Vaclav Smil: from wood to coal to oil, energy transitions take a long time

Preface.  This post has a shortened, reworded, and heavily commented on article from Scientific American in 2014.  You can also see two much longer articles about energy transitions by Smil from n 2008 and 2010 here:

Vaclav Smil: Our transition away from fossil fuels will take decades—if it happens at all.

Alice Friedemann  author of “When Trucks Stop Running: Energy and the Future of Transportation”, 2015, Springer and “Crunch! Whole Grain Artisan Chips and Crackers”. Podcasts: Practical Prepping, KunstlerCast 253, KunstlerCast278, Peak Prosperity , XX2 report


Vaclav Smil. January 2014. The Long Slow Rise of Solar and Wind. Scientific America.

  • The major global energy transitions—from wood to coal to oil—have each taken 50 to 60 years. The current move to natural gas will also take a long time.
  • There is no reason to believe that a change to renewable energy sources will be exceptionally fast. In rich countries, “old” renewables such as hydroelectricity are maxed out, so growth will have to come from new renewables such as wind, solar and biofuels, which provided only 3.35 percent of the U.S. supply in 2011.

Most people think that the 19th century was dominated by coal and the 20th century by oil and that the 21st will belong to renewable energy.  But it isn’t true.

Even with the rise of industrial machines, the 19th century wasn’t run on coal. It ran on wood, charcoal, and crop residues (mostly cereal straw) which provided 85% of all energy worldwide.

Coal didn’t provide 5% of energy until 1840, and didn’t reach 50% until 1900. This rise from five to 50% took about 60 years.  It wasn’t until 1885 when fossil energy (mostly coal, some crude oil, and a tiny amount of natural gas) surpassed the energy provided by wood and charcoal in the United States, in France this occurred in 1875, Japan 1901, and the U.S.S.R 1930, China 1965, and India the late 1970s.

Likewise, in the 20th century the biggest energy source was coal, not oil.  Crude oil didn’t surpass coal until 1964.

And yet because GDP and populations were growing exponentially, coal continued to be used in such huge amounts that it ended up being the 20th century’s most important fuel, contributing 5.3 yottajoules (YJ — 24 zeros) compared to oil’s 4 YJ.

Although natural gas is seen as the bridge to the future, only the USSR and UK use more gas than oil.

If wood to coal and coal to oil took roughly 50 to 75 years for each transition, the same can’t be extrapolated to renewable energy. First, consider the scale – about 450 exajoules (18 zeros) of fossil energy is used by the world, 20 times more than during the 1890s.  Generating this much energy without fossil fuels is daunting.

Second, wind and solar are intermittent, yet society needs a reliable, uninterrupted supply of electricity around the clock.

My comment to expand on Smil’s second point: In the United States in 2014, electricity was generated with 66% fossils, 20% nuclear, 6.3% hydropower, 1.6% biomass, 0.4% geothermal, 4.4% wind, and 0.5% solar.  Wind and solar will not only need to replace fossil fuels, but nuclear power as well, since far more plants are closing than being built.  If there were a way to store excess solar and wind energy renewables might have a chance of replacing fossil fuels, but there are no large scale energy storage systems that are even close to being commercial, except hydropower with very few places to put more dams, and compressed air energy storage where there are even fewer places to put them (only one exists in America), and not enough materials on earth to build a battery that could store just 12 hours of world generated power (Barnhart 2013).

Third, again my comment: electricity is only 18.7% of the overall energy consumed by society: transportation, manufacturing, industry, mining, heating, natural gas based fertilizers, the 500,000 products made with fossils as both feedstock and the energy source to make products, comprises another 67% of the energy used by society that renewables would need to replace.  They are not even close to replacing the fossils used in electricity generation and are growing at such a slow rate that it would take many centuries to replace just the electrical portion of energy use.

Fourth (again my comment) we don’t have an electric way to run for heavy duty trucks, neither batteries or overhead wires; and we don’t have electrical manufacturing processes to make cement or metals without fossil fuels.

Fifth, and finally back to Vaclav Smil, “the final factor that will lead to a prolonged shift to renewables is the size and cost of the existing infrastructure.” Even if renewables were free, it would be economically unthinkable for nations, corporations, or municipalities to abandon the enormous investments already made in the fossil fuel system, from coal mines, oil wells, gas pipelines, refineries, and hundreds of thousands of filling stations — infrastructure worth at least $20 trillion (Smil left out the existing billions of cars, trucks, and equipment that runs on combustion engines and much more). China has spent half a trillion dollars for 300 gigawatts of coal generation from just 2001 to 2010, and expects those plants to run for at least 30 years.

Smil suggests that the only way to speed up a transition is to use less energy.


Barnhart, C., et al. 2013. On the importance of reducing the energetic and material demands of electrical energy storage. Energy Environment Science 2013(6): 1083–1092.

This entry was posted in Alternative Energy, Coal, Dependence on Oil, Energy, Vaclav Smil and tagged , , , , , , , , . Bookmark the permalink.

One Response to Vaclav Smil: from wood to coal to oil, energy transitions take a long time

  1. Harry Fischer says:

    This, like almost everything you post ought to be widely read. It seems generally right on, sobering and depressing. But note a problem that it seems should have been addressed in editing Smil’s original article: I think Smil must be mistaking Yottajoules for Zettajoules. Current total global energy use is somewhat less than .6 Zettajoules (600 Exajoules), i.e. 6×10*20 joules per year. (Smil’s articles says 450 Exajoules in 2012, but of course it’s greater now.) Therefore, if oil contributed 4.3 Yottajoules (=4300 Zettajoules) of energy in the twentieth century, it alone would have contributed an average 43 Zettajoules per year for the entire century (i.e., about 80 times as great as our current total energy consumption). It’s likely, then, that when Smil wrote Yottajoules, he meant Zettajoules, despite his having correctly identifying a Yottajoule as 10*24 joules. If total oil consumption was 4.3 Zettajoules, that would mean oil averaged .043 Zettajoules (per year over the century, which seems low given current usage, but plausible given that its use expanded exponentially. Being off by a factor of a thousand may be close enough for government work, but I expect more of intrepid doomsayers. (Don’t ask what difference it makes.) Have I made a mistake?