The report is about heat demand and CHP (Building a roadmap for heat. 2050 scenarios and heat delivery in the UK) and was prepared by University of Surrey and Imperial College for the Combined Heat and Power Association (CHPA). On p. 18 it has the following Sankey diagram. I have shown a similar diagram for the UK in this post.

Data is from the Digest of UK Energy Statistics. All flows are in millions of tonnes of oil equivalent (MTOE). Primary energy demand in 2007 was 237 MTOE.

The second Sankey diagram presented (on page 23) is a scenario for 2050. It was calculated using the MARKAL model.

One must read all the assumptions made for the model to be able to interpret it, but you can see immediately that the “energy system in 2050 is signifcantly altered under the common assumptions presented in all-electricity scenarios. In particular, final energy consumption in 2050 will be reduced by 46% against 2007 figures under the assumptions used in the CCC 80% CO2 reduction scenario”.

I invite you to read chapters 3.3. and 3.4 of the report to better understand the 2050 Sankey diagram. Note that the overall primary energy demand is significantly lower, but power generation almost doubles compared to the current situation. Losses from oil refineries are omitted in this scenario due to lack of data.

A great Sankey diagram by the research group made up from researchers from ICEPT (Imperial Centre for Energy Policy and Technology) and Centre for Environmental Strategy at the University of Surrey.

The Sankey diagram shows the energy flows over a 30-month test period from July 1993 to December 1995. Out of the 4981 kWh of energy, 2124 kWh were produced by the diesel engine, while 2857 kWh came from the 18 sqm installed PV modules. A storage battery is installed, and the diesel generator can be used to recharge it if there is no sunshine. The overall efficiency of the system was 61% over the testing period, but could be higher after the energy measuring equipment (consuming 499 kWh itself) was removed after the testing period.

The Sankey diagram shows actually measured values in kWh. It has a top-to-bottom orientation. Diesel is on the left side, the photovoltaic system on the right. Losses branch out horizontally. The battery is the blue box labeled “DC-Ges.” in the middle.

This Sankey diagram is pretty much messed up, and definitely a candidate for the “Worst Sankey Diagram Contest” that has already been called for. It took me a few seconds to understand that the flows dangling vertically below the blue arrow are actually a breakdown of the 177 mtep consommation finale. Vraiment … j’ai vu mieux que ça!

This is more or less how I would do it. Less colors, a breakdown of the blue flow into the five consumption sectors.

Below are three image sections of a Sankey diagram showing Swedish energy flows in 2007. These were published on p. 54/55 in the report ‘Energy in Sweden 2008′ by Swedish Energy Agency (Energimyndigheten). Numbers are in TWh, source for the diagram was Statistics Sweden data.

The first diagram (which is the left part of the diagram) shows the total 624 TWh energy supplied and the energy sources. The second one is the breakdown of energy produced by sectors, and also shows the losses in energy conversion. The third on is a different breakdown of the 404 TWh by energy carrier.

Nothing more to add. They are simple, clean, beautiful.

The energy flow Sankey diagram below (created by Marshall University for WV Energy Division) was presented at last year’s energy summit, and is available along with the other presentations here.

Figures are for 2006 in trillion BTU. Energy carriers used in WV are displayed as flows entering from the left. Domestic energy sources are in orange, while imported supplies are in sand color. Overall energy in 2006 was 4,384 trillion BTU. The state exported 81% of the energy (blue) and consumed 19% within (836 trillion BTU). A breakdown of doemstic consumption by sectors is shown in pink.

This Sankey diagram looks nice, but violates the basic rule for Sankey diagrams: flows have to be to scale among each other. The magnitude of the stacked orange arrows (representing 4100 trillion BTU West Virginia production) should be 14.5 times larger than that of the sand color flow (representing 283.86 trillion BTU), however it is only about 9 times larger, overemphasizing external supply (or underrepresenting domestic energy supply).

Or, compare the two arrows fro “crude oil” (10.14) and “natural gas” (230.12). The latter should have 23 times the width of the other… The blue arrow for “international raw coal” (392) looks approximately as wide as that of the blue “natural gas” (106). The scale might still be somehow OK for the base of the arrow, but as the arrow becomes thinner towards the head, the 4:1 ratio is definitely not supported any more.

The two diagrams show energy consumption and use in Toronto (above) and Helsinki (below). Even though the absolute figures in GWh are given, one shouldn’t directly compare them. A per capita basis would be fairer (Toronto had a population of 2.5 mio in 2006, more than 5 mio. in the metro area, while Helsinki had 580.000 inhabitants in 2008 in the city, 1.3 mio in the greater Helsinki area — Toronto is today 4.5 times larger than Helsinki). Both are “cold-climate municipalities”.

The publication calculates a ‘community energy efficiency’ of 50% for Toronto and 68% for Helsinki. “A comparison of the two municipalities reveals that Helsinki significantly improved its efficiency by using the waste heat that is produced by local coal power plants to warm 90% of the buildings and homes in Helsinki. Further analysis has demonstrated that Helsinki’s energy system was able to achieve its overall level of 68% efficiency because the city’s compact land-use pattern made investments in energy-saving infrastructure, such as district heating and public transit, economically viable.”

Does anybody know other metropolitan energy flow studies? I am aware of research activities in the field of urban material flow accounts or urban metabolism (e.g. Lisbon) but have to check whether they show Sankey diagrams in their publications.

The Sankey diagrams were presented in May 2008 at the Seventh Annual Conference on CCS in Pittsburgh, PA on a poster titled ‘Strategies for Improving Efficiencies in Oxy-Combustion Retrofits’. See the PDF here.

The first diagram shows how heat losses occur at the boiler and at the stack. Only a part of the energy flow at the power generator can be used, while a large chunk is thermodynamic loss. In the second diagram, heat from the boiler and compression is recovered and fed back.

A colorful one for the weekend…

Unfortunately the diagram is too small to grasp the details. The Sankey arrows represent energy content. Losses are shown as black arrows. This seems to have been the result of a study project dealing with a fuel cell driven boat (ZEMSHIPS).

Sustainable Energy Ireland has published the report “Energy Ireland – Key Statistics 2008″.

This is the updated overall energy flow Sankey diagram from the report.

And here is another interesting one from the same report. It features the fuels used for transportation in Ireland with a breakdown by vehicle type.

All flows are in ktoe (1000 tonnes of oil equivalent, 1 ktoe = 41868 GJ). The flows seem to be mostly to scale, although there must be a problem with the flow ‘Refining Losses 6 ktoe’ which seems wider than the ‘Rail 47 ktoe’ flow, for example. Private cars consume most energy of all vehicle types (2184 out of 5685 ktoe or 38.4 %) , followed by trucks (road freight) and airplanes.

I won’t be commenting on the 3D design of the Sankey diagrams, which is definitely … eye-catching. Colm asked which software has been used to create these, but another reader had already pointed out in a comment that it is based on SankeyHelper, but has undergone retouching in Illustrator.

This is the U.S. Energy Flow diagram for 2008

Check the original PDF file with the accompanying footnotes for further details. Overall energy consumption in 2008 was 99 Quadrillion BTUs (preliminary value, slightly down from the 101 Quadrillion BTUs in 2007.

Among the other diagrams in the report, I chose to show the one for coal. 1121 mio. short tons have been consumed in the U.S. in 2008, mainly (1041 mio short tons) for electric power generation. The U.S. is a net coal exporter.

The diagram has a weird sinking downward feeling, caused by the fact that the main left-to-right orientation axis is not maintained. Looking at this it makes me want to shout out: “Hey coal Sankey, cheer up, life isn’t that black…!”

The original full AER report (7.5 MB) can be found here.

The diagram shows the main flows in gross PJ (1 petajoule = 10¹⁵ J) and is “to approximate scale”. Flows below 2.5 PJ are neglected.

Energy produced domestically from various sources comes from the left, imports of coal and oil enter in the leftmost column from the bottom and the top. The energy flows pass through transformation and conveyance phases, to be finally shown in the different use sectors. Losses in transformation and conveyance are visualized with downward arrows, while losses in the end use are not considered.

There are some design flaws, especially when you look at the arrow curves. Also the fact that flows are only “to approximate scale” is in my opinion not acceptable [an arrow representing a flow of 3 PJ has the same width as one for 6 PJ, and both are only half the width of the 30 PJ flow]. But the overall impression is much better than in the version 10 years before.

Click here for a larger version of the diagram (PDF).

The text under the diagram reads: “This energy flow diagram summarises New Zealand’s energy use. Primary energy sources are at the left. The flow of these through conversion processes to consumers is pictured, with final end-use classified by consumer type. The width of the bands is approximately to scale.”

Well, almost everything that can go wrong in information visualization goes wrong in this diagram… No quantities or units are given. Flows that are “approximately to scale” narrow down along the way. [The only explanation I have for this, is that this a novel way to account for transmission losses.] Streams meet, but don’t seem to merge. No idea what the spaghetti flows are good for…
In defense of the authors of this diagram I can say that in 1997 there probably weren’t any Sankey diagram software tools around.

I have a nicer one for NZ, which I will present in of my next blog posts. Better energy flow Sankey diagrams from other countries can be seen here, here or here.

Data is from an Energy Star flyer, that presents the data in a pie chart. The average yearly 1900 US$ for energy are spent as follows:

Heating and cooling is almost half of the spendings on energy, followed by water heating and lighting.

These energy balances are apparently produced annually by French DGEC (Direction Général de Énergie et Climat) and are different from the energy flow Sankey diagrams for various countries I have shown here in other posts.

In these French energy balances the left side shows the primary energy (expressed in mégatonne équivalent pétrole Mtep, mega tonnes of oil equivalent Mtoe in English) broken down into the different sources, the right side shows the final consumption of energy with the same breakdown. The differences are the energy losses and internal consumption in energy generation.

The large losses for nuclear energy are explained in a footnote for the 2007 diagram, where it says that these can be attributed to the fact that ‘Observatoire de l’Énergie’ (OE) accounting standard is based on heat value:

L’importance des pertes dans le domaine de l’électricité tient largement au mode de calcul adopté depuis 2002 par l’OE: l’électricité d’origine nucléaire est comptabilisée, au niveau de la production, en termes de chaleur, dont les deux tiers sont perdus lors de la conversion en énergie
électrique

Nevertheless the Sankey diagram gives a good idea about the efficiency of energy generation for each of the different fuels and the energy mix for France.

For those of you who wish to compare the developments over the last years, here is what I have dug up so far:
2003 is shown on Planete Bleue in this post.
2004 is shown above
2006 can be found here on page 25 (PDF)
2007 can be found here on pages 6/7 (PDF)

For the UK these Sankey diagrams are published by BERR (Department for Business, Enterprise and Regulation). Historic charts back to 1974 can be found on their website.

A commentor to Chris Vernon’s original post added the Swiss version of this Sankey energy flow diagram:

These Sankey diagrams have some nice details, which a worth a mentioning: The UK diagram shows stock increases and decreases with circles. The size of the circle has no significance, but the magnitude of the in and out flows seem to be to scale, thus allowing to see if the stock has increased or decreased in that year. Losses at transformation steps are shown with “hanging arrows” (the flows branching off to the bottom line of the diagram). The Swiss version also shows exports this way, but visualizes losses with a pin with a big round head.

Even though your run the risk of being overwhelmed by a gloomy feeling when your read through the comments to Chris Vernon’ post, I would like to draw your attention to a comment by “realist” on Sept 5. He writes: “Deceptive graph! Why show losses for electric power generation and not transportation? The heat losses from the internal combustion engine in most transport is 70-80%”. This is true, but I have always understood that losses explicitly shown in these energy flow Sankey diagrams are the losses occuring in the energy generation, conversion and grid, while losses in the energy consumption (such as use for transport) are not shown. This let alone that they are worthwile to discuss.