Occam’s Razor, The Null Hypothesis And Global Warming

  • Date: 06/02/13
  • John Hinderacker, PowerLine

For a sophisticated yet understandable introduction to the global warming debate, it is hard to beat this article by Professor Robert Carter at the American Institute for Technology and Science Education site. Dr. Carter notes that there is considerable common ground between the climate alarmists and the climate realists:

Though you wouldn’t know it from the antagonistic nature of public discussions about global warming, a large measure of scientific agreement and shared interpretation exists amongst nearly all scientists who consider the issue. The common ground, much of which was traversed by Dr. Hayhoe in her article, includes:

* that climate has always changed and always will,
* that carbon dioxide is a greenhouse gas and warms the lower atmosphere,
* that human emissions are accumulating in the atmosphere,
* that a global warming of around 0.5OC occurred in the 20th century, but
* that global warming has ceased over the last 15 years.

I would add one more: that the amount of warming attributable to increased CO2 in the atmosphere based on scientific experiment is trivial, and one can argue for a dangerous global warming only by assuming positive feedback effects, e.g. involving clouds, that magnify the otherwise-harmless impact of increased atmospheric CO2. Dr. Carter continues:

The scientific argument over [Dangerous Anthropogenic Global Warming] is therefore about none of these things. Rather, it is almost entirely about three other, albeit related, issues. They are:

* the amount of net warming that is, or will be, produced by human-related emissions,
* whether any actual evidence exists for dangerous warming of human causation over the last 50 years, and
* whether the IPCC’s computer models can provide accurate climate predictions 100 years into the future.

I would add that there is also disagreement over whether the net effect of significant global warming would be good or bad.

In framing these disagreements, Dr. Carter starts with first scientific principles:

Science deals with facts, experiments and numerical representations of the natural world around us. Science does not deal with emotions, beliefs or politics, but rather strives to analyse matters dispassionately and in an objective way, such that in consideration of a given set of facts two different practitioners might come to the same interpretation….

William of Occam

Which brings us to the matter of Occam’s Razor and the null hypothesis. William of Occam (1285-1347) was an English Franciscan monk and philosopher to whom is attributed the saying ‘Pluralitas non est ponenda sine necessitate’, which translates as ‘Plurality should not be posited without necessity.’ This is a succinct statement of the principle of simplicity, or parsimony, that was first developed by Aristotle and which has today come to underlie all scientific endeavour.

The phrase ‘Occam’s Razor’ is now generally used as shorthand to represent the fundamental scientific assumption of simplicity. To explain any given set of observations of the natural world, scientific method proceeds by erecting, first, the simplest possible explanation (hypothesis) that can explain the known facts. This simple explanation, termed the null hypothesis, then becomes the assumed interpretation until additional facts emerge that require modification of the initial hypothesis, or perhaps even invalidate it altogether.

Given the great natural variability exhibited by climate records, and the failure to date to compartmentalize or identify a human signal within them, the proper null hypothesis – because it is the simplest consistent with the known facts – is that global climate changes are presumed to be natural, unless and until specific evidence is forthcoming for human causation.

It is one of the more extraordinary facts about the IPCC that the research studies it favours mostly proceed using an (unjustified) inversion of the null hypothesis – namely that global climate changes are presumed to be due to human-related carbon dioxide emissions, unless and until specific evidence indicates otherwise.

So, what are the tests to which we can subject the anthropogenic global warming theory to determine whether it has more explanatory power than the null hypothesis?

The DAGW hypothesis that I want to test here is precisely and only “that dangerous global warming is being caused, or will be, by human-related carbon dioxide emissions”. To be “dangerous”, at a minimum the change must exceed the magnitude or rate of warmings that are known to be associated with normal weather and climatic variability.

What evidence can we use to test the DAGW hypothesis?

Many different lines of evidence can be used to test the DAGW hypothesis. Here I have space to present just five, all of which are based upon real world empirical data. For more information, please read both Dr. Hayhoe’s and my book.

Consider the following tests:

(i) Over the last 16 years, global average temperature, as measured by both thermometers and satellite sensors, has displayed no statistically significant warming; over the same period, atmospheric carbon dioxide has increased by 10%.

Large increases in carbon dioxide have therefore not only failed to produce dangerous warming, but failed to produce any warming at all. Hypothesis fails.

(ii) During the 20th century, a global warming of between 0.4O C and 0.7O C occurred, at a maximum rate, in the early decades of the century, of about 1.7O C/century. In comparison, our best regional climate records show that over the last 10,000 years natural climate cycling has resulted in temperature highs up to at least 1O C warmer than today, at rates of warming up to 2.5O C/century.

In other words, both the rate and magnitude of 20th century warming falls well within the envelope of natural climate change. Hypothesis fails, twice.

(iii) If global temperature is controlled primarily by atmospheric carbon dioxide levels, then changes in carbon dioxide should precede parallel changes in temperature.

In fact, the opposite relationship applies at all time scales. Temperature change precedes carbon dioxide change by about 5 months during the annual seasonal cycle, and by about 700-1000 years during ice age climatic cycling.Hypothesis fails.

(iv) The IPCC’s computer general circulation models, which factor in the effect of increasing carbon dioxide, project that global warming should be occurring at a rate of +2.0O C/century.

In fact, no warming at all has occurred in either the atmosphere or the ocean for more than the last decade. The models are clearly faulty, and allocate too great a warming effect for the extra carbon dioxide (technically, they are said to overestimate the climate sensitivity). Hypothesis fails.

(v) The same computer models predict that a fingerprint of greenhouse-gas-induced warming will be the creation of an atmospheric hot spot at heights of 8-10 km in equatorial regions, and enhanced warming also near both poles.

Given that we already know that the models are faulty, it shouldn’t surprise us to discover that direct measurements by both weather balloon radiosondes and satellite sensors show the absence of surface warming in Antarctica, and a complete absence of the predicted low latitude atmospheric hot spot.Hypothesis fails, twice.

Full story