@fneum
Three points before a real review:

• I am wondering, should we remove the rescaling function rescale_idees_from_eurostat or keep it in the script even though the function is not used right now for later, when we have updated eurostat data and JRC data only until 2021?
• The data for the agriculture sector for GB is missing. I am not sure, what to do about it. Take Eurostat final energy consumption for the agriculture sector and assume similar split between heat/electricity/oil as for the other countries?
• there are clearly some difference between the old and new JRC-IDEES data for e.g. 2015. I think they are small enough to not dig deeper into the reason of the changes, what do you think?

I am wondering, should we remove the rescaling function rescale_idees_from_eurostat or keep it in the script even though the function is not used right now for later, when we have updated eurostat data and JRC data only until 2021?

I would keep it for exactly this reason.

The data for the agriculture sector for GB is missing. I am not sure, what to do about it. Take Eurostat final energy consumption for the agriculture sector and assume similar split between heat/electricity/oil as for the other countries?

Is there an analogy for the other countries that were previously not in JRC-IDEES? Then, I would take that.

there are clearly some difference between the old and new JRC-IDEES data for e.g. 2015. I think they are small enough to not dig deeper into the reason of the changes, what do you think?

Perhaps let's do a few spot checks if the entries directly in the Excel files have changed to exclude that they have maybe disaggregated some entries further in the new version and we are missing them. I agree that the changes look mostly small.

For the missing agriculture data for GB eurostat data for the overall demand is used and split by EU average into electricity, heat and machinery. The data in Eurostat for GB for 2020 and 2021 is missing (as well as for some countries for other years). In this case, data is filled with country data averaged over all years.

One should be aware the Eurostat Agriculture final energy consumption is not always 1:1 the JRC IDEES final energy consumption

Just as an update on the data validation:

• Forecast seems to be using the old Eurostat data for iron and steel (13 447 ktoe for DE in 2015)
• AGB Bilanzen for DE 2015 have for primary energy consumption in blast furnaces ~181 PJ = 4 323 ktoe which is more in line with the new Eurostat data and JRC-2021 with around 6 418 ktoe for integrated steel works in DE

Forecast seems to be using the old Eurostat data for iron and steel (13 447 ktoe for DE in 2015)
(AGB Bilanzen for DE 2015](https://ag-energiebilanzen.de/wp-content/uploads/2023/11/EBD15e.xlsx) have for primary energy consumption in blast furnaces ~181 PJ = 4 323 ktoe which is more in line with the new Eurostat data and JRC-2021 with around 6 418 ktoe for integrated steel works in DE

One further check could be via the industry sector ratios and compare to other sources, e.g. http://www.energybc.ca/cache/coal/www.worldcoal.org/coal/uses-of-coal/coal-steel/index.html

Around 0.6 tonnes (600 kg) of coke produces 1 tonne (1000 kg) of steel, which means that around 770 kg of coal are used to produce 1 tonne of steel through this production route.

It's a bit curious that coke and coal demand are so drastically reduced, while electricity was not affected.

Will ask FORECAST team how they interpret the big changes in HVC and integrated steelworks.

Ok, great, then let's wait for their feedback before merging...

After the adjustments for the chemical feedstock and the integrated steelworks industrial_energy_demand_per_country_today looks like that for Europe 2015

@fneum maybe one final look from your site before the merging?