Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
×
Earth Power

A History of the American Energy System In One Chart (theatlantic.com) 64

Long-time Slashdot reader BoredStiff writes: An energy Sankey diagram [where the width of arrows is proportional to flow rates] was published today by the University of Chicago, and shows the history of the American energy system in chart form, from 1800 to 2019.
The Atlantic explains: It is the first attempt to put so much information about U.S. energy history in one place. This particular Sankey diagram shows the inputs and outputs for the U.S. energy system, measured in watts per capita. The left side of the chart shows where energy is coming from (coal, natural gas, or petroleum) and the right side shows what it's being used for (transportation, agriculture, or home lighting and heating)...

[I]t has a lot to teach us about how the energy system got to be the way it is today — and how it might change, and be made to change, in the future... The half century from 1800 to 1850 saw the country devour biomass, most of it in the form of firewood and animal feed. In the 1870s, biomass gave way to the first fossil fuels: coal and, to a lesser extent, petroleum... By the 1910s, coal was dominant.... In the 1920s, it began to fade from the economy, replaced by natural gas, electricity, and — in the transportation sector — petroleum (in the form of gasoline). This was the age of cars and electrified Sun Belt suburbs — and it lasted 50 years, until the energy crisis of the 1970s arrived and capped energy use. Since 1973, per capita energy use hasn't increased.

In recent years, you can see natural gas driving out coal from the electricity sector. It was getting a handle on that change, actually, that led the project's leader to start working on it in the first place. "The changes that are happening in the electricity sector now — changes that are as large as any energy transition we've seen — are difficult to grasp... without animating the data," Elisabeth Moyer, an atmospheric-chemistry professor at the University of Chicago who created the project, told me...

Emily Grubert, an engineering professor at Georgia Tech, noted that nearly all of the transitions depicted were accidental or the result of market forces. It's possible that the transition to zero-carbon energy could be faster, she said, because it will be intentional.

This discussion has been archived. No new comments can be posted.

A History of the American Energy System In One Chart

Comments Filter:
  • I can't imagine a more counterintuitive way to represent this relatively straightforward information.

    • Re: (Score:3, Insightful)

      by cellocgw ( 617879 )

      It would be a lot less misleading if they'd get rid of the stupid animation. Even then, I'm not sure Ed. Tufte would be pleased with this implementation of the Sankey method.

      • by ClickOnThis ( 137803 ) on Sunday December 13, 2020 @12:29PM (#60826010) Journal

        It would be a lot less misleading if they'd get rid of the stupid animation. Even then, I'm not sure Ed. Tufte would be pleased with this implementation of the Sankey method.

        There is a pause button for the animation. And I'm inclined to think Ed Tufte would like it. The information content is concentrated, clear, relational, informative, and beautiful. My only quibble is that the horizontal dimension is somewhat under-used.

      • by rbrander ( 73222 ) on Sunday December 13, 2020 @12:30PM (#60826014) Homepage

        Hit PAUSE. Then just click around, back and forth, in the timeline; it's an information browsing tool. The animation is just an automatic tour through the database in one direction.

        As to a Sankey diagram being a bad information visualization, give it time; a new graphic paradigm is like a new language, the brain needs to learn it to use it. I bet you didn't give it ten minutes. Sankey diagrams have been settled upon, for decades now, as the best shot you've got at apprehending this multi-variate data set; those who do this professionally, rely upon them.

        • As to a Sankey diagram being a bad information visualization, give it time; a new graphic paradigm is like a new language, the brain needs to learn it to use it. I bet you didn't give it ten minutes. Sankey diagrams have been settled upon, for decades now, as the best shot you've got at apprehending this multi-variate data set; those who do this professionally, rely upon them.

          I think the first example of what is now called a Sankey diagram was created not by Sankey, but by Charles Minard, [wikipedia.org] showing Napoleon's disastrous Russian campaign of 1812.

          Many years ago, I attended a short course on visualization conducted by Ed Tufte. He presented Minard's diagram as his favourite example of data-visualization, and perhaps the best one in history. It presents multiple dimensions of information (seven I think?) on a single page in a beautiful and informative way.

          • by rbrander ( 73222 )

            Right after the Minard map in Tufte, there's the centuries-old map of the wine trade, and that's even more like a Sankey diagram, because Minard has just the one source of men (France) and it only shows how they traveled to Russia and back, with a few diverging. The wine trade map shows multiple producers and multiple consumers, much more like Sankey.

            But both are *maps*, not diagrams, and while they thus offer that additional data, it makes flows harder to compare with each other, because one is arcing ac

            • Right after the Minard map in Tufte, there's the centuries-old map of the wine trade, and that's even more like a Sankey diagram, because Minard has just the one source of men (France) and it only shows how they traveled to Russia and back, with a few diverging.

              Minard's map of Napoleon's campaign starts and ends in Poland, not France.

      • It would be a lot less misleading if they'd get rid of the stupid animation.

        It's highly informative, but instead of automatically scrolling the year, there should be a user-moveable slider, like the ones used to compare two different versions of one picture.

        • It's highly informative, but instead of automatically scrolling the year, there should be a user-moveable slider, like the ones used to compare two different versions of one picture.

          Pause the animation. Then you can drag the slider to whatever year you want.

    • I can't imagine a more counterintuitive way to represent this relatively straightforward information.

      Then you must be retarded because this shows nine inputs and four outputs over a period of time. The standard way of showing this would be to have an area charts for each output or each input. These require the viewer to compare and contrast the information abstractly. This single chart shows all the information at once.

      Please share how you think they should have displayed the information to be more intuitive.

    • by ClickOnThis ( 137803 ) on Sunday December 13, 2020 @12:21PM (#60825986) Journal

      On the contrary. I think it shows a great deal in one place, over a short period of time using animation.

      It shows the use of energy in he USA over more than 200 years:

      - starting with almost exclusive use of biomass (i.e., firewood)
      - the dramatic rise of coal in the second Industrial Revolution starting in the late 19th century, and its gradual fall over the 20th century
      - the dramatic rise of petroleum in the early 20th century, and its continued strength despite alternate sources in the late 20th and early 21st centuries
      - the rise of natural gas and nuclear in the mid 20th century
      - the surprisingly small contributions from solar, hydro, and wind even now (watch this space)
      - the parts of the economy that use these different sources

      I think Ed Tufte would be impressed with this presentation.

      • There's more that I missed. For example, the display shows that solar and nuclear sources are used exclusively for electrical-energy production, whereas hydro, wind, and geothermal sources are not. There are also mouse-over messages that add details. I'm sure there's more I haven't discovered yet.

        • Those contributions are there. It's just that as of right now and despite all the media noise, the wind and solar fractions are so tiny that they are hard to identify as you mouse over the lines.

          • I think you misread my post. I said that solar and nuclear sources only produce electricity, whereas hydro, wind, and geothermal are used for other purposes besides producing electricity. As you said, this is clear in the diagram.

    • A the bottom of the page, under the real chart designed to give all the information, there was a simple chart that gave what you desired.

      The top chart gave the complex information that told the full story, including electricity conversion, energy lost to heat via the wires, etc.

      But if you can't understand the complex chart, just stick to the simple one.

      • The top chart uses animation to show changes over time, whereas the bottom chart uses the horizontal axis for time (at the expense of only being able to show inputs or outputs). Some would say this is a better use of the horizontal axis (which has the most resolution on a computer screen).
  • Too bad it was so impossible to read the chart. Thanks, I guess.
    • Too bad it was so impossible to read the chart.

      Try looking at it with your eyes open.

      • by gtwrek ( 208688 )

        Count me as one of those who's not sure what I'm looking at? Does it hurt to label each axis? I've no idea what the x-axis is. Y-axis I can infer as energy use per capita - but why not label it as such? I've still no idea what the x-axis is and what the sloped lines are telling me. Where the slope occurs on the x-axis - what information content is this telling me? It's not date. The point on the X axis where the slope changes as I scroll the years - what does this mean? Does the scale of the slope m

        • by bws111 ( 1216812 )

          The x-axis is not labelled because there is no x-axis. The left side of the page is energy sources. The right side is energy consumers. The thickness of the line indicates quantity of energy. For instance, pause the animation and move the slider to the left (1800). You can see that the bulk of our energy was biomass (firewood), and most of that energy was used by residential or commercial buildings, with much less being used in agriculture, and even less in industry. No biomass was used in transportat

  • by cellocgw ( 617879 ) <cellocgwNO@SPAMgmail.com> on Sunday December 13, 2020 @12:06PM (#60825924) Journal

    because where's the lines for pyramid crystal essential oils?

  • Energy is Joules, for power, the unit is Watt-hours.
    Chart shows Energy as W/capital
    The graphic however is nice.
    See this: https://www.eia.gov/energyexplained/us-energy-facts/

    • Well, you could multiply by the number of seconds in a year to get Joules per year per capita, but technically that would still be a unit of power, and less intuitive.

      Fair point, though.

    • It this really true? That just seems to unbelievable that the amount of energy used per person has remained constant. Does that means with all the additional energy people are using (computers, lots more a/c, more cars, more shipping and flying, cloud electronics, etc) that it is offset by increasing energy efficiency? If so, then all the energy efficiencies and conservation have NOT reduced our energy consumption and population increase is significantly more of the problem.

      • That just seems to unbelievable that the amount of energy used per person has remained constant.

        Well according to the link in TFS, it most definitely has not remained constant. You can see the per-capita watts change as the year changes.

    • Re: Units Issue (Score:5, Informative)

      by Linux Torvalds ( 647197 ) on Sunday December 13, 2020 @02:36PM (#60826386)

      A joule is simply a watt-second. Exact same thing as a kWh, just scaled differently.

  • What I find to be rather shocking is the amount of waste involved with the generation, distribution and use of electricity. I haven't reviewed their figures or methodology*, but this would seem to argue against the expansion of electrical use for transportation an other uses.

    *One big mistake I've seen in such analyses is the confusion with average loss (applicable to power use) vs peak loss. Which is proportional to the square of power flow.

    • Re: (Score:3, Insightful)

      by ClickOnThis ( 137803 )

      What I find to be rather shocking is the amount of waste involved with the generation, distribution and use of electricity. I haven't reviewed their figures or methodology*, but this would seem to argue against the expansion of electrical use for transportation an other uses.

      *One big mistake I've seen in such analyses is the confusion with average loss (applicable to power use) vs peak loss. Which is proportional to the square of power flow.

      What I think it argues for is the reduction of centralized power plants, and an increase in more local, or even in situ energy production. (Think solar panels.) IANAEE, but I think much of the energy waste may be I^2R losses in high-voltage power lines and other parts of the distribution network.

      • by PPH ( 736903 )

        much of the energy waste may be I^2R losses

        Yes. But they might be off by a factor of ten here. Losses being proportional to the square of the power flow, it makes a big difference how you calculate those losses w.r.t. the peak versus average power flows. Poor economic decisions are made based on bad data. If we take these numbers to be correct, then it may make no sense to rely on electrical power for transportation until after the loss problem is fixed. It may never make sense for people living where local solar isn't an option (think cities). It

        • I looked again and noted that the figure refers to it as waste, not transmission losses. A quick google reveals that transmission losses can be between 8% and 15%. You are right: the waste numbers in the figure are much higher than that.

          I don't see a definition on the website for waste, but there is a preprint cited in the "About" page that I skimmed quickly. As far as I can tell, they have calculated waste as a total of all losses in the creation, transmission, and usage of electrical power, not just trans

          • by PPH ( 736903 )

            Waste could include Carnot cycle efficiency. But then why isn't that evaluated for direct petroleum or natural gas inputs to transportation, industry or residential use?

            • Some of that waste is the heat produced when light bulbs are used to create light. Incadescent bulbs waste over 90% of the electricity as heat. Fluorescent bulbs waste about 30% of their electricity as heat. Even LEDs generate waste heat as do other electronic devices. Have you felt how hot a laptop can get?

              There are methods for converting waste heat into usable energy
              https://www.heatispower.org/case-study/waste-heat-to-power-natural-gas-compressor/
              https://www.sciencedirect.com/science/article/abs/pii/S

      • A huge amount of the losses are waste heat from generation, be it turbines or reciprocating engines. Perhaps this argues for more use of the waste-heat.

        I would also think that in some industrial locations, local Combined Heat and Power (CHP) Co-Generation plants might be very useful to utilize the waste heat, though this can have it shortcomings and is not applicable everywhere. A common problem being having non-correlated need for heat versus electricity. I have personally seen a shared residential locatio

    • What I find to be rather shocking is the amount of waste involved with the generation, distribution and use of electricity.

      As I read it, that's an artifact. It's mainly the carnot cycle efficiency of the heat engine that burns the fuel to heat, uses the heat to produce shaft horsepower, and uses the shaft horsepower to drive the generators. Though some of your "waste" is losses in the electrical machinery and grid, most of it is the heat engine "losses" of going from heat to

      If you look at usable power div

  • FANTASTIC Chart (Score:5, Interesting)

    by gurps_npc ( 621217 ) on Sunday December 13, 2020 @01:05PM (#60826116) Homepage

    The thing that got me was the waste electricity. That lighter yellow shade below the darker one is electricity lost to resistance.

    Frankly, we could eliminate all Coal if we had a room temperature superconductor. That was shocking to me. If we had one capable of going up to 120 degrees farenheit, we could eliminate coal AND natural gas.

    Really great chart, presented multiple inputs,multiple outputs, with the electricity conversion, over time, with a smaller, simpler chart underneath.

    • Frankly, we could eliminate all Coal if we had a room temperature superconductor.

      Not quite. Transmission losses include stepping losses. To combat that you need to generate at lower voltages, ultimately lower powers and more locally. That introduces its own inefficiencies. Room temperature superconductors would help, but it won't solve all the efficiency problems in a grid.

      • by tlhIngan ( 30335 )

        Not quite. Transmission losses include stepping losses. To combat that you need to generate at lower voltages, ultimately lower powers and more locally. That introduces its own inefficiencies. Room temperature superconductors would help, but it won't solve all the efficiency problems in a grid.

        The grid is remarkably efficient, and lower voltages cause great inefficiencies. Given a device taking constant power, running it at a higher voltage/lower current wastes less power than a lower voltage/higher current

    • The thing that got me was the waste electricity. That lighter yellow shade below the darker one is electricity lost to resistance.

      Nope. It looks to me like an artifiact: They're counting the carnot cycle efficiency of fuell driven heat engines. This is buried in the load boxes for things like fuel-powered transportation, but shows in the connector between power generation and power consumption.

      See my reply to the previous posting titled "Carnot cycle". [slashdot.org]

  • Right now electric cars being charged at home is too small to make any diff to a chart of this scale. But in the next decade the transportation sector will switch massively from petroleum to electric.

    The transportation at the bottom need to be moved just below residential/commercial that is in the down stream of fungible electricity gray square in the middle.

    Petroleum too should be moved close to that place, serving may be 50% of transportation.

  • Only a tiny fraction of US transportation is run on electricity. I guess it's more in countries that use trams and electrified-track trains.

  • by the Portola Institute, from 1974, page 113, has a tangentially-related diagram for (then) current energy flows possible from biofuels from one acre of land either from grain or woodlot -- but it was not interactive of course. They also have other diagrams and tables covering energy topics. The book was a bit like "The Whole Earth Catalog" packed with information and references to other books and resources. That book was a big inspiration to me when I saw it in the 1980s, to think about how so much inform

  • I have no idea what I'm looking at. Why did they do that?
  • The Atlantic mentioned biomass to feed animals but seems to miss the important factor of slavery.
    • I don't follow. Slavery was a factor in what?
      • by Teun ( 17872 )

        I don't follow. Slavery was a factor in what?

        A form of energy, just like many countries before steam had industry running on wind and water power.

        • The Atlantic article probably didn't talk about food as an energy input to human labor because the University of Chicago researchers excluded it from their visualization data. The researchers did look at food as an energy source for humans, and specifically looked at slave labor: [squarespace.com]

          5.1. Human Labor.

          We omit human labor in these statistics for two principal reasons. First, human labor is hard to apportion by sector... Second, humans comprise only a tiny fraction of the overall energy system.

          One question that may be asked of this data set is what insight energy data can shed on slavery...

          In 1850, there were a little over 3 million slaves in the US out of a population of 23 million total. Around that time, an estimated 8 million domestic horses were in the rural US. Horses require 8 to 10 times the caloric intake

  • by istartedi ( 132515 ) on Sunday December 13, 2020 @02:40PM (#60826398) Journal

    I get the people who are confused by this. I don't look at this kind of presentation very often either. Give it a chance and it grows on you.

    For example, see the way the coal line gets thinner and thinner towards transportation, then goes away? That's the end of steam trains there.

  • On the left show the money paid into the system by workers, on the right show the doctors, nurses & pharmaceutical companies. Between are the zillion middle men, insurance companies, lobbyists and other leaches. How much of every dollar I put into the pipe goes to the nurse who takes my blood pressure or the doctor who checks my eyes or prostate.
  • Kind of makes you doubt the whole thing if they don't understand energy

  • The Crescent Dunes Solar Energy Project is yet another waste of taxpayer money. It's pretty obvious that if an "alternative" energy source was so great, it would be able to pay for itself.

  • US Railroads are so long, it is not economical to electrify the tracks. Are the batteries anywhere close to be in the running? At what price point I wonder.

    Electrification has high minimum investments. Must electrify the dual tracks all the way (NY - SF) or at least part way (NY - Chicago) to become effective. Battery locomotives can come in lower minimums. A 32 MWh battery mounted on a flatcar would perform like a diesel locomotive with 3000 gal of fuel (assuming electric motors are 3 times more efficien

    • US Railroads are so long, it is not economical to electrify the tracks. Are the batteries anywhere close to be in the running? At what price point I wonder.

      I did a quick search for other countries with long railroads, China for example [wikipedia.org], and find that of 2019, the country has 139,000km of railways, the second longest network in the world of which 100,000km are electrified (71.9%).

      this [wikipedia.org] gives a handy table. If we ignore the E.U. because it tends to have very high population density, despite a huge length of track, we find Russia with 85,600km (51% electrified) and India 67,415km (51% electrified).

      So, while there are countries like the USA, Canada and Australia w

      • I am from India. India's motivation to electrify railroads is basically to save foreign exchange. India has lots of coal, but not much of oil. So it would rather burn coal, make electricity and move its transportation sector depend less on oil. That is the reason it spends/spent so much on track electrification. I would guess, China is also in the same boat.

        Countries rich in oil they do not mind their transportation sector depending so much on oil.

    • US Railroads are so long, it is not economical to electrify the tracks. Are the batteries anywhere close to be in the running? At what price point I wonder.

      No, batteries are not feasible for power to rail transport. Diesel fuel has 100 times the energy per kilo than any battery. Taking into account energy efficiency doesn't help much since even in a best case there my be a theoretical advantage from the lower energy loss in conversion to make the difference in mass closer to 10 time the mass. This means in stead of one or two fuel tankers on the tractor it would need more like 10 or 20 battery packs. These will have considerable mass that will count agains

      • Mass density is not an issue, trains would not haul 8 battery packs all the way to the coast. They will drop off a depleted battery car every 500 mile and pick a freshly charged one, these battery tenders can be self propelled and be switched as fast as a diesel loco is refueled. Tesla semi is talking about battery+motor coming within 1000 lb of engine+transmission+fuel+emission control system for a 80,000 lb class trucks. It can easily come to train, if the economies work out.

        Ultracapacitors are still i

        • You could also electrify the track in sections, and charge the batteries on the go. This'd be much cheaper than electrifying the entire route, and require no stopping at all. Charing a stationary battery car would probably be done using overhead wires anyway, since it's existing, reliable tech.
          • True. And it has got the added benefit of serving spur lines without electrifying it. I read about a project in Germany they were planning to use batteries to serve a 75 mile un electrified spur line branching off from an electrified route.

            This can also radically change the urban electrified street car. Add a battery, retract the pantograph, serve a few miles of roads and streets without overhead electric lines, come back and join the loop.

            Also hybrid buses with both flanged rail axles and rubber tires c

    • US Railroads are so long, it is not economical to electrify the tracks. Are the batteries anywhere close to be in the running?

      Do you have ANY IDEA how MUCH power is involved?

      A single typical diesel-electric locomotive unit (of which some trains have several) represents about 3,000 horsepower for traction, or about 2.2 MEGAwatts. (A horsepower is almost exactly 3/4 kilowatt, and you need more to cover motor losses.) Running that for a day or so is a LOT of batteries.

      In major power outages, cities (e.g. Tor

      • by Zak3056 ( 69287 )

        A horsepower is almost exactly 3/4 kilowatt

        Friendly mnemonic: "in Fourteen hundred ninety-two, Columbus sailed the ocean blue. Cut the sonofabitch in two, and that's how many watts are in a horsepower."

The use of anthropomorphic terminology when dealing with computing systems is a symptom of professional immaturity. -- Edsger Dijkstra

Working...