Rain can grow corn as well as trees, and corn can grow people. Thus, knowing something about trees, corn, and people, a team of scientists can start with tree rings and learn how many Ancestral Puebloan people could have lived in an area. Meanwhile, archaeologists are able to use their techniques of digging and dating to learn how many people actually lived in an area. Teams of archaeologists and tree-ring climatologists did this research at the “end of the road” in the small, remote Long House Valley, which was not a trading center.

What they learned is striking, as shown in the figure.

Next, take a look at a similar history, from Oklahoma over the last century. The Dust Bowl of the 1930s was a major drought, made worse by various economic decisions about land use. Wonderful literature documents the terrible economy and environment, as people suffered and died.

Every person I ever met who studied the ancestral Puebloan people of Mesa Verde and surroundings has come away deeply impressed with the resourcefulness and cleverness of the people. The difference between Puebloan and Oklahoman success during drought is not because one group was smart and the other wasn’t. But the technologies and trade are vastly different (for many reasons!), and the people who could call on more tools and more help were more successful. Some of those tools were wind-powered, but most ran on fossil fuels, and success has increased as the use of fossil fuels increased.

Want to know more?
Take a look at the Enrichment called Burning for Learning

Do you ever empty the lawn-mower bag to get your dinner? Or chew up a handful of wheat or leaves from the maple tree? How about raw meat?

Cows can succeed by eating grass, but they have four stomachs and spend a lot of time “chewing their cud” to help break down the grass to be digested. Caterpillars can eat wheat or maple leaves, but a whole lot of a caterpillar is a digestive tract. And many predators eat raw meat.

But, we don’t do any of these things. We have mastered the art of using fire to cook our food. This kills parasites, but it also starts the process of digestion. We don’t have the type of digestive system that would allow us to get enough energy out of leaves and grass or “raw” wheat and raw meat, to keep us active enough to grow, harvest or catch those foods in the wild. If you are dieting to lose weight, eating raw vegetables is a great idea; if you are trying to survive the winter as a fur trapper in some remote part of the Yukon, you might look for something that supplies a bit more energy.

Fire may be the big difference between humans and other primates. If we didn’t cook, we wouldn’t get enough energy from our food to supply our big brains. Instead, we’d need a bigger or longer digestive system to process leaves and seeds and roots and raw meat, but the extra digestive system would use up a lot of the energy it extracted from such things to keep itself alive, with not enough energy left over to support all the extra gray matter between our ears. We really may have needed to burn to learn!

Meat fillets being grilled

Credit: Meat fillets being grilled by La llama de la barbacoa is licensed under CC BY-SA 2.0. Accessed Feb. 25, 2025.

We’ll probably never know for sure whether fire was really required for us to survive as humans, but there is no question that it makes life easier in many ways. Staying warm in an Arctic winter is much, much easier with a fire than without one. Fire helps in scaring away predators, killing bad things in food, and more. For example, the native people of the eastern US grew corn, beans, and squash in clearings in the forest. Chopping down trees with stone axes is not easy; “girdling” by cutting the bark will kill the trees, and fire can then be used to clear the land and keep it clear. (Slash-and-burn agriculture is not a new invention!)

Burning wood is just one of the ways that we humans use to get someone or something else to do some of our work for us. Rather than being limited by the energy we can get from our metabolism (the food we “burn” inside of us), we get lots of extra energy by burning other things outside of us. We burn coal, natural gas, and petroleum to generate most of our electricity and power our machines. We all use this energy, and our share of it is something like 100 times as great as the energy we consume in the form of food! So our external energy use is far greater than our internal energy use from food.

Shortly after the last ice age ended, hunter-gatherers in many parts of the world began settling down and developing agriculture. This switch to growing food may not have been possible during the highly variable climate of the ice age. This switch helped fuel a major growth in population that continues today. But, by many measures, the switch also caused the new farmers to become less healthy, eating a less varied diet and suffering from more diseases-disease organisms and parasites enjoyed it when their human hosts settled down close together, making it easy to cause more sickness! You will find LOTS of ideas about why our ancestors settled down and started growing crops. One big possibility is that the world was nearly full of hunter-gatherers-the good places for finding something to eat were already taken, people died in marginal areas during bad years, people didn’t want their children to die, so they developed a new “technology” to feed themselves.

Very few people today have spent enough time with a shovel or hoe to know how difficult agriculture can be, even with modern tools. Plowing and cultivating are hard work. So, perhaps as early as 8000 years ago, people were figuring out how to get oxen to pull plows. This was NOT an easy undertaking, requiring selective breeding to domesticate wild creatures, then feeding those creatures and protecting them from predators and keeping them from running away, and inventing yokes and plows and convincing the oxen to wear the yokes and pull the plows. Yet all of this effort and more was easier for early agriculturalists than actually doing the digging themselves. Once again, people were getting ahead by getting something else to do their work for them.

Mistakes in unit conversions really can cost an immense amount of money. We are NOT going to turn this course into a worksheet on unit conversions, and we won’t require you to memorize unit conversions, but we will explain some of the key points next—enough to let you keep your hypothetical job with the power company...and maybe a real job someday.

Words such as energy, work, and power are tossed around in casual conversation but have very careful definitions in engineering and science, and for the people who buy and sell energy. You can think of energy as the ability to do something. Wind up an old-style alarm clock, and the spring has stored some mechanical energy, which is available to move the clock hands and make the ticking sound. Water above a dam has gravitational potential energy and can flow down under gravity, driving a generator to make electricity, or floating your boat to the sea. The chemical bonds in the gasoline in your car have chemical energy, and if you make the gas hot enough with a spark in your engine in the presence of oxygen, the bonds will change and make the car go.

When you are “using” the energy, it is doing work. Pushing you across the country, or moving the clock hands, or moving your boat down the river, require overcoming friction and wind resistance and such. So does using a plow to break the soil and turn it over so you can grow food and lots of other things. How fast you use energy, or how fast you do work, is power. You do some amount of work in climbing the stairs to the next floor, but doing it in 10 seconds requires more power for a shorter time than doing it in 10 hours.

In terms of units, and how you’ll answer your boss about the Quebec contract, energy can be measured in calories. A Big Mac has just over 500 calories, so 4 sandwiches provide just over the 2000 calories that a typical person would eat in a day. Most of the world measures energy in joules rather than calories, and those 4 Big Macs are just over 8 million joules, which is the same as 8 megajoules.

If you were on a starvation diet, you might make those 4 Big Macs last a month—a low-power diet! But if you eat 2000 calories per day and “burn” them inside you to make you go—normal power for a person—that is the same as 8 million joules per day, or roughly 100 joules per second, which is called 100 watts. Amazing as it may seem, all your skills and brilliance and good looks and charm use energy at the same rate as one old incandescent light bulb! Your energy use—your personal power—is a bit higher than 100 watts when you’re up and doing things, and lower when you’re sleeping, but averages out to 100 watts. We don’t usually define a “people power” unit, but 750 watts, or 7.5 people, is a usual definition of 1 horsepower.

So, energy can be measured in calories or joules and is the ability to do something, while power in calories per day or joules per second is how rapidly you do it, and a watt is a shorthand way of saying joules per second. But, suppose you have 10 old-style light bulbs turned on all the time, so you’re using 1000 watts or 1 kilowatt. Each hour, the computer at the power company says you have spent another hour using 1 kilowatt, so they add the price of 1 kilowatt-hour of electricity to your bill. At the end of 24 hours, you are billed for 24 kilowatt-hours. Kilowatt-hours, like calories and joules, provide a way to measure energy. The Quebec company uses joules, you use kilowatt hours, and you’ll keep your job because you know (maybe with some help from the internet) that 1 kilowatt-hour is 3.6 million joules, so those Québécois are not going to beat you in a deal!

You should also know that there are many more ways to measure these things, which you do not need to learn now, but which you should know exist. People often use British Thermal Units, or BTU's, for energy, and BTU's per hour for power, but occasionally, they get sloppy and say “BTU” when they mean “BTU per hour”. Or, they get lazy and say that one quadrillion BTU's is a “quad” and just quit talking about BTU's. People who sell natural gas have figured out how many BTU's, or joules, or calories, can be obtained by burning a particular amount of typical natural gas, and how much space that gas occupies at standard temperature and pressure, so they may measure energy in cubic feet of gas, or cubic meters of gas, even though they know that this depends on temperature and pressure and the particular gas. Barrels of oil can be used the same way. And, it goes on from here—refrigeration workers in the US talk about power in terms of “tons of cooling” linked to the power needed to freeze a ton of water in a day (one ton of cooling is approximately 3510 watts).

Now, we hope it is obvious that unless you are planning to work in cooling, or you have rather strange friends you would like to impress, it is probably not a good idea to clog your brain with the conversion factor between watts and tons of cooling! But you should know that a few fundamental ideas such as energy and power have been made to look very complicated by having a lot of names and units. And you also should know that many jobs you might be hired for will require you to figure out: 1) how things traditionally have been measured; and 2) how to convert to what other people are doing in their jobs. And if you can’t do that reliably, there is a high chance that you will be fired!

Energy use in the US is dominated by fossil fuels—oil (or more formally, petroleum), gas (or more formally, natural gas), and coal (which is generally just called coal). Recently, fossil fuels have been totaling about 85% of energy sales in the US (and more-or-less 85% worldwide), with the rest of US use split more-or-less equally between nuclear and renewables. (In 2010, the US Energy Information Administration gave US energy supply as Oil 37%; Gas 26%; Coal 21%; Nuclear 8%; Renewables 8%. This was used to move us around (transportation 28%), to build things (industrial use 20%), to heat and cool houses (residential 11%) and to power our plugged-in gizmos (electricity 40%).

Video: U.S. Energy Supply (0:52)

We’ll revisit these issues later. US usage per person is a little smaller than some countries, but (much) larger than many others. Per person, the world averages roughly 1/4 of US use. Most of the world's economy is dominantly fossil-fueled with people often getting about 85% of their energy from fossil fuels as in the US, and energy is often about 10% of the economy.

For now, though, it should be evident that if we spend 10% of our money on energy, it impacts everything—jobs and security and environment and more. As we saw in last week's Discussion, there are great options for making money and saving money by doing things better in the energy business. But, over the last few decades, we actually have doubled the amount of economic activity squeezed out of each barrel of oil or ton of coal—bright people have been working on this, and making or saving much more money might take a lot of effort or some new inventions.

Perhaps most importantly, the current system is grossly unsustainable. As we will see in upcoming content, the store of fossil fuels in the Earth is limited, and we are removing them much more rapidly than nature makes new ones. With essentially everything we do relying on energy use and 85% of the energy system relying on unsustainable fossil fuels, a lot of things will need to change.

We normally think that the world contains matter-stuff-and energy (the ability to get the stuff to do something). And we often measure how much stuff we have by its mass. (Weight is the mass multiplied by the acceleration of gravity.) A real physicist would remind you, however, that mass and energy are different aspects of something more fundamental.

Einstein’s famous formula says that the energy content of something, E, is equivalent to its rest mass, m, multiplied by the square of the speed of light in a vacuum, c². Because c is so large, a reaction that converts a little bit of mass can produce a lot of energy that is radiated away, as in an atomic bomb, for example.

The numbers are really wonderfully large. If you could somehow make an Einstein reactor to convert the matter in the food you eat directly to energy, just 1 gram (one-fifth of a teaspoon of water) would be enough to supply your 2000-calories-per-day diet for 30,000 years!

Laws of Thermodynamics

Suppose you don’t have an “Einstein reactor”, so you’re working in the ordinary world, where any changes between rest mass and energy involve too little mass to be measured. Then, as described in the main text, energy is neither created nor destroyed, but it is changed from one form to another. This is often called the First Law of Thermodynamics, and also can be written that the change of the energy in a system is the amount of heat added to it minus the amount of work it does on its surroundings.

The first law of thermodynamics, by itself, might leave you thinking that after you burn the gasoline to move your car to drive to Grandma’s house, heating the surroundings, you could just collect the heat and the carbon dioxide and the water from your tailpipe, put them all back together again, put that gas back in your tank, and drive home. The Second Law of Thermodynamics says that you will fail; it is possible to use the heat to recombine things to make more gasoline, but you’ll never get as much energy back into the gasoline as you started with. “Disorder”, or “entropy”, increases, and the concentrated energy that is useful to us becomes spread out and no longer useful.

Physicists often discuss a zeroth law of thermodynamics, which says that if two things are in thermal equilibrium with each other (not having a net flow of heat from one to the other), they are in equilibrium with a third. This leads to a definition of temperature, and other useful things. And, there is a third law of thermodynamics which says that you can’t actually cool something to absolute zero, the point at which a perfect crystal would have zero entropy. These can be approximated as (this is often attributed to the British thinker C.P. Snow): You must play the game, but you can’t win, you could break even on a really cold day, but it never gets that cold.

Water doesn’t hold much oxygen, so lakes and the oceans are relatively low in oxygen, especially if the water is warm. Oxygen made in the water by growing plants tends to form bubbles that rise and escape to the air above. Aquariums often need “bubblers” to add air to the water and give the fish enough oxygen to breathe. Running water, or fast currents in the ocean, do this job in nature, picking up a little oxygen at the surface and taking it down to fish and worms and other creatures. But if the currents are slow and a lot of dead plants are sinking, the bottom of the ocean or a swamp or lake may have more plants to be “burned” than oxygen to burn them.

Sometimes “dead zones” form in ocean water above the bottom, where the decay of sinking plants uses up almost all the oxygen so that fish and other large creatures cannot live. Such dead zones are especially associated with places where runoff of human fertilizer from fields on land causes huge blooms of algae.

Video: Gulf of Mexico Dead Zones (1:33)

More commonly, though, oxygen is present in the water but scarce in the sediments beneath. Almost everywhere in lakes and the ocean, sediment is piling up at the bottom. This may include large pieces of rock—sand and gravel—washed into the water by rivers, or carried across by melting icebergs and dropped. Smaller pieces are more common—silt and clay, sometimes just called mud—with most of the small pieces washing into the water in streams, but some blowing in, and even a tiny bit sifting down from meteorites. This sediment also includes organic matter (dead plants and animals).

Strong currents carrying plenty of oxygen tend to carry away the small pieces of mud and the dead plants, leaving sand and gravel without much organic material, and with big spaces between the big grains that oxygen-bearing water can move through. Where currents are slow, mud and dead plants accumulate, and the tiny spaces make it hard for water to move through, carrying oxygen. As worms and bacteria start to burn the dead plants, the oxygen is exhausted and the burning stops. So, where lots of plants grow in still, warm water, dead ones tend to pile up in the mud at the bottom without being burned.

Want to learn more?

Read the Enrichment called More on Oxygen in Water at the end of the module!

Oil and Gas Exploration

For over a century, exploration for oil and gas—finding the next big field full of valuable fossil fuels—has involved locating oil and gas traps and drilling into them. Most commonly, this has involved “seismic” exploration (see the figure and explanation below). Nature figured out how to use this technique long before humans did. For example, a bat flying around in the dark “looking” for a moth to eat will make a noise, and listen to the echo off the moth, using the time and direction to locate the flying dinner. Dolphins can find their food the same way.

Video: Air Gun Vessel to find Oil (0:55)

Oil explorers make noises, and listen to the reflections from layers in the Earth, using the time and direction to locate the oil-and-gas-filled traps. Then, drillers drill into the traps, and pump the oil and gas out. (Sometimes, the pressure is so high in the trap at the start that the oil comes out of the hole without being pumped, as a “gusher,” see figures below.)

But, soon, the pressure down there is reduced, and a pump is needed. Occasionally, a gusher catches on fire, with sometimes disastrous consequences, see the figure below.

Increasingly, a new technique is being used to recover oil and gas. Shale layers often have a lot of hydrocarbon left in them that did not escape in the past. Drillers have learned how to bore down to a shale layer, then turn the drill and bore along in the layer. When the hole is long enough, the drillers pump fluids at high pressure into the hole, breaking the shale in a process called “fracking” (from “fracturing”) that mimics the natural process by which oil and gas escaped the shale. Human use of this process was apparently first invented by a veteran of the US Civil War, Col. Edward Roberts, who saw the fractures in the ground caused by an exploding Confederate shell, and went on to patent the technique of using explosives to fracture rocks and allow more flow into wells. The technique has been improved in many ways since.

In many ways, fracking is not revolutionary but evolutionary from older techniques for recovering oil and gas. Under best practices, fracking probably isn’t inherently more risky or dangerous than those other methods. The biggest difference is that fracking is used to recover oil and gas that are spread out over large areas rather than having a large quantity concentrated in one place. So, fracking takes lots more drilling and pumping and installing pipelines in more places. Fracking is more likely to be in someone’s backyard, or near it, so there are more people seeing it and hearing it and complaining about it.

The more drilling there is, the more chances there are for mistakes to be made, contaminating groundwater or otherwise causing problems for neighbors. The drilling can also bring other problems, including lots of traffic. For example, back on Sept. 23, 2011, an article by Cliff White in the Centre Daily Times, State College, PA noted “A review of inspections performed by state police on commercial motor vehicles used in support of Marcellus Shale gas drilling operations in 2010 revealed 56 percent resulted in either the vehicle or driver being placed out of service for serious safety violations” but that “Thanks to heavy enforcement, the noncompliance rate has dropped to about 45 percent in the most recent study.” And, in the same article, “…a trooper in gas-rich Bradford County, said during the initial ramp-up of activity in that area a few years ago, almost all of the vehicles used for gas drilling-related purposes that he stopped had “some degree” of noncompliance.”)

Fracking is done with high-pressure fluids to which certain chemicals have been added, as noted above, and some of those chemicals may be dangerous to humans. The fracking fluids plus salty brines from the rocks “flow back” out of the wells, and these flowback fluids must be disposed of in some way. Much of that disposal recently has involved injecting the flowback fluids into the Earth in special deep wells. This has caused numerous earthquakes, some of them damaging. (See, for example, USGS: Induced Earthquakes.) Fluid injection for other reasons also has caused earthquakes; fracking is especially important in this only because it generates so much fluid that is being injected. Note that while fracking has probably triggered a few small earthquakes directly, the main cause of earthquakes is this injection of flowback fluids.

Fracking is likely to be with us for a long time. And, it is likely to remain at least somewhat controversial.

Video: Process of Fracking (1:01)

Modern fracking injects sand into the new cracks, so they don’t close up. “Frackers” probably also use some chemical to let gas move more easily through cracks. (People who are uncomfortable because bacteria are making too much methane inside their stomachs can take an anti-gas pill containing a chemical—often simethicone—that allows small bubbles to merge into bigger ones more easily, which makes it easier for the gas to escape from the body; some chemical with a similar function may be used in fracked gas wells.) Other chemicals may be added to prevent bacteria from “eating” the valuable gas before we can use it, or for other reasons.

What is unconventional Oil?

You may also hear about oil shales and tar sands (see image below). These are sometimes called unconventional petroleum or unconventional oil, or something similar, and represent opposite ends of a spectrum: oil shales haven’t been cooked enough to make oil yet, and tar sands are the leftovers after cooking and dining.

Oil Shale and Tar Sands

Credit: “Oil Shale and Tar Leasing Problematic EIS.” n.a., archived at Wayback Machine. Accessed February 25, 2025.

Tar Sands Open Pit Mining, Alberta, Canada.

Credit: “Oil Shale and Tar Leasing Problematic EIS.” n.a., archived at Wayback Machine. Accessed February 25, 2025.

Tar sands, such as the huge deposits of Alberta, Canada (see images above), are like the much smaller tar deposits in the pits at La Brea, mentioned earlier. Oil contains many different types of molecules. When oil seeps to the surface, the smaller ones tend to evaporate, or to be used preferentially by bacteria, leaving the larger molecules behind. These larger ones don’t flow as easily, so the result is a thick, almost solid mass of “tar” (technically called “bitumen”). Native Americans were waterproofing their birch-bark canoes with Alberta’s bitumen when the first Europeans arrived, probably with no knowledge that early peoples of the Fertile Crescent of Mesopotamia also used bitumen to waterproof boats.

Because the bitumen is so “thick” (viscous), normal drill-and-pump techniques don’t work well. Many techniques are in use or being tested to separate oil from the sand or gravel in which it occurs. For shallow deposits, the tar-soaked sands can be surface-mined and then heated or mixed with appropriate chemicals to free the oil from the sand. For deeper deposits, injection of steam or hot air or other hot fluids can warm the bitumen enough that it will flow. Oil companies are even experimenting with setting small fires in wells, to make heat and gases that drive liquid hydrocarbon to other wells. All of these techniques have associated costs, including water and energy use. For now, much more energy is obtained from the oil recovered than is used in recovery, but the ratio is not as good as for “normal” oil, and is likely to get worse as the easier-to-recover tar sands are used up.

In contrast to the tar-sand “leftovers” from normal oil after bacteria have eaten a lot, oil shales are undercooked not-yet-oil. In many places, dead plants and mud accumulated, but without being buried deeply enough to get hot enough to break down the dead plants and make oil. The dead plants have typically been changed enough to get a new name (“kerogen”), but not to make oil that can be pumped out easily. This sort of deposit is called oil shale (Figures 13-15). (The names are NOT the easiest to deal with. Oil pumped out of shale may be called shale oil, but the shale from which that oil is pumped is generally not called oil shale. Instead, that shale is an oil source rock. The name “oil shale” is saved for those shales that haven’t been heated enough to make oil, but that could be in the future. Given our choice, most of us who work in these areas would pick clearer names, but no one asked us!)

Oil shale generally can be burned without additional refining, although it does not yield a huge amount of energy because the clay in the shale interferes with the fire.

Credit: Credit: “Oil Shale Activities.” Department of Energy., archived at Wayback Machine. Accessed February 25, 2025.

Large weathered oil shale sample showing bedding planes and fissures. Uinta Basin, Utah.

Credit: “Oil Shale and Tar Leasing Problematic EIS.” n.a., archived at Wayback Machine. Accessed February 25, 2025.

Close-up photo of fractured oil shale specimen showing weathered and unweathered surfaces. Uinta Basin, Utah.

Credit: “Oil Shale and Tar Leasing Problematic EIS.” n.a., archived at Wayback Machine. Accessed February 25, 2025.

Oil shale can be burned as-is, but the organic matter is diluted by the clay in the shale, so just burning doesn’t work really well. Most plans for future use involve speeding up the natural process, heating the rock to “pyrolyze” the organic matter, releasing oil and gas while leaving some organic material behind in the rock. This may be done in the ground, or after mining the shale. Because energy is needed to heat the rock, costs tend to be higher, and energy recovered lower, than for conventional oil in which the heat of the Earth acting over millions of years did the cooking for us.

Proven Reserves Explained

But, because of the heating needed to get oil from tar sands and oil shales, and the extra effort to drill deeper and frack rocks, some of the resources will be used up recovering the rest of it, increasing the rate of use. And, we don’t really know very well what the resource is; serious investors and regulators tend to rely on the proved reserves for good reasons.

You may also recall from earlier in the course that peak whale oil production from the US fleet was followed almost immediately by a tripling of the cost, even though oil production continued at fairly high levels during the following decades—impacts of scarcity are felt long before the resource is exhausted. People often assume that production of a resource follows a sort of bell curve, starting slow, then rising rapidly, peaking, declining and tailing off to almost nothing. If that model is accurate, then the peak—peak oil, or peak coal, or peak-whatever, occurs when half of the resource has been used. The whale-oil experience suggests that scarcity shows up and starts to restrain the economy just about then. If so, then we now may be much closer to problems from fossil-fuel shortages than we realize.

Another point worth considering is that some countries have a lot of fossil fuel, and others not much (see the "World's Proven Reserves" figure just below). And, who has fossil fuel and who uses it are not always the same. For oil, for example, for most recent years including 2016, the US was the third-largest producer, although the US was #1 by a small margin in 2015, based on data from the International Energy Agency. But, because the US is the largest user, being third in production isn’t enough, and the US is the largest importer.

Video: Where is Oil, Coal, and Gas? (1:28)

Notice that the sorts of numbers here can be “spun” in many, many ways in the public discussion. Estimates of how much we have already discovered in the ground are at least somewhat uncertain; estimates of how much is in the ground that we haven’t discovered yet, or haven’t learned how to recover, are much more uncertain. Businesses and companies might have reason to report optimistic numbers, or pessimistic ones, depending on what they want to accomplish or sell or buy just now. How long the resource will last depends on how fast we burn. Should we estimate using modern rates of burning, or future ones, and if future, what will they be?

The rise of gas and oil fracking in the US has led to a rapid increase in reserves. You may hear people talking about a century of gas, although others use numbers as small as 25 years for the US reserve. But, at the start of fracking, gas was only about ¼ of the US energy use, so relying on gas as our main fuel could bump the low-end estimates down to only about 6 years. You could find some justification for bragging about a century or more of gas or warning that we may run out in a decade or less, by carefully choosing which estimates to adopt and how to use them. The module is to be very careful about the first numbers you hear, and think and compare before using a number to make decisions on, say, where to invest your retirement fund! (This applies to what you see in this class, too; no one can give you the absolute truth on this topic!)

Introduction

Oil and gas form at extremely slow rates — 10’s of millions of years — so we can consider the oil and gas present now to be all that is available. We can wait around all we want and there will be no significant increase in the oil and gas. The total amount of oil and gas in existence on Earth is sometimes called the oil in place. We can only guess at this (somewhere around 6 trillion barrels of oil equivalent), but regardless of its size, we can probably only get about 50% of it out of the ground (this recovery factor ranges from 10% to 80% for individual oil fields). The recoverable oil and gas can be divided into two types of reserves — proven and unproven. Proven reserves are the oil and gas that we know about (which means we have a 90% confidence level about them), while unproven reserves are the oil and gas that we are less certain of, but we have some indication of their existence. These reserves are usually expressed in terms of barrels of oil equivalent and include both oil and natural gas.

It is estimated that our proven reserves are on the order of 1.5 trillion barrels of oil, and unproven reserves are thought to be in the range of 3 trillion barrels. Last year, we consumed 31 billion barrels of oil, and at this rate of consumption, we’ve got less than 50 years worth of oil in the proven reserves, and about 97 years worth in the unproven reserves. Now, move onto the first part of this assessment 1. The Simplest Case.

In this first case, we’ll just consider the proven reserves, and we’ll assume that the oil produced is a constant percentage of how much remains in the proven reserves. The logic here is very simple — if there is more oil, you can produce more in a period of time, while if there is less oil, you produce less in the same time period — but the percentage remains the same.

Here is what the system looks like as a STELLA model:

Since this model is simply meant to illustrate the general pattern of oil/gas production resulting from an assumption of how production works, we’re not going to worry about the actual values, but you can think of the starting amount of Proven Reserves as 100% of what we have. Every year, we produce oil/gas at the rate of 2% of however much remains in the Proven Reserves reservoir. The production flow transfers oil/gas into the Produced Oil reservoir, so we can keep track of the total amount of oil/gas produced over time.

Let’s see if we can predict what will happen by doing a few simple calculations. When the model first begins:

Proven Reserves = 100
production = 100 x 0.02 = 2

This will reduce the Proven Reserves by 2, so it becomes 100-2=98. Then, in the next year:

Proven Reserves = 98
production = 98 x 0.02 = 1.96

This will reduce the Proven Reserves by 1.96, so it becomes 98-1.96=96.04. So, in the next year:

Proven Reserves = 96.04
production = 96.04 x 0.02 = 1.92

Notice that the production is declining as time goes on, and the amount of decline is getting smaller. If this pattern continues, the production will follow an exponential decline curve — like this:

Take a few minutes to watch the following video and learn about the browser-hosted STELLA model interface, before running the model.

Video: Peak Oil Model 1 (2:13)

Now, let’s run the model and see what happens. Follow this link to the Peak Oil Model which should be set up exactly the same as the diagram above. Answer the questions either on the worksheet you downloaded or on a piece of paper to be submitted later to the Module 3 Summative Assessment (Graded) quiz. If you didn't download the worksheet on the main page of this assessment, do it now.

Oil and gas companies have certainly become better at what they do over time. Originally, they drilled near natural oil seeps and hoped for the best, but now, a good team of geoscientists can “see” exactly where the oil/gas is, and engineers can drill with great precision and then “stimulate” the oil/gas-bearing rock formations to squeeze as much oil/gas as possible out of the rocks.

One way to incorporate this into the model is to change the rate of oil/gas production, r, so that it increases as time goes on. To do this, we make a simple equation that says r = 0.0005 x TIME, so then when TIME is 10 years, r will be 0.005 and when TIME is 100 years, r will be 0.05. Other than this change, the model is the same as in experiment 1. The value 0.0005 is called tech rate in the model, and we’ll see what happens if we change it.

Let’s see how this change affects the history of oil/gas production. Click this link to run the model and then answer the following questions on your worksheet or on a sheet of paper to be submitted to a Canvas Assessment later. As you can see, the production of oil peaks in this case. It rises because r is increasing, but as r increases, the Proven Reserves is decreasing and eventually a point is reached where the product of these two numbers (the production) starts to decline.

In addition to technology, economics also plays a role in the production of oil/gas, in the sense that higher prices will motivate greater production. Let’s assume that the as the supply of proven reserves drops, the price will rise. As long as there is a demand for oil and gas, as it becomes more scarce, it will become more valuable. This is a pretty simplistic view of what determines the price of oil and gas — reality is much more complex, which is why prices fluctuate quite a bit over time. But it is hard to escape the basic reality that as a desirable commodity becomes scarce, its value goes up.

To make this change in the model, we need to add something that will calculate the price. This new model looks like this:

STELLA Model of oil/gas production

Click for a text description of STELLA model.

A diagram illustrating a system with interconnected elements related to oil production. It features two large rectangles labeled "Proven Reserves" on the left and "Produced Oil" on the right, connected by a line labeled "production" with an intermediary circle containing a valve symbol. Below this main line, there are several interconnected circles representing variables within the system. These include "price," "r," "price slope," and "tech slope," each incorporated into the flow with arrows indicating relationship directions. Arrows are predominantly red, connecting circles with each other and with the main line, suggesting feedback loops and influences among the variables

Credit: David Bice @ Penn State is licensed under CC-BY-NC-4.0

As before, production is defined as Proven Reserves x r, and r in this case is defined as price x tech_slope x TIME, so it once again has the increase over time that our previous model had, but it also increases as the price goes up. The tech_slope is just the slope of the increase in technology over time and the default value is 0.0002. Price here is defined as 0.01 + price_slope x (100 – Proven Reserves); price_slope is the slope of price increase relative to change in Proven Reserves, and is originally set to 0.05. At the beginning, Proven Reserves is 100, so this gives a price of 0.01 — very small. But, when Proven Reserves has declined to 50, we get a price of 2.51. This equation is not meant to be anything more than a way to make the price increase as the Proven Reserves get smaller. The value 0.01 at the front end of this equation is just there so that the price is not 0 at the beginning, which would then make r be 0 and no oil would ever get produced.

What we have created here is a system with a feedback mechanism. Here is how it works:

System diagram showing effects of production increase

Credit: David Bice @ Penn State is licensed under CC-BY-NC-4.0

If the production increases, then the proven reserves must decrease; this triggers an increase in price, which in turns triggers an increase in production. Notice that the starting point (production increase) and the ending point (production increase) are the same. In other words, the change at the beginning of the mechanism promotes more of the same — this is what is known as a positive feedback mechanism. Positive feedback mechanisms tend to cause an acceleration of change, sometimes resulting in runaway behavior. In contrast, the are other feedback mechanisms that tend to counteract change, encouraging stability; these are known as negative feedback mechanisms. Note that in this context, positive is not necessarily good, and negative is not necessarily bad.

Take a few minutes to watch the video below to learn more about the positive feedback mechanism the oil production model before running the next model.

Video: Peak Oil Positive Feedback (1:19)

This model has two pages of graphs to look at; the first one shows the Proven Reserves, Produced Oil, price, and production, and r (which combines price and tech slope), while the second one shows just the production. The second graph retains the results from previous model runs, allowing you to make comparisons as you make changes to some of the adjustable model parameters. If you want to clear this graph, hit the Restore Graphs button.

For our next experiment, we’ll try a different assumption about what drives oil/gas production — demand. The demand for oil and gas has risen over time due to an increase in the global population and an increase in the per capita energy consumption. Here is what this modified version of the model looks like:

Production Driven by Demand From a Growing Population model created by David Bice

Click for a text description of the oil/gas production model.

The image is a diagram titled "Carbon Reservoirs and Fluxes in a Forest Ecosystem." It shows how carbon moves and is stored within a forest environment.

• Design: It’s a flowchart with black text, boxes, and arrows on a white background, illustrating a simplified forest carbon cycle.
• Components:
  • Atmosphere: A box at the top labeled "Atmosphere" contains the text "CO₂" (carbon dioxide), representing the air.
  • Photosynthesis Arrow: A downward arrow from "Atmosphere" to "Trees," labeled "Photosynthesis," shows trees absorbing carbon dioxide to grow.
  • Trees: A box labeled "Trees" includes a small sketch of a tree with a rounded canopy and trunk, symbolizing living vegetation.
  • Respiration Arrow: An upward arrow from "Trees" to "Atmosphere," labeled "Respiration," indicates trees releasing carbon dioxide back into the air.
  • Litterfall Arrow: A downward arrow from "Trees" to "Soil," labeled "Litterfall," shows fallen leaves and branches adding carbon to the soil.
  • Soil: A box labeled "Soil" represents the ground where carbon accumulates from litterfall.
  • Decomposition Arrow: An upward arrow from "Soil" to "Atmosphere," labeled "Decomposition," shows carbon dioxide released back into the air as soil matter breaks down
• Layout: The diagram uses straight black arrows to connect the boxes, forming a cycle of carbon flow between the atmosphere, trees, and soil.
• Style: Simple and uncluttered, with text labels placed directly on or next to arrows and boxes for easy understanding.

Credit: David Bice @ Penn State is licensed under CC-BY-NC-4.0

Here, the population increases according to pop pct, which is the net growth percentage per year derived from historical data and then extrapolated into the future — so it is a graphical function that changes over time. The population starts at the 1800 level of 1 billion; the net growth % drops to 0 in 2100, and at that point, the population will stabilize.

The demand for oil/gas is represented here by per capita demand, which is essentially a percentage of the proven reserves per billion people. The per capita demand is another graphical function of time, patterned after actual history up until 2010 and then extrapolated to 2100 — optimistically assuming that the per capita energy demands will level off at about 2100. Multiplying the population times the per capita demand gives us r, the fraction of the proven reserves produced in a given year, and then r multiplied by the Proven Reserves gives us the production. The fraction r will increase as the population grows and as the per capita demand grows, and if population and per capita demand level off, so will r. Recall from experiment 2 that if r is increasing over time, a peak in production is inevitable.

Because we are using real population values and real values for the per capita demand, it makes sense to use real numbers for the Proven Reserves. At the present time, the best estimates are that there are 1.5 trillion barrels of oil as proven reserves (this number includes natural gas too), and we have consumed about 1.2 trillion barrels from about 1900 to the present. This means that at the beginning of time, our Proven Reserves will be 2.7 trillion barrels.

This model also includes a component called per capita oil that keeps track of how much oil is actually available per person, by taking the production and dividing it by the population. As per capita oil increases, we can use more and more oil for our energy needs, but as it decreases, we will have to either reduce our energy consumption or turn to other sources to meet our energy demands.

For our last experiment, we’ll see what happens when we add two more reservoirs, Unproven Reserves (the oil and gas that, we think, is likely to be discovered in the future) and Unknown Oil (the oil and gas we don’t know about, but might be there). Discovery adds Unproven Reserves to the Proven Reserves reservoir, and another flow called discovery adds Unknown Oil to the Unproven reservoir. An example from the Arctic Ocean region helps us get a grasp of these unknown reserves. In this frontier region, less than half of the offshore sedimentary basins have been explored, but based on what is known from more serious exploration off the coast of Alaska, the USGS estimates that there might be ~130 billion barrels of oil and gas — so this is a resource that, we think, might exist, but not enough is known about it yet to put it into the unproven reserves category, which applies to oil reserves that we know exist, but we don’t know enough about them to put them into the proven reserves. For perspective, this Arctic Ocean oil might represent 10-15% of all the unknown oil/gas that remains, and it would be enough to last for 4 years at the current rate of global use.

The discovery of these new resources is a function of a rate constant that increases over time, dictated by something called the exploration slope. The discovery flow that leads from Unknown to Unproven Reserves is set to be 1/5 the rate of the other discovery flow, reflecting the fact that it is much harder to discover something we know little about. Both of the discovery flows are controlled by switches (they can be turned on or off) and they begin (if the switch is on) at a time that can be set using the explor start time control knob. Here is what this new model looks like:

Video: Peak Oil Intro (2:44)

Before launching the model and experimenting with it, take a few minutes and watch the video that explains how to operate the switches that can turn the discovery flows on and off.

Open the model here, and first make sure the switches are in the off position (down), disabling the two discovery flows. Run the model and you should see exactly the same thing you saw in experiment 4B, with the difference that it runs for a longer period of time. If you study the graphs #3 and #7 show comparative plots of the production (in billions of barrels per year) and oil per capita (in barrels). Make sure you watch the video above to get a general sense of what happens when you turn on the switches.

You will be presented with one of the following 4 sets of initial conditions. Your answers to the following 3 questions will depend on which case you are presented.

We’ve just completed quite a few experiments, so it is a good idea to try to summarize a few important points.

1. When we talk about “peak oil”, we’re talking about the rate of oil production — how much oil/gas is brought to market in a given year — not the total amount of oil/gas on Earth (which peaked as soon as we started to pump it out of the ground!).
2. Improvements in our ability to extract oil/gas (i.e., improving technology) lead to a distinct peak in oil production.
3. If we assume that production is motivated by price, and the price goes up as the oil becomes more scarce, this also leads to a peak in oil production.
4. If the demand for oil is related to population, and population increases, this also leads to a peak in oil production. This effect is enhanced if the per capita demand for energy increases, as it has during the last 100 years.
5. If we include unproven oil reserves and unknown reserves into the system, we can make the decline in oil production more gradual, but there is still a tendency for it to peak.
6. At the end of the day, there is a finite amount of oil and gas available to us. This fact, combined with improvements in technology, an increase in demand for production related to price, increasing population, and increased standard of living, makes a peak in oil production inevitable — we cannot have a sustainable supply of oil and gas to fuel our economy. Facing up to this reality is important because it leads us to be more serious about making plans for a future where our energy needs are met without relying on fossil fuels.

Oxygen in Water Enrichment

We oversimplified slightly in the text above. Even after oxygen is used up burning dead plants in mud beneath an ocean or lake, a little more burning may occur as bacteria use other chemicals in place of oxygen. For example, bacteria may use the sulfate in sea water. The reaction can be written this way:

Sulfuric acid + plant → hydrogen sulfide + carbon dioxide + water + energy

In reality, the sulfate (SO₄^-2) will also be reacting with other things in the ocean, but this isn't too far off. Hydrogen sulfide (H₂S) is the source of “rotten egg smell”. It also readily reacts with iron in mud to make iron sulfide minerals, which initially appear black in the mud but which later may recrystallize to beautiful fools-gold pyrite if they have enough time and a bit of heat and other help. You might have seen this if you have visited a salt marsh. The mud in the shallowest parts of salt marshes is often black just below the surface, and releases rotten-egg smell if stirred up, because the marsh is growing lots of plants, the mud has little oxygen, and bacteria are using sulfate to burn the organic matter.

As mud is deposited at the bottom of lakes, the sea floor and elsewhere, it buries older mud with its organic matter. If you dig a hole, the material farther down was deposited longer ago, and has had more time to run out of oxygen and the other chemicals that are used to burn dead plants. One often sees a sequence going down in the mud in which, at the top, oxygen is used to burn organic matter, and then nitrate, manganese oxides, iron oxides, and then sulfate. If organic matter still remains, the next step is for bacteria to produce methane, CH₄, which is the main component of natural gas.

Something really interesting may happen next. At the pressures and temperatures we commonly see under water, methane is usually a gas, although at high pressure it can be liquefied for storage or shipping. But, if the pressure is high enough, the temperature low enough, and there is lots of water around, instead of making bubbles, the gas will combine with the water to make a special kind of ice. This ice is often called methane hydrate or methane clathrate. When samples are brought to the surface, are brought to the surface, they actually will burn (see figures below).

Figure 3: Gas hydrate recovered in shallow layers just below the seafloor during piston coring in the Gulf of Mexico.

Credit: Gas Hydrate Recovered from Gulf of Mexico. Energy and Mineral Resources Mission Area (USGS) (Public Domain)

Figure 4: Methane hydrate is sometimes called “the ice that burns” because the warming hydrates release enough methane to sustain a flame.

Credit: Gas Hydrate Burning. Energy and Mineral Resources Mission Area (USGS) (Public Domain)

There is a lot of clathrate under the sea floor in many places, and more in the Arctic in permafrost. (Yes, we know that we told you that warmer conditions favor burial of plants without burning, but this burial can happen in cold places as well, and freezing may actually help it happen by keeping worms and other creatures from eating dead plants before they are buried in mud. The frozen soils of the Arctic are rich in dead plants, and much methane is produced from them where thawing occurs without much oxygen.)

As mud is buried deeper and deeper by more sediment, the Earth's heat warms it up. At some depth, the ice melts to release bubbles of methane. When this process was first discovered, some scientists were worried that undersea landslides or other accidents might release giant methane belches that would sink ships (if a huge bubble rose right where a ship was, the ship could fall into the bubble), and change the climate, and cause other problems.

Additional research has reduced these worries, although they haven't gone away entirely. It is still just possible that a bubble might endanger a boat in certain special conditions, but we are fairly confident that huge amounts of gas can't come out really rapidly. As the clathrate is buried by more sediment, trapping the Earth's heat, the deepest ice melts to make bubbles. But, making those bubbles requires pushing water out of the way, which requires that the gas have high pressure. Pushing more water away needs higher pressure. At some high enough pressure, the gas will fracture the icy layer above and bubble out gradually, before enough gas can build up to make a climate-changing belch. Also, as clathrate forms, it uses the water but not the salt in seawater, and that salt may build up in water remaining in mud nearby, lowering the melting point so that some water doesn't freeze even if a lot of methane is supplied, allowing gas to move up through unfrozen regions to leak out at the sea floor.

As we will discuss climate change next chapter, methane in the sea floor may be very important for amplifying warming over decades and centuries, as warmer conditions melt the ice and let methane escape to increase the greenhouse warming. But, conduction of heat through the sediments to cause melting is rather slow, so we don't think that giant methane belches will change the climate even faster than that.

You can think of energy in the Earth’s climate in a way that is similar to the materials entering and leaving the car factory. Almost all the energy for the Earth system comes from the Sun. About 30% of this is reflected from clouds and the land surface, and the other 70% is absorbed and heats the Earth. (The reflected fraction is called the “albedo”, so we say that the Earth’s albedo is about 30%. We don’t worry much about the heat coming up from the deep Earth because it is almost 4000 times smaller than the absorbed heat from the sun.)energy

You know that when the sun rises in the morning, the temperature in the air can go up a lot, quickly. If all of the Sun’s energy stayed on Earth, everyone would be dead from overheating in much less than a year.

But, warmer things lose energy to colder things. Suppose you turn on an electric stove. As the temperature of the heating element (the “burner”) rises, it begins to heat the pot of water on top to boil the water for your spaghetti. If there isn’t a pot of water on top, you can see the burner begin to glow, radiating energy.

This stove burner is actually reddish, but looks purple because of a fancy easy-to-clean stove top

Credit: Richard B. Alley @ Penn State is licensed under CC-BY-NC-4.0

The burner is “glowing” even before you can see the glow, as you could prove to yourself if you watched it while wearing special glasses that can “see” in the infrared, which is a longer wavelength of energy than visible light. As the burner gets hotter, it radiates more energy. And, while it continues to radiate long wavelengths such as the infrared you can’t see, a hotter burner shifts more of its energy to shorter wavelengths you can see, going to red and then orange and yellow as it warms up.

If you keep giving the burner the same amount of energy, its temperature will increase until the outgoing and incoming energy are equal, and then the temperature will stabilize. If you then supply energy more rapidly, the burner will warm to a new level that radiates the extra energy. Always, the burner tends to that temperature at which incoming and outgoing energy are equal, a balance like the stuff going into and out of the factory. But, electricity comes in, and electromagnetic radiation goes out, much the way car parts go in and cars come out of the factory.

For the Earth, energy from the very hot Sun comes in, mostly in the short wavelengths of light we can see, and we send back infrared radiation at longer wavelengths. But on average, the total amount of energy going out is just about the same as the total coming in. At the present time, this incoming and outgoing energy is not exactly equal — a bit less is leaving, which means that the Earth is warming.

When a car drives out of the factory, there may be tiny cracks in the pavement that the tires roll over easily. And, the road may go down into a small valley and up the other side, again causing no trouble for the car tires. But, if there is a pothole of the wrong size in the way, the tire may drop in, bending the rim, blowing the tire and getting the car stuck. Going really slowly might allow the car to ease through the pothole without damage, and going really fast might jump the pothole, but a car at the wrong speed in the wrong place can fall into the pothole and get into trouble.

Dr. Alley was careful putting his bike tire into this pothole for a picture, but real potholes really do break tires sometimes.

Credit: Richard B. Alley @ Penn State is lincensed under CC-BY-NC-4.0

We are all familiar with such situations, in which interactions happen when the size or energy is “right”, but otherwise there is almost no interaction. This is very common in the air. The shortwave radiation from the sun does interact with clouds, and the very shortwave (ultraviolet) interacts with ozone (which helps protect us from skin cancer caused by the high-energy radiation), but otherwise most of the light from the sun passes easily through the air. However, the infrared radiation going back up from the Earth does interact with certain gases in the air, which are often called greenhouse gases. Radiation tends to be absorbed if it is at or near those wavelengths with the right energy to make a particular molecule wiggle or spin in a particular way.

A molecule that is wiggling or spinning because it absorbed radiation has extra energy—it is hotter than it was. It usually will quit wiggling or spinning by colliding with a neighboring molecule and passing the extra energy along; occasionally, the extra energy will be sent out as radiation instead. Most of the energy absorbed by molecules in the air was going up from the surface, and if they “re-radiate” the energy it goes in a random direction, which has the effect of reducing the radiation going to space and sending some back to Earth. In the more common case of collisions, even a rare greenhouse gas can heat the atmosphere by repeatedly absorbing energy and then colliding with non-greenhouse molecules.

Without greenhouse gases, the Earth's average surface temperature would be well below freezing —about 18°C or -0.4°F.

Want to learn more?
Read over Enrichment titled The Simplest Climate Model to learn more about this.

The greenhouse gases now in the air do keep the Earth’s surface warmer than it otherwise would be, and adding more greenhouse gases will cause more warming. There is nothing new, surprising, or honestly controversial in any of this. With a calculation something like the one in The Simplest Climate Model (read more about it in the enrichments), the French scientist Jean Fourier discovered in 1824 that something was keeping the Earth’s surface anomalously warm, and among the hypotheses he considered was that the atmosphere is acting something like glass holding heat in a container (perhaps the origin of the comparison to a greenhouse; see The Discovery of Global Warming). The British physicist John Tyndall showed in 1859 that gases in the air, including water vapor and carbon dioxide, were contributing to the greenhouse effect. And, in 1896, the Swedish physical chemist and Nobel Prize winner Svante Arrhenius did a fairly good job of calculating the global warming from the carbon dioxide released by the human burning of fossil fuels. (Through history, scientists have actually been better at calculating the effects of greenhouse gases than at realizing just how incredibly skillful fossil-fuel companies would become at supplying large quantities.)

The science of the greenhouse effect thus is not some new discovery but has a long history compared to such “recent” science as relativity (Albert Einstein, 1905) or quantum mechanics (Max Planck, 1900). The pioneers who explored radiation in climate science were giants of physics, chemistry, and mathematics, who saw the strong interactions between laboratory studies and application to the atmosphere.

Much of the work on the details of the interaction between radiation and gases in the air was done by the US Air Force just after World War II and applied to topics such as sensors on heat-seeking missiles, as told in the introduction to this chapter. A missile uses a sensor to “see” the infrared radiation from a hot engine, but greenhouse gases such as carbon dioxide and water vapor block the view in some wavelengths by absorbing that radiation. Because the gases interact with radiation traveling in any direction, and there is much more energy in those wavelengths going up from the sun-warmed Earth than coming down from military bombers, the warming influence of the greenhouse gases is unavoidable.

Because warmer things begin to radiate more energy very quickly, the Earth’s climate is very strongly stabilized, as noted in The Simplest Climate Model. Other processes may stabilize the Earth system by reducing changes or destabilize by amplifying changes.

Some stabilizers can be very important but tend to be very slow. We saw in the last chapter that warming reduces oxygen in the ocean, which makes the burial of organic matter easier. And, because the organic matter grew from carbon dioxide in the air, burying rather than burning the dead bugs lowers atmospheric carbon dioxide. Thus, if something such as a brighter Sun causes warming, fossil-fuel formation reduces the size of the warming. However, we also saw that fossil-fuel formation is a slow process because most plants are still “burned” by bacteria or living things; fossil-fuel formation can be very important over a few hundred thousand years or longer, but not over a few thousand years.

Some carbon dioxide is also picked up from the air by rain, forming a weak acid that breaks down rocks in a process called “weathering” because the weather is involved. The chemicals released from the rocks are used to make shells, some of which contain carbon dioxide. (A coral reef or a clamshell is calcium carbonate, usually written as CaCO₃, but sometimes written as CaO•CO₂, showing more clearly that it contains carbon dioxide.) Chemists use Bunsen burners in their labs for good reasons; warming almost always makes chemical reactions go faster. So, if the temperature goes up, chemistry removes carbon dioxide from the atmosphere more rapidly. If something such as a brighter sun raises the Earth’s temperature, this “rock weathering feedback” can remove enough carbon dioxide to cool the climate back close to the starting temperature in approximately ½ million years.

The Earth's climate is a complex system, and like most other complex systems, it is, partially controlled by many feedbacks. Feedbacks can affect many things. If we think about temperature, if a warming or cooling affects other processes that in turn change the temperature, those other processes are called feedbacks. A feedback that works against the initial temperature change to reduce its size is said to be stabilizing or negative; a feedback that increases the size of the initial change is amplifying or positive. The most important stabilizing feedbacks for Earth’s temperature are the almost instantaneous increase in radiation leaving the planet when the temperature rises, and the faster removal of carbon dioxide from warmer air to form shells and fossil fuels over hundreds of thousands of years.

At the in-between times, however, the most important feedbacks are positive. As a result, climate changes over years to millennia can be almost as large as changes over much longer times.

The most important of these positive feedbacks is warmer air picking up more water vapor from the ocean and plants, and carrying that vapor along, thus strengthening the greenhouse effect (or, colder air picking up less water vapor…).

Want to know more?
Read the Enrichment titled Carbon Dioxide is more Important than Water Vapor as a Greenhouse Gas.

The air doesn’t know why it is warm, so anything that warms the air—brighter sun, or more greenhouse gas, or alien ray guns—will increase evaporation from the ocean, amplifying the warming.

Note that this does NOT mean that the warming “runs away” and the Earth burns up, but just that the total warming is made larger by the feedback. Suppose the sun becomes enough brighter to warm the planet by 1 degree, based on the simplest climate model in the Enrichment, which doesn’t include the water-vapor feedback. Including the effects of the extra water vapor would increase warming to almost 2 degrees.

Another important feedback is linked to snow and ice. Most surfaces (forests, grasslands, cities, oceans, even deserts) absorb most of the sunshine that reaches them, but snow and ice reflect most of the sunshine reaching them. Warming melts snow and ice, causing the Earth to absorb more sunshine, which causes more warming. This ice-albedo feedback is not nearly as strong as the water vapor feedback under modern Earth conditions because most of the snow and ice occur in places and at times without a lot of sunshine (mostly in the winter, near the poles, and often under clouds that already are reflecting the sunshine; note that this feedback would be much more important if the temperature were cold enough for the ice to extend near the equator).

But, the water-vapor and ice-albedo feedbacks interact with each other. If the sun becomes brighter or carbon dioxide is increased by fossil-fuel burning, the resulting warming melts snow and ice and picks up more water vapor. Each of these causes more warming. But, the warming from the extra water vapor also melts some snow and ice, and the warming from loss of snow causes more water vapor to be picked up. Under modern Earth conditions, this still doesn’t “run away”, but it amplifies the warming still more. (The warming did “run away” on Venus, evaporating the oceans and causing the surface today to be hot enough to melt the metal lead, and such a fate awaits Earth most of a billion years in the future as the sun slowly brightens, although if we hang around and keep learning, we could “geoengineer” our way out of the problem, perhaps using techniques that will be discussed later in the course.)

The best current estimate is that, including changes in vegetation and clouds as well as snow and water vapor, doubling the concentration of carbon dioxide in the air and letting the climate come into balance will cause a warming of roughly 3°C, with fairly high confidence that the number is not less than 1.5°C or more than 4.5°C (or, a most-likely warming of 5.4°F, with the range of possibilities primarily between 2.7°F to 8.1°F). This number is usually called climate sensitivity and is widely discussed in climate science. Of the roughly 3°C warming from doubled CO₂, the direct effect of the carbon dioxide on Earth’s radiation is just over 1°C (roughly 2°F), with the rest coming from the positive feedbacks. The stabilizing effect of warmer bodies radiating more energy is included here. Some additional amplifiers are omitted (melting of seasonal snow and sea ice are included, but not melting of the Greenland and Antarctic ice sheets, for example), so over many centuries or millennia, the warming may be somewhat larger than 3°C. The very slow stabilizers are also omitted, but they do not become important until even further into the future. Despite hopes that the climate sensitivity might be low, the most recent studies have made it less and less likely that sensitivity is as low as 1.5-2°C (2.7-3.6°F), with a value close to 3oC (5.4°F) looking fairly likely.

First, let’s start with Figure SPM-1 from the Fourth Assessment of the IPCC, showing the history of carbon dioxide and some other greenhouse gases over the last 10,000 years. Ice-core data from multiple cores and labs cover most of the history shown, and overlap with the recent instrumental record, all with very close agreement. The recent rise is unprecedented in the 10,000 years shown. Based on additional ice-core records not shown, the greenhouse-gas levels are now above anything seen in the last 800,000 years. And, data from other sources indicate that carbon dioxide has not been this high for millions of years. (Note that much further back in history, nature did cause higher CO₂ levels, a topic to which we will return later.)

The figure shows “radiative forcing” as well as atmospheric concentration. The Earth absorbs 240 W/m² from the sun. The extra warming from rising CO₂ is somewhat similar, although not identical, to the warming from a brighter sun, so the effect of the CO₂ can be discussed in W/m^2. CBy January of 2017, atmospheric CO₂ was at a concentration of 405 ppm, up from 280 ppm before the industrial revolution, with the extra CO₂ giving a radiative forcing of roughly 2 W/m², equivalent to the sun getting almost 1% brighter. The contributions from methane (from rice paddies, cow guts, and other sources) and nitrous oxide (especially produced by processes in soil stimulated by nitrogen fertilizers and animal waste) are significant but smaller.

The amount of extra CO₂ now in the air, and moving into the ocean to make it more acidic, closely matches the CO₂ we know has been produced from fossil-fuel burning. The human source is roughly 100 times as large as the natural volcanic source, and volcanoes have not done anything bizarre recently, so cannot be blamed for the recent rise. CO₂ is moving into the ocean rather than coming out, so oceans cannot be responsible for the rise.

Furthermore, the atmosphere confirms that humans are responsible, as discussed in the ETOM film clip below and the Enrichment linked below.

Want to learn more?
Read the Enrichment titles Humans are Primarily Responsible for the Rise in CO₂..

So, yes, humans are increasing the greenhouse effect, primarily by producing CO₂ by burning fossil fuels, with very little uncertainty.

Video: SPM2 (2:20)

Natural and Anthropogenic Warming (launch image in a new window)

Greenhouse gases are not the only things that affect climate. But, climate changes have causes; there are no magical “cycles” that somehow change the climate without letting us know why. (There are cycles that affect climate, but they have causes, such as features of Earth’s orbit, that we understand; they are NOT magical!) So, we can assess what things are affecting the climate.

More than a century ago, the Earth was a little on the cold side in what is sometimes called the “Little Ice Age” because the sun was a bit dim and volcanic eruptions were putting up dust that blocked the sun. The sun brightened early in the 20th century, contributing to warming, as shown by the little red bar extending to the right for natural solar irradiance down near the bottom of the figure. But, over the last 30 years when satellites have given us the best data, the sun seems to have dimmed just a bit. We humans have cut dark forests and replaced them with more-reflective grasslands, cooling the Earth a little, and we have put up a lot of particles to block the sun, with notable cooling influence (you can find blue bars for these, extending to the left, in the figure).

You may meet someone who agrees that the Earth is warming, but argues that much of the change is natural. This is wrong; over the last few decades, warming has occurred despite nature pushing a little toward cooling, and human particles and land-use changes pushing more strongly toward cooling. The most likely answer for how much of the warming has been caused by our greenhouse gases is “More than all of it”, because of warming despite these other cooling influences.

Video: SPM 3 (1:07)

Temperatures, Sea Level and Snow Cover (launch image in a new window)

The temperature is going up. The figure shows a few of the indicators, but many more are known. Consider the next figure, for example.

Video: Surface Temperatures (1:38)

Decadal Land-Surface Average Temperature (launch image in a new window)

The Berkeley Earth project is an interesting attempt by a group, involving a lot of physicists who were not primarily climate scientists through much of their careers, to use private as well as public funding to re-calculate the temperature record from thermometers. The Berkeley work follows efforts by NOAA and by NASA in the US, and by a British group at the Hadley Center and the University of East Anglia, and other efforts by others, to calculate global temperature changes from thermometer records. You can see clearly in the figure that over recent decades when the data are best, the different groups get the same answer despite having different funding sources and different techniques. The temperature is going up.

Furthermore, if you throw away the records from thermometers in and near the cities and just look in the country, you see warming. Thermometers in boreholes in the ground show warming. Thermometers taken aloft by balloons (radiosondes), and thermometers looking down from satellites and analyzed in different ways, show warming. So do thermometers in the ocean.

The temperature-sensitive snow and ice also show warming. You would not go searching for this effect in the coldest places; if you start off at -40 and warm by a couple of degrees, the snow and ice won’t melt yet. But, the effects of warming are seen in loss around the edges, in space and time, of seasonal snow cover, river, and lake ice, seasonally and perennially frozen ground, mountain glaciers and more. The melting of land ice and the expansion of ocean water as it warms are driving the rise in global sea level. And, the great majority of significant changes in where plants and animals live, and when they do things during the year, are in the direction of warming. So, warming is occurring, despite natural and human pushes toward cooling over recent decades.

Want to learn more? Read the Enrichment titled Global Warming Did Not Stop Recently.

Better Climate Models and Weather Forecasts End, but Climate Forecasts Continue

The energy from the sun that reaches the top of Earth’s atmosphere is sometimes labeled S, in units such as Watts per square meter (W/m²⁾, and is approximately S=1370 W/m². Most of the energy leaving the sun misses the Earth and goes streaming off into space, but we intercept a little of it. This total energy reaching the whole Earth is just the Earth’s cross-sectional area multiplied by S, or $\pi r^{2}S$, where r is the radius of the Earth.

But, because the Earth is a sphere rotating under the Sun, this energy must be spread around the whole surface area of the planet, including the side facing away from the Sun, with a total area of $4\pi r^2$. Hence, the energy available per square meter of Earth’s surface is $\pi r^{2}S/4\pi r^2=S/4$.

However, recall that some of this energy is reflected back to space without warming the planet. We call the reflected part the albedo, and for the whole Earth it is roughly 30%, or A=0.3. The absorbed energy is 1-A=0.7. The average energy going to warm the planet is then S(1-A)/4.

The Earth radiates energy back to space, and this can be approximated by “black-body” physics. In this approximation, the outgoing radiation increases with the fourth power of the absolute temperature T (which is how many degrees you are above absolute zero), so outgoing radiation is $\sigma T^4$, where the constant σ, which is often called the Stefan-Boltzmann constant, has a particular numerical value = $5.67\ast {10}^{-8}W/m^2/K^4$ (that is, 5.67 times 10 to the negative eighth power), with temperature in Kelvins (K).(Some people like to write “degrees Kelvin” or “^oK”, and the same for “degrees Fahrenheit or ^oF” or “degrees Celsius or ^oC”, but it is OK to just use K, F or C.)

Incoming and outgoing energy come into balance, so we have the equation $S(1-A)/4=\sigma T^4$. You can substitute the numbers given just above for S, A, and σ, and then calculate T, the average surface temperature of the Earth. This will give you about 255 K, or -18 C or 0 F, which is well below freezing; the actual average surface temperature is close to 288 K, or 15 C, or 59 F. Our very simple model omitted the greenhouse effect, which keeps the Earth’s average surface temperature above freezing.

Because radiation increases as the fourth power of absolute temperature, the climate is very strongly stabilized. A 1% increase in average temperature causes approximately a 4% increase in radiated power, which means that even a relatively large change in the brightness of the sun, or in other factors affecting the climate, will have a moderately small effect on the temperature. Without this strongly stabilizing effect giving us the climate we have, we might not even be here!

Better Climate Models

Climate models may be the part of the science that most people know the least about. Be very clear-scientists do not tell their computers to produce global warming, and then get excited when global warming comes out of the computer!

The simplest climate model we just discussed shows you a tiny bit of what goes into a real climate model. The starting point is physics. This includes the rules that mass and energy are not created or destroyed but just changed around. The physics also includes interactions between mass and energy-how much energy is needed to evaporate an inch of water per week, for example, or to warm the atmosphere by a degree. Interactions of radiation and greenhouse gases are specified from the fundamental physics worked out by the US Air Force after World War II, and other such studies.

The model also must “know” about the Earth-how much sunshine we get, how big the planet is and how fast it rotates, where the land and oceans are, how much air we have and what it is made of. (Climate models are applied to other planets, and very clearly give different answers because of the differences between the planets.)

All of this information is written down in equations, translated into computer language, and then the computer is turned on. What happens next is remarkable-the computer simulates a climate that looks like the real one. Air rises and rains in the tropics, then sinks and dries over the Sahara and Kalahari. Storms scream out of the west riding the jet stream, and snow grows and shrinks with the seasons across the high-latitude lands.

The model will not be perfect, of course. Suppose you are interested in wind speed. You know from personal experience that you can hide behind a windbreak for relief on a windy day. A forest can serve as a windbreak, giving weaker winds than on a prairie. So, the model must be “told” about the distribution of forests and grasslands (or else must calculate where they grow), and about the “roughness” of the forest and the grass. Scientists have conducted studies on the effects of forests and grasslands on winds, but all studies include some uncertainty. So, the modelers know that the surface roughness in this region must be about this much, but could be a little less or a little more within the range allowed by the data.

The modeler (or more typically, the modeling team) can now “tune” the model. If the winds in the model are a little stronger in some places than in the real world, the modeler may increase the roughness a little, although without going outside the uncertainties. To avoid any biases, different groups in different countries with different funding sources build different models, and tune them in different ways; when all of them agree closely, it is evident that the tuning hasn’t controlled the answer.

Some of the models are used for weather forecasting and for climate studies, and work fine for both. There are differences between weather and climate (see Weather Forecasts End, But Climate Forecasts Continue) - many climate models are simulating changes in vegetation, for example, but if you’re worried about the weather for next week, you don’t really care whether global warming endangers the Amazonian rainforest over the coming decades.

As a general rule, in talking to the public or policymakers, climate modelers rely especially on those results that:

• are exhibited by a range of models from simple to complex run by different scientific groups;
• are understood based on the physics;
• are observed in the history of climate; and
• are confirmed by recent instrumental observations.

No one has succeeded in forecasting the weather more than a week or two into the future, and we’re confident that such forecasts are impossible because of “chaos”. But, this difficulty does not interfere with the ability to project climate changes much, much further into the future.

For weather forecasting, you need to start with the current state of the atmosphere. If there is a cold front sweeping eastward across North Dakota in the US, areas just to the east in Minnesota are likely to experience the effects of that cold front soon. However, if the cold front has already passed across Minnesota, a different forecast will be more accurate.

This difficulty arises from the fact that no one can ever perfectly know the current state of the atmosphere everywhere (nor can we calculate perfectly, but let’s focus on the data here). If you give a good forecasting model the best available data, the model will produce a forecast that is demonstrably skillful for the next week or two, but the further you look into the future, the lower the skill, until the model is not able to predict the details of the weather. The model still produces “reasonable” forecasts—for summer in North Dakota, it will produce summertime conditions, not wintertime ones—but there is no skill for forecasting whether a cold front is coming in 26 days, or 27.

Suppose you now take your best data, and “tweak” them within the known uncertainties in the original measurements and the interpolations between the measurement stations. If the temperature in Fargo at noon on June 23, 2012 was 87.1 F, you don’t really know whether that was 87.100 or 87.102 or 87.009, nor do you know the exact temperature in all of the suburbs of Fargo that lack thermometers. So, take the 87.1 and try replacing it with the possible value 87.102, fully consistent with the available data. Make similar tweaks to other stations. Then, run the model again. What happens?

For the first few days, the forecast is almost unaffected. But, as you look further into the future, the forecast becomes more and more different from the original one. If you do this again, with different tweaks to the data (say, 87.009 rather than 87.100), you again will get almost the same forecast for a few days, but further out the forecast will differ from both of the prior ones. Do this a lot of times, and the odds are good that one of the runs will end up being close to what happens in the future, and that the average of the runs will be similar to the average behavior of the weather over a few decades (unless climate is changing rapidly!). But, you won’t know which individual run is the right one. This “sensitivity to initial conditions” is often called “chaos” in public discussions, and it means that weather forecasts can’t be accurate too far into the future. In the same way, you cannot predict the outcome of the roll of dice in a game until the dice have almost stopped moving.

Note that you can predict the average outcome of many rolls of dice, and you can predict the average behavior of weather over many years, which is climate. You may have met someone who argued that failure of a weather forecast casts doubt on climate-change projections, but that is like using one roll of dice to argue that if you keep gambling you’ll beat the casino. People who make that mistake at casinos are usually known as “poor” or “broke”.

Video: Chaos and Phil the Groundhog (PSUrockvideo) (3:50)

This may seem strange. If you track what happens to the radiation leaving the Earth's surface, some is absorbed on the way, and some goes straight out to space. Water vapor, carbon dioxide, and clouds dominate the absorption, with all the others (methane, ozone, chlorofluorocarbons, nitrous oxide, etc.) also enough to be important if taken together. (Clouds also have a slightly more important role in blocking the sun, with the net effect of clouds being slight cooling under modern conditions.) Because some radiation is blocked almost entirely by only one gas type, but other wavelengths may interact with both water vapor and carbon dioxide, there is a bit of uncertainty in the bookkeeping of the exact importance of a single type of greenhouse gas. Overall, though, it is fairly accurate to say that water vapor supplies close to half of the total greenhouse effect, clouds and carbon dioxide each a little under a quarter, and all others just under a tenth.

But, the amount of water vapor in the air is equal to the amount of rain that falls on the Earth in just over a week. As water vapor rains out very rapidly, it is replaced by evaporation of more water. Any extra water vapor we put in the air from burning of fossil fuels or irrigating crops just doesn't stay up there very long. And, because the natural source of water vapor is so huge (evaporation from a giant ocean and a lot of plants that together cover almost the entire Earth), the human source is actually tiny in comparison. The only practical way we know of to greatly change water vapor in the air is to change the temperature. A hair dryer has a heater for good reasons, and warming the air will allow it to pick up and carry along more water vapor, whether the warming is caused by carbon dioxide, or a brighter sun, or some sort of heat ray from space aliens, or anything else.

Some research has looked at what would happen if carbon dioxide were removed from the atmosphere. Loss of the carbon dioxide cools the planet, but that condenses some of the water vapor, which cools the planet more, and the Earth turns into an ice-covered snowball. If water vapor is removed, a lot more evaporates quickly before the Earth can freeze.

So, yes, water vapor is blocking more energy than carbon dioxide today. But, carbon dioxide is much more important for changing the climate than is water vapor. Carbon dioxide can be a forcing—add it to the air, and you force the climate to change. Carbon dioxide also can be a feedback—change something else (such as reducing oxygen in the ocean to allow more fossil-fuel formation), and that changes carbon dioxide in the air, which in turn changes the temperature. But, water vapor is almost entirely a feedback because there aren't any natural or human processes other than changing the temperature that can put water vapor up fast enough to make a big difference to climate.

Bookkeeping by itself shows that humans are responsible. We produce roughly 100 times more CO₂ than volcanoes do (maybe only 50 times, maybe closer to 200 times, if you include the uncertainties, but something like 100). Nature was producing its CO₂ for a long time, but humans have increased from being a very small source to being much more important than volcanoes.

Furthermore, several tracers in the atmosphere confirm the bookkeeping. These include:

• Dropping oxygen. The level of oxygen in the air is dropping as CO₂ rises. You'll still be able to breathe with no trouble, but this shows that the rise in CO₂ is caused by fire, which uses oxygen, and not by melting beneath a volcano or CO₂coming out of the ocean, which don't use oxygen.
• Falling carbon-13. Plants preferentially use CO₂ containing the lighter and faster-diffusing carbon-12 rather than the heavier carbon-13, so CO₂ formed by burning living plants or fossil fuels is more enriched in carbon-12 than is CO₂emitted by volcanoes or coming from chemicals dissolved in the ocean. The CO₂ in the air is becoming more enriched in carbon-12 and less enriched in carbon-13, which shows that the extra CO₂ comes from living or dead plants.
• Falling carbon-14. Carbon-14 is made in the atmosphere by cosmic rays, then taken quickly into living plants, but decays radioactively over about 50,000 years. (Half of the carbon-14 decays in just about 5730 years, half of the remainder in the next 5730 years, and by 50,000 years or so almost none remains.) Thus, living plants contain carbon-14, but fossil fuels don't. Carbon-14 is becoming less common in the atmospheric CO₂, much more rapidly than can be explained by radioactive decay, showing that the extra CO₂ in the air is coming from plants that have been dead a long time.

Taken together, bookkeeping says that the rise in atmospheric CO₂ is coming from human burning of fossil fuels. And, the atmosphere says that the rise in CO₂ is coming from burning of plants that have been dead a long time. The agreement is beautiful, confirming that we are responsible for what is occurring.

There is a bit more complexity to this, linked to our burning of forests, but also letting some forests grow back and fertilizing others, and linked to us releasing some CO₂ while making cement. But overall, the biggest source of CO₂ is our fossil fuels, and this will become more and more important in the future if we continue on our present path.

The main text presented some of the evidence that temperature is rising. But, the climate is influenced by the 11-year sunspot cycle, the occasional sun-blocking influence of particles from big volcanic eruptions, and also by the sloshing of water in the tropical Pacific Ocean associated with El Nino and La Nina-when the hot waters spread along the equator in an El Nino event, some heat moves from the ocean to the air, and when the cold waters of La Nina follow, heat flows back into the ocean. An extra El Nino, or an extra-strong one, in a decade can make global warming look very fast, whereas an extra La Nina can temporarily slow the upward march of temperature from the rising CO₂. This sort of sloshing cannot ultimately change the warming of the planet, but can make it appear more variable, and control whether the air warms fast and the ocean much slower, or whether faster warming of the ocean slows the atmospheric warming a bit.

Think for a minute about a neighbor taking a very active dog for a walk. Watch the person, and you can see steady progress down the street. Watch the dog, and you may have to study carefully for a while to even know which way they are going. You may find it useful to think of the year-to-year temperature changes as the dog, and the average behavior as the person. Please click the image below to watch the animation.

Animated gif demonstrating trends.

Click here for a text description of the Global mean temp Celsius vs year graph.

Frame 1 (Early 1960s to 1970s)

• The graph initially shows data only from the 1960s to late 1970s.
• The red temperature line fluctuates, but appears to decline slightly.
• A black trend line slopes downward, suggesting cooling.
• Text displayed: "Dr. Alley was born in 1957, at the start of cooling?"

Frame 2 (Extending into the 1980s)

• The graph expands, now including temperature data up to the early 1980s.
• The red line still fluctuates, but some warmer years start appearing.
• The black trend line slightly flattens, though cooling is still implied.
• Text changes to: "Oh wait... we forgot some data."

Frame 3 (More 1980s Data Appears)

• Additional data from the mid-to-late 1980s appears.
• The red temperature line rises more visibly.
• The black trend line flattens further, reducing the cooling trend.
• Text now reads: "Oops... forgot some more data."

Frame 4 (1990s Data Added)

• The timeline now extends through the 1990s<.
• The red temperature line shows a noticeable warming trend.
• The black trend line is nearly flat.
• Text changes again to: "Hmmm... looks like it's getting warmer."

Frame 5 (Early 2000s Data Appears)

• Data from the early 2000s appears, reinforcing warming.
• The red temperature line climbs further, surpassing earlier fluctuations.
• The black trend line now slopes slightly upward.
• Text now reads: "Maybe the cooling was just a fluke?"

Frame 6 (Late 2000s Data Added)

• The graph extends into the late 2000s.
• The red temperature line is now clearly trending upward, confirming warming.
• The black trend line now visibly slopes upward.
• Text changes to: "Yeah, definitely a fluke."

Frame 7 (Final Frame - Full Data to 2010)

• The graph completes the dataset, extending into 2010.
• The red temperature line reaches its highest points, solidifying the warming trend.
• The black trend line fully reverses, now clearly showing long-term warming.
• Final text displayed: "Never mind!

Credit: Richard B. Alley @ Penn State is licensed by CC-BY-NC-4.0

Next, take a look at the figures, which highlight events from Dr. Alley’s career. In each case, the jagged red line connects the temperatures from year to year, using data from NASA’s Goddard Institute for Space Studies, and the smoother black line is the best fit to the data over the interval selected. You will see that in each case, Dr. Alley has carefully picked the end points so that the best-fit line slopes downward, indicating a cooling trend. For the last 20 years, Dr. Alley has met important people in Washington, DC who declared that global warming stopped. It is very easy to do so; be quiet during a year with strong warming, and then the next year go back to claiming that global warming stopped.

Over a century ago, the Guinness brewery in Ireland hired an Oxford mathematician, W.S. Gosset, to develop ways to separate actual trends from short-term variability. The techniques were published with a pseudonym (A. Student), presumably to help people without telling competitors how valuable it was for a business to avoid self-delusion. If you apply techniques derived from that research, global warming has not stopped; all time intervals long enough to show a statistically significant trend do show warming.

By 2016, the temperature had risen enough that it barely fit in the chart above, aided by the ongoing human warming and by a strong El Nino event. This strong El Nino was warmer than the previous one, which was warmer than earlier ones, mostly because of human CO2. But, temperature was dropping a bit in late 2016 as the El Nino faded. And, some inaccurate voices were already, again, declaring that global warming had stopped.

Some Better Climate Models

Climate models may be the part of the science that most people know the least about. Be very clear — scientists do not tell their computers to produce global warming, and then get excited when global warming comes out of the computer!

The simplest climate model we just discussed shows you a tiny bit of what goes into a real climate model. The starting point is physics. This includes the rules that mass and energy are not created or destroyed, but just changed around. The physics also includes interactions between mass and energy — how much energy is needed to evaporate an inch of water per week, for example, or to warm the atmosphere by a degree. Interactions of radiation and greenhouse gases are specified from the fundamental physics worked out by the US Air Force after World War II, and other such studies.

The model also must “know” about the Earth-how much sunshine we get, how big the planet is and how fast it rotates, where the land and oceans are, how much air we have and what it is made of. (Climate models are applied to other planets, and very clearly give different answers because of the differences between the planets.)

All of this information is written down in equations, translated into computer language, and then the computer is turned on. What happens next is remarkable — the computer simulates a climate that looks like the real one. Air rises and rains in the tropics, then sinks and dries over the Sahara and Kalahari. Storms scream out of the west riding the jet stream, and snow grows and shrinks with the seasons across the high-latitude lands.

The model will not be perfect, of course. Suppose you are interested in wind speed. You know from personal experience that you can hide behind a windbreak for relief on a windy day. A forest can serve as a windbreak, giving weaker winds than on a prairie. So, the model must be “told” about the distribution of forests and grasslands (or else must calculate where they grow), and about the “roughness” of the forest and the grass. Scientists have conducted studies on the effects of forests and grasslands on winds, but all studies include some uncertainty. So, the modelers know that the surface roughness in this region must be about this much, but could be a little less or a little more within the range allowed by the data.

The modeler (or more typically, the modeling team) can now “tune” the model. If the winds in the model are a little stronger in some places than in the real world, the modeler may increase the roughness a little, although without going outside the uncertainties. To avoid any biases, different groups in different countries with different funding sources build different models, and tune them in different ways; when all of them agree closely, it is evident that the tuning hasn’t controlled the answer.

Some of the models are used for weather forecasting and for climate studies, and work fine for both. There are differences between weather and climate (see Weather Forecasts End, But Climate Forecasts Continue) - many climate models are simulating changes in vegetation, for example, but if you’re worried about the weather for next week, you don’t really care whether global warming endangers the Amazonian rainforest over the coming decades.

As a general rule, in talking to the public or policymakers, climate modelers rely especially on those results that:

• are exhibited by a range of models from simple to complex run by different scientific groups;
• are understood based on the physics;
• are observed in the history of climate; and
• are confirmed by recent instrumental observations.

Short version: Increasingly strong evidence shows that natural changes in carbon dioxide have been the main control on Earth's climate history and that the climate changes have greatly affected living things.

Friendlier but longer version: During the late 1700s and early 1800s, scientists were building the geologic time scale, drawing “lines” to separate history into blocks of time that could be given names. Fossils showed the species that lived at different times, and the lines were usually drawn when many species became extinct before new species evolved to take over the “jobs” left vacant by the extinctions. Those early geologists didn’t know why the species went extinct, but they knew that something big happened.

Since then, an immense amount of effort has gone into learning what happened. In one case about 65 million years ago, a giant meteorite impact killed the dinosaurs and ended the Mesozoic Era, to start the Cenozoic Era. Changing climate was responsible in other cases, and climate changes may prove to have been the main drivers in most of the big extinctions. Climate change was probably very important in how the meteorite killed the dinosaurs, too; for most of them, it didn’t fall on their heads but instead blocked the sun with dust it kicked up, causing great cooling for a few years, among many changes.

We’ll look briefly at three big changes, and then see what they say when viewed with the rest of climate history. Don’t worry about memorizing names and dates we’ve already given or the ones coming unless you’re really into that; just get the sense of the story.

Video: Continental Glaciation (3:27)

(launch image in a new window)

At the end of the Permian Period, which also is the end of the Paleozoic Era about 252 million years ago, approximately 95% of the species known from fossils went extinct. This is the same time, with very little uncertainty, as the greatest volcanic outpouring on Earth in the last 500 million years.

The rise in CO₂ from the volcanic eruptions caused warming. (Volcanoes generally cause cooling over short times, such as their role in causing the Little Ice Age of a couple centuries ago, but volcanoes raise temperatures over longer times, such as their role in warming the end of the Permian.

Do you want to learn more?
Read the Enrichment titled Volcanoes Cool and Warm, without Doubletalk.

The volcanic eruptions are estimated to have raised CO₂ much more slowly than humans are doing, but the volcanoes didn't run out of CO₂ as rapidly as we will run out of fossil fuels, so the event back then lasted longer. Our understanding indicates that the extra warmth from the CO₂ accelerated rock weathering, providing extra fertilizer reaching the ocean. This would have helped make extensive “dead zones” as parts of the ocean ran out of oxygen, aided by the lower oxygen level in the water caused by the higher temperature. Sediments from that time contain special “biomarker” molecules made by green sulfur bacteria that photosynthesize with the poisonous-to-us gas hydrogen sulfide, indicating loss of oxygen and rise of hydrogen sulfide in the ocean. New data also suggest the Earth became so hot that the few remaining large creatures could not live in the tropics immediately after the extinction, but only closer to the poles.

Trilobites (specimen from the Grand Canyon) and eurypterids (the State Fossil of New York) lived for well over 100 million years, but did not survive the great dying at the end of the Permian.

Source: Pictures from the US National Park Service, trilobite and eurypterid.

We do not expect the warming in our near future to produce anything nearly so bad, but fertilizer runoff from our fields and warming from our CO₂ can contribute to oceanic “dead zones”. And, we cannot rule out the possibility that beginning or near the end of this century, we could make the Earth so hot that living unprotected in the tropics becomes difficult or even impossible for us and some other large creatures.

PETM

In case you find yourself getting confused by the geologic names, or you just want to see whether you remember the names you learned sometime in the past, here is a fantastic image from the United States Geological Survey showing the geological time scale.

Click here for a text description of the Geologic time scale image.

The image is a colorful, spiral-shaped diagram illustrating the geologic time scale and the history of life on Earth. The spiral starts at the center and winds outward, representing billions of years of Earth’s history from its formation to the present day.

Overall Structure

• The spiral is three-dimensional, resembling a coiled ribbon, with each loop representing a major era or period in Earth’s history.
• The background is a deep blue, symbolizing oceans or space, while the spiral itself is mostly white and light brown, with blue sections indicating water.

Text and Labels

• Along the spiral, there are labeled sections for different geologic eras and periods, including:
  • Precambrian Era (innermost coil)
  • Paleozoic Era (includes Cambrian, Ordovician, Silurian, Devonian, Mississippian, Pennsylvanian, Permian)
  • Mesozoic Era (Triassic, Jurassic, Cretaceous)
  • Cenozoic Era (Tertiary and Quaternary periods, with epochs like Pliocene, Pleistocene, Holocene)
• Numbers like “251 million years ago” and “65 million years ago” mark major transitions.

Visual Elements

• Center of the spiral: Represents early Earth, mostly barren and oceanic.
• Moving outward, the spiral shows:
  • Marine life such as trilobites and fish in early periods.
  • Plants and forests appearing in later Paleozoic.
  • Dinosaurs dominating the Mesozoic era.
  • Volcanoes and continental landscapes scattered throughout.
  • Mammals and humans near the outermost edge, representing recent history.
• Each era includes illustrations of characteristic life forms and environments:
  • Precambrian: Simple organisms, mostly microscopic.
  • Paleozoic: Fish, amphibians, early reptiles, lush vegetation.
  • Mesozoic: Dinosaurs, flying reptiles, coniferous forests.
  • Cenozoic: Mammals, ice age scenes, humans.

Additional Details

• Clouds and volcanic eruptions appear near transitions, symbolizing major events like mass extinctions.
• The outermost edge shows modern landscapes with humans and large mammals.
• The spiral narrows toward the center, emphasizing the vast time span of early Earth compared to recent history.

Credit: U.S. Geological Survey General Interest Publication 58—Version 1.1, United States Geological Survey

Video: Paleoclimate Concepts (1:31)

During the Cenozoic, about 55 million years ago, an extinction event wiped out many sea-floor foraminifera, small shelly critters, at the time dividing the Paleocene and Eocene Epochs. Starting with an already-warm world, the temperature went up several degrees in roughly 10,000-20,000 years (with some uncertainty) as CO₂ rose and then cooled over the next 100,000-200,000 years as CO₂ fell. The Arctic was ice-free during the event. Plants and animals migrated rapidly. Many large animals became “dwarfed” during peak warmth, possibly because high temperatures cause greater trouble for larger animals. (We generate heat over the volume of our bodies and lose heat from the surface, and the ratio of surface area to volume is generally smaller in larger animals, making heat loss harder.) Insect damage to leaves spiked and patterns of rainfall and drought shifted. The ocean became more acidic, and that extra acidity was then neutralized in part by dissolving shells.

The source of the CO₂ remains somewhat uncertain but most likely was volcanic eruptions linked to rifting of the North Atlantic cooking organic material including oil in rocks, amplified by the loss of carbon from soils and sea-floor methane clathrates. The event is unique over tens of millions of years in its size and speed, so may have involved a coincidence of some sort, or else more such events would have occurred.

Wherever the CO2 came from in detail, it warmed the climate as much or more than models generally calculate and had very large impacts on living things. And, the effects lasted a long time. For example, although corals did not go extinct, coral reefs disappeared as functioning ecosystems and did not come back for millions of years.

Over the last million years or so, ice has grown and shrunk on the Earth’s surface, with a main spacing of 100,000 years, and lesser wiggles at about 41,000- and 23,000-year spacing. Early geologists identified and named many of the times of large and small ice, and eventually developed tools that allow quite precise estimates of when events occurred.

Remarkably, astronomers had predicted the measured timings decades before they were observed because they arise from cycles in Earth’s orbit. These cycles have very little effect on the total amount of sunshine reaching the whole Earth, but they move sunshine around on the planet, with large effects (more than 10%) on the sunshine reaching a particular latitude during a particular season.

Even more remarkably, when sunshine has dropped in the far north, especially in summer, the whole world has cooled, including places getting more sunshine. And, when sunshine has risen in the far north, especially in summer, the whole world has warmed, including places getting less sunshine. The explanation is that when northern sunshine dropped, a whole lot of ice grew on the lands of the north (enough to lower sea level about 400 feet), many other things in the Earth system changed, and some of these changes caused some CO₂ to move out of the air into the oceans; when ice melted in the far north, those other changes reversed and moved the CO₂ from the oceans back into the air. Several processes may have contributed, including northern ice changes shifting winds that shifted ocean currents that controlled how rapidly deep ocean waters came back to the surface, bringing CO₂ released by ecosystems living on the sinking organic matter from the surface.

Want to learn more?
Read the Enrichment titled The History of the World.

Ice growth lowered CO₂, which cooled the regions getting more sunshine; ice melting raised CO₂, which warmed the regions getting more sunshine. The known physics of CO₂ explain what happened, and nothing else has succeeded in doing so.

Let's return to the figure showing the broad histories of atmospheric CO₂ (with estimates from different techniques shown by different lines plus the shaded band at the bottom), and of ice on the planet (glaciers extending farther toward the equator are shown by longer bars hanging down from the top). Clearly, CO₂ and ice moved in opposite directions, with rising CO₂ occurring with melting ice. The figure has been “smoothed”, and so doesn’t show the details of the shorter-lived events discussed just above.

Estimates of the history of CO₂ over the last 400 million years (MA), with today on the right. The shaded band is output from a model calculating CO₂ from inputs such as volcanic eruptions and outputs such as fossil-fuel formation; the other curves represent estimates of atmospheric CO₂ from different techniques applied to sedimentary deposits. The bars hanging down from the top show the extent of ice at low elevation (very high mountains can have ice even in times with warm temperatures at low elevations where most people live, but low-elevation ice requires cooler temperatures in most places); times with no bar had no ice at low elevation.

Click for a text description of the history of CO2graph.

The image is a line graph showing the atmospheric CO2 levels over time, measured in parts per million (PPM), with time on the X-axis ranging from 400 to 0 million years ago (MA) and CO2 levels on the left Y-axis, with a scale from 0 to 8,000. The right Y-axis indicates continental glaciation (paleolatitude), ranging from 0 to 90. There are several data sets plotted: "GEOCARB III," shown by a green shaded area, "Palaeosols" in a pink line, "Phytoplankton" in a light green line, "Stomata" in a dark green line, and "Baron" in a red line. The graph depicts fluctuations in atmospheric CO2 across different millions of years, with visual overlaps of the data sets and glaciation trends highlighted by purple vertical bars. The color legend on the chart identifies each data set along with the sample size in parentheses.

Source: From CCSP, 2009: Past Climate Variability and Change in the Arctic and at High Latitudes. A report by the U.S. Climate Change Science Program and Subcommittee on Global Change Research [Alley, R.B., J. Brigham-Grette, G.H. Miller, L. Polyak, and J.W.C. White (coordinating lead authors)]. U.S. Geological Survey, Reston, VA, 257 pp. modified from IPCC (2007).

By themselves, the correlations just discussed between CO₂ and temperature do not prove that CO₂ caused the warmth. But, straightforward physics shows the warming effect of CO₂. And, although warming can raise CO₂ over short times, as at the start of the PETM or the ends of the ice ages, over long times warming lowers CO₂ by causing faster rock weathering and fossil-fuel formation. Thus, the prolonged high levels of CO₂ during warm times were not caused by the hot climate; instead, such high levels were caused by faster volcanism, or thicker soils slowing access of CO₂ to react with rocks, or other geological reasons.

The physics, and the lack of other plausible causes despite major efforts to find something, show that the warmth was caused by the CO₂. Testing our understanding by “retrodicting” what happened—starting with the causes and simulating the effects of the climate changes—shows that our models work well. If there is a problem, the world has changed a little more in response to CO₂ than expected from the models.

The past confirms much more about our understanding. The major events in Earth’s history were identified first by their influence on living things, including extinctions. A huge amount of additional research was required to learn that changing climate was responsible for many of those events, and perhaps for almost all of them. This long history of climatically caused extinctions supports our scientific expectation that continuing climate change risks extinctions in the future. We also expect that the CO₂ we put up will continue to affect the climate for a long time, based on models and understanding that are well-confirmed by the geologic history.

The biggest of the climate changes of the past were much larger than the changes humans have caused so far. But, if we continue to burn the available fossil-fuel resource, we can cause a change that is as more-or-less as large as, and much faster than, the biggest natural events (except for the meteorite that killed the dinosaurs, which caused large changes very very rapidly).

The geologic record highlights another major issue. Science always involves uncertainty. All measurements have some “plus and minus”—Dr. Alley is within an inch of 5’7” and weighs within a few pounds of 145, for example, but he surely is not known to be exactly those measurements. And, when measurements are used to drive models that project climate changes that are used to estimate economic impacts, many sources of uncertainty are involved, and we cannot in any way be exactly certain what the future holds.

In assessing those uncertainties, though, we find evidence of an asymmetry that you probably could have figured out from common sense. In ordinary life, breaking things is almost always easier than building them. If you want to build a new house, you will need a lot of different materials and tools and know-how. But, if you want to tear down a house, you can do it with just a wrecking ball or an exploding stick of dynamite.

When we survey the history of climate, we see something similar. We don’t find evidence of Eden, a time when changing CO₂ and climate had turned the whole Earth into paradise. Deserts and ice have grown and shrunk, so some times may have been “nicer” than others, with no guarantee that we now live in the best of all possible worlds. But, hazards existed at all times.

We do find evidence of occasions that were much closer to Hell, with up to 95% of the known species becoming extinct. A species might survive from just a single pregnant female or a few eggs or seeds even if all other individuals are killed, so the extinction times were very bad indeed.

If we continue to rapidly change the atmospheric concentration of CO₂, we have a best estimate of likely impacts, which will be discussed further just below and in additional material later in the course. Uncertainties are real, and the future may be somewhat better than expected, or somewhat worse. But, we don’t see any reasonable chance that the changes will be much better than expected—cranking up CO₂ is very, very unlikely to make Eden. And, the history of climate suggest the possibility that things will be much worse than expected—cranking up CO₂ might break things we really care about.

If you drive somewhere, you face a similar situation. What you expect is very far on the "good" side of what is possible, as shown in this short piece...

Video: Possibilities (5:10)

Video: Global Surface Warming (2:23)

The figure shows the past warming, which is just under 1°C or roughly 1°F, together with the future warmings for the different scenarios. The lowermost future line assumes that the atmospheric composition had been stabilized in the year 2000, with no further rise of CO₂. Warming continues in that scenario because some heat is now going into the ocean, keeping the air cooler than it will be as ocean warming catches up. Note that it is already too late for us to follow that path because we have raised CO₂ since 2000. Also, we are committed to some additional warming if we choose to stabilize the atmospheric concentrations at any point in any of the scenarios, again because of the slow warming of the ocean.

In all the other scenarios, if we don’t make major efforts to reduce future CO₂ emissions, the future warming is projected to be quite large compared to the past warming, and the temperature is still going up as the next century starts. Also notice the uncertainty bars on the right, showing that warming may be a little less than the most-likely estimate, or a little more, or somewhat more than that.

Video: Projected Surface Temperature Changes (2:55)

The figures here show the projected warming, and uncertainties. The maps are the projected warming for the next decade (2020-2029, center) and the last decade in this century (2090-2099, right), for different possible emissions scenarios, with more CO₂ emitted as you go down through the maps. The estimates were made with Atmosphere-Ocean General Circulation Models (AOGCMs), the big climate models of the world. And, the maps here are the averages of the projections from all of the models participating in this effort—tests in the past have shown that this average across all the models generally does better than any single model (the “wisdom of the crowd” in models).

Warming is projected to be especially slow in those places where ocean water sinks into the deep ocean, and especially fast in the Arctic. Projected warming is generally larger over land than over the ocean. Because the Earth is mostly ocean, the numbers usually given for “global warming” are closer to ocean than to land values. But, almost everyone lives on land, so the great majority of people are expected to experience above-average warming!

The panels on the left show the uncertainties in the projected warming. Notice for the 2090-2099 projections (the larger warmings, in red), that the most-likely warming tends to be towards the low end of the possible warmings. We have already seen that the most-likely impacts of a specified warming are on the low-damage side of the possible impacts, and now we see that the most-likely warming is on the low side of the possible warmings. Both of these have the same effect: the less you trust climate scientists to get the most-likely estimate correct, the more worried you probably should be about climate change, because the numbers most frequently quoted by scientists are on the optimistic side of the possibilities.

Video: Relative Changes in Precipitation (1:21)

This slightly complex figure shows projections of future precipitation. In general, the models project that wet places will get wetter, dry places dryer, and the dry subtropics will expand somewhat into currently wetter regions toward the poles. Evaporation is also expected to go up with warming, and many of the models find summertime drying in places we grow much of our food, so agriculture may be reduced more than you might think from looking at this, as we discuss after a quick look at sea level.

LOTS of other issues come up because climate affects so many things that we care about. A few of the larger issues include sea level rise, more floods and droughts, agriculture impacts, and impacts on people.

Warming causes ocean water to expand and melts mountain glaciers. (Despite a few outliers or oddities, the great majority of mountain glaciers are melting.) The big ice sheets of Greenland and Antarctica are also losing mass. With continuing warming, we expect more sea-level rise. The recent rise has been about 3 millimeters per year, or just over an inch per decade, and sea level has risen almost a foot (just under 1/3 m) over the last century or so. We expect sea-level rise to continue and probably accelerate moderately, with at least a slight chance of a large acceleration if the big ice sheets change rapidly. A foot of sea-level rise might not seem like a lot when the biggest hurricanes can have storm surges of 10 or rarely even 20 feet (3 to 6 m). But, the last foot may be the one that goes over the levee or into the subway tunnels, so even a relatively small change in sea level can have large consequences for cities and other human-built structures.

Video: Sea Level Rise (1:17)

As noted above, there is a tendency for wet places to get wetter and dry places to get drier, with the subtropical dry zones expanding somewhat. When conditions are right to rain, warmer air holds more water (by roughly 7% per degree C or 4% per degree F), so all else equal, a warmer climate can deliver more rain in a hurry. But, evaporation speeds up with warming, too. All winter, Dr. Alley’s tomato patch is damp or frozen; in the summer, just a week or two after a downpour, he needs to water the plants again. A more summer-like world is likely to have more variability in the water cycle, with more floods and more droughts.

Video: Potential for Drought by the End of this Century (1:01)

Plants need CO₂ to grow, and higher CO₂ levels will give faster plant growth. But, plants need many other things, too; in experiments with extra CO₂ added to natural ecosystems, an initial growth spurt lasts a few years before settling down to only slightly faster growth than before the CO₂ addition because the plants need more of those other things to sustain fast growth. If CO₂ is added to farm plants that also are supplied those other things, faster growth can continue, but the gain is still not huge.

Working against this fertilization effect of CO₂, the projected increase in floods and droughts would make farming more difficult. Farmers have learned to handle the bugs and weeds that now annoy them, but changing climate allows new ones to invade.

Perhaps the biggest concern is heat stress on crops. At present, anomalously hot weather reduces crop growth in many agricultural regions even if the plants have enough water, fertilizer and protection from bugs and weeds. For much of the world, continuing our present path until late in this century is projected to give average summer conditions hotter than the hottest summer up to 2006 (the last data available for an influential study). Record highs are rising with average temperatures, and expected to continue doing so. Thus, unless crop breeders become highly successful at developing heat-resistant varieties, heat stress may become quite damaging if we cause large warming.

Video: Net Photosynthesis (1:19)

Note also that the tropics are the big belt around the middle of the Earth, the polar regions are the small caps on the ends, and mountain ranges taper to points at the top, so simply moving poleward or up the mountains to follow cooler conditions involves losing ground. In addition, we now grow mid-latitude crops in soil that was transported by glaciers from higher latitudes or altitudes, so moving poleward in at least some places leaves most of the soil behind. Greenlanders are doing a little farming in special places such as on raised beaches from the ice age, but much of Greenland is too rocky for good farming, as shown below. So if Greenland's ice melts, raising sea level about 7.3 m (24 feet) averaged around the globe, and flooding valuable coastal property, the land revealed beneath the former ice sheet is not likely to be a wonderfully fertile replacement.

Video: South Greenland (1:14)

Overall, the effects of the rising CO₂ and the changing climate are expected to be mildly positive for farms for the near term, switching to negative and becoming increasingly worse beyond a few decades. One study found losses for US corn and soybeans of 30% to 82% by late in this century, depending on the scenario used and other factors.

Too hot or too cold cause problems for people. But, we have largely mastered the art of putting on coats, boots, hats and gloves, whereas personal air conditioning is not well-advanced. Thus, in too-hot places we tend to stay inside air conditioned places or be unhappy, whereas in cold places we go skiing or snowmobiling. As warming reduces the snowy season in some places, fewer automobile accidents and airports closed by blizzards will be beneficial. But, the arrival of unexpected heat can be dangerous—the highly anomalous European heat wave of 2003 is estimated to have killed 70,000 people. Adaptations such as expansion of air-conditioning tend to reduce the health impacts when heat continues.

Still, humans and other animals risk damage or death when conditions are too hot. How hot is too hot depends on humidity (we can take higher temperatures when it is drier), and on exercise level. A recent study found that, averaged across the world’s human population, heat already here reduces the ability for people to work outside in the hottest months by about 10%. If we continue to release CO₂ rapidly, this is projected to rise to a 20% reduction in work by 2050, 40% by 2100 and perhaps 60% by the end of the next century. These losses are concentrated in the warmer parts of the world, where they can be very large.

Climate affects almost everything somehow, so a great number of other issues can be raised, from huge to tiny. Vines seem to like carbon dioxide, for example, so poison ivy is expected to grow well, and vines may out-compete large trees in tropical rain forests.

Poison Ivy: More vigorous poison ivy may not be the end of the world, but is among the many, many influences of changing climate.

Credit: Richard B. Alley @ Penn State is licensed by CC-BY-NC-4.0

More broadly, almost all ecosystems will be perturbed, often in major ways. Rare and endangered species may have difficulty migrating, especially if they are persisting in a park or preserve surrounded by human-controlled landscapes, or if they are migrating up a mountain and eventually having nowhere further to climb. Acidification of the oceans, and loss of oxygen with warming, will affect marine species and those of us who eat them. Loss of wintertime cold doesn’t mean that everyone in the high latitudes is about to get malaria, but one line of defense will go away. Changes in hurricane frequency are still highly uncertain, but the strongest storms seem likely to get stronger, and so much of the damage is done by the strongest storms. Cooling towers for power plants expect enough, and cold enough, water, and may experience troubles. And on, and on.

Very generally, we are adapted to the climate we have. In the short term, almost any change has associated costs. If two regions with different climates simply swapped their climates, for example, both would have wrong-sized air conditioners and heaters, too many or too few snow plows and swimming pools, less-than-optimal seeds for crops, etc. All of these can be fixed, but not for free.

If changes remain small, there are likely to be winners and losers. Warming may make beach resorts happier, but ski areas less happy. Rare and endangered species, and people trying to live traditional lifestyles, may be pushed to the edge by even small changes. For most people, if you have winter that interferes with travel and other activities, air conditioners so you can work in the summer, and bulldozers to build walls against the rising sea, a little warming is not especially costly; if you lack winter, air conditioners and bulldozers, even a little warming is likely to make your life at least a little harder.

But, if the temperature continues to rise, and the big hotter-than-we-like belt around the equator expands towards the poles, life is projected to get harder for most people in most places.

There are real uncertainties, so things really may end up better than this. But recall that, because breaking is easier than building, we don’t see how raising CO₂ greatly and rapidly will create Eden, but we do see at least a slight chance of huge and damaging changes.

Video: Reducing Risks of Climate Change Damage (2:59)

One way to look at the future is shown in the figure. Different things you might care about are shown by the different columns, and the risks from warming are shown by the increasingly orange-red (saturated) color going up in the columns. Another way to look at the issue is that damages are projected to go up faster than temperature; the first degree of warming is nearly free, but each degree beyond that costs more than the previous one did. The first degree has allowed us to test our models and learn that they are doing well; the next degrees really matter.

Please recognize that these projections do not include major human efforts to reduce emissions of CO₂ and other greenhouse gases. And, we are certainly capable of making such reductions, or of adopting other approaches that might reduce the warming while supplying plenty of energy.

So, in the next Unit, we’ll look at some of the options. Then, we’ll return to Economics, Ethics and Policies that might address the paired issues of getting valuable energy for many people while reducing damages from climate change.

Right after World War II, when industry powered up in peacetime and started cranking out consumer goods, emissions increased rapidly for both CO₂ and particles from smokestacks. If emissions are suddenly ramped up like that, and then held constant, the number of sun-blocking particles in the air increases for a week or so. Then it stabilizes because particles are falling out of the air as rapidly as they are added. But, for a given rate of emission, the CO₂ concentration of the air will rise for a few hundred thousand years, until the rate of rock weathering balances the new, raised rate of emission. (Human emissions did not remain constant, but this may help you think about things.) For industry after the war, the particles emitted in the first week had a cooling effect that was much larger than the warming effect of the CO₂ emitted during that week. But, as years became decades, the particles fell down, much of the CO₂ stayed up, and the warming grew to outweigh the cooling.

Volcanoes emit CO2, but they also put up many sun-blocking aerosols, including ash such as seen in this USGS photo of the 1980 eruption of Mt. St. Helens.

Credit: MSH80 eruption mount St Helen plume 05-18-80 by USGS Cascades Volcano Observatory from Wikimedia (Public Domain). Accessed Feb. 26, 2025.

Volcanic eruptions have essentially the same story. Over short times, the sun-blocking cooling from particles exceeds the warming of the CO₂. The volcanic particles typically get thrown into the stratosphere, above most rainfall, and so stay up a year or two rather than a week or so, but then fall out. So if extra volcanism continues long enough, the particles fall down, the CO₂ builds up, and warming results. Exactly how long you have to wait for “short time” to become “long time” depends on the types of volcanoes and many other issues. In general, an increase in volcanic activity (typically involving many volcanoes, or huge volcanic provinces) will cause cooling over times of years to centuries that most economists worry about, but with warming over longer, geologic time.

To help you “see” some of the material we just discussed, here are some data from an ice core at a place called Vostok in East Antarctica. It is probably not the coldest place on Earth, but it’s close. There is a Russian base there with good measuring equipment, and it observed the lowest reliably documented natural temperature ever at Earth’s surface: −89.2°C or −128.6°F. Snow accumulates very slowly there, and an ice core contains a long, accurate record of the temperature at Vostok, and of the atmospheric composition because air bubbles trapped in the ice are little samples of the old atmosphere. Several long ice-core records have been collected in Antarctica, with the longest continuous one about 800,000 years, and older ice found in other places but disturbed by ice-flow processes so that a complete, continuous record beyond 800,000 years is not yet available from ice cores. (Other sedimentary records go much further back in time, but don’t trap bubbles of old air, so estimates of older atmospheric concentrations rely on indirect indicators and are slightly less certain.)

The temperature record, from the isotopic composition of the ice, is what happened in the Vostok region, not the whole world. But, if you take records from elsewhere, and smooth them a good bit, they all look similar to Vostok; the whole world cooled and warmed together through the ice-age cycles. And as explained in the next clip, this is primarily because of changes in CO₂.

As noted on the previous page, the ice ages were caused by features of Earth’s orbit. The spacing between ice ages actually was predicted decades before it was measured accurately, based on astronomical calculations from the orbits. The prediction and test are explained in the clip just below, and shown in the figure below it. The figure is from a fancy way (called a Fast Fourier Transform, or FFT) to figure out the spacing between wiggles in a curve, such as the climate record—the arrows are the predicted peaks, and you can see that the actual peaks line up beautifully.

The story is wonderfully complicated but can be made fairly simple, again as noted on the previous page. When the summer sun has dropped in the north over thousands of years, ice grew, forming vast ice sheets that have bulldozed across Scandinavia, Boston, New York and Chicago. (Antarctica is already glaciated, and it doesn’t really get cold enough to get ice onto Australia, Africa, or most of South America, so sunshine in the south isn’t so important). The ice sheets were made from water from the ocean, which dropped more than 100 m (about 400 feet). Many other changes occurred as the ice grew, and these shifted some CO₂ into the ocean. Then, the whole world cooled, including places getting more sunshine. When sunshine rose in the far north, this reversed. The temperature of the whole world changed together, even though half of the world got less sunshine when the other half got more, and CO₂ is the main explanation.

The last figure is then important, showing where CO₂ may go this century if we don’t change our energy system.

Video: Data from the Vostok Ice Core (5:45)

This is a “minitext” that was originally written by Dr. Richard Alley for Geosciences 320, Geology of Climate Change, at Penn State. It provides a short history of the world’s climate over the last 4.6 billion years. It assumes a little more background knowledge than is expected in the rest of the class, but may be useful.

When most people think of “solar power,” they think about one of two things – vast arrays of solar collectors laid out in hot deserts (the left-hand panel below) or smaller arrays on rooftops or highways (the right-hand panel below). This is perhaps the most ubiquitous method of converting solar energy into electricity, but it is not the only method. These arrays of solar collectors are known as “solar photovoltaic” installations or Solar PV for short.

Solar PV installations consist of individual collectors called cells, which are packaged together in bundled modules. An individual cell does not generate enough electricity to power much of anything, which is why they must be bundled together. A single module might be enough to provide electricity for a single parking meter or roadside telephone. A number of modules can be further bundled together to form an array (see below). Multiple arrays might be needed to provide electricity for a building or a house.

There are many different kinds of solar PV cells in existence (and even more being developed in research laboratories), but they all work in more or less the same way. Unlike virtually any other type of power plant (be it coal, natural gas or wind), there is no turbine in a Solar PV cell. In fact, there are basically no moving parts at all. .

Solar PV cells harvest solar energy through a phenomenon called the photovoltaic effect, discovered in 1839 by the French physicist Bequerel. Photons of solar energy interact with electrons to “excite” them, causing them to move through conductors, thus producing an electric current. The first solar PV module was made at Bell Labs in the 1950s, but was too expensive to be more than a curiosity; in the 1960’s NASA started to use PV modules in spacecraft and by the 1970s, people started to explore their use in a wider range of terrestrial applications.

This following video explains a bit more about how Solar PV cells work and describes the different Solar PV technologies in use today. One of the potentially most important evolutions in Solar PV technology is the use of semiconducting materials other than silicon in Solar PV cells. These materials are of interest because they could, in concept, allow more of the sun’s energy to be captured on a single array. But they face barriers in the form of high costs and, in some cases, questions about the availability of raw materials.

Video: Photovoltics, a Diverse Technology (4:26)

Most modern Solar PV technologies are relatively inefficient compared to other forms of electricity generation. Remember here that “efficiency” refers to how much of the fuel that is injected into an electricity generation system is actually converted into useful electricity, versus being rejected as waste heat or otherwise escaping from the generation system. While modern coal-fired and gas-fired power plants can have efficiencies as high as 60% (or sometimes even higher), most Solar PV cells convert sunlight to electricity with an efficiency of 20% or less (see below), though this number has been rising over time.

Whether the efficiency of Solar PV cells is all that important is a matter of some debate. On the one hand, higher-efficiency cells would require less land or space to produce a given amount of electricity. Land use (or the number of rooftops) can be a significant limiting factor in the deployment of Solar PV. On the other hand, fuel from the sun is free and there is no scarcity of sunlight, so whether Solar PV cells can achieve 30% efficiency versus 20% efficiency may not be such a big deal, and may not be worth the extra economic cost to produce such high-efficiency cells.

If you have ever left a cold drink out in the sun during the summertime (or if you have children, if you have ever left water in the kiddie pool out in the sun for a long time), you would notice that the formerly cold water gets warm – maybe even hot. If it happens to be summertime where you are living right now, try it! Whether you realize it or not, this little science experiment is the basis for a second way of harnessing the sun’s energy to produce electricity, called “concentrated solar power” or CSP. (This technology is also sometimes called “solar thermal.”)

The following video explains how CSP works. The basic idea is that a collection of mirrors reflects the sun’s light (and heat) onto a large vessel of water or some other fluid in a metal container. With enough mirrors reflecting all of that sunlight, the fluid in the metal container will get hot enough to turn water into steam. The steam is then used to power a turbine just like in almost any other power plant technology

In addition to being free as a source of energy (it does cost money to harness it and turn it into electricity), energy from the sun is practically limitless. The surface of the Earth receives solar energy at an average of 343 W/m². If we multiply this times the surface area of the Earth, about 5x10¹⁴ m², we get 1715x10¹⁴ W. But, 30% of this is reflected, and only 30% of the Earth is above sea level, so the usable solar energy we receive on the land surface is about 360x10¹⁴ W. We need to reduce this further because not all of the land surface is suited to installation of solar PV panels — we don't want to cut down forests, and ice-covered areas are not suitable, so we reduce the area by about one half. Over the course of a year, this amount of solar energy adds up to 66x10²² Joules. In 2018, we used about 600x10¹⁸ Joules of energy, which is just a shade less than 0.1% of the harvestable solar energy we receive on the land. This means that even if we got all of our energy from the Sun, we would not make a dent in the total! The potential is vast — 10,000 times what we need!

Let’s consider what it would mean for us to get all of our energy from Solar PV — how much of the Earth’s surface would we need to cover with panels? The black dots (radii of 100 km) in the figure below represent areas that could generate enough energy from sunlight to completely power the planet for an entire year. Practically, there are barriers to running the planet entirely on sunlight (everything would need to be electrified, we would need very large quantities of battery energy storage, and so forth), but the dots are useful as a demonstration of just how vast the energy production potential from solar is.

One of the important differences between Solar PV and CSP is that CSP requires more intense sunlight, and as such, it is not a viable option in many places. In contrast, Solar PV works just about everywhere — it is more versatile. Another important difference is in scale — CSP is really suited to utility-scale power plants, whereas Solar PV works at both the utility-scale and the very small scale.

The map below shows the PV potential for the world. The variability in the map is mainly a function of cloudiness and latitude. Many of the big, utility-scale solar PV plants are located in the red areas, but there is a surprising amount of Solar PV energy being harvested in places like Germany and Japan, both of which are fairly cloudy. But, even in a fairly cloudy place like Pennsylvania, you can see from the map that we could expect about 1460 kWh per year from a 1 kW PV array. From this, you can calculate how many square meters of PV panels you’d need to provide the electricity for a house that uses the typical 10,800 kWh per year. If you divide 10,800 kWh by 1460, you see that you’d need about 7kW of solar panels, which would fit on a typical house roof. The main point here is that Solar PV is a viable energy source in most parts of the world where people are living. In contrast to Solar PV, energy from CSP is only viable in places where the daily totals in the map above are higher than 6 kWh/day. Nevertheless, there are many regions where CSP viability and human population coincide, so it too can be an important energy resource in the future.

The generation of solar energy – primarily through Solar PV – is a story of exponential growth. Since 2000, the global Solar PV industry has grown by around 25% per year on average, so installed capacity has been doubling every 2.7 years (see below). Even so, solar represents a very small sliver of total global power generation — for now.

This shows the history of global solar energy generation in Exajoules (EJ) of energy each year (1 EJ = 10¹⁸J). This growth history is following an exponential curve. In 2018, this represents a small fraction of global energy consumption, which is almost 600 EJ.

Click for a text description of the Global Solar Energy Generation History graph.

The image is a line graph illustrating the historical trend of global solar energy generation from 1985 to 2020. The x-axis represents the years, marked from 1985 to 2020, while the y-axis indicates the energy generated per year in exajoules (EJ), ranging from 0 to 6. The graph depicts a sharp upward trend starting around 2005, indicating significant increases in solar energy production. The blue line that represents the data forms a steep curve upward after 2010, emphasizing rapid growth. A text box on the graph annotates that the average growth rate is 25% per year. The title of the graph is "Global Solar Energy Generation History."

Source: David Bice, data from BP Statistical Review of Energy, 2018

The nice thing about exponential growth is that it is easy to project it into the future. Over the time period shown in the graph above, solar energy generation has grown by 25% per year; if we continue that into the future, we find that before long, we would have enough solar energy to make up a substantial portion of the global energy needs by 2030 (see figure below). By the year 2040, this growth would rise to 1360 EJ, more than twice the global energy consumption of the present. Of course, that makes no sense — we would not produce more energy than we need, and this reminds us of an important fact, which is that exponential growth cannot continue forever.

This shows the history of global solar energy generation in Exajoules (EJ) of energy each year (1 EJ = 10¹⁸J). This growth history is following an exponential curve. In 2018, this represents a small fraction of global energy consumption, which is almost 600 EJ.

Click for a text description of the Global Solar Energy Generation Projection graph.

The image is a line graph titled "Global Solar Energy Generation Projection," depicting projected solar energy generation from 1985 to 2030. The x-axis represents the year, ranging from 1985 to 2030, and the y-axis represents the energy generated per year in exajoules (EJ), ranging from 0 to 120. The graph starts with a nearly flat blue line from 1985 to around 2015, indicating little to no growth in solar energy production. After 2015, the line turns orange and begins to curve sharply upward, showing a significant increase, projecting exponential growth in solar energy generation up to 2030. A text box within the graph notes "Projection into the future assuming 25% per year growth."

Source: David Bice, data from BP Statistical Review of Energy, 2018

One reason to trust this projected future growth is that the price of solar energy has fallen dramatically over time as can be seen in the graph below. In fact, if the generation of solar PV energy has been growing exponentially, the price has been dropping exponentially.

The price history of solar PV cells in $/watt shows an incredible decline in price. This is due to technological advances and the economy of scale as more and more PV cells are manufactured.

Click for a text description of the graph: Price History of Silicon PV Cells.

This bar chart illustrates the decline in the cost of silicon photovoltaic (PV) cells from 1977 to 2015, measured in US dollars per watt ($/Watt).

Key Observations:

1977:

•  
  • The price of silicon PV cells was $76.00 per watt, the highest recorded in the dataset.
  • Marked by a tall blue bar at the far left.

1980s:

•  
  • Significant price reduction as PV technology improved.
  • By 1985, the price dropped to around $10 per watt.
  • The bars become progressively shorter, showing a steep decline.

1990s:

•  
  • The price stabilized around $5–10 per watt.
  • The bars maintain similar heights, indicating slower reductions.

2000s:

•  
  • Continued gradual decline, reaching around $2 per watt by mid-2000s.
  • More efficient manufacturing and increased adoption contributed to lower costs.

2015:

•  
  • The price reached $0.30 per watt, as indicated by a highlighted blue label at the bottom right.
  • This marks a 99.6% reduction from 1977 prices.

Additional Details:

• A text box in the center of the graph reads:
  “Price history of silicon PV cells in US$ per watt”
• Data Source: Bloomberg New Energy Finance & pv.energytrend.com
• The trend follows Swanson’s Law, which states that solar PV prices drop by 20% for every doubling of cumulative shipped volume.

Source: Price history of silicon PV cells since 1977 by Rfassbind from Wikimedia Commons

The price decrease is following a pattern that has been given a name: Swanson’s Law, which states that the price drops by about 20% for each doubling in the number of PV cells produced. This law suggests that the prices of solar PV energy will continue to decline in the future.

This brings us to an important question — how does the cost of solar energy compare to other sources of energy? Energy economists have come up with a good way of comparing these costs by adding up all of the costs related to producing energy at some utility-scale power plant (a big wind farm, a big solar PV array, a CSP plant, a nuclear plant, a gas or coal-burning power plant). This is called the levelized cost of energy, and you get it by taking the sum of construction costs, operation and maintenance costs, and fuel costs over the lifetime of a plant and then dividing that by the sum of all the energy produced by the plant over its lifetime. This cost provides us with a way of comparing the energy from different sources. Since the boom in natural gas production due to fracking, natural gas has been the lowest cost form of energy (which is why coal is being used less and less), but energy from solar and wind have been decreasing rapidly, as can be seen in the following graph. When a renewable electrical energy resource such as solar or wind becomes equal in cost to the cheapest fossil fuel source of electricity, we say that the renewable resource has reached "grid parity". Once grid parity is achieved, the renewable resource makes sense from a purely economic standpoint, and as it drops below the grid parity point, it is the smartest electrical energy resource.

The levelized cost of solar energy in the US has been dropping fast in recent years and has been slightly cheaper than electrical energy from natural gas since 2015. The point in time where the two are equal is when solar energy achieved what is called grid parity. Wind energy reached grid parity by 2012, so at this point in time, both of these renewable sources provide cheaper energy than the cheapest fossil fuel source.

Click for a text description of the graph: Levelized Costs of Solar, Wind, and Natural Gas (2008–2018).

This line graph displays the levelized cost of energy (LCOE) for solar, wind, and natural gas from 2008 to 2018 in 2018 dollars per megawatt-hour (MWh). The LCOE represents the total cost of generating electricity from each source, considering installation, maintenance, and fuel costs.

Key Observations:

Solar Energy (Red Line)

•  
  • In 2008, the LCOE for solar was about $180 per MWh, the highest of the three energy sources.
  • The cost declined rapidly over the years, reaching approximately $40 per MWh by 2016.
  • By 2018, solar became the cheapest energy source at under $30 per MWh.
  • The decline reflects advances in solar panel technology, manufacturing efficiency, and increased adoption.

Wind Energy (Green Line)

•  
  • The cost of wind energy fluctuated between $60–80 per MWh from 2008 to 2010.
  • After 2010, wind costs began to decline, reaching a low of about $20 per MWh in 2018.
  • The drop can be attributed to larger and more efficient wind turbines, better grid integration, and falling installation costs.

Natural Gas (Blue Line)

•  
  • The LCOE for natural gas remained relatively stable, fluctuating between $40-60 per MWh over the period.
  • Natural gas was cheaper than solar for most of the timeline but was surpassed by both solar and wind by 2018.
  • The stability of natural gas prices is due to fuel costs, infrastructure maintenance, and policy factors.

Overall Trends:

• Solar energy saw the most dramatic cost reduction, making it one of the cheapest electricity sources by 2018.
• Wind energy also experienced a significant drop, becoming the least expensive energy source by 2018.
• Natural gas remained relatively stable, but was no longer the most cost-effective option by the end of the period.

Source: David Bice, data from Lawrence Berkeley Labs.

Part of the reason that solar and wind have expanded in recent years has to do with government policies — a number of countries have instituted subsidy and incentive programs that offset a large portion of the construction/installation costs of solar and wind technologies or devise rules that otherwise give advantages to electricity generation from renewables. Subsidies enacted in various countries have included feed-in tariffs (which guarantee an above-market sales price for solar power); rebates (which directly offset capital and installation costs); and favorable tax treatment (which is like an indirect feed-in tariff). Germany has one of the world’s largest Solar PV markets not because it has the best solar resource on earth but because it has been willing to support a generous feed-in tariff on solar power. (For many years the tariff was over 30 cents per kilowatt-hour, or more than five times the average power price in the United States; in recent years the tariff has been reduced.) These government policies have effectively stimulated the growth of these renewable energy resources, which has, in turn, resulted in lower prices.

In a conventional power plant (fueled by coal or natural gas), combustion heats water to steam and the steam pressure is used to spin the blades of a turbine. The turbine is then connected to a generator, which is a giant coil of wire turning in a magnetic field. This action induces electric current to flow in the wire. The workings of a wind turbine are much different, except that instead of using a fossil fuel heat to boil water and generate steam, the wind is used to directly spin the turbine blades to get the generator turning and to get electricity produced.

The inner workings of a wind turbine consist of three basic parts, seen in the figure below. The tower is the tall pole on which the wind turbine sits. The nacelle is the box at the top of the tower that contains the important mechanical pieces – the gearbox and generator. The blades are what actually capture the power of the wind and get the gears turning, delivering power to the generator. The direction that the blades are facing can be rotated so that the turbine always faces into the wind, and the pitch of the blades (the angle at which the blades face into the wind) can also be adjusted. Pitch control is important, especially in very windy conditions, to keep the gearbox from getting overloaded.

The amount of power (in Watts) collected by a wind turbine is explained in the following equations:

The Kinetic Energy (KE) of the wind is:

$$K E = \frac{1}{2}m v^{2 }$$

Where m = mass, and v = velocity of wind.

Power (P) in the wind is the KE per unit time, so we replace the mass(m) with the mass flux rate dm/dt:

$$P = \frac{1}{2}\frac{d m }{d t }{v}^{2 }$$$$\frac{d m }{d t }= \rho A v$$

Where p = air density, and A = swept area of blades.

So the wind Power(P) is:

$$P = \frac{1}{2}\rho A v^{3 }$$

If the wind turbine collected all of this power, the wind would have to stop and the blades would stop spinning. If you want the blades to keep spinning, it turns out that you can collect about 60% of the power (called the Betz limit).

So, collectible Power(P) is:

$$P = 0.3 \rho A v^{3 }$$

How much power could we get with a turbine whose blades are 100m long, with a wind speed of 10m/s (about 22mpg>, with an air density of 1.2kg/m2?

$$P = 0.3 \times 1.2 \times \pi \times {100}^{2 }\times {10}^{3 }= 11 , 304 , 000 \text{ Watts } = 11.3 \mathrm{MW}$$

This is clearly a lot of power! But, mechanical inefficiencies related to the gears and the generator mean that we might only get 30% of this figure, but that is still a lot of power from one turbine.

All wind turbines have a minimum wind speed that differs depending on the size but is typically about 4-5 m/s (10 mph) and maximum wind speed above which they shut down to avoid damage, usually around 20-25 m/s (about 50 mph). Most wind turbines have a maximum spinning rate, reached a bit above the minimum velocity, and when the wind speeds up, the pitch of the blades is adjusted so that the rate of spinning remains more or less constant. The figure below shows a typical "power curve" for a small wind turbine.

The wind, as you may have noticed, is highly variable in any given place, but as a general rule, it is stronger and steadier as you rise up above the ground. This is because friction between the wind and the land surface slows the wind. But there is also a lot of regional variation in the wind velocity. Both of these factors (elevation above the ground and location) can be seen in the maps below, showing the average wind speed in the US at two different heights.

The graphs above show annual average wind speeds in the US at 2 different heights above the ground surface. For reference, 10 m/s is 22.3 mph. You can see that the wind speeds at 100 m are far greater than at 30 m — this is the friction effect of the land surface (which is minimal above large water bodies). As you can see, the Great Plains have great wind potential, as do the Great Lakes and offshore areas on both coasts.

The area covered by the turbine’s blades is another important factor in determining power output. While wind turbines are available in a wide variety of capacities, from a few kilowatts to many thousands of kilowatts, it’s the larger turbine sizes that are being deployed most rapidly in wind farms. Several years ago the image on the right side of the figure below of a Boeing 747 superimposed on a wind turbine gave an astonishing representation of the scale of the state-of-the-art wind technology. Now, turbine rotor diameters are approaching the size of the Washington Monument!

Over the last 20 years, growth in the total installed capacity of wind energy generation across the globe has been growing rapidly. Germany was the first country to lead the development of wind power, but the US and China have dominated the growth since 2010. China is especially impressive in terms of its recent growth.

Part of the reason for this growth is the steady decline in the cost of wind energy, as discussed in the previous section on solar energy. But government policies are another important factor. The United States has one of the most volatile markets for wind energy in the world, while those in Europe and China have been among the most stable. This is due in part to differences in how governments in these countries treat wind energy. In many parts of Europe, wind energy (and other renewable generation technologies) enjoy subsidies and incentives known as feed-in tariffs. The feed-in tariff is essentially a long-term guarantee of the ability to sell output from a specific power generation resource to the grid at a specified price (typically higher than the prices received in the market by other generation resources). The United States, on the other hand, has favored a system of tax incentives called the “Production Tax Credit” (PTC) to encourage renewable energy deployment. In theory, a tax incentive should not work much differently than a feed-in tariff (both are just payments based on how many kilowatt-hours are generated). But the PTC has historically needed to be re-authorized frequently by the US Congress – this “on-off” policy strategy has been a major factor in the volatility of wind energy investment in the US as shown in the figure below. It is worth noting that the PTC was recently renewed for 2013, but will lapse again at the end of 2019, so it is difficult to say what impact it will have on wind investment going forward.

The above clarifies that government policies are important to the growth of renewable energy production (both wind and solar). In a very real way, you can think about these policies (feed-in tariffs or tax credits) as a form of investment. Governments can also provide investments in the form of funding for basic research related to these technologies. In general, these investments do not add up to a huge amount when seen in the context of a country's gross domestic product (GDP), which is a measure of the size of the economy, as seen in the figure below.

A quick look at an annually-averaged wind map of the world (below) shows the regions of the world that are best suited for the production of wind energy in colors ranging from yellows to red (where the average winds are at least 9.75 m/s or 20 mph). The offshore regions are clearly the best in terms of the energy potential, but not all of these offshore regions are close to where people live. Even for onshore portions of the world, the wind energy potential does not always coincide with where the people are concentrated. This points to the necessity of new transmission lines to deliver this wind energy to major population centers.

So, just how much energy could be produced by the wind? In 2009, a group of scientists makes some calculations to estimate the potential for the world and the US, using wind data and some assumptions about the size and spacing of the turbines. They assumed 2.5 MW turbines on land, and 3.5 MW turbines offshore, which were big for that time. They assumed that you could only place the turbines in unforested, ice-free, nonmountainous areas away from any towns and that the turbines had to be spaced by several hundred meters so they do not interfere with their neighbors. They further assumed that each turbine generated just 20% of its rated capacity to account for mechanical problems and intermittent winds. What they came up with is summarized in the table below, and it is pretty remarkable. The units here are exajoules (EJ = 1 x 10¹⁸ Joules) of energy over the course of a year. For reference, in 2018, the US total energy consumption (not just electrical energy) was 106 EJ and the global consumption was about 600 EJ. So, with just onshore wind energy, the potential is more than twice what we consume in the US, and more than 4 times the global consumption. But getting there is a matter of installing a lot of wind turbines!

Now let's consider a more practical question — how much wind energy have we managed to produce, and can we somehow project the past trends into the future? The figure below shows the global history of wind energy (solar is plotted too just for comparison), and you can see that it is growing fast.

Both of these curves are growing exponentially, and the history so far suggests a growth of about 25% per year on average. If we assume that they continue to grow in the further following this exponential growth, we can project where we'll be at any time in the future. Below, we see where we might be in the year 2030, just eleven years from now. What you see is that we end up with vast amount of wind energy by 2030 — if it grows at the same rate it has been growing at, we end up with almost 300 EJ per year, about half of the current global energy consumption, and if it grows at a smaller rate of 20% per year, we still end up being able to supply about 20% of the total global energy demand.

It is worth noting that, as with solar, wind investments are not always happening in the windiest areas. The reality is that there are a large number of factors that influence the development of wind energy globally. As the technology for wind energy has improved, other factors have also come together to create market drivers for wind power. These drivers include:

• Declining Wind Costs
• Fuel Price Uncertainty
• Federal and State Policies
• Economic Development
• Public Support
• Green Power
• Energy Security
• Carbon Risk

We have already mentioned the US Production Tax Credit, which is responsible for a good amount of the trend in US wind energy investment – both up and down! A decline in wind investment in 2010 and 2011 was due in part to the global financial crisis. A drop in natural gas/wholesale electricity prices has made some planned projects less competitive than originally expected and halted development. There has also been a slump in the overall demand for energy. Another factor that limits the growth of wind power capacity is the constraint on the transmission infrastructure. As can be seen in the wind capacity map on the previous page, many of the locations that experience the windiest conditions are not close to coastal population centers. The cost of upgrading this infrastructure is significant — perhaps \$30 to \$90 billion in the US by the year 2030 according to some estimates. This seems like a huge amount, but consider that our government spends about \$20 billion each year in direct subsidies to the fossil fuel industry, which would sum up to \$200 billion by the year 2030. In light of that, the upgrade cost for better transmission lines is a bargain!

Remember how your basic steam turbine works in a power plant that uses fossil fuels: Fuel is burned to heat water in a boiler, to create steam. The steam is used to drive a turbine, which generates electricity. What if you could get all that steam without burning a single ounce of coal, oil or natural gas? That is the appeal of geothermal electricity production. In certain locations (primarily near active or recently active volcanoes) there are very hot rocks deep under the earth’s surface. In these "geothermal" regions, the temperature may rise by 40-50°C every kilometer of depth, so just 3 km, the temperature could be 120 to 150°C, well above the boiling point for water. The rocks in these regions will typically have pore spaces filled with water, and the water may still be in the form of liquid water since the pressure is so high down there (in some very hot areas, the water is actually in the form of steam trapped in the rocks). If you drill a deep well into one of these "geothermal reservoirs", the water will rise up and as it approaches the surface, the pressure decreases and it turns to steam. This steam can then be used to drive a turbine that is attached to a generator to make electricity. In some regards, this is very much like a coal or natural gas electrical plant, except that with geothermal, no fossil fuels are burned, which means no carbon emissions.

There are three basic types of geothermal power plants, depending on the type of hydrothermal reservoir:

• Dry steam plants, which draw steam directly from deep underground (a la Old Faithful);
• Flash steam plants, which draw hot water under high pressure up towards the surface. As the pressure decreases, the water boils, which generates steam to power the turbine;
• Binary steam plants, which utilize hot water (perhaps around 150 degrees Celsius) to vaporize another fluid (one with a lower boiling point). This hot vapor then drives the turbine, generating electricity.

The oldest geothermal plant (1904) in the world is Lardarello, in Italy, which is a dry steam plant. The Geysers, in California, is the largest geothermal installation in the world and the only accessible dry-steam area in the United States (other than Old Faithful and the rest of Yellowstone, which is off-limits). Most modern geothermal plants are “closed-loop” systems, which means that the water (or steam) brought up from the surface is re-injected back into the earth, as shown in the figure below. If the water is not replaced, then eventually, the geothermal reservoir will dry up and cease function.

On a global scale, the potential for geothermal energy is quite large. The IPCC estimates that even though just a fraction of the total heat within the Earth can be used to generate geothermal power, we could nevertheless generate about 90 EJ of energy per year, and this is energy that is constantly renewed from within the Earth. Keep in mind that at present, we generate just over 2 EJ per year, so this energy source can definitely expand, but by itself it cannot meet the total global energy demand of 600 EJ.

To harness geothermal energy to generate electricity using any conventional technology (dry steam, flash steam or binary steam), you’ve got to be in the right place, where there is just the right amount of hot fluid or steam in an accessible reservoir. Unfortunately, those places are few and far between. The figure below shows a map of geothermal resources in the U.S., with identified conventional sites marked with dots on the map. All are located in just a handful of western states, plus Alaska.

Most places do not have that right combination of an accessible, large reservoir of underground heat. Instead, reservoirs are more dispersed, in geologic formations with less permeability (this naturally inhibits the flow of hot fluid towards the surface). Engineers have discovered how to alter the subsurface to create man-made reservoirs of hot water that could be tapped to produce electricity, in either a flash steam or (with higher potential) a binary steam technology configuration. The process of engineering a geothermal reservoir underground is known as “enhanced geothermal systems” or EGS. As the resource map in Figure 2 shows, EGS could be done in a lot more places than conventional geothermal. Hundreds of thousands of gigawatts of power, basically enough to run the United States several times over, could potentially be harnessed through EGS.

The basic idea behind EGS is to fracture hot rocks deep within the earth to create channels or networks through which water could flow. When water is injected into these networks, the heat from the rocks boils the water directly, or the now-hot water is transported to the surface where it is used to boil a working fluid, much like a binary steam plant. Fracturing of the rock occurs via “hydraulic fracturing,” under which water is injected into the rock formation at high pressures, causing the rock to fracture. This is actually very similar to the way that natural gas and oil is being extracted from shale. So we can “frack” for geothermal in much the same way that we frack for oil and gas.

Only a few countries use geothermal resources as a major source of electricity production –Iceland, El Salvador, and the Philippines all use geothermal for more than 25% of total electricity generation within those countries. New Zealand is the next (but distant) largest at 10%. Where hydrothermal resources are easy to access, they have often been utilized. The trouble is, there just aren’t that many Old Faithfuls in the world.

EGS represents the most significant potential for geothermal electricity production, but other than a few small military or pilot projects, the systems have not really caught on commercially. One of the big reasons is cost – like many low-carbon electricity technologies, EGS is inexpensive to run but very costly to build. Drilling geothermal wells is much more expensive than drilling conventional oil or gas wells, so electricity prices would probably need to increase by 25% or more (relative to current averages) to make EGS a financially viable technology.

Perhaps a more serious challenge for EGS is “induced seismicity,” which is a fancy term for causing earthquakes. EGS wells that were drilled below Basel, Switzerland caused over 10,000 small tremors (less than 3.5 on the Richter scale) within just a few days following the start of the hydraulic fracturing process. In Oregon, a test EGS well is being monitored for induced seismic activity – you can see some neat real-time earthquake data at Induced Seismicity (U.S. Department of Energy: Energy Efficiency and Renewable Energy.

Induced seismicity occurs whenever hydraulic fracturing (related to EGS or developing a natural gas well) takes place, but in most cases, the earthquakes are so small they are not felt. However, if the hydraulic fracturing occurs near pre-existing faults (which are often not visible at the surface), then larger earthquakes can and do occur, and some of these are strong enough to cause minor damage to buildings nearby.

There are three main types of conventional hydropower technologies: impoundment (dam), diversion, and pumped storage.

Impoundment is the most common type of hydroelectric power plant. An impoundment facility, typically a large hydropower system, uses a dam to store river water in a reservoir. Water released from the reservoir flows through a turbine, spinning it, which in turn activates a generator to produce electricity. Generation may be used fairly flexibly to meet baseload as well as peak load demands. The water may also be released either to meet changing electricity needs or to maintain a constant reservoir level. The layout of a typical impoundment hydropower facility is shown below in the first figure. One of the world’s most famous impoundment dams, the Hoover Dam, is shown in the second figure (although it’s worth noting that on a global scale, the Hoover Dam is more famous than it is large).

A diversion, sometimes called run-of-river, facility channels a portion of a river through a canal or penstock. It may not require the use of a dam but also has limited flexibility to follow peak variation in power demand. Thus, it will mainly be useful for baseload capacity. This scenario results in limited flooding and changes to river flow. In the United States, many of the dams in the Pacific Northwest (on the Columbia and Snake Rivers) are diversion or run-of-river dams, with limited or no storage reservoir behind the dam. The figure below shows a picture of a diversion hydropower facility. Compare what that facility looks like with the picture of Hoover Dam, the impoundment facility shown above.

A “pumped storage” hydro dam combines a small storage reservoir with a system for cycling water back into the reservoir after it has been released through the turbine, thus “re-using” the same water to generate electricity at a later time. When the demand for electricity is low (typically at night), a pumped storage facility stores energy by pumping water from a lower reservoir to an upper reservoir. During periods of high electrical demand (typically during the day), the water is released back to the lower reservoir to generate electricity. The figure below shows a schematic of a pumped storage hydro facility. Pumped storage facilities are typically smaller in terms of generation capacity than their impoundment or diversion counterparts, but are sometimes combined with impoundment or diversion facilities to increase peak power output or flexibility.

Water in the oceans is constantly in motion due to waves and tides, and energy can be harvested from these kinds of motions. Waves, driven by the winds, make the water oscillate in roughly circular orbits extending to a depth of one half of the wavelength of the wave (distance between peaks). Tides, related to the gravitational pull of the Moon and Sun on the oceans, are like very long-wavelength waves that can produce very strong currents in some coastal areas due to the geometry of the shoreline. In terms of power generation technologies, wave and tidal power have both similarities and differences. Both refer to the extraction of kinetic energy from the ocean to generate electricity (again, by spinning a turbine just as hydroelectric dams or wind farms do), but the locations of each and the mechanisms that they use for generating power are slightly different.

Wave energy projects extract energy from waves on the surface of the water, or from wave motion a bit deeper (a few 10s of meters) in the ocean. Surface wave energy technologies capture the kinetic energy in breaking waves – these provide periodic impulses that spin a turbine. The US Department of Energy has a nice description of different types of surface wave projects as follows:

• Oscillating Water Columns: Oscillating water columns consist of a partially submerged concrete or steel structure that has an opening to the sea below the waterline. It encloses a column of air above a column of water. As waves enter the air column, they cause the water column to rise and fall. This alternately compresses and depressurizes the air column. As the wave retreats, the air is drawn back through the turbine as a result of the reduced air pressure on the ocean side of the turbine.
• Tapchans: Tapchans, or tapered channel systems, consist of a tapered channel that feeds into a reservoir constructed on cliffs above sea level. The narrowing of the channel causes the waves to increase in height as they move toward the cliff face. The waves spill over the walls of the channel into the reservoir, and the stored water is then fed through a turbine.
• Pendular Devices: Pendular wave-power devices consist of a rectangular box that is open to the sea at one end. A flap is hinged over the opening, and the action of the waves causes the flap to swing back and forth. The motion powers a hydraulic pump and a generator.

Offshore wave energy systems are typically placed deeper in the ocean, though not too deep – perhaps a few hundred feet below the ocean’s surface. The periodic wave activity at this depth is typically used to power a pump that feeds into a turbine, generating electricity.

Tidal energy projects typically work by forcing water through a turbine or a “tidal fence” that looks like a set of subway turnstiles. The systems depend on regular tidal activity to generate power. Because this tidal activity is predictable (each coast sees at least one tidal cycle per day – high tide and low tide – and some areas actually see two tidal cycles on a daily basis), tidal energy projects have the advantage of being able to provide a fairly predictable source of electricity. The use of tidal power, globally, has been quite limited because there are only a few sites in the world that see sufficiently large variations in tides to produce enough power, as shown in the table below.

When rivers are utilized to produce electricity, that is usually accomplished by building some sort of hydroelectric dam, like the three we discussed earlier. It doesn’t have to be that way, however. Many of the technologies used to extract energy from the tides (or similar technologies) could be deployed in freshwater river systems rather than the saltwater ocean, effectively acting as very small run-of-river facilities. These “hydrokinetic” power generation systems are typically individually small (each generating about 100 kilowatts or less of power) and could be situated in two ways. First, a propeller-like or turnstile-like turbine could be deployed directly into the riverway, operating much like a small-scale tidal power system. Second, a “micro-hydro” type of system could be employed, where river water is channeled to a turbine housing via a channel or pipeline, as shown in the figure below.

Globally, hydroelectricity is a major electricity resource, accounting for more than 16% of all electricity produced on the planet. More electricity is produced globally using hydropower than from plants fueled by nuclear fission or petroleum (natural gas and coal do produce more electricity globally than hydropower does). More than 150 countries produce some hydroelectricity, although around 50% of all hydropower is produced by just four countries: China, Brazil, Canada, and the United States. China is by far the largest hydropower producer on the planet, as shown in the figure below. Hydroelectricity production in China has tripled over the past decade, with the completion of some of the world’s largest dam projects, in particular, the Three Gorges Dam (the world’s largest), which could produce nearly enough electricity to power all of New England during a typical summer and left an area roughly the size of San Francisco flooded underwater.

Once hydroelectric dams are built, they run very cheaply and generally provide reliable supplies of electricity except during times of extreme drought. Developed countries that have substantial hydro resources have, by and large, already utilized those resources to produce electricity. In these countries, hydropower dominates the electricity supply system as shown in the chart below. Norway leads the pack here – the amount of hydropower that it produces is not large in an absolute sense (it is the world’s seventh-largest producer) but nearly all electricity generated in Norway comes from hydro-power. Brazil and Canada are also highly dependent on hydropower. Other large hydro producers, such as China and the United States, produce much less hydroelectricity relative to the size of their overall power sectors.

It is often said that developed countries like the United States have little potential for growth in hydroelectric power generation – the US, in particular, has dammed so many rivers, that what could possibly be left? In some areas of the Pacific Northwest, the US is removing some dams that produce electricity due to environmental concerns. It is certainly the case that there is relatively little potential for new hydro mega-projects in the US. This does not mean, however, that there is nowhere left to build new hydroelectric projects. There are, in fact, several hundred megawatts of planned hydroelectric generation in the Mid-Atlantic US alone (see the map at PJM), though most of the projects would be small pumped storage facilities) We’ll walk briefly through a few examples of the hydro resource potential in the United States before taking a more global look.

First, not all dams in the US are equipped with turbines to generate electricity. There are, actually, quite a few that aren’t (see the interactive map at Energy.gov – potentially enough power generation to supply more than ten million homes. Many of these are located along major shipping routes, like the Ohio and Mississippi Rivers. Others are located in areas where development might not make economic sense because the power would need to be shipped across large and expensive transmission lines. Another unconventional technology – hydro-kinetics – could potentially supply enough electricity to power the state of Virginia, although these resources are highly concentrated in the Lower Mississippi River and in more remote areas such as Alaska.

Globally, the picture is very different. Developing nations with abundant hydro resources, like China, Brazil, and other South American countries, are rushing headlong into planning and building new dams. So globally, hydroelectricity is alive and well, and growing rapidly (no pun intended) – although this growth comes in fits and starts since new dams each represent a big chunk of capacity and take a long time from project start to finish. Nearly half of the world’s fifteen largest dams were built since the year 2000, with only one of those in a country other than China or Brazil (the Sayano Shuskenskaya dam in Russia was recently upgraded, although the original went into place in the 1980s).

Energy from hydropower has been growing at a steady annual rate of 0.3 EJ per year since the year 2000, faster than the previous decades, but overall, hydro still accounts for just 16 EJ of the total 580 EJ we used in 2018. Some estimates suggest that if we really developed all of the economically viable hydroelectric sites, we might be able to generate as much as 50 EJ of hydroelectric energy — still a far cry from the +600 we will need in the near future. But even though hydro cannot solve all of our energy problems, it is nevertheless very important in that it supplies a relatively low-emissions, dispatchable (on-demand) energy source that can help smooth out the variability in wind and solar energy production.

While hydro development is still growing in several regions of the world, many countries, including the United States, will not likely pursue larger hydro projects due to two main factors — first, we have already built hydroelectric dams in most of the best places, and secondly, there are concerns over the environmental and societal impacts of building more dams. These environmental impacts have even been used to justify dam removal in some cases, though weighing those environmental impacts against the societal benefits (such as irrigation, flood control, recreation, and so forth…not just electricity) has always been controversial. Some of these specific environmental and societal impacts include:

• Impacts on fish migration – fish migrating up or downstream may find passages blocked by dams or may get killed going through turbines.
• Reduced water quality – changes in natural stream-flow may, in some cases, muddy waters. This affects the river ecosystems that aquatic plants and animals utilize.
• Land inundation – reservoirs behind dams can flood large areas. There is not necessarily a relationship between the size of the dam and the amount of area flooded behind the dam. Two impacts can arise here. First, building large hydro projects often involves the displacement of many people who used to live in the area surrounding the river. The Three Gorges Dam in China involved the relocation of more than one million people.
• Methane release to the atmosphere — plant life can decay in flooded areas, which over time releases large quantities of methane into the atmosphere – enough that it can potentially offset the avoided CO₂ emissions from many years worth of fossil fuel use. A recent estimate places the global emissions of methane high enough to make hydropower be just a little bit less than natural gas in terms of CO2 (equivalent) per unit of energy generated.

Maggie Koerth-Baker has a really great article on how nuclear power plants work, with a focus on the nuclear fission reaction and what mechanisms in a nuclear power plant keep the reaction from spinning out of control. It was written right after the incident at Fukushima Daichi. Before continuing on, please have a look at the article, and pay some attention not only to how plants work, but how the nuclear reactions inside the plants are controlled. The article also has a really nice description of how reactions at nuclear power plants can keep cascading even after the plant has been “shut down,” which is basically what causes meltdowns like those that happened in the Three Mile Island and Chernobyl power plants.

The basics of a nuclear power plant aren’t actually all that complicated. In fact, there is a remarkable similarity to fossil fuel plants, in that what ultimately happens in a nuclear power plant is that steam is produced, to drive a turbine inside a generator, which produces electricity. But unlike fossil fuel plants, which heat water by burning fuel, the water in nuclear power plants is heated through an atomic reaction.

There are two basic types of atomic reactions. The first is nuclear fusion, with which we are all intimately familiar, whether we know it or not. It is nuclear fusion that keeps the sun hot. In nuclear fusion, atoms are joined together. The word “joined” here is a bit of scientific jargon. In reality, the energy is released when atoms collide together at really high speeds. If you have ever seen two cars collide at high speed, you have some idea of how energy could be released when things hit each other. Despite years of research into nuclear fusion, scientists have never been able to engineer a controllable reaction in a laboratory environment. If they could, most of the world’s energy problems would basically be solved overnight, since the amount of energy released through a fusion reaction would be massive. But for now, fusion goes in the “maybe someday” pile.

The second type of nuclear reaction is fission, which is the opposite of fusion – atoms are broken apart, which also releases energy. U-235 is naturally radioactive, meaning that the nucleus is unstable, and it will eventually give off some energy and parts of its nucleus to get to a stable atom, but this takes a long time — the half-life is 700 million years.  The figure below illustrates roughly how this works. An atom (in the case of a nuclear power plant, a uranium-235 atom) is bombarded with neutrons, some of which are absorbed by the nucleus, so the U-235 becomes U-236 — this makes it even more unstable, so the atom splits apart into two lighter atoms called the daughter products.  U-236 splits into krypton (Kr-92) and barium (Ba-141), and it also releases energy in the form of heat, gamma radiation (bad for us) and 3 neutrons. (Note that if you add up the weight of the daughter products and the neutrons,  92+141+3, you get 236, the weight of the U-236 that split apart).  These neutrons come hurtling out of the original atom and smash into other uranium-235 atoms, triggering 3 more U-235 fission reactions, each of which generates 3 more neutrons.  As you can see, before long, there are a lot of neutrons and thus a lot of reactions and thus a lot of heat, which heat the water surrounding the fuel rods, creating steam, spinning the turbine — just like many of the other systems for making electricity.

The nuclear fission reaction described above is an example of a positive feedback mechanism that will naturally tend to speed up until all of the fuel (the U-235) is used up.  This means that it has a tendency to create more and more heat, and if left unchecked, this would cause the water in the reactor vessel to get too hot and build up too much pressure for the reactor to contain — then you would have a big steam explosion, such as happened at Chernobyl.  To control this reaction, the reactor core has a series of control rods, made of materials that absorb the neutrons emitted during a fission reaction.  So the control rods allow the operators to adjust the rate of the reaction and thus the rate of heat production.

There are two basic types of nuclear power plants that are in operation today. The first, and most common, is the Pressurized Water Reactor (PWR), which is illustrated in the animation below. In a PWR, hot water passes through the reactor core (where it absorbs the heat from the nuclear fission reactions) and is then pumped through a heat exchanger, where it heats another fluid that produces steam, powering the turbine. The primary advantage to this type of design is that the water in the primary loop (which passes through the core) does not actually come into contact with the fluid in the steam generator, so unless pipes or valves break there is no risk of contamination or radioactive water leaking from the plant. The Boiling Water Reactor (BWR), illustrated in the next figure, utilizes a somewhat simpler design, where the water that runs through the core is allowed to vaporize to steam, thus powering the turbine to generate electricity. While the design is simpler, it does mean that the steam entering the turbine can be radioactive.

Whether one design is inherently more advantageous than another is difficult to say. Both types have been involved in major nuclear power plant incidents. The reactor at Three Mile Island was a PWR while the reactor at Fukushima was a BWR, so the potential exists for problems at either type of plant. It is perhaps worth mentioning that the Three Mile Island incident was likely due as much to human error and poor design of the reactor’s control system at least as much as to the reactor design itself. The reactor at Chernobyl was an unusual Soviet design called a “light water graphite reactor” that was not really designed for use as a commercial nuclear power plant but was adapted for that use anyway. The World Nuclear Association has a nice description of the Chernobyl plant technology with a description of what went wrong (here too, human error played a central role).

Advanced PWRs have been developed that use more passive designs to keep the reactor from overheating, without any pumps or offsite power required. Westinghouse has developed one such design, the AP1000, which is currently being deployed in China. For those who are interested, more information on passive PWR designs can be found at Westinghouse Nuclear.

Nuclear fuel rods typically last for 3-5 years, and when a rod is "spent" it still contains some fissionable 235-U along with a host of other radioactive elements. So, what do we do with these spent rods? Many people would argue that recycling is a good thing. In the nuclear energy industry, recycling of spent nuclear fuel is a somewhat contentious topic. Many countries, including those European countries that still use nuclear energy, recycle spent fuel into new fuel for re-use. The United States does not do this, preferring a "once-through" fuel cycle for reasons of security as well as economics. Understanding the pros and cons of recycling nuclear fuel requires some understanding of how fuel for nuclear power plants is mined and fabricated.

The figure above outlines the many steps necessary to get uranium out of the ground and into a nuclear power plant. After extraction and processing (“milling”), uranium ore is transported to conversion facilities to remove impurities. The next step in the nuclear fuel cycle is enrichment. Owing to security concerns, all enrichment for the US commercial nuclear industry takes place at one government-owned gas diffusion facility in Paducah, Kentucky. Enriched uranium is then transported to one of several commercial fuel fabrication facilities where the fuel rods are manufactured. In the U.S., fuel fabrication is a competitive industry; private firms compete to provide finished fuel to nuclear power plants. Nuclear fuel rods are generally not purchased directly from the government. Nuclear fission and disposal of spent fuel rods constitute the final steps of the nuclear fuel cycle in the US

The US is heavily dependent on the global market for uranium and nuclear fuel. n 2017, 90% of uranium oxide supplies used to develop nuclear fuel in the US come from outside of the country. The main suppliers for the US are Canada (24%), Kazahkstan (20%), Australia (18%), and Russia (13%). Proposals to open new uranium mines in both the western and eastern United States have been met with resistance, primarily on environmental grounds.

Current US policy prohibits the reprocessing of spent nuclear fuel, for two primary reasons. First is economics – the fuel costs for nuclear power plants are already among the lowest of any non-renewable power generation resource. Once nuclear power plants are built, if they are well-run they cost very little to operate. While the recycling of spent nuclear fuel would eliminate the need for virgin uranium ore to be mined or for additional fuel to be purchased on the world market, it is not at all clear whether the benefits of doing so outweigh the costs of reprocessing. The other reason is nuclear security. The process of recycling nuclear fuel involves the separation of uranium and plutonium from the spent fuel rods. There have been concerns regarding plutonium falling into the wrong hands and contributing to the proliferation of nuclear weapons.

One very serious concern with nuclear power has to do with the highly radioactive waste from the process. Much of the waste needs to be isolated for at least 10,000 years. All civilian nuclear waste was intended to be stored permanently at a repository in Yucca Mountain, Nevada. Yucca Mountain was chosen as a waste repository site back in 1987 and we have spent over $15 billion investigating the site and developing 65 km of tunnels deep underground to store the waste. Currently, it could hold 65,000 tons of waste, but we have 94,000 tons of radioactive waste in temporary storage at nuclear plants. The Yucca Mountain facility is not currently operational and significant uncertainties exist as to whether it will ever be used. In the interim, spent nuclear waste will continue to be stored on-site at the power plants.

There are currently several hundred operating nuclear power plants in the world, spread over a few dozen countries, with over a hundred more “proposed” nuclear power plants (these may or may not get built, depending on economic and political factors in the relevant countries). The US still has the largest number of plants, with about 100 currently operating. France’s economy is the most dependent on nuclear energy, with more than 75% of electricity in that country coming from nuclear power plants. Countries with fleets of nuclear power are primarily wealthier nations, such as the US and European countries, but developing nations are really the biggest growth area, particularly China. Prior to the Fukushima incident, other Asian nations besides China had plans to grow their nuclear fleets, but whether that growth will materialize is highly uncertain. In response to concerns regarding the safety of nuclear power plants and waste disposal/management issues, some European countries have enacted various policies mandating the phase-out of nuclear energy, including Austria, Sweden, Germany, Italy, and Belgium. Other countries, including Spain and Switzerland, have imposed a moratorium on the construction of new nuclear power plants. Of the countries that have decided to phase out nuclear energy, Germany has been among the most aggressive following the Fukushima incident. Because of concerns over electricity supply and costs, however, some countries have delayed or back-stepped on plans to phase out nuclear energy.

Before going ahead, we need to make sure we all have a clear picture of the various units we use to measure energy.

Joule — the joule (J) is the basic unit of energy, work done, or heat in the SI system of units; it is defined as the amount of energy, or work done, in applying a force of one Newton over a distance of one meter. One way to think of this is as the energy needed to lift a small apple (about 100 g) one meter. An average person gives off about 60 J per second in the form of heat. We are going to be talking about very large amounts of energy, so we need to know about some terms that are used to describe larger sums of energy:

In recent years, we humans have consumed about 518 EJ of energy per year, which is something like 74 GJ per person per year.

British Thermal Unit— the btu is another unit of energy that you might run into. One btu is the amount of energy needed to warm one pound of water one °F. One btu is equal to about 1055 joules of energy. Oddly, some branches of our government still use the btu as a measure of energy.

Watt— the watt (W) is a measure of power and is closely related to the Joule; it is the rate of energy flow, or joules/second. For instance, a 40 W light bulb uses 40 joules of energy per second, and the average sunlight on the surface of Earth delivers 343 W over every square meter of the surface.

Kilowatt hours— when you (or you parents maybe for now) pay the electric bill each month, you get charged according to how much energy you used, and they express this in the form of kilowatt hours — kWh. If you use 1000 Watts for one hour, then you have used one kWh. This is really a unit of energy, not power:

$$\text{1[kWh]=1000[W]}\times \text{1[hr]=1000}\left[\frac{J}{s}\right]\times \text{3600[s]=3,600,000[J]}$$

In other words, one kilowatt hour is 1000 joules per second (kW) summed up over one hour (3600 seconds), which is the same as 3.6 MJ or 3.6 x 10⁶J or 3.6e6 J.

The energy we use to support the whole range of human activities comes from a variety of sources, but as you all know, fossil fuels (coal, oil, and natural gas) currently provide the majority of our energy on a global basis, supplying about 81% of the energy we use:

The non-fossil fuel sources include nuclear, hydro (dams with electrical turbines attached to the outflow), solar (both photovoltaic and solar thermal), and a variety of other sources. These non-fossil fuel sources currently supply about 19% of the total energy.

The percentages of our energy provided by these different sources have clearly changed over time and will certainly change in the future as well. The graph below gives us some sense of how dramatically things have changed over the past 210 years:

There are a couple of interesting features to point out about this graph. For one, note that the total amount of energy consumed has risen dramatically over time — this is undoubtedly related to both population growth and the industrial revolution. The second point is that shifting from one energy source to another takes a long time. Oil was being pumped out of the ground in 1860, and even though it has a greater energy density and is more versatile than coal, it did not really make its mark as an energy source until about 1920, and it did not surpass coal as an energy source until about 1940. Of course, you might argue that the world changed more slowly back then, but it is probably hard to avoid the conclusion that our energy supply system has a lot of inertia, resulting in sluggish change.

We are all aware of some of the ways we use energy — heating and cooling our homes, transporting ourselves via car, bus, train, or plane — but there are many other uses of energy that we tend not to think about. For instance, growing food and getting it onto your plate uses energy — think of the farming equipment, the food processing plant, the transportation to your local store. Or, think of manufactured items — to make something like a car requires energy to extract the raw materials from the earth and then assembling them requires a great deal of energy. So, when you consider all of the different uses of energy, we see a dominance of industrial uses:

Since we are going to be modeling the future of global energy consumption, we should first familiarize ourselves with the recent history of energy consumption.

Question: Why has our energy consumption increased over this time period?

Here, we will explore a few possibilities, the first of which is global population increase — more people on the planet leads to a greater total energy consumption. To evaluate this, we need to plot the global population and the total energy consumption on the same graph to see if the rise in population matches the rise in energy consumption.

The two curves match very closely, suggesting that population increase is certainly one of the main reasons for the rise in energy consumption. But is it as simple as that — more people equals more energy consumption?

If the rise in global energy consumption is due entirely to population increase, then there should be a constant amount of energy consumed per person — this is called the per capita energy consumption. To get the per capita energy consumption, we just need to divide the total energy by the population (in billions) — so we’ll end up with Exajoules of energy per billion people.

Today, we use about 3 times as much energy per person than in 1900, which is not such a surprise if you consider that we have many more sources of energy available to us now compared to 1900. Note that at the same time that the population really takes off (see Fig. 5), the per capita energy consumption also begins to rise. This means that the total global energy consumption rises due to both the population and the demand per person for more energy.

Let’s try to understand this per capita energy consumption a bit better. We know that the global average is 74 EJ per billion people, but how does this value change from place to place? There are some huge variations across the globe — Afghans use about 4 GJ per person per year, while Icelanders use 709 GJ per person. Why does it vary so much? Is it due to the level of economic development, or the availability of energy, or the culture, or the climate? You can come up with reasons why each of these factors (and others) might be important, but let’s examine one in more detail — the economic development expressed as the GDP (the gross domestic product, which reflects the size of the economy) per capita.

The obvious linear trend to these data suggests that per capita energy consumption is a function of GDP, while the fact that it is not a tight line tells us that GDP is not the whole story in terms of explaining the differences in energy consumption. Not surprisingly, we are near the upper right of this plot, consuming more than 300 GJ per person per year. Iceland’s economy is not as big per person as ours, and yet they consume vast amounts of energy per person, partly because it is cold and they have big heating demands, but also because they have abundant, inexpensive geothermal energy thanks to the fact that they live on a huge volcano. Many European countries with strong economies (e.g., Germany) use far less energy per person than we do (168 GJ compared to our 301 GJ), in part because they are more efficient than us and in part because they are smaller, which cuts down on their transportation. A big part of the reason they are more efficient than us is that energy costs more over there — for instance, a gallon of gas in Italy is about $8. Our neighbor, Mexico, has a per capita energy consumption that is just about the global average.

Pay attention to the two red squares in Fig. 7 — these show the global averages in terms of GDP and energy consumption per person for two points in time. The trend is most definitely towards increasing GDP (meaning increasing economic development) and increasing energy consumption per person. Economic development is definitely a good thing because it is tied to all sorts of indicators of a higher quality of life — better education, better health care, better diet, increased life expectancy, and lower birth rates. But, economic growth has historically come with higher energy consumption, and that means higher carbon emissions.

Now that we’ve seen what some of the patterns and trends are, we are ready to think about the future.

There are many ways to meet our energy demands for the future, and each way could include different choices about how much of each energy source we will need. We’re going to refer to these “ways” as scenarios — hypothetical descriptions of our energy future. Each scenario could also include assumptions about how the population will change, how the economy will grow, how much effort we put into developing new technologies and conservation strategies. Each scenario can be used to generate a history of emissions of CO₂, and then we can plug that into a climate model to see the consequences of each scenario.

Emissions per unit energy for different sources

The global emission of carbon into the atmosphere due to human activities is dominated by the combustion of fossil fuels in the generation of energy, but the various energy sources — coal, oil, and gas — emit different amounts of CO₂ per unit of energy generated. Coal releases the most CO₂per unit of energy generated during combustion — about 103.7 g CO₂per MJ (10⁶ J) of energy. Oil follows with 65.7 g CO₂/MJ, and gas is the “cleanest” or most efficient of these, releasing about 62.2 g CO₂/MJ.

At first, you might think that renewable or non-fossil fuel sources of energy will not generate any carbon emissions, but in reality, there are some emissions related to obtaining our energy from these means. For example, a nuclear power plant requires huge quantities of cement, the production of which releases CO₂ into the atmosphere. The manufacture of solar panels requires energy as well, and so there are emissions related to that process because our current industrial world gets most of its energy from fossil fuels. For these energy sources, the emissions per unit of energy are generally estimated using a lifetime approach — if you emitted 1000 g of CO₂ to make a solar panel and over its lifetime, it generated 500 MJ, then its emission rate is 2 g CO₂/MJ. If we average these non-fossil fuel sources together, they release about 5 g CO₂/MJ — far cleaner than the other energy sources, but not perfectly clean.

So, to sum it up, here is a ranking of the emissions related to different energy sources:

*Hydro, Nuclear, Wind, Solar

**This will decrease as the non-fossil fuel fraction increases

Our recent energy consumption is about 518 EJ (10¹⁸ J). Let’s calculate the emissions of CO₂ caused by this energy consumption, given the values for CO₂/MJ given above and the current proportions of energy sources — 33% oil, 27% coal, 21% gas, and 19% other non-fossil fuel sources. The way to do this is to first figure out how many grams of CO₂ are emitted per MJ given this mix of fuel sources, and then scale up from 1 MJ to 518 EJ. Let’s look at an example of how to do the math here — let r_1-4 in the equation below be the rates of CO₂ emission per MJ given above, and let f_1-4 be the fractions of different fuels given above. So r₁ could be the rate for oil (65.7) and f₁ would be the fraction of oil (.33). You can get the composite rate from:

$$r_{c }= r_{1 }{f}_{1 }+ r_{2 }{f}_{2 }+ r_{3 }{f}_{3 }+ r_{4 }{f}_{4 }$$

Plugging in the numbers, we get:

$$65.7 \times .33 + 62.2 \times .27 + 103.7 \times .21 + 6.2 \times .19 = 61.4 \left[ \frac{{\mathrm{gCo} }_{2 }}{\mathrm{MJ} }\right]$$

What is the total amount of CO₂ emitted? We want the answer to be in Gigatons — that’s a billion tons, and in the metric system, one ton is 1000 kg (1e6 g or 10⁶ g), which means that 1Gt = 10¹⁵ g (1e15 g).

$$1 [ \mathrm{EJ} ] = 1 \mathrm{e} 12 [ \mathrm{MJ} ] , \text{so} , 518 [ \mathrm{EJ} ] = 518 \mathrm{e} 12 [ \mathrm{MJ} ]$$

$$518 \mathrm{e} 12 [ \mathrm{MJ} ] \times 61.4 \left[ \frac{{\mathrm{gCO} }_{2 }}{\mathrm{MJ} }\right] = 31.7 \mathrm{e} 15 \left[ {\mathrm{gCO} }_{2 }\right] = 31.8$$

So, the result is 31.8 Gt of CO₂, which is very close to recent estimates for global emissions.

It is more common to see the emissions expressed as Gt of just C, not CO₂, and we can easily convert the above by multiplying it by the atomic weight of carbon divided by the molecular weight of CO₂, as follows:

$$31.8 \left[ {\mathrm{GtCO} }_{2 }\right] \times \frac{12 [ \mathrm{gC} ] }{44 \left[ {\mathrm{gCO} }_{2 }\right] }= 8.7 [ \mathrm{GtC} ]$$

And remember that this is the annual rate of emission.

Let’s quickly review what went into this calculation. We started with the annual global energy consumption at the present, which we can think of as being the product of the global population times the per capita energy consumption. Then we calculated the amount of CO₂ emitted per MJ of energy, based on different fractions of coal, oil, gas, and non-fossil energy sources — this is the emissions rate. Multiplying the emissions rate times the total energy consumed then gives us the global emissions of either CO₂ or just C.

We now see what is required to create an emissions scenario:

1. A projection of global population
2. A projection of the per capita energy demand
3. A projection of the fractions of our energy provided by different sources
4. Emissions rates for the various energy sources

In this list, the first three are variables — the 4^th is just a matter of chemistry. So, the first three constitute the three principal controls on carbon emissions.

Here is a diagram of a simple model that will allow us to set up emissions scenarios for the future:

In this model, the per capita energy (a graph that you can change) is multiplied by the Population to give the global energy consumption, which is then multiplied by RC (the composite emissions rate) to give Total Emissions. Just as we saw in the sample calculation above, RC is a function of the fractions and emissions rates for the various sources. Note that the non-fossil fuel energy sources (nuclear, solar, wind, hydro, geothermal, etc.) are all lumped into a category called renew, because they are mostly renewable. The model includes a set of additional converters (circles) that allow you to change the proportional contributions from the different energy sources during the model run.

This emissions model is actually part of a much larger model that includes a global carbon cycle model and a climate model. Here is how it works — the Total Emissions transfers carbon from a reservoir called Fossil Fuels that represents all the Gigatons of carbon stored in oil, gas, and coal (they add up to 5000 Gt) into the atmosphere. Some of the carbon stays in the atmosphere, but the majority of it goes into plants, soil, and the oceans, cycling around between the reservoirs indicated below. The amount of carbon that stays in the atmosphere then determines the greenhouse forcing that affects the global temperature — you’ve already seen the climate model part of this. The carbon cycle part of the model is complicated, but it is a good one in the sense that if we plug in the known historical record of carbon emissions, it gives us the known historical CO₂ concentrations of the atmosphere. Here is a highly schematic version of the model: