A Very Big Picture Indeed

Posted on Thu, 05/01/2014 - 09:10

Visualizing Canada’s energy systems helps us to understand the challenges we face and then identify alternative pathways to a more sustainable energy future. Within CESAR, we have created some exciting new visualization tools, powered by the Canadian Energy Systems Simulator (CanESS), with its unique capacity for looking both forward and backward in time.

Sankey diagrams provide perhaps the biggest, and most comprehensive picture of the energy systems that drive the Canadian economy.  First created by Charles Minard in 1869 to map Napolean’s invasion of Russia, Sankey diagrams provide a graphical representation of large quantities of quantitative data.  They are now widely used to depict energy systems. The line widths are proportional to the flows of energy from primary resources, through their conversion to fuels and electricity, to final consumption.

Click here to see an interactive Sankey of the energy systems of Canada in 2010 that is similar to the one shown here (click on the picture to enlarge).

The left-hand side of the diagram reflects the richness and diversity of Canada’s energy resources – biomass, hydroelectricity, coal, oil natural gas, and uranium. The production levels, like the other numbers in the diagram, are presented in energy units – Petajoules (PJ) – but the physical magnitudes are staggering. In more familiar units, the 1,446 PJ of primary hydropower translates into nearly 400 billion kilowatt-hours, more than 10% of the world’s hydropower production and enough to provide over half of the nation’s electricity consumption.    

It is a prodigious amount of energy, and yet it is only about 10% of the energy content of Canada’s fossil fuel production. In 2010, the 6,680 PJ of primary oil production represents over a billion barrels, and the 7,035 PJ of primary natural gas represents 189 billion cubic metres. Even the much smaller quantity of primary coal production, 1,486 PJ, represents some 67 million tonnes, enough coal to fill a train that would extend from Vancouver to Halifax and back again. 

And then there is uranium. Although often overlooked in tabulations of Canada’s energy production, the energy content in the uranium Canada produces (8,263 PJ in 2010, estimated by counting the heat generated when the uranium is used in nuclear power reactors) exceeds the energy content of both total oil and total gas production.  In energy terms, uranium ranks as our largest energy resource, both in terms of production and exports.

All totaled, Canadian primary energy production in 2010 was nearly 25,600 PJ, and after including 3,700 PJ of imports, total primary energy availability was 29,500 PJ. As the Sankey diagram shows, 58% was exported, with the remaining 42% or 12,500 PJ being used domestically, 910 PJ for non-energy applications and 11,652 PJ for the provision of energy end use services to Canadians. The overall efficiency of energy end use Canada is about 32%, so in 2010 the actual useful energy delivered to Canadians, after conversion losses, was 3,712 PJ, just 13% of the total primary energy supply of 29,500 PJ.

Following the Sankey flows from left to right tells us a great deal about the use of fuels and electricity and Canada.  For example:

  • The transportation sector (1) runs mostly on petroleum (gasoline and diesel) and petroleum products are no longer widely used for energy outside of the transportation sectors.
  • Natural gas (2) is the preferred fossil fuel for almost all stationary applications, where it is available, and is the fuel of choice for residential and commercial building heating and industrial process heat.
  • Most domestic coal use in Canada is for power production (3); outside of steel and a few other industrial applications, coal is no longer widely used as an end use fuel in Canada.
  • While wood (4) is a significant energy source of home heating in many parts of Canada, 4-5 times more biomass is used by the pulp and paper industry.
  • As noted above, hydropower (5) is the largest source of electricity on a national scale; solar and wind and other primary renewable energy forms while growing in recent years are still a very small part of the national electricity supply.

These observations are made at the national level; there are large and important interprovincial variations.  Also, the patterns revealed by the Sankey are constantly evolving.  A comparison of the current situation with the Sankey for 1978, for example, shows the extent to which oil and gas exports have grown as share of total production, the decline of petroleum fuels in all the non-transport end use sectors, and the growth of natural gas as a feedstock in the non-energy sector.

There is a story behind every region and every line in the Sankey diagram; we will be using these visualization techniques in future blogs to tell these stories as a way to identify and understand the opportunities and challenges presented by the coming energy system transition.


Thank you for the comment David! It is a good point; these Sankey diagrams depict the energy content of resources and end use fuels and electricity, and the conversion losses in the system. These particular Sankey diagrams do not portray the different environmental attributes of the fuels and their associated technologies, risks, costs and benefits. But they help point to the policy issues, and they are great visualization tools for better understanding some important aspects of our energy systems.

The Sankey diagram tracks the flow of measured commercial energy carriers; but it says nothing whatever about what they are flowing through. An energy system is made of physical artefacts, and these artefacts - buildings, fittings, appliances, industrial plant, vehicles and so on - are what actually deliver the services we desire. The Sankey diagram approach focuses on supply of carriers so completely that it diverts attention from the parts of the energy system most amenable to improvement. It is an instance of the mindset that says 'energy' but really means 'fuel+electricity'.

The 'uranium' bar on the Sankey diagram is an egregious example of the limitations of that representation. It identifies the 'energy content' of the uranium, as if it could be accessed like that of a stick of wood. Of course it cannot. It requires a vast array of physical artefacts to mobilize the energy. You might as well talk about the 'energy content' of, say, the boreal forest, or all the rivers of Canada. It's more accessible than that of uranium, if you put in place all the physical artefacts to mobilize it. - Walt Patterson

When energy resources (fossil fuels, uranium, solar, wind) are converted to energy commodities (gasoline, diesel, pipeline natural gas, fuel pellets, electricity) and then these commodities are converted to energy services, there are significant 'losses'. Some of these losses are associated with the conversion of energy from one form to another, but we also count as 'conversion losses' the energy that is invested to move the energy commodity from one part of Canada to another (e.g. through pipeline, wires or on trains).

These amounts accumulate in the grey 'Conversion Losses' flows on the Sankey, at the lower right hand. Note that the losses account for about two thirds of Canada's domestic energy flows. So yes, these are "...parts of the energy system most amenable to improvement". No argument there.

Regarding the energy content of the uranium in our Sankeys, this represents the thermal energy of the uranium processed each year (NOT the uranium in the ground, the biomass in a forest or the water in a river) if the uranium where to be put through a Candu reactor. Nuclear reactors like the Candu only capture a small portion of the nuclear energy in the fuel; our calculations and this Sankey does not show that unused nuclear energy. In a Candu reactor/generator, the electricity output has an energy content equivalent to only about 32% of the thermal energy extracted from the uranium. That conversion loss is shown in the 'electricity generation' vertical bar.

Thanks for the comments Walt. I agree that Sankeys are limited, as are all diagrammatic depictions of complex systems. But they do provide an instructive illustration of the energy content of primary energy resources and secondary energy commodities, and how that energy content moves from primary production to final end use, about the role of exports of fuel and electricity in Canada, and about the overall First Law efficiency with which the energy content is applied to end uses. But I dont agree that the depiction of the energy content of the uranium is any more an illustration of the Sankey's limitation than the depiction of the energy content of crude oil, or oil sands, or sticks of wood for that matter. They all require physical artefacts and technologies of varying complexity in order to apply the energy content to human needs, and uranium is not unique in this regard. But the key point you make is important: the Sankey, at least this particular Sankey, does not reflect the various other attributes of the resources and end use fuels and electricity, or the technological context in which they are exploited, or (as David Katz has pointed out) the relative environmental stresses generated by their mobilization.

Kate - there is no single data source for these sankey diagrams. The CanESS model is used to "simulate in history" (a.k.a. calibration) using data from a variety of sources including Statistics Canada, Natural Resources Canada, Environment Canada and various scientific and engineering studies. The result is a complete and coherent time-series database which descibes the evolution of Canada's energy system. Further information on the CanESS calibration process and data sources will be available soon - stay posted for updates.

I am a fan of Sankey diagrams for showing energy flows. And I think these visualizations are very powerful and valuable additions to the energy discussion in Canada. Some of the historical patterns are extremely interesting and the ability to see changes over time for different energy resources and different energy jurisdictions is very useful.

I also agree with David Katz and Walt Patterson that it would be great to see other dimensions of the energy system reflected in these visualizations, but I see that more as a comment as to where this could go than a critique of what is here. It would also be nice to think about how second law efficiencies and patterns of energy services (and generally the energy/activity link) could be represented.

Hi Ralph,

As we discussed by email, i think it would be great if there was a way for municipalities with community energy use data (from PCP GHG inventories and/or energy mapping work) to add in their data to get "customized" versions of these visualizations for their community.

I recognize that for tools like your Sankey diagrams that this would be a crude approximation (based on provincial averages) for things like industrial energy use. But, it still provides a good "order of magnitude" breakdown useful for illustrating the potential for actions like CHP and waste heat utilization.

Fantastic visualization tool!

Now it would be nice to "drill down" and be able to display, say, the residential end-uses of homes in Alberta and their associated GHG emissions, then export the data as an excel file. I understand that this fine granularity is already present in the backend data base system?

Hello Ralph et. al. great project here. Thank you so much for it. I am asking direction to good reasonably priced Sankey software, hoping to map and model City controlled urban forest tree residues for potential City of Toronto  fuel supply to bio-energy projects. Any help appreciated by the Toronto Renewable Energy Office (Sean Cosgrove, M. Env. Design, Calgary'90)

Hi Sean,

Thank you for your interest. Your project sounds very exciting, and given the level of   detail you are envisaging, it should provide valuable insights.

If you are looking for a desktop software, I can recommend e!Sankey whose Pro version allows live links with Excel files (great feature to automatically update Sankeys).

There are also options for online tools that are free to use, but you will need to code (and host) the online interface.  I suggest that you check out the following links:

I hope this helps. Good luck with your project and do not hesitate to keep us informed of the outcome.


Log in or register to post comments.