Tuesday, August 4, 2015

Causality Problems With Global Warming

This graph shows atmospheric CO2 from 1820-1960.  Most of the consumption of fossil fuels takes place in the 20th century, generally after 1940.

This graph show temperature change:
Note that much of the warming took place on the left side of the graph, with a decline after 1940, when CO2 concentrations increase and fossil fuel consumption increases.  When I pointed this out to an acquaintance who BELIEVES, his response was to call me a pragmatist instead of a believer in science!  This is fast becoming a religiously based system of thought.


  1. I tend to be a "warmist", one who believes there is some warming going on, and that messing with CO2 levels can't be something that has zero effect, but I need serious convincing before I'll accept that we need to turn off our power plants in order to save the planet. However, I noticed that the charts you present have different X-axis values, and thus cover different periods of time.

    I'm sending you a graphic that shows the results when the two graphs are adjusted to the same scale (years per inch) and so that the same years overlap.

  2. Hmmm... that does not agree with the Mauna Loa measurements. I read the original article where this appears, but it isn't clear to me what 'background CO2 level' means.

    The Mauna Loa measurements show a much smoother curve that rises monotonically. I believe those measurements are quite accurate, and Mauna Loa was chosen because it is the least likely place on earth to get local contamination that is also reasonably accessible. It is very high, and way out in the middle of the ocean, so the CO2 it measures is most likely a good proxy for the whole earth.

    Beyond that, there are solid physical reasons to believe that increasing the CO2 concentration will move the CO2 trajectory upwards from its natural trend. That really is settled science, unlike the various alarmist predictions. The radiative physics of CO2 radiation is well understood in theory, verified by experiment (the one dimensional radiative balance model).

    That is not the same as saying that climate alarmists are right about the amount of deflection of that curve. It also does not mean that the very complex climate models are right. Those models are (roughly) finite element, parameterized weather models run to simulate a very long time (decades). When forecasting from the corresponding weather models, which I do frequently, we know that they are worthless beyond 10-15 days, and often very wrong just a few days in advance.

    I suspect the alarmists are off by a substantial factor.