Contributed by Robert Lyman © 2018

Robert Lyman is an Ottawa energy policy consultant who was a public servant for 27 years; a diplomat for 10 years prior to that.

Dr. Ross McKitrick of the University of Guelph and Dr. John Christy of the University of Alabama in Huntsville have jointly published an important paper that assesses the validity of the global warming rate that is predicted by climate models used by the Intergovernmental Panel on Climate Change (IPCC). The paper, while extremely well written, uses several technical terms that may be confusing to the non-specialist. It is possible that people who read only short summaries will draw incorrect conclusions. Due to the paper’s importance, I thought I would take the risk of furthering that confusion by offering one person’s interpretation of what the paper says and does not say.


Some background is necessary. The IPCC uses 102 Global Circulation Models (GCMs) runs to assess the probable impact on the global climate of rising carbon dioxide concentrations in the atmosphere.  A GCM employs a mathematical model of the general circulation of a planetary atmosphere or ocean. GCMs are extremely complex, involving hundreds of equations and thousands of hypotheses about the relationship between the atmosphere, the oceans, and climate. One of the most important hypotheses concerns the representation of thermodynamics in the troposphere (the lowest region of the atmosphere, extending from the earth’s surface to a height of about 3.7–6.2 miles (6–10 km) in the tropics.


The McKitrick/Christy paper is mostly devoted to explanation of its methodology, and why it selected the tropical troposphere as the variable that most clearly serves as an indicator of whether GCMs are producing results that match experience. The period used to assess this is 1958 to 2017.


To be clear, the paper does not challenge or test the hypothesis that increased concentrations of carbon dioxide in the atmosphere causes or will cause increased global warming. It does not attempt to distinguish the comparative influence of increasing carbon dioxide concentrations and of other factors (e.g. the sun) on global temperatures.  It, rather, uses data from actual experience to test whether the results of GCMs are broadly correct. It measures temperature rises in terms of degrees Celsius per decade.


According to the models, the average warming rate based on increased carbon dioxide concentrations is 0.33 degrees per decade (i.e. 3.3 degrees C. per century). Based on this rate, the warming experienced over the period 1958-2017 should have been about 2.0 degrees C. Actual observations, however, show an average decadal warming of 0.17 degrees C, about half the rate projected by the models.


All 102 models have projections that run warmer than the observations. In most individual model cases, the discrepancy is significant, and on average the discrepancy is significant. McKitrick and Christy conclude that the major hypothesis in contemporary GCMs (related to the feedback response of the climate to increasing carbon dioxide emissions, is flawed.

*Updated to include the word ‘runs’ after (GCMs)



Ross McKitrick has a commentary on the paper and its findings posted on Dr. Judith Curry’s “Climate, Etc.”

A Test of the Tropical 200-300 mb Warming Rate in Climate Models

McKitrick on Climate Change – A layman’s guide to carbon pricing (Social Cost of Carbon) 

McKitrick on Climate Change – 14 short videos from 2015

See Dr. McKitrick’s full presentation – power point and video – from 2015. 

Dr. Christy spoke at Friends of Science Society’s 2010 event:

Dr. Christy’s 2017 written presentation to the US House Committee on Science, Space and Technology:

All of Friends of Science policy and science speaker events since 2004 are available on-line (some limitations on content/video in the earliest years):