Every EU country has a renewable energy target to be met by 2020 where the target is set as a percentage of gross final energy consumption. Since most countries are using combinations of hydro, solar and wind electricity (primary electricity) to achieve their targets, one needs a way of comparing primary electricity with energy from coal, oil and gas etc. The standard way adopted by the EU and by BP is to convert all forms of energy to tonnes of oil equivalent (toe). If coal, oil or gas is used to make electricity then there are large thermal losses doing so. BP account for this by grossing up renewable electricity by a factor of 100/38 (2.63) to account for “thermal gain” when converting from primary electricity to a fossil fuel equivalent. The EU does not do this, hence the toe figures reported by the EU and BP differ. Which methodology is correct?
In September last year, Roger Andrews had a post called EU Renewable Energy Targets: The Compliance Statistics are Suspect which raised a number of interesting issues without getting to the root of the problem. I decided to follow up with a more detailed analysis and quickly wished I hadn’t since I had to venture into the chaotic world of Eurostat – the EU energy statistics portal. Click a link in Eurostat with the expectation of learning something and you will be disappointed in a link merry-go-round. But every now and then you may get lucky and actually find some data.
If you get really lucky you stumble upon SHARES where it is possible to actually download a spreadsheet with RE data from 2004 to 2014 (note that I still don’t know how to find this page through the Eurostat portal and only came upon it by chance on Google with some help from Roger). But then compare this data with BP and you will find there is no semblance of similarity between the data sets as illustrated for hydroelectric power (Figure 1). One of the data sets is clearly wrong, but which one? Solar and wind data show the same discrepancy.
Figure 1 Hydroelectric power for EU28 as reported by BP 2015 and by the EU in SHARES. How can the two sources of data be so different? And which is correct?
The answer lies in how primary electricity (hydro, solar, wind etc) is normalised to the common datum of tonnes oil equivalent (toe).
BP give the following conversion factor:
1 toe = 12 MWh of electricity
If we take BP’s figure for EU28 hydroelectric production in 2014 of 370.3 TWh and convert using the above factor we get:
370.3 TWh / 12 MWh = 30.9 Mtoe
But if we then check what BP report on their Mtoe tab we find that they report 83.8 Mtoe for EU28 hydro in 2014 compared with 30.0 Mtoe reported by SHARES, which is the answer one gets converting using 12 MWh / toe (see above). At face value it appears that BP have made a mistake.
Its easier to think of this problem going in the other direction from fuel to electricity. Let us imagine that instead of normalising to toe that all energy was normalised to TWh instead. If we were to take our tonne of oil and burn it somehow in a generator we know that a large portion of the energy it contained will be lost as waste heat. We will not get our 12 MWh of electricity. We may get something closer to 4.6 MWh instead assuming 38% efficiency. Now, if we had 4.6 MWh of hydroelectric power that has no thermal losses associated with its production, we would need to gross that up by 100/38 if we were to compare it to oil or oil equivalents to account for thermal gain going in the opposite direction. And this is exactly what BP do noting at the foot of every spreadsheet the following:
* Based on gross primary hydroelectric generation and not accounting for cross-border electricity supply. Converted on the basis of thermal equivalence assuming 38%.
So let’s see if we correct the EU SHARES figures in this way if we resolve the issue (Figures 2, 3 and 4).
Figure 2 Hydroelectric power for EU28 expressed as toe. BP already gross up production for “thermal gain” going from primary electricity to fossil fuel equivalent using a factor of 100/38. Applying the same correction to the Eurostat SHARES data brings the BP and EU data into line.
Figure 3 See caption to Figure 2. Applying the same correction to solar provides a very close alignment between the BP and EUR data.
Figure 4 See caption to Figure 2. Applying the same correction to wind provides a close alignment between the BP and EUR data.
From Figures 2, 3 and 4 it should be clear that grossing the EU SHARES data up by 100/38 resolves the discrepancy between the two data sets. We can conclude this with a high degree of certainty. It is less certain whether the EU data should be adjusted up or the BP data adjusted down. I have been convinced in the past that BPs methodology was correct but based on opinions from experienced reviewers I am now undecided. This, therefore is a topic open for debate.
There are compelling reasons for wanting to compare the amounts of energy produced from different sources but this is a technical and philosophical challenge and the rules need to be clearly set out. The Eurostat SHARES page says this:
The use of renewable energy sources is seen as a key element in energy policy, reducing the dependence on fuel imported from non-EU countries, reducing emissions from fossil fuel sources, and decoupling energy costs from oil prices. Directive 2009/28/EC on promotion of the use of energy from renewable sources established accounting criteria for the 2020 targets on renewable energy sources.
Directing us to Directive 2009/28/EC that has 97 clauses, 29 Articles and 7 annexes we find the rule for normalising hydroelectric and wind power in Annexe II:
This is opaque as mud and while I have not close read the whole directive I cannot see reference to how one should convert from primary electricity to a fossil fuel equivalent. In a vast ocean of blether the vital information is either lost or missing.
If one adopts the BP methodology then EU primary electricity production may actually be much higher than everyone thinks and needs to be increased by a factor of 2.63 when expressed as toe.
The consequences of these findings are potentially far reaching and one point merits further investigation and that is this conversion factor from BP:
1 toe = 12 MWh of electricity
If that has built in conversion losses in power stations then it would be BP that was at fault. And so some further checking is in order this time using conversion factors from The World Energy Council. I want to check this conversion using thermal units.
1 toe = 10.03 Gcal
1 KWh = 860 Kcal
hence 1 toe = 10.03 cal^9 / 860 cal^3 = 11.7 MWh
BP are correctly using thermal equivalence. The question is, should imaginary thermal losses be factored into converting from one imaginary toe of hydro electric power to electricity?