Cascading Fallacies in Climate Risk Assessment

Published August 14, 2020

As a logician, I am always on the lookout for fallacies and there is no lack of them in climate change alarmist policies. New Zealand’s newly released climate risk assessment not only has multiple fallacies, they build on one another in a cascade.

This is not about New Zealand. The authors of the assessment make clear that theirs is a new approach which they hope will be used globally. So this is about the world, including America.

The massive report is titled “First national climate change risk assessment for New Zealand.” Under New Zealand’s climate law, these assessments are supposed to be done every five years and this is the first.

The scope is breathtaking. The idea is to identify all of the significant risks due to human caused climate change that will be present in 2050 and 2100. Moreover, these supposed risks are prioritized.

Unfortunately this elaborate procedure is just a cascade of fallacies. Some of the major ones are listed below.

First, they use computer models to say precisely what the average weather will be in 2050 and 2100. This includes short and long term temperatures, precipitation patterns, and other climate features.

The fallacy is that there is no computer model today that can accurately make such forecasts. Different major models disagree strongly in predicting all of these features. For example the model sensitivity to doubling CO2 ranges from 1 to 6 degrees C, which is a huge range.

Second, they use the average of what is called the CMIP5 climate model runs. These are runs on a large number of climate models that are made to feed into the IPCC process. (Here there are several problems. In particular the models are all constrained so that all the significant forces are human, but that is a different issue.)

The fallacy is that there is no reason to believe that the average of a bunch of bad models is good. In fact the CMIP5 average has been shown to run very hot compared to observed warming. (CMIP6 is even worse.)

Third, they then choose to use the modeling of a wildly worst case emissions scenario called “RCP 8.5”. This scenario for future emissions is so high that it has been criticized as impossible. Using RCP 8.5 is certainly a fallacy.

Fourth, they do what is called “down scaling” of these questionable modeling results. Down scaling means taking the crude modeling results for a large area and somehow generating results for specific places. There is no scientific way to derive fine scale forecasts from the model’s large scale ones. The data simply is not there. However it is done is arbitrary.

New Zealand is geographically pretty small with a land area of just over 100,000 square miles, roughly the size of Colorado. The risk assessment divides New Zealand into 8 tiny zones, with a unique climate forecast for each. This is a glaring fallacy.

Fifth, these impossible fine scale forecasts were then discussed by a large number of people, in a variety of ways, to define all the significant risks. This is an exercise in imagination, not science. It is well known in decision theory that the results of group gropes like this depend heavily on who is there, what they are given and how they are guided.

The fallacy here is to pretend that this is a systematic inventory of risks, suitable for policy making.

Sixth, the supposed risks were ranked based on polling the participants. In addition to the group grope problem there is the pesky fact that risk is a two dimensional concept so risks cannot simply be ranked in one dimension. Each risk has both a severity and a probability.

Generally speaking, high severity but low probability risks are not worth addressing. Meteor strikes are a standard example (the impact really is an impact). The same is true for high probability but low severity risks. What one looks for are risks that combine relatively high severity and probability.

This 2-value ranking was not done, giving the fallacy of the single ranking of risks.

The seventh fallacy is yet to come. This wrongly ranked list of imagined risks based on arbitrary down scaling of an average of questionable computer model results running an impossible scenario is supposed to lead to a National Adaptation Plan in two years. That would be a mega-fallacy.

On the amusing side, I think they got the highest ranked risk right. This is the risk that the government will do the wrong thing. I agree completely, especially if they use this risk assessment.

Also very funny is the “Give us a lot more money” risk. It goes like this:

Risk of delayed adaptation and maladaptation due to knowledge gaps resulting from under-investment in climate change adaptation research and capacity building.
Risk summary:
Under-investment in research and capacity building to inform understanding of climate change risks and impacts is undermining New Zealand’s ability to develop evidence-based adaptation policy. Critical research gaps relate to:
–atmospheric processes
–hydrological cycle impacts
–ecosystem responses
–biodiversity and biosecurity
–New Zealand’s rural and urban communities
–the economic costs of climate change
–impacts on the primary sector
–impacts on heritage
–effects on health and health services
–use of mātauranga Māori to inform adaptation
–cascading impacts
–how to govern climate change adaptation at a number of scales.
These research gaps are a critical barrier to informed decision-making. While these gaps remain, maladaptive actions are a key risk.
” (Page 188)

Given all these significant gaps one would think that an accurate assessment would conclude that no risk assessment is possible at this time. That is my assessment.


The New Zealand climate risk assessment is a cascade of fallacies, unfit for policy making.

[Originally posted at CFACT]