Loopbacks in Sankey Diagrams

My previous post on circular links got a number of responses. Indeed it seems as if drawing loopbacks (as I prefer to call them) is one of the tougher challenges in Sankey diagrams.

  • Loops back to the same node are typically not required in relationship diagrams (bands depicting relationship between categories, see here), but they may be necessary if you want depict actual physical flows (e.g. recycling of material)
  • Direct loops back, where the output of a process leads directly back to the same node, are not very common (one example can be seen here, called “functional recycling”). There could be in reality a node along the way back (e.g. a pump that pumps cooling water cycling the process). If you have a node in the loopback, then the input and output side are flipped.
  • Loops can “go back via several nodes, they may even branch on the way back” like in the example below


If you take a left-to-right column oriented approach when setting up the diagram working with tabulated data as the source (sth like “source”: “process3”, “target”: “process8”, “value”: 20) then you have to consider the column depth to identify whether you have a back loop. All sorts of routing issues for the arrow come up and you need to create room to not produce overlaps. Drawing the Sankey diagram manually (like I did i the figure above) rather than programmatically gives more freedom in that respect.

Sankeys with circular links

Guus, a reader of this blog, DMed me to ask whether I “knew of any (open source) JavaScript libraries that can generate circular Sankey diagrams”?

Many of the Sankey diagrams I see on the web are created using d3.js by Mike Bostock. Typically these diagrams are left-to right oriented and have a column structure. What is less common are feedback loops or circular Sankey arrows, like the pink ones in the figure below.

The place to look for is Tom Shanley’s Block.

Here you can find many samples for:

Guus, I hope you can find what you are after there. Enjoy!

Sankey Diagram for Air Compressor

A comment from a reader pointed me to this Sankey diagram for energy flows in an air compressor system. It can be found as a sample on the web page of VHK Research Engineers. VHK uses Sankey diagrams to “get the message across”.


(Image copyright: European Union, author: VHK)

This is a rare example of a top-down oriented Sankey diagram. There are no absolute numbers, rather we see the 100% energy being turned into useful work, and off-heat, which could potentially be recovered.

Nice piece of art, and I invite Rene to comment on this…

Energy Consumption Logistics Center

Here is another Sankey diagram out of Germany. Found this in a 2017 doctoral thesis titled ‘Wechselwirkungen und Auswirkungen von Planungsalternativen auf die Gesamtenergiebilanz und die CO2-Emissionen von Logistikzentren’ by Julia Freis, Lehrstuhl für Fördertechnik Materialfluss Logistik, Technical University Munich (TUM).

From what I understand using Google Translate this seems to be one of the energy scenarios (maintaining a 17°C temperature) for a logistics hub. Flows are in kWh per year.

China Carbon Emissions from Energy

Energy generation in China is dominated by the use of hard coal. This Sankey diagram is from an article titled ‘A Method for Analyzing Energy-Related Carbon Emissions and the Structural Changes: A Case Study of China from 2005 to 2015’ by Honghua Yang, Linwei Ma and Zheng Li (Tsinghua University) in: Energies 2020, 13, 2076; doi:10.3390/en13082076. It shows carbon flow and emissions (I take that as CO2 only, although it might include CH4 if biogas was used)


Published under the terms and conditions of the Creative Commons Attribution(CC BY 4.0) license

Depicted are energy-related carbon flows in China in 2015. Unit of flow is 10 Mt C, which in the last column also translates into Mt CO2.

The sectors Transport (“Vehicle”), Industry (“Factory”) and Buildings are further broken down into the individual services the energy provides, like illumination, thermal comfort, hygiene.

There is another energy flow diagram for China in 2015 in this article, and it shows that there are also other energy sources (hydro, wind, nuclear, solar), but these don’t show up in the carbon flow diagram.

Global Energy Flows 2018/2050, DNVGL

Fresh off the press last week is DNVGL’s Energy Transition Outlook 2020. This report is a “forecast of the world’s most likely energy future through to 2050” and in my opinion this is really good information. The new edition already factors in the effects of the pandemic.

The spread on pages 122/123 has the following Sankey diagrams, a comparison of the global energy landscape in 2018 and the forecast for 2050:


The individual flows are not labeled with quantities, but we get an idea from the blue stream, that represents 27 PWh/yr electricity in 2018 and 60 PWh/yr in 2050. “One striking change on the supply side of the picture is the emergence of solar PV and wind at the expense of coal and oil. Electrification more than doubles through to 2050, which leads to an increase in the overall system efficiency.”

Make sure to download your copy of the DNVGL Energy Transition Outlook 2020, to study the global energy flow Sankey diagram in more detail.

Material Flow Analysis for Almaty

Almaty is the largest city in Kazakhstan. It is also the first city in Central Asia that did a circular economy opportunities analysis. Numerous ideas were proposed during a project with local stakeholders.
The project is described on the Shifting Paradigms blog, and the full report ‘Metabolic analysis and circular economy strategies for Almaty, Kazakhstan’ describing the project can be downloaded there.

Visuals turned out to be crucial in this project for communicating information and to be able to oversee the amount of data. “Mapping out the metabolic system of a city, helps understand how a city uses material resources to deliver valuable services to its inhabitants, like nutrition, shelter and mobility, and identify opportunities for improvement.”

This is the Sankey diagram depicting the material flows for Almaty, Kazakhstan covering minerals, metals, biomass, fossil fuels, energy and water used in Almaty’s industry. It is shown on pages 36/37 of the report. Flows are in kT (per one year ?) with imports/exports across the city’s boundaries.

Blended into this material flow analysis (MFA) diagram are greenhouse gas emissions (GHGs). “The red flows at the lower section of the graph show the embedded gas emissions in imported goods and materials.” These could be considered hidden quantities associated with the production of products or the import of fuels, “piggybacking” on the actual physical material flows that enter and leave the industrial sector in the city.

Note that it seems as if flows are not always to scale or parts are hidden behind other flows (see, for example, the yellow stream representing 968 kT of fuel being turned into 2,975 kT GHGs, a flow that is not three times as wide). Also the recycling flow width seems to overblown, probably to point out that recycled materials can loop back through the system multiple times, and to focus the circular economy perspective.

There are two other Sankey diagrams in the report, so make sure you have a look at it.

France’s Grand Est Region Energy Flows

Le Grand Est is a French administrative region in the north-east of the Hexagone, comprising Alsace, Champagne-Ardenne, and Lorraine. ATMO Grand Est is a not-for-profit, government-backed association that is monitoring air quality in the region and looking at ways to improve it.

As part of their work they have prepared energy flow diagrams, not only for the whole region, but also on the communal level. More than 100 Sankey diagrams, depicting the energy flows and the use of energy are available on the ATMO website. Visit the resources section and check “Diagrammes de Sankey” in the filter list at the right.

Here is the figure for the overall region with energy consumption and use in Grand Est in 2017. Flows are in GWh.


On the left we see primary resources, with the largest source being nuclear fuel. With an efficiency of only 33% this still delivers most of the secondary energy consumed in the Grand Est region. The grey stream from the top are “imports” to the region, most likely from other parts of France (or from neighboring Germany). The energy consuming sectors are shown on the right. A table next to them points out the GHG emissions linked to the energy.

Each of the individual Sankey diagrams for the CCs (communauté de communes, rural communities) and some CAs (communauté d’agglomération, semi-urban communities) available have exactly the same structure, but they can look strikingly different.

Here is one example from CC Portes de Meuse, a community with just over 17,000 inhabitants


CC Portes de Meuse can apparently cover three-quarters of their energy demand from wind and wood, with only 25% being imported. (Not sure though how they can have 260 GWh energy use in road transport, unless most vehicles are already electric).

Note that although both Sankey diagrams have a similar structure and flows are both in GWh in 2017, this diagram mustn’t be compared directly to the one above in regard to the width of the arrows, since they are on a different scale.

Great work by ATMO Grand Est, and I am sure these visualizations are useful when discussing energy consumption, GHG and air pollution among stakeholders in the communities or the region. They have already announced the Invent’AIR v2020 version to be published in the near future.