Climate Etc.

The ‘threat’ of climate change

by Judith Curry

A major disconnect in the discourse surrounding climate change is interpretation of the ‘threat’ of climate change.

Last week I attended the annual meeting of the American Meteorological Society in Seattle.  It was a very good meeting; once the talks are online, I will discuss a few of them.

One of the best things about such conferences is the opportunity for extended face to face discussions with other scientists.  I had one such discussion that triggered the theme for this post.  This scientist (who will remain unnamed) does not disagree with me about climate change science in any significant way, although he has more confidence in climate models than I do.  In particular, he has publicly discussed the uncertainty issue.

He doesn’t take the ‘heat’ that I do largely because, in spite of these substantial uncertainties, he makes statements about the ‘serious threat’ of climate change, substantial risk of dangerous or even calamitous impacts,  reducing this risk requires a reduction of carbon emissions.

We both agree that there is the ‘possibility’ of extreme impacts if the warming is on the high end of the model projections.  We agree that we can’t quantify the probability of such impacts; it is best to regard them as ‘possibilities.’

So, what is the differences in reasoning that lead us to different conclusions regarding policy responses?

Definitions of ‘threat’ and ‘risk’

Some definitions of ‘threat’:

Two of the definitions imply something that has a high probability of occurring: ‘approaching’, ‘imminent’, ‘impending’.  A third definition includes the term ‘likely’, which (at least in IPCC parlance) implies a probability > 66%.  The last definition uses the word ‘possibility’.

The ‘possibility’ definition seems to be used for military threats and for threats to computer security.  For issues related to extreme weather events, food and water shortages, the ‘imminent’ or ‘impending’ definitions are arguably the more common parlance.

As per the Wikipedia, ‘risk’ has connotations of ‘probability’ and ‘quantifiable damage’:

A probability or threat of damage, injury, liability, loss, or any other negative occurrence that is caused by external or internal vulnerabilities, and that may be avoided through preemptive action.

The probability or threat of quantifiable damage, injury, liability, loss, or any other negative occurrence that is caused by external or internal vulnerabilities, and that may be avoided through preemptive action.

The words that are used — threat and risk — provide connotations of impending damage and that this is quantifiable and avoidable.

I think that use of these words mislead the public debate on climate change — any damages from human caused climate change are not imminent, we cannot quantify the risk owing to deep uncertainties, and any conceivable policy for reducing CO2 emissions will have little impact on the hypothesized damages in the 21st century.

‘Threats’ or ‘reasons for concern’?

I do not question that the possibility of adverse impacts from human caused climate change should be under consideration.  However, the human caused impacts of climate change have been overhyped from the beginning — the 1992 UNFCCC treaty on avoiding dangerous human interference on the climate.  This implied warming was dangerous before any work had actually been done on this.

Some much needed clarification is presented in a recent article published in Nature: IPCC reasons for concern regarding climate change risks.  This article provides a good overview of the current IPCC framework for considering dangerous impacts.  A summary of the main concerns:

The reasons for concern (RFCs) reported in AR5 are:

The eight overarching key risks are:

I think that qualitatively, these are the the appropriate risks to consider.  Where I don’t find this analysis particularly convincing is their links of ‘undetectable’, ‘moderate’, ‘high’, ‘very high’ to specific levels of temperature increase.

The confounding societal effects on all of these risks are overwhelming, IMO, and very likely to be of greater concern than actual temperature increase. Apart from (vii) and (viii) related to ecosystems, these risks relate to vulnerability of social systems.  These vulnerabilities have put societies at risk for extreme weather events throughout recorded history — adding a ‘delta’ to risk from climate change does not change the fundamental underlying societal vulnerabilities to extreme weather events.

The key point IMO is one that I made in a previous post Is climate change a ‘ruin’ problem?  The short answer is ‘no’ — even under the most alarming projections, human caused climate change is not an existential threat on the timescale of the 21st century.

JC reflections

So what are the words that we should use to talk about the potential harm from human caused climate change?  I think that the following phrases are appropriate:

I think that ‘threat’ is overly alarmist, since it implies imminent harm.  ‘Risk’ is not overly alarmist, but it does imply that the harm is quantifiable and mitigable — which I have argued that it is not.

How do we deal with potential harm and possible catastrophic impacts?  This puts us in the domain of decision making under deep uncertainty — a topic I have written about many times at CE.

I have been planning a full post on this, but I am way behind, so I will point to this here:  Society for Decision Making Under Deep Uncertainty — deep — under the leadership of Robert Lempert (who has been featured in several previous CE posts).  This society and its website is a gold mine of information that can be used for thinking about how we should respond to the wicked climate change problem.