This is quite an interesting Sankey diagram from the World Energy Outlook 2014. It visualizes international spending on energy efficiency measures in the transport sector under a hypothetical ‘New Policies Scenario’.

A total of 14.5 trillion US$ would be spent until 2040 to improve energy efficiency in the transport sector. The largest chunk (37%, 5.3 trillion US$) on improving private cars. This amount is further broken down to four geographic regions. The money would be spent mainly on improving the power train, and on development of light-weight components.

The underlying scenarios are described in detail at the beginning of the WEO-2014 study. The authors point out that “[f]or each scenario, we offer a set of internally consistent projections to 2040. None should be considered forecasts.”

“The New Policies Scenario is the central scenario of WEO-2014. It takes into account the policies and implementing measures affecting energy markets that had been adopted as of mid-2014, together with relevant policy proposals, even though specific measures needed to put them into effect have yet to be fully developed. These proposals include targets and programmes to support renewable energy, energy efficiency, and alternative fuels and vehicles, as well as commitments to reduce carbon emissions, reform energy subsidies and expand or phase out nuclear power.”

The article ‘Aprovechamiento de la energía procedente del frenado regenerativo en ferrocarriles metropolitanos’ by Álvaro López López published in the Spanish journal ‘Anales de Mecánica y Electricidad (May/June 2013)’, pp 12-18 has the following Sankey diagram.

No absolute numbers are given here. Still, we understand that from the motion energy during braking of the train a part (green flow) can be recovered and is being used for secondary systems (‘SSAA’) as well as being fed back into the overhead wire (‘cantenaria’).

Not sure though whether this Sankey diagram is a representation of the energy recovery during braking action only, or of the energy flows on a typical train ride.

South China Sea has recently garnered increased media attention due to China reclaiming land and building an airfield on Fiery Cross Reef. The territorial dispute regarding Spratly Islands has been simmering since the 1970ies when oil was discovered in the region. South China Sea is also “one of the busiest shipping lanes in the world” with “more than half of the world’s supertanker traffic, by tonnage, pass[ing] through the region’s waters every year” (Wikipedia).

The Department of Energy has two interesting maps on their beta website showing LNG and crude oil transport for 2011.

Transport of liquefied natual gas (LNG) in trillions of cubic feet in the South China Sea:

Transport of petroleum in millions of barrels per day in the South China Sea in 2011:

(both maps from website)

These are ‘Sankey-inspired maps’ rather than exact Sankey diagrams. Arrow widths are not maintained where the shipping routes pass through narrow straits. Nevertheless, transport volumes are generally on a correct scale.

I was asked if Sankey diagrams could meaningfully be used to visualize passenger loads on a tram or bus line. Here is what I came up with:

These are fictitious values. I just labeled the stops A, B, C, … and decided to go for a short feeder line. At the last stop all passengers get off (e.g. to transfer to a train).

At each stop there are passengers that get on (green) and get off (red). The number of pax on the bus is shown by the blue arrows.

The profile would probably look differently at different times of day, so depending on the data availability one would have to create diagrams for off-peak/peak hours, weekdays/holidays and so on.

Your thoughts?

From a project summary on the webpage of the Fuel Cell Research Lab at University of Delaware’s Department of Mechanical Engineering comes this Sankey diagram.

This is for a bus operating on the University of Delaware campus. The Sankey diagram shows energy flow and losses in the hybrid power train for a typical drive cycle. Unit is Wh, percentages are given in the labels as additional information. Energy is recovered when braking and is fed back to the battery (see upstream arrow ‘Energy Recovery’).

“The fuel cell system balance of plant consumes a significant fraction of the energy of the hydrogen supplying the stack, so efficiency gains there are potentially quite useful. Most of the balance of plant energy feeds the air compressor, so efficiency could be increased by improving air humidification to allow lower air system backpressure”

Simple black and white diagram with a top-down orientation. The only extra that does not serve to carry information is the schematic road figuring at the bottom….

For the full publication check Bubna P., Brunner D., Gangloff Jr. J.J., Advani S.G., and Prasad A.K., “Analysis, operation, and maintenance of a fuel cell/battery series-hybrid bus for urban transit applications,” Journal of Power Sources, Vol. 195, pp. 3939-3949, June 15, 2010. doi:10.1016/j.jpowsour.2009.12.080

John Cochran blogs about his coursework at University of Virgina. His project on ‘Urban Metabolisms’ has this Sankey diagram of food being transported to New York City. Data is from The Federal Highway Administration (USDOT) Freight Analysis Framework.

The first Sankey diagram shows transports to New York (excluding the Northeastern States and transports within NY). The food supplied by other US states becomes relatively insignificant:

The second one includes food transports within NY state (still excluding the Northeastern States):

John, however has not been satisified with the results of his work. He writes (scroll down to his September 21, 2011 notes):

“Neither produced effective graphics, but what they did demonstrate was the inability of the information to be able to represent food going to New York. (…) As a result, the data “revealed” that we already have a very local food system, when in reality this is not the case; instead, it does indicate how many extra miles are traveled for food around the location of purchase. (…) The images below demonstrate just how disproportionate the amount of miles traveled in New York are to the miles traveled bring food to New York from the rest of the country.”

It remains unclear whether the flows displayed in the diagram are for payload (e.g tonnes of food) or payload distance (e.g. tonne-kilometres). Also, it is not mentioned, whether, for example, water and drinks (typically sourced locally) are included.

I think the idea of thie Sankey map overlay is great, but the issue of spatial representation of (dense) data points has not been adequately adressed. A zoomed NY state would maybe help.

Dr. Vino on his blog presents a Sankey diagram of wine that was originally shown in National Geographic. To be exact, it is a diagram of greenhouse gas emissions associated only with the transport of wine from certain wine producing areas (Australia, Bordeaux, Napa Valley, Chile) to consumers in three U.S. cities (Los Angeles, Chicago, and N.Y.C). So the title should rather read as “Carbon Footprint of Wine Transport”. Neverhteless, an interesting Sankey diagram:

The rounded Sankey arrows are definitely not very common, but are used nicely here. The arrow magnitudes represent weighted emissions potentially contributing to climate change, measured in pounds of CO2-equivalents. The values are for an average 0.75l bottle being shipped. When an arrow get’s wider at a certain point (e.g. Bordeaux to L.A.), this means a change in transport mode (e.g. from ship to truck). The comments to Dr. Vino’s post are well worth reading to understand the diagram better. I am not sure whether it has been taken into account where the wine, typically being shipped in tanks, is filled into glass bottles (adding to the weight, and consequenty to the transport related emissions).

So from this Sankey diagram we can learn that Californian wine being consumed in New York has the highest (transport) carbon footprint, while the French rouge being savoured in the same city comes with the smallest footprint.

BTW, I heave heard that there are studies that look into the life cycle assessment and carbon footprint associated with wine production, e.g this one. Would be interesting to find a carbon footprint Sankey Diagram that combines both wine production, transport, and end-of-life climate change impacts, in order to compare the different phases and their carbon footprint.


A reader of the blog pointed me to some Sankey diagrams available on the U.S. Department of Transport (DOT) Federal Highway Administration (FHWA) website. Sankey arrows are shown as a U.S. map overlay.

The first transport Sankey diagram shows the net tons of goods being transported on flatcars (either as trailer-on-flatcar, or container-on-flatcar) on the U.S. railway systems. The transport volume quantities are clustered into four groups shown with four different arrow widths. It is nice to see how in the eastern part you still have many railroad tracks, but with significantly less transport on them, while to the west coast you basically have 3 main lines, two of which carry more than 25 mio tons of freight per year.

The second one represents freight transport on railroad (bright red), inland waterways (blue) and national highways (dark red). Values also in million tons per year. I am not sure whether flows are to scale, or if transport quantities are also clustered into groups as in the first diagram. I can distinguish at least five different arrow magnitudes, even though only three sizes are given in the legend. A great transport Sankey diagram, as it shows that the East has most of the cargo, and has a much denser transport infrastructure. The reason why railroad transport on this second Sankey diagram differs so much from what is shown the first one, is probably due to the transport of bulk materials not shown in the intermodal transport quantity map.

Flows are not directional in both diagrams, so I assume that quantities for both directions have simply been added.

Can anybody confirm that the massive stream out of Wyoming by rail is coal?