The Reality of Fossil Fuel as it Accounts for 80% of Our Global Energy Supply

On April 22nd we had the pleasure of hearing a presentation by Ripu Malhotra, a researcher with the Stanford Research Institute, who was speaking at AlwaysOn’s “Nordic Green” conference. Mr. Malhotra has taken the unusual step of converting all forms of energy use into cubic mile of oil equivalents – a measure that is quite helpful if one is trying to really grasp the scope of fossil fuel compared to alternatives.

Using Malhotra’s figures, which he has garnered from the British Petroleum statistical database of global energy usage, among other sources, it is clear that today, at least, the world’s economy is utterly dependent on fossil fuel.

Global energy production expressed in “cubic miles of oil” (CMO),
billion barrels of oil, quadrillion BTUs, and gigawatt-years.

While there is impressive percentage growth in the solar category – including wind power in this analysis – and while these figures have already changed dramatically since 2005, it is clear that alternative energy still contributes well below 1% of total energy supply, whereas fossil fuel contributes well over 80% to global energy supply.

In terms of choosing between fossil fuel development and alternative energy development, another point which should be put to rest is the notion we are running out of fossil fuel. The next three charts show the potential reserves of the primary fossil fuels – oil, coal, and gas. In order to develop estimates for unconventional sources of these fuels, we have taken the midpoint between the high and low estimates.

If oil provided 100% of global energy, and we used twice as
much as we do today (1,000 Quad BTUs per year), there
would be a 59 year supply of oil based on known reserves.

If coal provided 100% of global energy, and we used twice as
as much as we do today (1,000 Quad BTUs per year), there
would be a 218 year supply of coal based on known reserves.

If gas provided 100% of global energy, and we used twice as
much as we do today (1,000 Quad BTUs per year), there
would be a 45 year supply of gas based on known reserves.

So when you add it all up, at twice the current energy consumption overall, oil, gas and coal could potentially supply all the energy we need in the world for the next 300 years – not including gas hydrates.

The other question of course is how do the alternatives stack up in terms of affordability and short-term feasibility? For this analysis, let’s return to the total energy production goal – if you assume 1,000 quadrillion BTUs of energy for 10 billion people, you achieve a per capital energy production of 100 million BTUs per person. In reality, global population will probably stabilize somewhat under 10 billion people, but 100 million BTUs per person is not enough – it really needs to be as much as twice that.

As we demonstrate in our feature “The Good, the Bad, & the BTUs,” citizens in the nations where per capita income exceeds $15,000 per year, consume on average 216 million BTUs per year per person. In the USA, per capita energy consumption is 327 million BTUs per year per person. If we assume the planet’s population will stabilize at 8.5 billion people, at 1,000 quadrillion BTUs of global energy production, per capita energy consumption will be 117 million BTUs per person. Even with extraordinary developments in energy efficiency, it is unlikely we can expect to deliver less than this amount of energy per capita, and still allow the world to achieve universal prosperity – global energy production will need to double.

The next chart shows the potential costs of adding another 500 quadrillion BTUs of energy to global energy production using non fossil fuel means – remember, to eliminate fossil fuel, you would have to add nearly 1,000 quadrillion BTUs to global energy production.

At current costs, adding 500 quadrillion BTUs of energy using
alternative energy sources would require well over $100 trillion.

Wind is the surprise winner in this survey – we’ve made some huge assumptions regarding cost per megawatt, however – using various sized production units across the energy sources hydro, nuclear, wind, roof PV, and utility CSP (concentrated solar thermal). At a cost of $2.5 million per 1.0 megawatt (installed), wind energy looks pretty good. Is this an accurate cost?

The key variables affecting these results are, along with the cost per megawatt, the operating availability percentages. Nuclear has at least a 90% up-time – it goes full bore almost constantly, which greatly lowers the cost for nuclear as a scaleable replacement to fossil fuel. But do we want 20,000 new nuclear power plants – actually these power stations can be 5-10x larger than 900 MW each – to get our 500 quad BTUs?

None of these solutions are cheap. Annual global economic output is in the 50-100 trillion range today, probably creeping up on $100 trillion. This figure is more subjective than you might think when simply compiling World Bank data. Using purchasing power parity metrics and devalued dollars, $100 tr. is probably not far off. But even at that, it would cost well over 100% of our entire global economic output for a full year to develop 500 quadrillion BTUs of new annual alternative energy capacity – using very favorable assumptions. Meanwhile global energy production needs to double as soon as possible. Pick your poison. Adapt.

Most of the data referenced here is based on a presentation by Ripu Malhotra, who is authoring a book on these topics to be published by Oxford University Press.

4 Responses to “The Reality of Fossil Fuel as it Accounts for 80% of Our Global Energy Supply”
  1. I am in the final stages of finishing my manuscript and wanted to double check my estimates for the cost of building capacity to produce a CMO worth of electricity by various sources against what you had estimated in your blog ( Your estimate for the number of nuclear plants seems much too high, even after I correct it by the CMO to 500 Quads ratio (about 3.76). I also want to go over the estimates for installing 1 MW capacity of different power production systems.

    Can you give me a call sometime? I want to make sure I understood your analysis correctly.



    Ripudaman Malhotra
    Chemical Science and Technology Lab
    SRI International

  2. Ed – Thanks for the spreadsheet. Although it was the nuclear number that had caught my attention, all your numbers for the alternatives (Hydro, Nuclear, Wind, PV, and CSP) are higher than mine by about a factor of 3. The reason is that because the rated capacity for all these sources is for electricity, and not thermal BTUs. Coal plants also provide electricity, but to get to that we consume ~10,000 BTUs for a kWh, and not 3412 BTUs, which is the primary energy equivalent). Since the EIA, DOE, BP and all others quote us the the consumption numbers (100 quads, or MTOE), which all end up telling us the thermal content of the fuel, I have used 10,000 BTU per kWh in all my calculations whenever considering “end-use” electrical energy.

    So, 1.0 CMO, which is 26.2 billion barrels of oil or 153 Quads, is
    3.27 times smaller than your your favorite unit, 500 quads. However, where you equate 33.4 gigawatt-years to 1 quad, I would use 2.93 quads, and lower the number of required units by that factor. With that correction, our numbers for hydro, nuclear and wind are in agreement.

    For rooftop PV systems you used a 2.1 MW system as the example. Your numbers are consistent, but a 2.1 MW PV system? That’s huge! (Google’s project covering 4.5 acres is only 1.6 MW) I had used a commercial 2.1 kW system for my analysis. Similarly, 900 MW for a Solar Thermal is larger than anything we have seen. They are in the 100 MW range. So, for these two if I apply the factor of 2.93, our numbers would be in agreement.

    Now for the costs. What was your source? $10 MM for 10 MW nuclear? I see a lot lower numbers, 2-7 depending on the length of construction and interest rates. Most hydro is being built in China for which $1.6 MM/MW is the number from the Three Gorges dam experience ($28B for 18 GW).

  3. Ed Ring says:

    Ripu – Your email is very, very interesting. Your point about “end-use” energy consumption is clearly valid, and I really need to adjust the numbers or make a clarifying comment in my post. If you agree, I would like to publish your remarks here as a corrective comment in the post.

    So here are some thoughts: I just use 500 quads because that appears to be roughly how much total worldwide energy consumption occurs per year at this point in history.

    I agree that 10,000 BTUs coal is the proper conversion to one kWh, but for natural gas I think the number probably is closer to 5,000 BTUs, at least for the modern plants we have in California. And I think there is some variation between the types of coal – and I don’t know to what extent electricity is generated from oil in other parts of the world, or what that efficiency averages.

    In general, end use of fossil fuel derived energy varies greatly in terms of the difference between the energy in the fossil fuel input and the energy delivered to the end-user. But if we normalize to units of electricity, the equation is probably not completely inpenetrable since once the generated power hits the transmission lines, all fuel sources are on an equal playing field. Local PV and wind of course even complicates this analysis since they may have minimal transmission losses.

    A conclusion one can draw – I haven’t read you book yet! – is that going to alternative electricity can essentially reduce the amount of energy we need source to deliver equivalent end user energy. I.E., if alternative electricity generation means 33.4 gigawatt-years equals 2.34 quads of fossil fuel input, not 1.0 quads – put another way 1.0 quads of alternative electricity offsets 2.34 quads of fossil fuel – it improves the picture in terms of how soon alternative electricity can make a dent in fossil fuel consumption. The task is still daunting.

    The high cost for nuclear power was simply a large number to avoid controversy. What is interesting is it still presents as an extremely cost competitive alternative, which was my point – especially since the costs for surge management with wind were not included in my estimates. Frankly I agree with your numbers, probably the lower ones at that, for the costs of nuclear power. The French have gotten the costs way down.

    The figures I used for a PV unit are obviously huge, I was just trying to keep the number of digits down in the figures on the table. Since my analysis was mainly to look at cost, it didn’t matter anyway. If the purpose of the analysis is to look at the number of installations required, the size you assume for an average system is much more consequential.

    In any event I like the conclusion – alternative electricity use offsets more fossil fuel consumption than one might initially calculate (as I did), by a factor of 2-3.

    Thank you for sharing these insights. Please let me know when your book is out.

  4. Ed – Your conclusion – alternative electricity use offsets more fossil fuel consumption than one might initially calculate (as I did), by a factor of 2x or even 3x – must be followed by the fact that this reduction is more than wiped out because of the need for higher installed capacity by a factor between 3-to-5 depending on the availability of the source.


Leave a Reply

You must be logged in to post a comment.