Data indices for European Gas Prompt

DataHub is capable of handling LEBA (London Energy Brokers Association) European Gas Prompt dataset. The European Gas Prompt dataset comprises indices for the following hubs: Austrian VTP, French PEG, Italian PSV, German THE, Dutch TTF and Belgian ZTP.

These data have the following format :

Column name Format Data description
DATE date - yyyymmdd date of the forecast computation
INDEX [hub] [data index] eg : TTF Day Ahead Window => define the "type" of data. This value must be used as mapping config for the variable which is expected to hold the data
DELIVERY date - See accepted format below the date of the datapoint to load for this row
WA double the value of the datapoint to load for this row

Datahub configuration

To be able to load and access Leba data via DataHub, you need to create and configure some metadata in your account.

Every datapoints (which represents a value at a specific datetime) is linked to/hold by a variable. Every variable is linked to a source which is itself linked to a site.

Tip

Before continue this reading, we recommend you to be familiar with DataHub Data Model

Create a source

European Gas Prompt data are available for different European countries. There should be one source configured by country for which you want the data.

To be recognized by DataHub as a Leba source, the source must has a serialNumber following this pattern : LEBA_[hub]

Examples :

  • Leba European Gas Prompt data of hub "TTF" will be loaded on the source with a Serial Number = "Leba_TTF"
  • Leba European Gas Prompt data of hub "PEG" will be loaded on the source with a Serial Number = "Leba_PEG"
  • Leba European Gas Prompt data of hub "VTP" will be loaded on the source with a Serial Number = "Leba_VTP"

If the file contains data for a hub and no corresponding source exist, data won't be uploaded.

Create variables

On each Leba Source, a variable with a mapping config equals to the index of the data you want to load (see file format);

Example :

DATE,INDEX,DELIVERY,WA
20250307,TTF Day Ahead Window,05-Mar-25,45.025

For this row to be uploaded as a datapoint in DataHub, you need :

  • to have a source with a serial number = Leba_TTF
  • to have a variable on that source, with a mapping configuration = TTF Weekend Window

Friday Only variable

The weekend data have special processing rules. You have the possibility to load the weekend data only if their values have been calculated the Friday before this specific weekend. See Weekend Data for details.

In this case, the mapping config of the variable should be the index + 'Friday Only'

Example : PEG Weekend Window Friday Only

Create a Leba trigger

Finally, to react to the leba file uploading, you need to create and activate a 'Leba' trigger in your account.

Learn more on Trigger Management

File Format

The file loaded into DataHub must follows the following rules for the leba data to be ingested :

  • the file must be a csv
  • with the following columns and format :
Column name Format Data description
DATE date - yyyymmdd date of the forecast computation
INDEX [hub] [index] define the "type" of data (see Index). This value must be used as mapping config for the variable which is expected to hold the data
DELIVERY date - See accepted format below the date of the datapoint
WA double the value to load

Other columns will be ignored.

Index

The index column can have a whole range of values but for this implementation, we use only the following ones:

  • [hub] Day Ahead Window (ex : TTF Day Ahead Window)

  • [hub] Weekend Window

  • [hub] Christmas

  • [hub] New Year

where [hub] can be any of the European hub handled by Leba :

  • Austrian VTP,
  • PEG,
  • PSV,
  • THE,
  • TTF,
  • ZTP

At this point, all other indexes are ignored.

Example of valid index :

PEG Day Ahead Window TTF Weekend Window Austrian VTP Christmas PSV New Year

Date Handling

The delivery date in the csv file represents the date of the datapoint that will be loaded for this row.

Date format

The format of this input can be one of the following :

  • dd-mmm-yy => 20-Jul-19
  • yyyy-mm-dd => 2019-07-20
  • dd-mmm-yy to dd-mmm-yy => 25-Dec-19 to 26-Dec-19 : in this case, the value should be stored on both dates (2019-12-25 AND 2019-12-26). Data should be stored as is (no division or anything else).
  • Week xx - yyyy (where xx represents the week number) => Week 30 - 2019 => See week numbering
  • mmm-yy => Mar-24

Week numbering

The date format "Week xx - yyyy" (where xx represents the week number) represents the number of a week in the related year. The data should then be uploaded on the first day of this week.

There are several week numbering systems. In Europe, the most used seems to be the ISO week date system; That's the one used in the app to determine the first day of the week.

Gas Day

When the date of the datapoint is defined through the previous step (parsing date), it needs to be converted to a date following the gas day specifications.

The gas day is defined as

The Gas Day establishes the trading window for exchanging day-ahead and daily gas products on the EU trading platforms. The Gas Day has a harmonized start and end from 5.00 to 5.00 UTC the following day for winter time, and from 4.00 to 4.00 UTC the following day when daylight saving is applied.

The timestamp of the datapoint loaded into DataHub follows this rule :

  • in winter time : DELIVERY date + 05:00:00Z
  • in summer time : DELIVERY date + 04:00:00Z

Examples:

DATE,INDEX,DELIVERY,WA
20250303,TTF Day Ahead Window,04-Mar-25,45.304

=> generates a datapoint:

{
 "date" : "2025-03-04T05:00:00Z",
 "value" : 45.304
}

DATE,INDEX,DELIVERY,WA
20250804,TTF Day Ahead Window,05-Aug-25,45.304

=> generates a datapoint:

{
  "date" : "2025-08-05T04:00:00Z",
  "value" : 45.304
}

Weekend Data

The weekend data is estimated multiple days ahead of the weekend. You have the possibility

  • to load the data every time it is received on a file, on a variable having a mapping config matching the Index value (as for every other index)

  • to load the same data only, on another specific variable (with a mapping config equals to the INDEX column + "Friday Only"), only when the Date in the DATE column (the date of the computing of the forecast) is the Friday just before the weekend represented by a the date range in the DELIVERY column.

Examples :

DATE,INDEX,DELIVERY,VOLUME,WA,STATUS,FLAG
20250303,TTF Weekend Window,08-Mar-25 to 09-Mar-25,30,45.025,4,N
  • the 03-03-2025 is a Monday
  • the data is loaded on the variable with the mapping config "TTF Weekend Window" but not on the variable specific for the Friday (with a mapping config "TTF Weekend Window Friday Only")
DATE,INDEX,DELIVERY,VOLUME,WA,STATUS,FLAG
20250228,TTF Weekend Window,08-Mar-25 to 09-Mar-25,30,45.025,4,N
  • the 28-02-2025 is a Friday
  • the 28-02-2025 is NOT the Friday just before the dates in the Delivery columns (08-03-2025 and 09-03-2025)
  • the data is loaded on the variable with the mapping config "TTF Weekend Window" BUT NOT on the variable with mapping config "TTF Weekend Window Friday Only"
DATE,INDEX,DELIVERY,VOLUME,WA,STATUS,FLAG
20250307,TTF Weekend Window,08-Mar-25 to 09-Mar-25,30,45.025,4,N
  • the 07-03-2025 is a Friday
  • the 07-03-2025 is the Friday just before the dates in the Delivery columns (08-03-2025 and 09-03-2025)
  • the data is loaded on the variable with the mapping config "TTF Weekend Window" and on the variable with mapping config "TTF Weekend Window Friday Only"

Examples

DATE,INDEX,DELIVERY,WA
20250303,TTF Day Ahead Window,04-Mar-25,45.304
20250303,TTF Weekend Window,08-Mar-25 to 09-Mar-25,45.025
20250307,VTP Weekend Window,08-Mar-25 to 09-Mar-25,45.025
20250303,TTF Month Ahead,Apr-25,45.653

First data row

=> will produce a datapoint on the variable with a mappingConfig = 'TTF Day Ahead Window' on a source with a SerialNumber = "Leba_TTF"

{
 "date" : "2025-03-04T05:00:00Z",
 "value" : 45.304
}

Second data row

=> will produce two datapoint to be loaded on the variable with a mappingConfig = 'TTF Weekend Window' on a source with a SerialNumber = "Leba_TTF";

{
 "date" : "2025-03-08T05:00:00Z",
 "value" : 45.025
}

{
  "date" : "2025-03-09T05:00:00Z",
  "value" : 45.025
}

Third data row

=> will produce two datapoint to be loaded on the variable with a mappingConfig = 'VTP Weekend Window' AND two datapoint to be loaded on the variable with a mappingConfig = 'VTP Weekend Window Friday Only' on a source with a SerialNumber = "Leba_VTP"

{
  "date" : "2025-03-08T05:00:00Z",
  "value": 45.025
}

{
  "date" : "2025-03-09T05:00:00Z",
  "value" : 45.025
}

Fourth data row

=> will be ignored as the index '[hub ]Month Ahead' is not a index handled by DataHub at this time.