Global Warming
Shortly after Al Gore's film and book An Inconvenient Truth came out in 2006 into Vaho's group happened to be having our annual get together with all of the team from across the several Boeing sites in the country. We were walking back to our hotel after dinner on the sidewalks just across the street from the ocean beaches. I happened to mention to some of the group as we walked along that if Gore is right, then where we were walking was going to be under water eventually. On the spot, Vaho gave me an assignment to dig into it and find out what, if anything, it meant for Boeing IT. So, for the next six months, my primary assignment was researching global warming. My co-researcher for this project was Conrad Kimball.
One of the first things I discovered, was that this was not something new. Scientists had started talking about it in the 1890s, and for obvious reasons. In the 19th and early 20th centuries many towns and cities in the northern hemisphere were suffering from fog and soot, especially in the winter months. Much of this was caused by the burning of coal for heat. These blankets of fog were well known to just about everyone who lived at the time. At times, they would become a serious health hazard and kill a lot of people. One such fog in 1952, actually caused the end of Winston Churchill's political career when he failed to treat it seriously, despite the fact that it was killing a lot of people. When I was in college in Kalamazoo in the early 1970's, the university and much of the town still heated with coal. Within a few hours after a snowfall, the surface of the snow would be grey from the coal dust settling out of the air.
That humans burning coal and other things had increased the amount of carbon or what we now call greenhouse gasses in the atmosphere was well known. That it would probably lead to a warming in of the planet was assumed, although there was no data to support that notion, nor to suggest how fast it might happen, or what its impacts would be. The first reputable scientific paper on the topic was written by the third recipient of the Nobel Prize in Chemistry, Svante Arrhenius. In his paper, he speculated that the warming might actually be a good thing, and lead to more abundant crop yields. He called it the "hot house effect." That was in 1896.
There was another paper published in 1925 that purported to provide some data on the topic, but the methods used to collect that data were deemed to be badly flawed. But an interesting thing happened during the Cold War between the Soviet Union and the western democracies. The political leadership of the 1950's got into a bragging contest about what their societies were accomplishing. Scientists on both sides of the iron curtain took advantage of this and goaded their leaders into a year of competition to see which side could do the most science in one year. The primary subject would be the earth, and all we could learn about it. So half of 1957 and all of 1958 were declared by the politicians to be International Geophysical Year (IGY) and funding for project proposals that had been around for decades suddenly fell like manna from heaven.
Roger Revelle, the director of Scripps Institution of Oceanography was one of the scientists involved in the goading campaign that led to IGY. He managed to snare some of the funding for Scripps to develop the instruments necessary to study the chemical interaction between the atmosphere and the oceans. It is important to note that his goal was merely to understand this aspect of atmospheric and oceanic chemistry. He was not looking for data on the greenhouse effect. That wasn't part of the proposal.
It took the whole of IGY just to develop those instruments and setup research stations for their use on Mauna Loa in Hawaii, and at the "Little America" research station in Antarctica. The bulk of the work was done by Charles Keeling. Revelle was able to secure enough funding from other sources after IGY was over, to keep the station at Mauna Loa going, so Keeling was able to start collecting data. By 1962 he had enough to sound an alarm. Atmospheric CO2 levels were rising. Global warming was real.
As Conrad and I read about this history and poured through the early reports of the Intergovernmental Panel on Climate Change (IPCC) it became clear that the panel's work was enmeshed in a political filtration process. Each panel member had to get permission from their home government's sponsoring agency, which in most cases was also their employer, to sign off on the reports. So, there was some watering down and wordsmithing going on. The scientists on the panel were faced with trying to report what they were discovering in a way that would get through this censorship process. Conrad and I had to be very careful in our reading of their reports. We found the equivalent of the sounding of an alarm in the Second Assessment, which had been published in 1995.
The key discussions dealt with the non-linear nature of climate systems, and the difficulty, if not impossibility this creates for any attempt to accurately project such things as ice melt rates or sea level rise. The following statement is in section 2.12: "This implies that future climate changes may also involve "surprises." In particular, these arise from the non-linear nature of the climate system. When rapidly forced, non-linear systems are especially subject to unexpected behavior." In other words, accurate projections are impossible.
One of the sad things about the state of the human condition is that only a relatively small percentage of the general population is comfortable with the way scientific information is presented. And when it comes to the math parts, the situation is much worse. Humans, in general, are just awful when it comes to understanding the math of probability and anything involving large numbers. John Allen Paulos wrote about this eloquently in his famous book Innumeracy.
So let me explain what we are talking about when we say something exhibits either linear or nonlinear behavior. We'll touch on three ideas, not two.
Scientist collect data. It is comment to present data as points on a graph with an x and a y axis, showing two variables. For example, one could plot chart a person's height versus their age from birth to when they stop growing taller, typically sometime around 18. Here is an example from the Britannica online edition.
If it is meaningful, a trendline will be plotted through data presented in a chart. Here is an example of a stock market trendline from Investopedia.com. Anytime something is highly variable over time or some other variable, understanding the trend can be more important that what is happening over a short range of data. This is often the case when trying to build a long range plan or make wise public policy decisions.
Anytime a trendline that is meaningful can be generated, we say the system being described behaves in a linear manner. It has a generally describable trend, which over a sufficiently large sample size can be usefully described by a nice straight trendline. That said, there is that variability or noise level in all data, so it is useful to also understand how much flexibility is required when trying to estimate where things will be after so many cycles or data collection points that are not yet available.
There are two basic formulas for R-Squared. One is super simple, but limited to the data available. A better adjusted version will tell you whether or not you really should get more data before putting to much stock in either your trendline or the R-Squared value you have for it. In both cases, a value of 1.0 is deemed to be perfect. And as that value goes toward zero, you've got problems, or basically that the trendline is a complete work of fiction along the lines of Ray Bradbury's The Martian Chronicles.
If a dataset is so volatile as to render a trendline meaningless (i.e. its R-Squared value is approaching zero, we say that the attributes which generated the data are nonlinear. Predictions are just wild guesses, and one might as well be consulting the local palm reader or a deck of Tarot cards. Many aspects of climate change behave in a nonlinear manner.
Ok, now that third definition. Every once in a while, a scientist will be asked for something absurd, such as a prediction of sea level rise due to global warming over the next century. They may respond with some charts if they have been forced to make some, and then apply a footnote stating that all nonlinear elements have been ignored. That's what it says in the the IPCC reports about the trend predictions they were forced to make. When asked why they bothered, the typical response is that they were forced to do so, and then they roll their eyes and say that the person doing the asking was nonlinear. That's not a compliment.
This is an ancient problem for scientists. Both Socrates and Galileo had to endure nonlinear people in power.
I did not show the formulas for R-Squared here. The reason is that as I was looking at them and thinking about whether or not to include them, I happened to recall a remark often repeated by Woodie Guthrie, but inconsistently. Sometimes he said two and sometimes he said one. The version with three makes a bit more sense musically. It went like this:
"Anyone who used more than three chords is just showing off"
So I decided to leave out the formulas.
Now all of that said, the state of things when it comes to math literacy, or innumeracy as John Allen Paulos calls it, things are just terrible. To give one a feel for just how awful humans are at basic math concepts involving large numbers and probability, let's talk about print newspapers. Who buys them? It's a rapidly declining percentage of the public, and the demographics of newspaper readers is interesting in that it is literate and reasonably well educated. But the pain newspapers are feeling is real. One impact is that no newspaper publisher knowingly purchases content for inclusion their papers unless it helps sell the paper. One bit of content they all carry is some version of a daily horoscope. The horoscope is based on the proposition that somehow the position of the planets and moon relative to the backdrop of the stars other than the sun somehow has an impact on the course of life and events on earth. To say that such an impact cannot be measured is giving it too much credit. The numbers are so large as to be absurd. It's right up there with the nonsense of homeopathic medicine's dilution proposition, in which it is asserted that a preparation that is diluted to the point that it has nothing of the original ingredient, will none-the-less retain the curative properties of that ingredient. I would like to pay for my groceries like that. Give the clerk the money, take it back, and pretend they have been paid. And yet otherwise literate and educated people believe such nonsense.
Conrad and I finished our research and shared what we had found. Several Boeing sites, including Plant II, Renton, the "beaches" locations south of LA, and the new site in South Carolina would eventually be under water. When, we could not say. It could be in as little as a century, or it could be a thousand years away. All we could recommend was to establish some indicators for what decision triggering events would look like, and then develop some long range plans to have available should one or more of those indicator events occur. That was a pretty bland recommendation. Our more salient observation was that almost certainly, the future of energy supplies was going to change in some unknown ways, and since computing is super power hungry, that might prove to be a more serious problem for folks working in IT. Again, a pretty bland observation. And yet, one of the Associate Technical Fellows started arguing with all of it spouting all sorts of pseudo science nonsense.
Scientific illiteracy is seemingly nearly as common as innumeracy.