One terawatt (TW) is equal to one trillion watts. It is an enormous quantity of energy, equivalent to the output of a thousand Bonneville Dams. Nearly impossible to comprehend, but it happens to be a useful unit for measuring energy on a global scale. Human metabolism, at 100 watts per person and seven billion people, comes to 0.7 TW worldwide. Our energy use – all of the coal, oil, gas, uranium, and renewable sources together – stands at 17 TW and growing. We use about 25 times more external energy than we eat with our food. Here in the United States, we use about 3.5 TW, or 20% of the world’s energy with 4% of the world’s population.
Solar energy arrives on the Earth’s surface at a rate of 89,000 TW. It is sometimes said that this equals more energy in one hour than humanity uses in a year, but such a comparison ignores the realities of how difficult and potentially damaging it would be to convert this energy to usable forms. This is the sum total of the power of the planet: searing desert days warmed from freezing nights, daily storms drenching tropical rainforests, hurricanes in the remote southern ocean, and blinding glare reflected back into space from Greenland’s endless ice. We could not tap even 1% of this energy without throwing weather and other natural cycles into disarray, nor could we afford to build any infrastructure on that scale.
How much of this energy, then, could we realistically harvest? While energy can be harvested from waves, tides, salinity gradients, and other flows, I will focus here on the four biggest players: hydropower, wind, direct solar, and bioenergy.
Hydropower is the cheapest and most-developed of the three, with a little over 1 TW of installed capacity providing about 6% of the world’s energy demand. This could be tripled by building new dams and power plants around the world, though even with all of our rivers dammed hydropower would supply a relatively small fraction of our demand.
Available estimates of the power available in land and near-shore winds vary widely, but a recent conservative estimate suggests 18 TW (Adams and Keith 2014), with 0.3 TW currently installed. Wind turbines have a capacity factor (the ratio of power produced to the rated power of the turbine) of around 1/3, mainly because the wind is intermittent. This means that we would need to install 50 TW of generation capacity to capture all 18 TW of available energy. At one megawatt per turbine, that would require 54 million turbines, and at roughly $1/watt installed cost, that would be $54 trillion or nearly two thirds of global annual economic activity.
To generate 18 TW of power from 15% efficient photovoltaic panels, we would need to cover 0.4% of global land area. If we subtract mountains and glaciated lands, this equates to an average of four acres of panels per square mile, and it is similar to the amount of land (0.6%) covered by roads in the United States. Solar panels have capacity factors around ¼ and currently cost at least $2/watt to install, so the cost of a solar-only solution would be substantially higher than the wind scenario above.
Global photosynthesis – global biological conversion of sunlight to chemical energy – has been estimated at 80 TW, of which 40 TW occurs on land and 40 TW occurs in the oceans. Plants store this energy in chemical forms (starch, fatty acids, proteins) that are of little direct use to us. Producing a useful fuel like ethanol or biodiesel entails substantial energy losses and energy inputs, such that turning 100% of annual land primary production into ethanol would yield something on the order of 10-20 TW. Furthermore land-based photosynthesis includes all of the world’s food production, most of which is essential, and all of the world’s ecosystems, which we value both intrinsically and extrinsically. With this in mind, it is fair to say that biofuels cannot reasonably supply a large fraction of our 17 TW global energy demand.
These scenarios assume that we would have enough raw materials to build 50 million wind turbines or 225,000 square miles of photovoltaic panels. In reality, some of the rarer materials (such as the neodymium used in generator magnets) will prove limiting, requiring substitution with more readily available elements at a cost of lower efficiency. The take-home message in terms of global-scale sustainable energy is that while it would be theoretically possible to meet our current energy demand from renewable sources, such a transition would come with a huge expense both financially and ecologically, in terms of converted habitats, dammed rivers, mining of raw materials, and manufacturing.
It is thus imperative that as we shift toward sustainable energy sources, we also reduce our energy demand. Before I discuss practical ways to reduce energy use, however, I need to touch on three important aspects of sustainable energy: EROEI, fossil fuel dependency, and conversion/storage.
EROEI (Energy Returned On Energy Invested)
To tap any source of energy, from oil in the ground to wind on the prairie, some energy must first be invested. This is the energy to drill the wells, to refine the oil to gasoline, to build and maintain the turbines, and to build the transmission lines linking the turbines to areas of power demand. For any energy situation, it is thus possible to calculate a ratio of Energy Returned On Energy Invested, or EROEI.
As fossil fuel reserves are depleted, EROEI decreases as the remaining reserves require more energy to extract. The first oil wells had EROEI close to 100:1. This value declined to 30:1 by 1970 and 15:1 by 2005, and the much-touted shale oil “revolution” has an EROEI around 5:1. Theoretically any value above 1:1 is an energy source, but because society cannot afford to devote more than about 25% of its total labor and energy expenditure to the procurement of energy, the cutoff value for viability is in reality higher than 1:1 – at least 3:1 or 4:1.
In most cases, the EROEI of an energy source is closely reflected in its cost. The rising cost of gasoline over the last 50 or so years directly parallels the decreasing EROEI of crude oil. Similarly, wind energy (EROEI around 18:1) is now sufficiently profitable to stand on its own feet while solar photovoltaics (EROEI around 7:1) still require subsidies to compete with electricity from higher-EROEI coal. However, this cost-EROEI linkage can fail in a complex political economy, as exemplified by the corn ethanol boom in the US.
Depending on who did the study, the EROEI of corn ethanol is somewhere between 0.9:1 and 1.5:1, which if we take the average means we get slightly more energy out in the form of ethanol than we invest in the form of fossil energy. The energy in the ethanol, of course, comes from the sun via photosynthesis, but the inputs are nonetheless necessary. Growing corn on an industrial scale requires synthetic fertilizer (natural gas + oil) and tractor fuel (diesel), and only the kernels are harvested with the rest of the photosynthetic biomass left in the field. Then these kernels must be transported to an ethanol plant (more diesel), ground into flour (electricity), fermented, and finally distilled (coal for heat).
Corn ethanol did not come into existence as a logical way to convert biomass to liquid fuel. If that were the goal, we would be using a different feedstock – Brazil produces ethanol from sugarcane with an EROEI of around 8:1. Corn ethanol appeared on the market because US farmers were growing too much corn and needed additional demand to boost the price. The government got on board with subsidies allowing ethanol to compete with gasoline despite a much lower EROEI, boosters hyped up the new “green fuel,” and pretty soon everyone was burning 10% ethanol fuel with almost zero improvement with regard to our energy sustainability.
Fossil Fuel Dependency
The EROEI values for sustainable sources remain constant with time and may even improve somewhat with manufacturing innovations, while EROEI values for fossil resources are on an inexorable decline toward 1:1, the point at which more energy is required to extract and process the fuel than can be obtained by burning it. Optimistic economists use these numbers to forecast a smooth transition; as fossil fuels become unprofitable, sustainable sources will take their place. In a smaller, simpler society that would be true, but such a smooth transition is far from guaranteed today.
The problem derives from the following facts:
- Sustainable energy technologies all require significant up-front energy investments.
- Until sustainable energy makes up a larger proportion of our total supply, this energy investment must come from fossil sources.
- Scarce, essential commodities like fossil fuels fluctuate wildly in price.
The fluctuations in fossil energy prices create a challenging environment for investment in sustainable energy. When oil prices are high, it is politically difficult to prioritize sustainable energy manufacturing over meeting basic needs. When oil prices drop, sustainable technologies cannot compete on a price basis with the result being bankruptcy or government subsidy. This leads to an “energy trap” – a situation in which a transition to sustainable energy requires a large initial investment in fossil energy – and as fossil energy becomes increasingly scarce and valuable, this hurdle grows ever higher.
The solution to fossil fuel dependency is a “breeder” program that uses sustainable energy to produce more sustainable energy, thus freeing additional investment from the political and economic volatility associated with fossil energy. Such projects have been proposed, particularly for solar installations, but to date nothing of significance is in operation.
Conversion and Storage
Liquid fuels are in many ways ideal energy sources. Gases, like hydrogen, are too light to permit adequate storage density. Solids, like coal or wood, are burdensome to handle. Electricity, while easy to move, cannot readily be stored and must be produced at the rate it is consumed. Fossil fuels come in gaseous (methane), liquid (oil), and solid (coal) forms. We use the gases and solids primarily to generate electricity and heat, while we use the convenient liquids for transportation.
The three main sustainable energy sources – hydropower, wind, and solar – all generate electricity. Reservoirs permit dams to match generation to demand, but wind and solar have no such luxury. Furthermore, liquid fuels are in extremely short supply. It would be theoretically possible – using grid-scale batteries, compressed air storage, and electrolysis – to replicate our existing infrastructure with sustainable sources. However, all of these conversion and storage steps would require enormous investment of labor and energy, decreasing EROEI and limiting overall investment in energy generation.
Rather than attempt to “map” a sustainable energy infrastructure onto the status quo, it makes sense to adapt ourselves to a solar-, wind-, and hydropowered world. What might this look like in practice?
- Smart grids, manipulating discretionary electricity consumption (e.g. electric car charging) to match supply and demand.
- Surprise vacations for students and factory workers on cloudy days, or days with insufficient region-wide wind.
- A return to wind transport sails on the seas.
- Much less (liquid-fuel-dependent) air travel.
- Land commerce transition from diesel trucks and trains to electric railroads with overhead wires.
This can seem harsh in a world where we are accustomed to using as much energy as we want whenever we want it, but I personally look forward to a time when we must tune our energy use to match the cycles and patterns of our planet.
For a more thorough treatment of sustainable energy on a global scale, see my 2012 essay “Our Energy Future.” A few of my thoughts have changed since then, but I still stand behind it.
It is easy to philosophize about how a world powered by sustainable energy might look. Charting a path from here to there is much more challenging and will require dramatic change on both personal and collective levels. Next week I will focus on the personal level, examining which of our energy expenditures can be most easily reduced, eliminated, or replaced by small-scale sustainable sources.