Food losses and food waste has been addressed in a number of scientific research papers in recent years. Peter Alexander et.al. write about ‘Losses, inefficiencies and waste in the global food system’ (In: Agricultural Systems, Volume 153, May 2017, Pages 190-200, doi.org/10.1016/j.agsy.2017.01.014)

The article contains two beautiful Sankey diagrams. The first depicts the global food system in 2011. Flows are shown as dry mass. Flows are not individually labelled with the underling quantity, but rather a scale at the bottom shows 5 representative flow quantities and their corresponding width.


(under terms of Creative Commons Attribution 4.0 License (CC BY 4.0))

Crop (yellow) and grassland (green) net primary production (NPP) are shown as sources for the global food system. Losses are branching out as grey arrows. These “inefficiencies” of the system are described in detail in the article. The authors observe that “44% of harvested crops dry matter are lost prior to human consumption” and that “the highest loss rate can be found in livestock production”.

The second Sankey diagram shows a section of the above figure, just the dry matter flows from crop harvest and processing, without any losses. This is interesting because it allows us seeing the share of processed and non-processed food being consumed by humans worldwide, and the the share of crop-based food intake (dark blue) compared to animal-based food intake (red). You could call this the veggie / non-veggie split. Based on dry matter that is.


(under terms of Creative Commons Attribution 4.0 License (CC BY 4.0))

If you want to see the corresponding global food system wet mass, protein and energy Sankey diagrams check out this interesting article. A recommended read for all of us eaters.

Back from a summer break … sorry for neglecting the blog for a couple of weeks. The FIFA Worldcup 2018 prediction from my last post has already become outdated … 😉

Here is a black and white hand drawn Sankey diagram from a doctoral thesis (Giovanni Angrisani:’Experimental and Simulative Analysis of a Micro Trigeneration System based on an Air Handling Unit with Desiccant Wheel’. Doctoral Thesis University of Naples, year unknown).


Instead of color-coded Sankey arrows the different types of energy are shown with different patterns of the arrow head. We can see some typical flaws: width of the arrow changes when going from horizontal to vertical (see ‘Losses 3.24 kW’ arrow branching out to the top), flows not to scale (compare ‘Losses 3.24 kW’ arrow to ‘Losses 6.25 kW’ arrow), overemphasis of some arrows through massive arrow heads.

A retro-style Sankey diagram from the times you would hand-draw such figures.

Every four years soccer related Sankey diagrams pop up. For the 2010 FIFA worldcup in South Africa I featured this figure. In 2014 there was the beautiful “The Road to Rio” diagram from an inflight magazine.

For the upcoming 2018 FIFA worldcup in Russia, two researchers have taken a more scientific approach. Their prediction model uses mathematical methods to determine who will most likely be handed the gold trophy on July 15 in Moscow. If you are into Monte Carlo simulations, bivariate and nested Poisson regression models, Brier score and Rank-Probability-Score (RPS) then you will enjoy the paper ‘On ELO Based Prediction Models for the FIFA Worldcup 2018’ by Lorenz A. Gilch and Sebastian Müller.

All others can just skip and go directly to page 22 of the paper to find this Sankey diagram based on 100.000 simulation runs:


via Twitter user @ggojedap

Groups and teams are color-coded, and the wider the band in the Sankey diagram, the higher the probability. So, according to this model, which takes into account performance of the teams since 2010, a nation from the green group will become the world champion. Purple runs with a probability of 18% and red with 14% (for the detailed values that form the basis for this diagram please see table 12 on page 15 of the paper).

Well, we’ll see, and in five weeks we will know the outcome. Whether you trust this more scientific approach, or whether you would rather go with a straight-forward Paul the Octopus divination … I hope you enjoy watching the matches!

The French region Auvergne-Rhône-Alpes in the south-east of the Hexagone borders with Switzerland and Italy. Lyon and Grenoble are located in this region, known for skiing, lush pastures … and great cheese!

Auvergne-Rhône-Alpes Énergie Environnement (AUR-EE) is a regional agency that works to bring together players in the renewable energy field and to promote RE projects.

Given the agricultural character of Auvergne-Rhône-Alpes, biomass use for energy generation has been going strong in recent years. The agency has created energy flow Sankey diagrams for existing biogas installations, as well as a projection for the ones being under development.

Data is for 2017 and for the scenario where all projects currently under development would already completed. The yellow stream (‘déchets ménagers’) is household waste, providing 374 GWh of energy. Manure and other side-products from agriculture (green arrow) contributes another 260 GWh.
The stacked bar on the left hand side of the diagram indicates the potential availability of biomass by 2035, and one can see that only a small fraction of it is currently being taken advantage of.
Biogas is produced in anaerobic digesters (‘méthanisation’) and the region yields some 271 GWh electricity and 200 GWh heat per year from cogeneration plants. Already almost 100 GWh of biogas could be injected to the natural gas network, allowing for storage of the energy.

Note that smaller or even negligible flows are still shown with a minimum width in order to make them visible (these thinner arrows are not to scale with the others).

This Sankey diagram depicting the energy balance of Chile for 2015 can be found on the website Gestiona Energía MiPyMEs (MiPyMEs is the Spanish term for ‘small and medium-sized enterprises’, SMEs).

Flows are in TCal (teracalories), a unit for energy we don’t get to see very often (1 TCal = 4,205 Joules). What surprised me most in this figure was that ‘Biomasa Leña’ (biomass firewood) is the third most used primary energy source. The accompanying pie chart on the same page confirms that crude oil (25%) and coal (20%) are the most important sources, followed by biomass and oil derivates (each 19%). I guess this should read ‘biomass AND firewood’ rather than ‘biomass firewood’.

Some design shortcomings, in particular where the downward sloping stacked Sankey arrow turns to run horizontally to join the node ‘Electricidad’, and at the input side of the primary energy box, where the flows for ‘Petróleo Crudo’, ‘Carbón’ and ‘Biomasa Leña’ overlap and somehow don’t seem to hold their width all the way. My guess is that this is owed to the wish to keep the figure as compact as possible.

As part of the Canadian SPRUCE-UP research project one activity is dedicated to Genomic, Ethical, Environmental, Economic, Legal or Social (GE³LS) aspects of this applied genomics project. As part of their work the scientists have developed the Canadian Forest Service – Fiber Cascade Model (CFS-FCM) simulation model.


(see high res image here)

This Sankey diagram shows one specific scenario for a downstream flow of wood fibre from Canadian forests to products. Flows are in metric tonnes (probably for one reference year), with the exception of the ‘Bioenergy’ flow, shown in terajoules (TJ).

Another Sankey diagram from the article ‘Exergoecology Assessment of Mineral Exports from Latin America: Beyond a Tonnage Perspective’ by Jose-Luis Palacios I discussed in this recent post.

Non-fuel minerals exported in 2013 from Latin America to other continents. Flows are in Mtoe (for the reason why these flows are measured with a typical energy unit and to learn about the ERC approach read the article). Due to the scale, some minerals can not be seen as individual flows in the Sankey diagram and are thus grouped as ‘Rest of Minerals’ (black stream).

Up on the EUR-Lex, the European Union’s database on laws, regulations, publications and reports is a staff working paper ‘Measuring progress towards circular economy in the European Union – Key indicators for a monitoring framework’ meant as accompanying background text for a ‘Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions on a monitoring framework for the circular economy’.

And it shows this beautiful Sankey diagram on material flows in the EU economy (2014).

Beautifully crafted, this diagram shows that “8 billion tonnes of raw materials were processed during 2014 in the EU: of this 1.5 billion (i.e. around 20%) are imported, which indicates the EU dependency on imports of materials. Out of the 8 billion tonnes of processed materials, 3.1 billion tonnes are directed to energetic use, 4.2 to material use and 0.6 are not used in the EU but exported.”

Flows are in Gt/yr (billion tons per year. The composition of the flows is presented at certain points in the diagram as bar charts on top of the dark blue bands: metal ores, non-metallic minerals, fossil energy materials/carriers and biomass. For each of those four groups individual Sankey diagrams can also be found in the working paper.

The EU never stops to surprise me! In this case in a positive way, as Sankey diagrams seem to have arrived at the top echelons of European policy making (or at least with their staff).