The global climate change debate is currently wide ranging, will be historic, and may lead to costly solutions. Participants include school children, scientists, journalists, and politicians. Only a few are well informed, and even some scientists in the relevant fields have forgotten what it means to be a scientist. I wrote this essay to provide background information about how science works, introduce the two leading theories behind climate changes, and offer my thoughts about the breakdown of scientific discourse.

I am not a scientist, I'm more useful as a software engineer. My brother and sister are scientists, and I've subscribed to Science News, a wonderful magazine, continuously since 1969 and it keeps me pretty current.

One of the great achievements of mankind is the development of the scientific method. I'll start with a look at science and the scientific method, but the focus will be on four types of data: anecdotal, empirical, statistical, and theoretical. All have their place and quality.

Then we'll look at the two leading climate change theories and how the informal rules of scientific discourse have broken down. Indeed, the arguments have gotten so vitriolic that I all but gave up paying attention to science and tried to avoid the politics whenever I could.

In just the last blink of an eye, weather and solar patterns have changed and open up an avenue to repair the discourse, at least among some scientists, and even better, may allow us to gauge the relative impacts of two competing theories.

Science and the Scientific Method

In Aristotle's time, science was approached more as thought experiments, in part because key physical concepts were not recognized. Aristotle knew that heavy objects fell faster than lighter objects, it took a millennium before Galileo proved otherwise with the experimentation, measuring, and reproducing that are the hallmarks of modern science. To that base, many refinements have been made, for example, the best drug testing on humans involves a "double blind" test. Neither the test subjects nor the people administering the drugs know if the recipient is getting the drug being tested, what strength, or whether it is a placebo.

Central to any scientific study is data, and I see five kinds:

1. Anecdotal evidence
   This is the weakest, and the weakest of the weak is something like "Joe says that if he had been wearing a seat belt, he would have hit the rock instead of being thrown into a snowdrift." Better are stories of "cancer clusters" where several people in a neighborhood are diagnosed with cancer in the course of a couple years. Sometimes a detailed look finds a reason, sometimes it doesn't. Given the millions of neighborhoods in the country, some will have cancer clusters simply due to random chance. Some of the best anecdotal evidence comes from truisms passed down across generations, e.g. "Red sky in the morning, sailor's take warning, red sky at night, sailor's delight." Given our relatively new understanding of storm systems and weather fronts, this should work and often does.

2. Empirical evidence
   This is evidence that is collected through observations. By itself it means little and will not explain how the world works. For example, I collect daily snowfall and snow depth data from home and from a few like-minded individuals elsewhere in New England. Without people using that data for something else, it's just a pile of numbers and usually not very interesting.

3. Statistical evidence
   One of the best things to do with empirical evidence is to manipulate it. Compare snow depths to other data separated by time or location, sum it up into a single number that measures how "snowbound" a location is each year. Bend, fold, and spindle it, look for relationships that haven't made it into old wives' tales yet. While statistical results won't tell you how something works, they can often let you predict future events very accurately. By the way, the season 2007/2008 in central New Hampshire smashed pretty much all the records that could be broken. Even NWS snowfall records stretching back to the snowy decade of the 1880s were broken.

4. Theoretical evidence
   This is the first place you need someone who understands the meaning behind the data. Empricists and statisticians may do their job better if they understand some of the science, but they may also be blinded by the current understanding and miss something odd or ignore it as measurement error.

   The high energy nuclear physics institutes have developed into two groups. Theoreticians figure out how various particles should interact, and empiricists build the experiments that will prove or disprove the theory. Once in a while an experiment will come up with something unexpected and the theoreticians have to play catch up.

5. Model result evidence
   In some fields of science you can take the theory and readily apply the math behind it to solve relevant questions. For example, a car moving at at 20 meters per second will travel 100 meters in 5 seconds. Often secondary effects can be ignored, e.g. a ball thrown with a horizontal speed of 20 meters per second will be affected by wind resistance, but perhaps not significantly so. Some apparently simple problems, e.g. determining the position of a satellite in an elliptical orbit, require a multi-step convergence procedure because a key equation involved can't be integrated into one needed to solve the question directly. There is no chance that systems as complex as the Earth's climate can be followed with a single application of the fundamental equations, so projections must be computed by iterating over small time periods. This is fine, it's how everything from buildings to airplane engines are designed.

   There are two requirements for a model's output to be accurate. First, the data that goes into the model must be good. The classic sound bite is "Garbage In, Garbage Out." Second, the physics in the model must be accurate. If the physics is well understood, then there's a good chance the model will work. When it's not, models can be tuned by feeding them historical data and adjusting parameters until you get results that match what really happened. The parameters may be backed by theory or statistics, but more theory provides more confidence.

   One controversial parameter in global climate models is CO2 forcing. While we might accurately anticipate the effect of additional CO2 in cloudfree atmosphere before any warming takes place, warming leads to increased evaporation, and water vapor is a greenhouse gas. Therefore, models try to account for this by giving CO2 a bigger feedback than what one would expect from just CO2. If you model the Earth over the length of the satellite temperature record, you can adjust the feedback value until the model's output matches recent temperatures. This may be fine if CO2 is the only thing that affects the results. If there are other things, e.g. ocean currents, changes in clouds, or other external forcings like solar changes, they must be identified and included. If the authors lose track of the uncertainties, the result may be "Gospel Out."

Hypothesis vs. Theory

What is a theory, anyway? Basically it's the end result of collecting and studying the four forms of data. Anecdotal evidence, e.g each bounce of a ball doesn't go as high as the previous bounce, suggests that there is something interesting going on. Making measurements of each bounce is likely to disclose a pattern that the bounce heights follow an exponential decline. That in turn suggests that on each bounce a fixed percentage of potential or kinetic energy is lost. A good guess would be that the deformation of the ball on each bounce results in energy being lost to noise, heat, etc. Now we have a hypothesis which is tested against balls made of other materials, we can look for speed changes on either side of a bounce and during the time between bounces. We could test against temperature, time of day, lighting, or the presense of small furry mammals. Ultimately we might predict that no material will result in a ball that bounces higher unless it releases stored energy. If that is held up by other researchers who replicate the work and make their own extensions, the hypothesis morphs into a theory. The key things a theory has over a hypothesis are it explains how the hypothesis works in better detail and makes predictions that can be tested or studied.

Generally, a theory gets stronger as its predictions are verified. Eventually it can be elevated to a law, e.g. Newton's three laws of motion. When a field is new, theories come and go as scientists gradually learn what's important. Sometimes even laws can be disproved. For example, the discovery that light always moves at a fixed speed in a vacuum implies objects can't move faster than the speed of light because all objects except those at absolute zero emit photons. Einstein's Special Theory of Relativity extended Newton's laws to very fast objects. Starting with the observation that objects gain mass as they approach the speed of light, Einstein realized that a little algebra would show E = mc^2. Who would ever have guessed that experiments to measure the speed of light would yield such a fundamental relationship? Such times are rare, and scientists involved in them are truly lucky.

Sometimes a new theory receives more than skepticism, it can receive rejection or even ridicule. This happened with Alfred Wegener's theory of Continental Drift he developed to explain the existance of very similar fossils from both sides of the Atlantic and the apparent fit of South America's coastline with Africa's. It wasn't until ocean floor core samples provided overwhelming evidence that the sea floor has spread and is still spreading before he was proven right and the improved theory of Plate Tectonics is now unquestioned truth and explains much of the volcano and earthquake activity throughout the world.

A lot of evidence takes the form of numbers. Some data are wonderfully clean and predictive. The CO2 measurements from Mauna Loa are one of the best examples anywhere. Other data is so noisy that even teasing out a signal may leave you with not much more than a trend. Climate has a lot of that. The global average temperature needs many years for trends and cycles to become evident, and local data is far worse. My ten years of snow data for my home has essentially no predictive value by itself or any reasonable statistical manipulation.

How much data should we gather before looking for insights? The Colorado State Klotzach/Gray hurricane forecasts use 1950-2000, some satellite-based data only exists for recent decades. Having data that covers at least a couple cycles of relevant variables is nice, having at least one cycle is important. Some long-lived natural cycles include:

• 11 year solar sunspot cycle
  When the sun is active, this is about 10.3 years long, when the sun is inactive, it's about 12.3. years long. Only rarely is the cycle about 11 years!

• 50 year Pacific Decadal Oscillation (PDO)
  This has only been recognized since 1997, but an historical record has been assembled from sea surface temperatures and air pressures.

• 45 year Atlantic Hurricane Active Multi-Decadal Signal
  A very active period of hurricanes between 1950-1969 was followed by and inactive period from 1970-1994, and then the current active period.

For climate research, various periods are used. Directly measured temperature data covers 100 years or so over a large portion of the world. Ice core data and other proxies can provide thousands of years of lower quality data. Satellite data reaches back no more than 50 years and often quite a bit less. Often 30 years is used, which is unfortunate because 30 years ago was the end of a period of cooling. That means the recent record likely includes both natural change and man-made change.

A good scientist is a skeptical creature. He is skeptical about everything from the quality of data to the claims behind a new theory. He is skeptical of old science, new science and even his own science. While a good scientist will humbly accept proof that his analysis is wrong, he will check and recheck his work before publishing. Having to retract work means that other scientists have wasted time working with the bad results and that progress in learning how the world works has been delayed. Sometimes the data is good and other scientists can combine it with other data to discover the conclusions are wrong but a new interpretation works with both sets of data.

Climate Theories

The Mauna Loa CO2 record was not the first data showing increasing CO2, but was and remains the best. The first reports presenting it came out in the late 1970s, soon after a period of cooling that included significant weather events and spawned several papers and books about the coming ice age and how to stop it. (My favorite was spreading soot on the arctic snowpack.) The Mauna Loa data, and the return of milder winters, helped change climatologists' attention back to global warming. There weren't many options, solar output had long been considered constant and even referred to as the "Solar Constant." While there were intriguing relationships between the Maunder Minimum (a period of very little sunspot activity) and the Little Ice Age, the sun seemed to be behaving itself and attempts at finding variations in solar output were inconclusive due to weather induced noise in the data. It was clear that any effect had to be too small to readily explain the warming that was observed. Later on, satellite measurements of the sun's output finally measured changes through an eleven year sunspot cycle, confirming that the variance was too small.

So CO2 was the obvious culprit and pretty much held sway.

Along the way, scientists discovered the ozone hole and that turned into a political and popular movement that eventually resulted in banning chloroflorocarbons (CFCs, including various types for Freon), the first case of science finding a problem and triggering an international effort to fix it.

Eventually as data showing the Earth's warming became more and more clear, the general public and politicians got involved, much as happened with CFCs. The biggest differences are that banning CO2 is basically impossible, so all that can be done is to limit it or sequester it, or develop and embrace alternatives. The Kyoto Treaty was the first major step in that direction, it also implied an endorsement of the greenhouse gas theory. Along the way research funding increased to study warming, other greenhouse gases, and what we might do about them.

There are some problems with the theory. CO2 blocks a small "window" of wavelengths, and current levels block nearly all light in that window. Additional CO2 acts by blocking light at the edge of the window where the atmosphere isn't quite opaque. That requires a lot of additional CO2 for declining impact. No problem - there are other greenhouse gases to study, like methane (CH4), and CFCs. CFCs were under control, but CH4 was increasing, though lately it has nearly leveled off. Water vapor is also a greenhouse gas, but hard to study. It varies quickly and widely and ultimately so while it's clear that CO2 is increasing, scientific skepticism reminds us that we don't understand all the details. Still, the graphs that show CO2 and temperature anomalies increasing helped start and maintain CO2 as the leading theory.

Other scientists felt that there had to be more to the warming seen and that people were putting too much faith in the CO2 theory. In the geologic record, our current conditions are atypical and occupy just the barest amount of the Earth's history, even after the Earth formed, cooled, developed life, and then photosynthesis that produced our oxygen rich atmosphere. There have been times when the atmosphere had 10 times the CO2 it does today, but conditions weren't extremely hot. The best candidate for controlling recent climate may not have been CO2, but the Sun. Exactly how was not at all clear. Besides the Maunder Minimum and the Little Ice Age, other events have suggested a linkage. The Dalton Minimum was a briefer period that included cold weather, especially after Mt. Tambora exploded in 1815 and brought the Year Without a Summer in 1816. Some temporary warm periods occurred, e.g. during the Medieval Warm period when the Vikings created fishing and farming settlements in Greenland that lasted centuries before the climate returned to average. That cold period in the 1970s was preceded by a less active solar cycle.

So, there is empirical evidence that supports solar influence, is there theoretical evidence? You don't really need theory to understand that hot stoves burn you, but claiming that sunspots are like hot stoves won't advance climatology very far.

An inconstant Solar Constant was the best first guess, but when satellite data finally provided solid results, it was clear that the variance wasn't great enough. Research began into looking for something that amplifies the effect. For example, UV radiation greatly increases with solar activity, and attempts have been made to determine the heating it causes in the upper atmosphere couples down to the lower atmosphere. The best candidate so far is not sunlight, but the Sun's magnetic field and solar wind. This hypothesis says that an inactive Sun produces weak magnetic fields, and that allows more cosmic rays to reach Earth. The cosmic rays hit the atmosphere, smash into nitrogen and oxygen molecules spraying ions and subatomic particles like muons down to the lower atmosphere. The muons ionize more molecules and the released electrons help clump sulfuric acid molecules together to act as "nucleation sites" where water vapor can condense and trigger cloud formation.

That sounds like a really long reach. However, cloud seeding is done by adding nucleation sites, and a device called a Wilson cloud chamber has long been used to demonstrate cosmic rays by letting radiation make condensation nuclei in saturated alcohol vapor. Visible streaks show the path of the ion trail. An experiment named SKY demonstrated the sulphuric acid link and showed that a single electron would trigger several condensation nuclei to form. An atomic physics site, CERN, is developing an experiment to further explore the hypothesis. If it holds up, it has some very attractive features. Clouds both trap heat below the clouds and reflect light from above. Ultimately, reflecting light means the Earth absorbs less heat and cools. We do not have good data about cloud cover before the satellite era, we may not even have good proxy data to reconstruct the cloud cover record, it would be really nice to correlate each step of the process. We do have good proxy data for cosmic ray activity, it affects the amount of a beryllium isotope that occurs in the ice core record.

However, remember our current focus on what to do about anthropogenic CO2 levels? We want to compare solar forcing against the CO2 released through burning fossil fuels and deforestation, things we can do something about. The real focus is not on warming since the start of the Industrial Revolution, but on the warming since the 1970s, when the world's population and energy use has resulted in enough CO2 to have a measurable impact. Umm. There's a problem here. Weren't the 1970s a time of cooling? Isn't this assuming that what we want to determine is fact? Shouldn't a good skeptical scientist looking to compare CO2 and solar forcing consider that the population was great enough and energy use high enough to impact climate then? Well, ignore that, we've been looking at the time since then so much and we didn't have good satellite measurements that we don't want to struggle with the uncertainties from back then.

Well, okay, but there's a bigger problem, because both the CO2/greenhouse theory and solar forcing theories call for warming. How do we decide which describes the truth, or better, how much does each theory contribute in case both apply? Well, science and better data can handle that, and has handled that, but something happened in climatology where normal scientific discourse has broken down. The CO2 "Proponents" call everyone with alternative theories or concerns about the data "Skeptics" and do so in a pejorative way. How can this be? All scientists are skeptics. All scientists are proponents of learning the truth about nature more so than promoting their theories. The schism in climatology is so extreme and the field so large that things look more like a fight between gangs of soccer hooligans instead of the refereed soccer match that is closer to scientific ideals.

It is so bad that each side comes up with lists of supporters on petitions to governments to act on their concerns or to impress the public with the wide support for their stand. It is so bad that scientific conferences are essentially segregated. The Proponents go to the UN Climate Change Conferences, a recent conference attended by the Skeptics was dismissed with a blog article What if you held a conference, and no (real) scientists came? and described by one of the attendees with It has been all about science, and science policy. It is so bad that much of what should be scientific discourse is missing, replaced with rancor, insults, and virtually everything short of aspersions on each other's upbringing and genetic heritage.

Cooling the Climate Implies Cooling the Debate

However, there is hope for a return to sensible discourse! There is enough hope that I've written this web page. Things may change quickly, quickly enough that I have set aside several duties to get this done because things are already changing. After the personality, political, and funding issues, the basic problem is that both the CO2 and solar theories call for warming. The change is that some people, not all, are forecasting that the Sun is entering a period of less activity, perhaps for one cycle, perhaps for two, perhaps longer. Those forecasts imply that the solar forcing theory is now calling for global cooling and (Hurrah!) we may see data that clearly favors one theory or the other. Or perhaps it will disclose that climatology is even more complex than we thought and everyone needs to go back to the drawing board.

Even better, there is evidence that says the next solar cycle will be weak. As mentioned above, the 11 year solar cycle is really shorter than that when the sun is active, and longer when the sun is inactive. The last cycle, actually the last solar minimum, was in May 1996. If the next cycle were active, the minimum should have been in mid 2006, if it is weak, expect the minimum in mid 2008. We're almost there, so let's assume we'll have a weak cycle. Perhaps coincidentally, the Pacific Decadal Oscillation (PDO), switched to a negative phase in Sepember 2006. It's a bit early to know if it will stay switched, but the trend over the last few years has been negative. A negative PDO correlates with more La Ninas instead of El Ninos, and that means a general cooling. The global temperature anomaly for the last several months has been declining to the long term average.

The recent months have been interesting, and if things keep up then not only will we definitely be in an inactive solar cycle, we may be at the start of a protracted period of global cooling. While that would be great for climate science, it could be a disaster for people and agriculture that have to live with the consequences. Turning crops into fuel was never a very good idea, hopefully the work on sensible systems will be deployed before the growing season gets too short.

Concluding Thoughts

It distresses me mightily that scientific discourse has been trashed by people who don't know better, and moreso by people who should know better. I guess that there are some aspects of human personality that just won't change. Until the last few years I've avoided the field, as the sniping and ad hominem attacks were too distracting and annoying. At one point I considered writing a web page listing the major players and which "team" they were on, but concluded I had better things to do. I'm still tempted to create a page of quotes to act as a reminder that people deserve better, but that feeling is passing, especially as some of the worst offenders may be finding crow on the menu. Perhaps I'll collect interesting predictions and note which verify and which fail. I might create a list of comments I receive about this essay, however, I won't be creating a blog. In the end, this Golden Era will be just one tiny snippet within an interglacial warm period and where I live will return to being under thousands of feet of ice. That will not be a bad thing, especially since I won't have to deal with the inconvenience, it will be a very natural thing to be studied and learned from, just like any other event on our home planet.

References and Additional Reading

Continuing the theme

Atlanta meteorologist Kirk Mellish wrote an essay on Global Climate Change, which is much more comprehensive than mine. He introduces many other hypotheses, phenomena, and today's dysfunctional dialog. It makes a very good followup.

Multi-subject reports

1975: Chilling Possibilities
The mid-1970s saw a flurry of articles about the cooling climate. Even then, science was split between continuing cooling, cooling due to particulates and contrails versus warming due to CO2 increase. Once the first paper on CO2 measurements at Mauna Loa came out, cooling was quickly forgotten and most people jumped on the warming bandwagon. As did the climate - around 1979 cooling became warming and things "snowballed."

Fire and Ice
This article looks at media reports on the imminent Ice Age (1895), the benefits of ongoing CO2-related warming (1938), how the cooling since 1940 ... will not soon be reversed (1975), "Our ability to live is what is at stake" (2006). One of its recommendations is "Don't stifle debate."

The myth of dangerous human-caused climate change
This is a paper by Robert Carter that expands on the You Tube video mentioned in the CO2 section below.

IPCC Fourth Assessment Report (AR4)
This is the 2007 report from the Intergovernmental Panel on Climate Change and is the fundamental document supporting the science behind global warming. Unfortunately it is also political document and biased to the point that changes requested by IPCC reviewers were often ignored and other changes were made after the review was complete. Despite the 2007 date, the IPCC may not have looked at new data since the 3rd report in 2001, the emphasis was on refining previous work.

Summary for Policymakers of the Report of the Nongovernmental International Panel on Climate Change
The NIPCC is a group of scientists brought together by their disagreements with the IPCC process and its reports. Critics have harsh words for them, and harsher words for Heartland Institute, but their report is quite good and is free of the political pressure within the IPCC.

Natural Phenomena

CO2 and other Greenhouse Gases

Climate Change - Is CO2 the cause? This is a You Tube video of a talk by Prof. Robert Carter. Some of Carter's commentary on the lack of understanding of the scientific method by members of the Australian Parliment is one reason the first half of this essay covers method and data.

Sunspot activity

Current Solar Cycle progression, updated early each month.

A long statistical look at sunspot cycles with an interesting tie-in with Jupiter.

Solar Cycle 24: Implications for the United States
David Archibald looks at solar activity from several angles and the potential effects on climate for the United States. After looking at CO2 data and why it has a small effect on current climate, Archibald takes a speculative turn and gets a little carried away about how wonderful life will be with more CO2.

Raw data of sunspot cycles, lengths, etc.

Influence of Solar Activity on State of Wheat Market in Medieval England
This furthers some research started by Royal Astronomer Sir William Hershel in 1801.

An (apparently busted) prediction that cycle #24 will be much larger than normal.

Solar mediated cosmic ray flux

An early look at "Cosmoclimatology"

A look at the design of CERN's CLOUD experiment to reproduce and extend earlier research showing the formation of cloud condensation nuclei.

Pacific Decadal Oscillation

PDO index over time, kept current.

Pacific_decadal_oscillation an okay Wikipedia entry.

A fairly technical paper.

Raw data listing months and the PDO index.

Global average temperature

University of Alabama at Huntsville has their Earth System Science Center. One of the studies is a satellite based Microwave Sounder Unit. I don't see images for temperature anomalies. (Raw data)

Remote Sensing Systems report satellite data. I don't see images for temperature anomalies, but they have others that are interesting. (Raw Data)

Met Office Hadley Centre for Climate Change in the UK has several plots, including one that is updated monthly. (Raw Data)

NASA Goddard Institute for Space Studies This site processes data from individual ground stations in the United States Historical Climatology Network and other sources. This data shows more warming than from satellite measurements. (Raw data)

Other Phenomena

2006 Kerry Emanuel vs. William Gray on hurricane cycles.

Alfred Wegener's Continental Drift theory.

Credits and acknowledgements

Some images above are from Global Warming Art, thank you Robert A. Rohde and the Creative Commons license.

The Leaning Tower of Pisa cartoon is from The Open University under the Creative Commons license.

I've asked for permission for the rest, I wanted to get this page out as soon as I could.

Contact Ric Werme or return to his home page.

Written 2008 March, last updated 2008 July 20.