A Convenient Supposition

Or, how to confuse correlation with causality, part 34,289,767.

Some friends and I went to see Al Gores recent movie "An Inconvenient Truth" a few weeks ago. It was really quite a good show, in marked contrast to his political voice through the failed campaigns.

Let me start my review, lest my yelping liberal wife slay me out of hand, by saying that I DO believe that there should be serious and growing concern regarding global warming. I also believe that there should be more movement in federal policy circles to address these issues sooner rather than later. I also think that Gore put together a very approachable and emotionally effective presentation on why we should all take the global stakes of failing to act on such an awesome threat to our ecosystem much more seriously.

I even saw a news article briefly appear on Yahoo entitled something to the effect of "Gore gets science right in global warming movie." Unfortunately though, the blogosphere dug in over the next few days to discover that the reporter who wrote the article only contacted 35 scientists, only 5 of whom had actually seen the movie, none of whom were climatologists.

To Gore's and the movie producers' credit, they did get me to go and do a little technical research regarding the data presented in the movie. In the process, I had a vaguely familiar feeling that I was able to trace back to my distaste when watching the last round of presidential debates. It was the discomfort that came from only hearing a convenient part of the story while the rest was left out to spin it towards one side or the other.

I found the global warming link at Wikipedia to be very useful, as regardless of your affiliation in this so very politicized issue, all of the recent experimental data was readily accessible through compilations of web links. I found a couple of links to be quite interesting. Probably the biggest "oooh----ahhh" moment of the movie was around Gore's dramatized revelation of the historical atmospheric CO2 concentration data. It looked like a sexed-up version of the following chart:

Now this all looks pretty serious. Gore really hammers the point that nobody really disagrees that the industrialization of the world is causing the rather precipitous CO2 rise. He plots the CO2 levels on a chart just below a similar temperature chart and notes the obviously correlation. He is careful to say that they are correlated and the relationship is complex, but grim looks and head shaking abound. I then goes on to project what might happen to our CO2 rates over the next 100 years and then make some really fantastic predictions about melting ice caps and deluged and disappearing coast lines. All very dramatic, but a little misrepresentative.

When I went to dig in on the actual climatological data, I found that interesting things become apparent when you actually plot the CO2 variations against the temperature on the same chart and time scale.



Note, in particular that time runs backwards, with the present along the y-axis, going back in time towards the right. A closer look at this representation certainly supports the correlation argument. But the causality proposal that our increasing CO2 generation will undoubtedly cause global ruin if unchecked takes a real hit. Note that for most of recorded history, the temperature changes PRECEEDED the CO2 changes. Most likely this variation follows what we see seasonally, that when the temperatures get hotter, flora abounds, and increases CO2 respiration under photosynthesis, and likewise, when the earth cooled through periods of glaciation, flora died, and so did the CO2 respiration that went away as a result.

Also note, that there is clearly some sort of larger limiting effect which causes the temperature to plummet rather drastically even though the CO2 has rising to quite high levels, and then the CO2 levels descend in turn. So there was clearly SOMETHING that was causing the temperature to vary which then affected vegetation growth. A little more research into insolation and solar forcing functions offers the following support.



Note here that the eccentricity, precession, and obliquity of the earth's orbit combine in their effect to change the amount and duty-cycle of sunlight reaching the earth. And when you combine all of these into an effective "solar forcing function" it becomes pretty clear that we have found the zeroth-order temperature regulation mechanism for the Earth's surface which seems to override even the CO2 concentration levels. It's also worth noting that these rather drastic-looking temperature variations are charted in such a way that the approximately 1-2 degree variations around a mean temperature seems rather drastic, and it occurs to me that the likely thermal mass of the ENTIRE EARTH has to a significant damping influence on any atmospheric boundary effects on overall earth cooling.

So in short, yes there is more atmospheric CO2 than there has ever been, and that might actually raise the temperature a bit. But will it be a relevant forcing function relative to the insolation variations? I don't think we really have the foggiest clue. We would need to see the temperature rise to follow the CO2 variations, and meager increases to date are not particularly well correlated to the dramatic sudden and recent increases in greenhouse gasses.

Should we take the problem seriously and even take legislative and regulatory steps despite the uncertainty? Given the potential magnitude of loss in the event there really is a problem, I think we have a moral imperative to act. But let's make sure we keep track of what is causing what, rather than what is correlated with the other.