All cluster members for each field
III-iii	Probabilities	A single field
III-iv	Time series	Please go to the section time series
III-v	Extreme Forecast Indexes	A single field
III-vi	Ensemble means	A single field
III-vii	Ensemble standard deviations	A single field
IV-i	Wave fields	All members for each field including perturbed and control forecasts and the equivalent deterministic product (same resolution).
IV-ii	Probabilities	A single field

Product Set		Definition of item for ENS extended (monthly)
VI-i	Weekly means of ensemble means	A single field and the equivalent reforecast data
VI-ii	Weekly mean anomalies of ensemble means	A single field
VI-iii	Weekly means of individual ensemble member forecasts	All the weekly means for each field and the equivalent reforecast data.
VI-iv	Weekly mean anomalies of individual ensemble member forecasts	All the weekly means for each field.
VI-v	Individual forecast runs (real-time and reforecast)	All the ensemble members including perturbed and control forecasts for each field and the equivalent reforecast data.
VI-vi	Probabilities	A single field
VI-vii	Time series	Please go to the section time series

Product Set		Definition of item for SEAS
V-i	Monthly means of ensemble means	A single field
V-ii	Monthly mean anomalies of ensemble means	A single field
V-iii	Monthly means of individual ensemble member forecasts	All the monthly means for each field and the monthly means of ensemble means (V-i).
V-iv	Monthly mean anomalies of individual ensemble member forecasts	All the monthly means for each field and the equivalent monthly mean anomalies of ensemble means (V-ii).
V-v	Individual forecast runs (real-time)	All the ensemble members within one day for each field.

Price of Time Series

For a given dataset, the cost of a time-series is set to the cost of 8 parameters from the dataset, at full resolution, at all time steps in the time-series over 6 square degrees.

Time unit

The costing of data will be based on a time unit of :

hours for medium-range forecasts, and for ENS extended where individual time-steps are available
days for ENS extended and SEAS
weeks for weekly mean data from ENS extended
months for monthly mean data from SEAS

3.4 Unified costing formula

The price of a requested item is based on the following formula:

EPU per item = B * A * R * E * V * D

where

B is the base cost, set to 20 EPUs;
A is the area factor, which is the ratio between the area provided and the area of the globe. For ocean wave fields, this factor will in addition take into account the land/sea ratio over the provided area
R is the resolution factor, defined as the ratio between the resolution of the Centre's high-resolution operational model and the resolution provided (please note: 0.1 degrees for spectral truncation, 0.1125 degrees for lat/long grid);
E is the ensemble factor, set to 1.4 see table below;
V is the valid factor, set to 1 for real-time data;
D is the dataset factor, defined as follows:

Range	Set Number	Product Set	D	E
Medium-range	I-i	Fields from HRES	1	1
	II-i	Fields from WAM HRES	1	1
	III-i	Fields from ENS	1	1.4
	III-ii	Clusters from ENS	1	1
	III-iii	Probabilities from ENS	1	1
	III-v	Extreme Forecast Indexes from ENS	1	1
	III-vi	Ensemble means from ENS	1	1
	III-vii	Ensemble standard deviations from ENS	1	1
	IV-i	Fields from WAM ENS	1	1.4
	IV-ii	Probabilities from WAM ENS	1	1
Extended Range	VI-i	Weekly means of ensemble means from ENS extended	15	1
	VI-ii	Weekly mean anomalies of ensemble means from ENS extended	30	1
	VI-iii	Weekly means of individual ensemble member forecasts from ENS extended	15	1.4
	VI-iv	Weekly mean anomalies of individual ensemble member forecasts from ENS extended	30	1.4
	VI-v	Individual forecast runs (daily real-time and hindcast products) from ENS extended	1	1.4
	VI-vi	Probabilities from ENS extended	1	1
Long range	V-i	Monthly means of ensemble means from SEAS	40	1
	V-ii	Monthly mean anomalies of ensemble means from SEAS	80	1
	V-iii	Monthly means of individual ensemble member forecasts from SEAS	40	1.4
	V-iv	Monthly mean anomalies of individual ensemble member forecasts from SEAS	80	1.4
	V-v	Individual forecast runs (daily real-time) from SEAS	2	1.4