What are the most common vehicle repairs that are going away with electric vehicles?

An internal combustion engine drivetrain contains about 2,000 parts, while an electric vehicle drivetrain contains about 20.  All other things being equal, a system with fewer moving parts will be more reliable than a system with more moving parts. In 2006, the National Highway Transportation Safety Administration estimated that the average vehicle, built solely on internal combustion engines, lasted 150,000 miles. Current estimates for the lifetime today’s electric vehicles are over 500,000 miles.

Tesloop, a Tesla-centric ride-hailing company has already driven its first Model S for more 200,000 miles, and seen only an 6% loss in battery life. A battery lifetime of 1,000,000 miles may even be in reach. This increased lifetime translates directly to a lower cost of ownership: extending an EVs life by 3-4 X means an EVs capital cost, per mile, is 1/3 or 1/4 that of a gasoline-powered vehicle.  Better still, the cost of switching from gasoline to electricity delivers another savings of about 1/3 to 1/4 per mile.  And electric vehicles do not need oil changes, air filters, or timing belt replacements; the 200,000 mile Tesloop never even had its brakes replaced.  The most significant repair cost on an electric vehicle is from worn tires. For emphasis: The total cost of owning an electric vehicle is, over its entire life, roughly 1/4 to 1/3  the cost of a gasoline-powered vehicle. The IRS allows charges of 53.5¢ per mile in 2017, a number clearly derived for gasoline vehicles.  At 1/4 the price, a fleet electric vehicle should cost only 13¢ per mile, a savings of 40¢ per mile.

An autonomous vehicle will cost about $0.13 per mile to operate, and even less as battery life improves.  By comparison, your 20 miles per gallon automobile costs $0.10 per mile to refuel if gasoline is $2/gallon, and that is before paying for insurance, repairs, or parking.  Add those, and the price of operating a vehicle you have already paid off shoots to $0.20 per mile, or more. And this is what will kill oil:  It will cost less to hail an autonomous electric vehicle than to drive the car that you already own.


 ~ SETH This is how Big Oil will die, Perspicity

It’s 2025, and 800,000 tons of used high strength steel is coming up for auction.

The steel made up the Keystone XL pipeline, finally completed in 2019, two years after the project launched with great fanfare after approval by the Trump administration.  The pipeline was built at a cost of about $7 billion, bringing oil from the Canadian tar sands to the US, with a pit stop in the town of Baker, Montana, to pick up US crude from the Bakken formation.  At its peak, it carried over 500,000 barrels a day for processing at refineries in Texas and Louisiana.

But in 2025, no one wants the oil.

The Keystone XL will go down as the world’s last great fossil fuels infrastructure project.  TransCanada, the pipeline’s operator, charged about $10 per barrel for the transportation services, which means the pipeline extension earned about $5 million per day, or $1.8 billion per year.  But after shutting down less than four years into its expected 40 year operational life, it never paid back its costs.

The Keystone XL closed thanks to a confluence of technologies that came together faster than anyone in the oil and gas industry had ever seen.  It’s hard to blame them – the transformation of the transportation sector over the last several years has been the biggest, fastest change in the history of human civilization, causing the bankruptcy of blue chip companies like Exxon Mobil and General Motors, and directly impacting over $10 trillion in economic output.

And blame for it can be traced to a beguilingly simple, yet fatal problem:  the internal combustion engine has too many moving parts.

Cummins diesel engine
The Cummins Diesel Engine, US Patent No. 2,408,298, filed April 1943, awarded Sept 24, 1946

Let’s bring this back to today:  Big Oil is perhaps the most feared and respected industry in history.  Oil is warming the planet – cars and trucks contribute about 15% of global fossil fuels emissions – yet this fact barely dents its use.  Oil fuels the most politically volatile regions in the world, yet we’ve decided to send military aid to unstable and untrustworthy dictators, because their oil is critical to our own security.  For the last century, oil has dominated our economics and our politics.  Oil is power.

Yet I argue here that technology is about to undo a century of political and economic dominance by oil. Big Oil will be cut down in the next decade by a combination of smartphone apps, long-life batteries, and simpler gearing.  And as is always the case with new technology, the undoing will occur far faster than anyone thought possible.

To understand why Big Oil is in far weaker a position than anyone realizes, let’s take a closer look at the lynchpin of oil’s grip on our lives: the internal combustion engine, and the modern vehicle drivetrain.

BMW 8 speed
BMW 8 speed automatic transmission, showing lots of fine German engineered gearing.  From Euro Car News.

Cars are complicated.

Behind the hum of a running engine lies a carefully balanced dance between sheathed steel pistons, intermeshed gears, and spinning rods – a choreography that lasts for millions of revolutions.  But millions is not enough, and as we all have experienced, these parts eventually wear, and fail.  Oil caps leak.  Belts fray.  Transmissions seize.

To get a sense of what problems may occur, here is a list of the most common vehicle repairs from 2015:

  1. Replacing an oxygen sensor – $249
  2. Replacing a catalytic converter – $1,153
  3. Replacing ignition coil(s) and spark plug(s) – $390
  4. Tightening or replacing a fuel cap – $15
  5. Thermostat replacement – $210
  6. Replacing ignition coil(s) – $236
  7. Mass air flow sensor replacement – $382
  8. Replacing spark plug wire(s) and spark plug(s) – $331
  9. Replacing evaporative emissions (EVAP) purge control valve – $168
  10. Replacing evaporative emissions (EVAP) purging solenoid – $184

And this list raises an interesting observation: None of these failures exist in an electric vehicle.

The point has been most often driven home by Tony Seba, a Stanford professor and guru of “disruption”, who revels in pointing out that an internal combustion engine drivetrain contains about 2,000 parts, while an electric vehicle drivetrain contains about 20.  All other things being equal, a system with fewer moving parts will be more reliable than a system with more moving parts.

And that rule of thumb appears to hold for cars.  In 2006, the National Highway Transportation Safety Administration estimated that the average vehicle, built solely on internal combustion engines, lasted 150,000 miles. Current estimates for the lifetime today’s electric vehicles are over 500,000 miles.

The ramifications of this are huge, and bear repeating.  Ten years ago, when I bought my Prius, it was common for friends to ask how long the battery would last – a battery replacement at 100,000 miles would easily negate the value of improved fuel efficiency.  But today there are anecdotal stories of Prius’s logging over 600,000 miles on a single battery.

The story for Teslas is unfolding similarly.  Tesloop, a Tesla-centric ride-hailing company has already driven its first Model S for more 200,000 miles, and seen only an 6% loss in battery life. A battery lifetime of 1,000,000 miles may even be in reach.

This increased lifetime translates directly to a lower cost of ownership: extending an EVs life by 3-4 X means an EVs capital cost, per mile, is 1/3 or 1/4 that of a gasoline-powered vehicle.  Better still, the cost of switching from gasoline to electricity delivers another savings of about 1/3 to 1/4 per mile.  And electric vehicles do not need oil changes, air filters, or timing belt replacements; the 200,000 mile Tesloop never even had its brakes replaced.  The most significant repair cost on an electric vehicle is from worn tires. For emphasis: The total cost of owning an electric vehicle is, over its entire life, roughly 1/4 to 1/3  the cost of a gasoline-powered vehicle.

Of course, with a 500,000 mile life a car will last 40-50 years.  And it seems absurd to expect a single person to own just one car in her life.

But of course a person won’t own just one car.  The most likely scenario is that, thanks to software, a person won’t own any.


Here is the problem with electric vehicle economics:  A dollar today, invested into the stock market at a 7% average annual rate of return, will be worth $15 in 40 years.  Another way of saying this is the value, today, of that 40th year of vehicle use is approximately 1/15th that of the first. The consumer simply has little incentive to care whether or not a vehicle lasts 40 years.  By that point the car will have outmoded technology, inefficient operation, and probably a layer of rust.  No one wants their car to outlive their marriage.

But that investment logic looks very different if you are driving a vehicle for a living. A New York City cab driver puts in, on average, 180 miles per shift (well within the range of a modern EV battery), or perhaps 50,000 miles per work year.  At that usage rate, the same vehicle will last roughly 10 years.  The economics, and the social acceptance, get better.

And if the vehicle was owned by a cab company, and shared by drivers, the miles per year can perhaps double again.  Now the capital is depreciated in 5 years, not 10.  This is, from a company’s perspective, a perfectly normal investment horizon. A fleet can profit from an electric vehicle in a way that an individual owner cannot.

Here is a quick, top-down analysis on what it’s worth to switch to EVs:  The IRS allows charges of 53.5¢ per mile in 2017, a number clearly derived for gasoline vehicles.  At 1/4 the price, a fleet electric vehicle should cost only 13¢ per mile, a savings of 40¢ per mile.

40¢ per mile is not chump change – if you are a NYC cab driver putting 50,000 miles a year onto a vehicle, that’s $20,000 in savings each year.  But a taxi ride in NYC today costs $2/mile; that same ride, priced at $1.60 per mile, will still cost significantly more than the 53.5¢ for driving the vehicle you already own.  The most significant cost of driving is still the driver.

But that, too, is about to change.  Self-driving taxis are being tested this year in PittsburghPhoenix, and Boston, as well as SingaporeDubai, and Wuzhen, China.

And here is what is disruptive for Big Oil:  Self-driving vehicles get to combine the capital savings from the improved lifetime of EVs, with the savings from eliminating the driver.

The costs of electric self-driving cars will be so low, it will be cheaper to hail a ride than to drive the car you already own.


Today we view automobiles not merely as transportation, but as potent symbols of money, sex, and power.  Yet cars are also fundamentally a technology.  And history has told us that technologies can be disrupted in the blink of an eye.

Take as an example my own 1999 job interview with the Eastman Kodak company.  It did not go well.

At the end of 1998, my father had gotten me a digital camera as a present to celebrate completion of my PhD.  The camera took VGA resolution pictures – about 0.3 megapixels – and saved them to floppy disks.  By comparison, a conventional film camera had a nominal resolution of about 6 megapixels.  When printed, my photos looked more like impressionist art than reality. However, that awful, awful camera was really easy to use.  I never had to go to the store to buy film.  I never had to get pictures printed.  I never had to sort through a shoebox full of crappy photos.  Looking at pictures became fun.

Wife, with mildly uncooperative cat, January 1999.  Photo is at the camera’s original resolution.

I asked my interviewer what Kodak thought of the rise of digital; she replied it was not a concern, that film would be around for decades.  I looked at her like she was nuts.  But she wasn’t nuts, she was just deep in the Kodak culture, a world where film had always been dominant, and always would be.

This graph plots the total units sold of film cameras (grey) versus digital (blue, bars cut off).  In 1998, when I got my camera, the market share of digital wasn’t even measured.  It was a rounding error. By 2005, the market share of film cameras were a rounding error.

Analog Cameras
A plot of the rise of digital cameras (blue) and the fall of analog (grey).  Original from Mayflower via mirrorlessrumors, slightly modified for use here.

In seven years, the camera industry had flipped.  The film cameras went from residing on our desks, to a sale on Craigslist, to a landfill.  Kodak, a company who reached a peak market value of $30 billion in 1997, declared bankruptcy in 2012.  An insurmountable giant was gone.

That was fast.  But industries can turn even faster:  In 2007, Nokia had 50% of the mobile phone market, and its market cap reached $150 billion.  But that was also the year Apple introduced the first smartphone.  By the summer of 2012, Nokia’s market share had dipped below 5%, and its market cap fell to just $6 billion. In less than five years, another company went from dominance to afterthought.

Nokia market share
A quarter-by-quarter summary of Nokia’s market share in cell phones.  From Statista.

Big Oil believes it is different.  I am less optimistic for them.

An autonomous vehicle will cost about $0.13 per mile to operate, and even less as battery life improves.  By comparison, your 20 miles per gallon automobile costs $0.10 per mile to refuel if gasoline is $2/gallon, and that is before paying for insurance, repairs, or parking.  Add those, and the price of operating a vehicle you have already paid off shoots to $0.20 per mile, or more. And this is what will kill oil:  It will cost less to hail an autonomous electric vehicle than to drive the car that you already own.

If you think this reasoning is too coarse, consider the recent analysis from the consulting company RethinkX (run by the aforementioned Tony Seba), which built a much more detailed, sophisticated model to explicitly analyze the future costs of autonomous vehicles.  Here is a sampling of what they predict:

  • Self-driving cars will launch around 2021
  • A private ride will be priced at 16¢ per mile, falling to 10¢ over time.
  • A shared ride will be priced at 5¢ per mile, falling to 3¢ over time.
  • By 2022, oil use will have peaked
  • By 2023, used car prices will crash as people give up their vehicles. New car sales for individuals will drop to nearly zero.
  • By 2030, gasoline use for cars will have dropped to near zero, and total crude oil use will have dropped by 30% compared to today.

The driver behind all this is simple: Given a choice, people will select the cheaper option.

Your initial reaction may be to believe that cars are somehow different – they are built into the fabric of our culture.  But consider how people have proven more than happy to sell seemingly unyielding parts of their culture for far less money.  Think about how long a beloved mom and pop store lasts after Walmart moves into town, or how hard we try to “Buy American” when a cheaper option from China emerges.

And autonomous vehicles will not only be cheaper, but more convenient as well – there is no need to focus on driving, there will be fewer accidents, and no need to circle the lot for parking.  And your garage suddenly becomes a sunroom.

For the moment, let’s make the assumption that the RethinkX team has their analysis right (and I broadly agree[1]): Self-driving EVs will be approved worldwide starting around 2021, and adoption will occur in less than a decade.

How screwed is Big Oil?


Perhaps the metaphors with film camera or cell phones are stretched. Perhaps the better way to analyze oil is to consider the fate of another fossil fuel: coal.

The coal market is experiencing a shock today similar to what oil will experience in the 2020s. Below is a plot of total coal production and consumption in the US, from 2001 to today.  As inexpensive natural gas has pushed coal out of the market, coal consumption has dropped roughly 25%, similar to the 30% drop that RethinkX anticipates for oil.  And it happened in just a decade.

Coal production
Coal consumption has dropped 25% from its peak. From the Kleinman Center for Energy Policy.

The result is not pretty.  The major coal companies, who all borrowed to finance capital improvements while times were good, were caught unaware.  As coal prices crashed, their loan payments became a larger and larger part of their balance sheets; while the coal companies could continue to pay for operations, they could not pay their creditors. The four largest coal producers lost 99.9% of their market value over the last 6 years.  Today, over half of coal is being mined by companies in some form of bankruptcy.

Coal market cap
The four largest coal companies had a combined market value of approximately zero in 2016.  This image is one element of a larger graphic on the collapse of coal from Visual Capitalist.

When self-driving cars are released, consumption of oil will similarly collapse.

Oil drilling will cease, as existing fields become sufficient to meet demand.  Refiners, whose huge capital investments are dedicated to producing gasoline for automobiles, will write off their loans, and many will go under entirely.  Even some pipeline operators, historically the most profitable portion of the oil business, will be challenged as high cost supply such as the Canadian tar sands stop producing.

A decade from now, many investors in oil may be wiped out.  Oil will still be in widespread use, even under this scenario – applications such as road tarring are not as amenable to disruption by software. But much of today’s oil drilling, transport, and refining infrastructure will be redundant, or ill-fit to handle the heavier oils needed for powering ships, heating buildings, or making asphalt.  And like today’s coal companies, oil companies like TransCanada may have no money left to clean up the mess they’ve left.


Of course, it would be better for the environment, investors, and society if oil companies curtailed their investing today, in preparation for the long winter ahead.  Belief in global warming or the risks of oil spills is no longer needed to oppose oil projects – oil infrastructure like the Keystone XL will become a stranded asset before it can ever return its investment.

Unless we have the wisdom not to build it.

The battle over oil has historically been a personal battle – a skirmish between tribes over politics and morality, over how we define ourselves and our future.  But the battle over self-driving cars will be fought on a different front.  It will be about reliability, efficiency, and cost.  And for the first time, Big Oil will be on the weaker side.

Within just a few years, Big Oil will stagger and start to fall.  For anyone who feels uneasy about this, I want to emphasize that this prediction isn’t driven by environmental righteousness or some left-leaning fantasy.  It’s nothing personal.  It’s just business.


1 Thinking about how fast a technology will flip is worth another post on its own. Suffice it to say that the key issues are (1) how big is the improvement?, and (2) is there a channel to market already established? The improvement in this case is a drop in cost of >2X – that’s pretty large. And the channel to market – smartphones – is already deployed. As of a year ago, 15% of Americans had hailed a ride using an app, so there is a small barrier to entry as people learn this new behavior, but certainly no larger than the barrier to smartphone adoption was in 2007. So as I said, I broadly believe that the roll-out will occur in about a decade. But any more detail would require an entirely new post.


What climate skeptics taught me about global warming


Long before research exposed evidence that humans cause global warming, science made another sensational claim – that smoking caused lung cancer.

That case has been proven beyond doubt. But there is a science story from this era that is mostly forgotten: The battle against cigarettes taught science how to prove.

Before linking cigarettes to lung cancer, science had no established method to prove that one thing caused another.  The fields of epidemiology and statistics were new, and while they had some prior successes, the questions were so evident – think about mercury causing madness – that proof did not require the level of meticulousness that modern science expects.  The need to establish a link between cigarettes and lung cancers – and the backlash that ensued – changed this.  Epidemiology and cigarettes grew up together.

Today’s debate over global warming echoes that era.  Because of politics, a post like this, intended to inform, will sway few minds.  But I have spoken with skeptics who honestly want to understand, but don’t have the tools to grapple with such a large, complex field of science. And they have a point – while we talk a lot about the data, we rarely describe the path to a conclusion.

Provoked by their questions, I began to dig.  And I unearthed a notion that is rarely mentioned in the global warming debate: Science actually has a method for establishing that one thing causes another.  Scientists don’t have to vote on the issue – the 97% consensus of climate scientists who believe that humans cause warming is telling, but only one part of a broader process.  And for those who want to honestly weigh their skepticism in context of the evidence, there is a way.

Here’s the story.

Data collected by Gallup from 1954 to 2013.  Image Credit: Gallup

In the 1950s, Professor Bradford Hill kept a box of cigarettes in his desk at the London School of Hygiene and Tropical Medicine.

Hill led the school’s Statistical Research Unit, and like most men of the establishment, he would open his box to respected visitors.  This was hardly remarkable, save for one detail:  Hill was lead statistician on the British Doctor’s Study.  This was one of the two large studies that, when published in 1954, led the American Cancer Society to declare that ‘the presently available evidence indicates an association between smoking, particularly cigarette smoking, and lung cancer’.

Between 1930 and 1940, the lung cancer rate among men tripled.  Between 1940 and 1950, it doubled again.  Between 1950 and 1960, it nearly doubled again.  To quote the Surgeon’s General’s famous 1964 report linking smoking to lung cancer, “This extraordinary rise has not been recorded for cancer of any other site.”

Yet in the 1950s, even with data against smoking amassing, it was still considered rude not to offer a cigarette.

There was no singular moment when scientists realized that smoking caused lung cancer.  Beginning in 1912, when the first suggestion was made, scientists slowly built multiple strands of evidence, refining experiments as they learned. As the case grew in strength, each scientist, looking at the evidence before him (it was almost all men), individually concluded the causal connection was irrefutable.

Hill embarked on the Doctor’s Study because his previous research, performed with his longtime collaborator Professor Richard Doll in 1950, found a substantial correlation between cigarettes and lung cancer in a small patient population.  The 1950 link was so striking, in fact, that Doll gave up his cigarettes immediately.

Yet Hill himself held on to his pipe until the Doctor’s Study completed in 1954.

And while Hill and Doll and the American Cancer Society were in agreement by 1954, even in the late 1950s high level critics remained, including the esteemed statistician Ronald Fisher, who pronounced himself “extremely skeptical of the claim that decisive evidence has been obtained.”

Fisher was not a man to be taken lightly.  As a scientist, he has been described by the scientist Richard Dawkins as “the greatest biologist since Darwin”.  He provided a mathematical basis for evolutionary theory.  He single-handedly created most of modern statistics and the design for the randomized controlled trial, which went on to become the primary tool of medical research.

Fisher also smoked, preached libertarian political views, and was an advisor to the Tobacco Manufacturers Standing Committee.  He was not happy about the idea that scientists should inject “propaganda” onto an unsuspecting public.  Especially because he believed the science was wrong.

To illustrate Hill and Doll’s folly, Fisher tore apart their data, highlighting discrepancies between cancer rates in cigarette, cigar, and pipe smokers.  He described how much of the increase in cancers could be ascribed to improved methods of detection.  And he inaugurated the study of spurious correlations, showing that Doll and Hill’s methods would directly tie an increase in the import of apples to an increase in the divorce rate.

Hill would eventually be proven right.  But needed to develop better tools to show this. And by creating these tools, he would define the rules for epidemiology for the next fifty years.

Data credit: cancer.org, original image slightly reformatted by me in Tableau

It is simple to show correlation.  But how can one prove causation?

This problem is not limited to studies of smoking – it extends through all of science.  In fact, if you were to ask scientists outside epidemiology what process they use to “prove” causality, I’d wager that most would either change the subject, or stare back blankly at you.  Different scientists evidently maintain different standards for proof.  No one is working with a standard process.

This is vaguely unsettling.

The problem certainly infects the global warming debate.  “Proof” gets thrown around by different people in different way, leaving everyone confused.  Politically-driven skeptics leave the term undefined to sandbag the discussion for their own purposes – it’s easier to claim “not enough is known” when you never define “known”.  Yet the lack of definition hovers like a fog over anyone trying honestly to parse out the answers for themselves. It’s hard to have faith in something so ill-defined.

And this brings us back to Professor Hill.

The battle against smoking was the first bare-knuckles public policy debate driven by science.  So over years of defending his work, Hill had to think deeply about what constitutes ‘proof’, and how to overcome the intelligent rebuttals of the world’s Ronald Fishers.

In 1965, he formally proposed a solution.

Hill recognized that there are more ways to support causation that finding that two variables track.  In fact, Hill identified nine separate strands of ‘proof’, each of which makes an independent case for or against causation.  The list of nine aspects – and I’ll go into details below – are now called Hill’s Criteria.

You don’t need strong support from all of the strands to prove a result.  But when independent strands tell the same story, with no contradictions, the case is strong. Perhaps as importantly, by using fixed criteria, we can categorize not just data we have, but identify what data are missing as well.  And with all of the possible evidence in mind, we can effectively draw a conclusion using classic, human judgment.

Ronald Fisher passed away before Hill published his criteria, so he never had a chance to render his last judgment.  But the field of epidemiology has.  Hill’s Criteria have effectively ended the debate over how to analyze cause, and have been used largely unchanged for the last fifty years.  Fisher’s contribution was not to prove Hill wrong, but to make Hill’s arguments stronger.  While Fisher’s skepticism did much damage to the public (some of whom might have stopped smoking sooner but for his efforts), the battle forced Hill to structure his thinking, to the benefit of all of science.

And while Hill’s Criteria are not commonly used outside epidemiology, they should be. The criteria take an impossibly large and complex pile of data and break them up into chunks. They make the evidence understandable.  And they make the case for causality transparent – each piece of evidence is categorized, and weighed in the context of the whole.  If evidence is challenged, it becomes clear just how devastating or inconsequential that challenge is.  We lose any presumption that somehow a single set of data could prove the entirety of scientific understanding to be in error.

So from here, we go off from the history of cigarettes and heath, and drive to the weeds of global warming.[1]  What happens when we apply Hill’s criteria to the question:

 Are humans, by adding CO2 to the air, causing the planet to warm?


Hill’s Criterion #1:  Strength.  How strong is the relationship between CO2 and temperature?

As the old saying goes, “correlation is not causation, but it’s a damn good place to start.”  All other things being equal, a strong correlation is more likely to hold up as causal.  The correlation between temperature and carbon dioxide concentration over the last thousand years looks something like this.

co2-vs-tempImage credit: Wikimedia

This does not look like a coincidence.

But knowing that there is a strong correlation is not enough.  We do not know if carbon dioxide causes temperature to rise; temperature causes carbon dioxide to rise; or some third, independent factor is causing both to rise.  Many, many scientific papers outside climate science offer up a correlation as if it were meaning.  Many, many scientific papers have been wrong as a result.

To get more insight into this, we need to look deeper.

Criterion #2: Consistency.  Is the data consistent across multiple measurements, at multiple places and times?

I harped on consistency a lot in my last blog post.  Science should never rely on a single type of measurement, because single measurements can have unexpected flaws.  Multiple strands of data are needed to confirm a hypothesis.

When looked at through that lens, how does the above graph hold up?

Thermometer records have only existed since the 1850s, and were only recently distributed throughout the globe.  As a result, scientists have had to get creative to reconstruct the temperature record, developing proxies such as grape harvest times in Europe, or the compositions of sea shells in the ocean.  A 2012 paper collated 173 different measurements, and their average accurately tracked thermometer measurements over the past century, yet extend backwards even further.

An example that confirms that the globe is warming is a measure of the growing season in the US, which has on average extended by about two weeks over the last century.  It doesn’t match temperature exactly – there is more to farming than temperature – yet has a recent rise that looks familiar.

growing-season-vs-timeImage credit: epa.gov

Again, this data simply confirms that we are not kidding ourselves that the climate is changing, and that this change correlates with CO2.  It still does not say why.

Criterion #3: Specificity.  Is the change that we are seeing specific to this point in history?

Claiming that humans are causing climate change by burning fossil fuels makes a very specific kind of prediction: You should see nothing like this change at any other point in the Earth’s history.

The climate has varied continually throughout the Earth’s geological past through simple patterns such as periodic changes in the Earth’s distance from the sun.  By drilling miles deep through the Antarctic frost and measuring the nuclear composition of the ice, scientists can infer the average temperatures during the time the ice was deposited.  By measuring air trapped inside the ice, they can also infer carbon dioxide concentrations.  The resulting graph looks like this:

400000 year record3.pngImage credit:  Climate Outcome

You can see a natural cycle of ice ages due to variations in the planet’s orbit.  In fact if you believe the graph, sometimes CO2 spiked before the warming, and sometimes the warming started before CO2.  In fact, both may be true: There is a feedback loop, and a warming climate releases CO2 from the oceans; the increased CO2 in the air in turn warms the climate more.  As one climatologists joked, arguing that one of the two has primacy is like arguing that chickens can’t create eggs, because we have proven conclusively that eggs create chickens.

But that blue line at the end sure is interesting.

While there is no question that climate varies naturally, that giant spike of CO2 stands out. It does look specific to today.

There is also one other useful piece of information you can glean from this graph.
That last increase in the temperature (the red line) is not human-caused warming –  it’s the end of the last ice age, the one that allowed humans to cross the Bering Strait from Russia to North America.  The x-axis of the graph stretches 400,000  years, so each of these vertical-looking jumps in temperature represents a change that occurs over about 10,000 years, at rates of about 1°C every 1,000 years.

How does that compare with today?  Since global temperature records started being kept in 1880, temperatures have risen just under 1°C, with most of that increase happening since 1970.

temperature-rise-rateImage credit: Nasa.gov

That’s at least 10X faster than a typical ice age warming.

Whatever is going on, it seems eerily specific to the time the world was industrializing.  We have identified a cause that is unprecedented in the Earth’s history,[2] and we see a result that is similarly unprecedented.

It’s a highly suggestive relationship.

Criterion #4: Temporality.  Which came first in modern times, the CO2 or the warming?

This one is should be pretty easy.  The hypothesis that human-created CO2 causes climate change yield a simple prediction: emissions should come before warming.  But the data looks like this:

Image source: Tableau public data posted by Jonathan Wilkendorf of UT Austin, reformatted by me in Tableau.  CO2 emissions are plotted in millions of tons/year, so current emissions are around 30 billions tons/year.  Temperature is plotted relative to pre-industrialization average.

What gives?

Before you panic, let’s talk a bit about the purpose of investigating temporality, and then on why this graph looks a certain way.

We already know from other data that industrialization caused our rise in CO2, independent of the Earth’s climate.  That is accounting, not science.  Given that pre-existing knowledge, we don’t have to worry about getting causality backwards.  But that doesn’t eliminate the possibility that some unexpected coincidence is causing temperature to rise with CO2, so it’s still helpful to know if one came first.

In this particular case, the simple chart is inconclusive.  There was a move to higher temperatures just as industrialization began, and that temperature rise preceded emissions.  However, once the burning of fossil fuels for transportation and energy really took off after WWII, the effects of humans became more pronounced. From then, emissions precede warming.

Our natural climate is not perfectly stable on its own, so it’s not surprising that when we were emitting very little CO2, there was a natural uptick or downtick that swamped out the effects of man-made change. Climate scientists point out that once emissions started their exponential climb in the 1950s, emissions clearly precede warming.

Skeptics, meanwhile, point to this graph as something that should make the whole edifice of climate science crumble.

But as I try to highlight in this blog and everywhere else, no single graph will make or break a theory.  We use Hill’s checklist to enforce discipline – to make sure we are looking at the problem from all directions, and to highlight places where we should look harder. If a single bullet point doesn’t deliver unequivocal support, that’s ok.  Reality is sometimes complicated.  As long as it doesn’t unequivocally contradict, the hypothesis should survive.

This bullet point flags a place where we need to look harder.  To understand a complex system, you need to build more complex models, and I’ll come back to this again below. But meanwhile, this discussion lead directly to the next important criterion:

Criterion #5: Dose-response.  Does the temperature increase scale with CO2 increase?

Smoke more cigarettes, and you are more likely to get lung cancer.  This simple relationship – an increased dose yields an increased response  – is a hallmark of a causal connection.  It’s easy to imagine one experiment going awry.  It’s much harder to imagine a series of experiments going awry in a well defined, orderly process.

The link between CO2 and temperature has feedback loops  – an increase in temperature will raise atmospheric CO2 levels as the gas moves from ocean to air.  So the historical correlations that I have shown above aren’t really relevant here – we know the climate is not so simple.  To deconvolute the two, we have to look at data taken in modern times, when we know the CO2 rise has been driven by the burning of fossil fuels.

Now, the modern temperature increase does correlate with CO2, but it’s just one data set.  One data set is suggestive, not convincing.

But scientists are nothing if not resourceful.  Below is a measurement in the amount of infrared energy reflected (more technically, absorbed and re-emitted) back from the atmosphere to earth — the “Greenhouse Effect”.  This measurement isolates wavelengths where CO2 is the sole contributor of the reflection.  And lo, the amount of reflected energy not only tracks the long-term trend in CO2, but it also tracks the seasons, as atmospheric CO2 decreases in the spring as plants grow, and increases in the fall when they go dormant.  It clearly shows a dose-response relationship.

co2-forcingImage from “Observational determination of surface radiative forcing by CO2 from 2000 to 2010”, doi:10.1038/nature14240

Pretty damn impressive.  That’s an awfully complex relationship to be a coincidence.

Very similar measurements (with less pretty graphs) have been made for outbound radiation as well – we can measure the amount of energy radiated from the Earth using satellites, and find that it has gone down since the 1970s, when the first satellite measurements were made.

Adding CO2 leads to more energy staying on the planet.  And that retained energy manifests as heat.

Criterion #6: Plausibility.  Does the causal relationship make physical sense?

The idea that the Earth’s atmosphere functions as a sort of insulating blanket was first proposed by French mathematician Joseph Fourier in the 1820s, while he was developing a formal theory of heat flow (one that is still taught to engineering students today).  He calculated that, given the Earth’s distance from the sun, the planet should be colder than it actually is.  To solve this dilemma, Fourier postulated that the atmosphere traps heat just as a glass wall of a greenhouse does.

In 1859, British physicist Joseph Tindall teased out the degree to which each atmospheric gas should contribute to warming, calculating that CO2 indeed participated.  And in 1897, the Swedish chemist Svante Arrhenius published a rough calculation that doubling the amount of CO2 in the Earth would increase its temperature by 5-6°C.

Back in the 1890s, human emissions were so small that it would have taken several hundred years to reach this threshold, so his calculation was seen more as a parlor trick than call to action.

But then came industrialization.

In 1938, the engineer and amateur meteorologist Guy Callendar dug into CO2 records from the 1800s to the (then) present day, and found that atmospheric CO2 had increased by 10%.  This was much faster than Arrhenius had anticipated, because industrialization consumed increasing amounts of fossil fuels each year.  Based on these measurements, Callendar estimated that warming from humans was already under way.

first-plot-of-warmingCallendar’s 1938 plot of temperature and CO2, including an estimation of the contribution of CO2 to temperature rise. Image credit: Climate Lab Book

Of course, none of these scientists would be remembered save that their early guesses, based on insufficient data and absurdly immature models of the climate, turned out to be roughly in line with modern assessments.  Callendar was in fact wrong in his estimate that warming in the 1930s was being cause by CO2; modern models find that warming to be primarily from natural causes.  History remembers both the good and the lucky.

But if we are trying to assess whether human-induced warming is plausible, then the answer is clearly yes.  It has been for generations.

Criterion #7:  Coherence.  Do the data fit in with current theory and knowledge?

This is where the much-discussed scientific consensus comes in: Of climate science papers that take a position on the issue, 97% support the concept of anthropogenic (human-induced) global warming.

And to be clear, this is not a consensus generated by a monolithic group of nerds.  It includes over 10,000 scientists from an astonishing range of sub-disciplines, from computer modeling to atmospheric spectroscopy to paleobiology.  They hail from seventy-four different countries.  They represent people who get their funding from different sources, publish in different journals, fraternize with different cliques, and generally have nothing in common with each other besides their desire to understand the climate.

Even when you break down the published literature by subfield, over 97% of the literature of each subfield supports the hypothesis of man-made climate.

To wit: A search of the literature of over 12,000 papers containing the phrases “global climate change” or “global warming” shows that only 77 papers reject the hypothesis of human-induced climate change.

That’s a coherent story.

consensusAGW stands for Anthropogenic Global Warming.  Image from skepticalscience.com

Criterion #8:  Experiment.  Can we alter, prevent, or improve the situation with an intervention?

The metaphor with epidemiology breaks down slightly here, since it is not possible to take a handful of Earth-like planets and test an intervention on them.  We have one Earth, one climate, and one fate.  We cannot – yet -intervene to engineer a new, calmer climate.

However, nature has provided us with a recent, natural experiment in climate engineering: The eruption of Mount Pinatubo in 1991.  The largest eruption of the last century, the volcano shot rocks over 40 kilometers high, and spewed out about 17 megatons of sulfur dioxide, which rapidly reacted with water in the atmosphere to form aerosols of sulfuric acid.  Within weeks, the aerosol plume had spread over the globe, and within a year formed a uniform layer around the atmosphere.

Aerosols have the opposite effect of carbon dioxide – instead of heating the planet by trapping radiation, they reflect the radiation away, cooling temperatures.
Seventeen megatons of material may sound like a lot, but it represents about the volume of Boeing’s Everett Factory, where they assembly 787s.  A small amount of material, dispersed evenly through the atmosphere, can have global impact.

mount-pinatuboImage credit: Vermont State Colleges

Fortunately for us, aerosols disperse quickly, allowing climate to revert to normal in just a few years.  With carbon dioxide, by contrast, normalcy will not return for centuries, or even millions of years if species that capture carbon dioxide are rendered extinct by the changes.  That last scary bit has happened before.

Criterion #9: Analogy.  Is there an analogous, better-understood system that makes the CO2 climate hypothesis plausible?

The idea behind this criterion is that an explanation is more likely to be valid if there is another system that behaves similarly.

The most accessible example is a greenhouse.  Another example might be that Venus is hotter than Mercury, even though Mercury is closer to the sun.

Pick your favorite.

Where did the climate models go?

It’s somewhat disconcerting to have a conversation about global warming that doesn’t involve climate models – they are what gets most of the press.

Yet we have seen above that you don’t need models to make an effective case that humans are causing climate change.

Models are nonetheless profoundly important, both for understanding our past and for predicting our future. So let’s return to this question we asked above in Temporality:  How do we know that the rise in temperature at the end of the 19th century was natural, while subsequent rises are man-made?

To address this, scientists run “experiments” on computers, adjusting their models to consider only natural changes (with no human contribution) to see if they can get a nature-only model to match the data.  They fail.  When scientists consider only human additions to the climate, with no natural forcings, those models fail too.  Only when they combine human and natural contributions to climate do the models fit the data.

Image credit: IPCC

The models – and there are a lot of them – all vary slightly from each other.  The average model may overestimate warming by a few degrees, or underestimate it.

But together, the models are unequivocal in their predictions.  The climate will continue to warm.  Much faster than it ever has before.  With enormous consequences.

Why I like Hill’s criteria.  A magical thing about structure is that it gives you no place to hide.

When placed in Hill’s criteria, the strong points and weak points of the argument leap out.  You know exactly what data you’d like to gather, if you had the chance. (Go find another planet to test our hypothesis on, for one).  If there are holes in the plot of your story, the truth is laid bare for all to see.

If you believe alternatives to human-caused global warming, test them in this structure.  See if the story holds.

The fact that we rely on stories to judge the legitimacy of an idea may strike some as lacking scientific rigor.  So be it.  I would love to put a probability behind the declaration that humans are causing climate change.  But the world is too complex for us to reduce inquiry to a single number.  Part of being a good scientist is to understand your limits.

All scientific work is incomplete, and at risk of being toppled by tomorrow’s discoveries.  That does not give us leave from acting today.

The evidence supporting man-made global warming creates the one of strongest science stories I have ever seen.

And this process is how I know.


If you liked this blog post, please like my page at perspicacity.xyz and get the next blog delivered right to your inbox!


1  I want to cite skepticalscience.com as an amazing repository of explanations of our current understanding of climate.  If you have a specific question, I highly recommending searching through their archives for the answer.  And if you are a professional climate scientist and want to refine any of the explanations I’ve dug out in this research, please leave a comment!
2  There are several unusual, natural changes that have been competitive.  At the end of the last ice age, Greenland warmed 10°C in a decade when ocean currents changed. But that was a local effect; Antarctica did not warm at all.  Since we are talking about global change, that’s a poor comparison (although I have seen it used by denialists).
The global event that looks closest to man-made warming was the Permian extinction, when supervolcanos spewed huge amounts of CO2 and methane into the air, and caught Siberian coal reserves on fire.  CO2 rose by perhaps 10X over just a few thousand years.  The CO2 dissolved in the oceans, and the resulting acidity dissolved calcium carbonate, killing all shellfish.  The planet heated so such an extreme that the tropics became too hot for complex life to survive.  90% of the world’s species went extinct.
But there is no supervolcano going off today. Just us.