More Studies Find Lower Climate Sensitivity

Secy. of State John Kerry last week exhorted all State Department officials to conclude a new international climate change agreement, integrate climate change with other priorities, and, in general, “elevate the environment in everything we do.” In the same week, climate researchers produced two more studies undercutting Kerry’s opinion that climate change is “perhaps the world’s most fearsome weapon of mass destruction.”

The studies address the core scientific issue of climate sensitivity — the question of how much warming results from a given increase in atmospheric carbon dioxide (CO2) concentrations.

There are two types of sensitivity estimates. Equilibrium climate sensitivity (ECS) is an estimate of the increase in ‘steady state’ surface temperature after the climate system has fully adjusted to a doubling of CO2 concentrations — a process assumed to take centuries or longer due to oceanic thermal inertia. Transient climate sensitivity (TCS) is the estimated increase in surface temperature during the 20-year period when CO2 doubling occurs, presumably during the final decades of this century.

ECS is the key variable in both climate model predictions of future global warming and model estimates of the “social cost of carbon” – the damage allegedly inflicted on society by an incremental ton of CO2 emissions.

The IPCC’s 2007 Fourth Assessment Report (AR4) estimated a “likely” ECS range of 2°C-4.5°C, with a “best estimate” of 3°C. Since 2011, however, the warming pause and the growing divergence of model predictions and observed global temperatures have been the impetus for several studies finding that IPCC sensitivity estimates are too hot.

Cato Institute scientists Patrick Michaels and Chip Knappenberger maintain a growing list of such studies, which totaled 18 as of February 2014.

The average sensitivity estimate of the 18 studies is just under 2°C. In other words, the AR4 “best estimate” of 3°C is 50% higher than the mean estimate of the new studies. That may be why the IPCC’s 2013-2014 Fifth Assessment Report (AR5) declines to offer a “best estimate.”

A new “best estimate” of 2°C would deflate the scary climate change impacts featured elsewhere in AR5, but recycling the same old 3°C “best estimate” would deflate the IPCC’s claim to be an honest broker. So instead the IPCC chose to lower the low end of the “likely” sensitivity range. Whereas the “likely” range in AR4 was 2°C-4.5°C, in AR5 it is 1.5°C-4.5°C.

That small concession, however, does not dispel the growing challenge to consensus climatology. As indicated in the chart above, the average sensitivity of the climate models used in AR5 is 3.2°C. That is 60% higher than the mean of recent estimates (<2°C). Let’s take a quick look at three studies that have come out this year.

In February, the journal Ecological Modeling published A minimal model for estimating climate sensitivity by Craig Loehle of the National Council for Air and Stream Improvement. Using the historical surface temperature record, Loehle subtracted out the periodic temperature oscillations associated with ocean cycles to identify the “linear trend” of the “anthropogenic signal.” Comparing that trend with increasing CO2 concentrations from 1959 to 2013, he calculates an ECS of 1.99°C and a TCS of 1.093°C. 

Last week, the UK Global Warming Policy Foundation published Oversensitive: How the IPCC Hid the Good News on Global Warming by Nicholas Lewis and Marcel Crok. Based on the average of “observationally-based” estimates, the two scientists arrive at a best sensitivity estimate of 1.75°C – “more than 40% lower than both the best estimate in AR4 of 3°C and the 3.2°C average of GCMs [general circulation models] used in AR5.”

Lewis and Croc accuse the IPCC of hiding the good news by refusing to assess the evidence and “declare that some studies are better than others or to adjudicate between observational and model-based lines of evidence.” Michaels and Knappenberger offer commentary here. Georgia Tech Prof. Judith Curry wrote a foreword to the Lewis and Croc paper and discusses the study on her blog.

Also last week, the NASA-alumni Right Climate Stuff Research Team published Bounding GHG Climate Sensitivity for Use in Regulatory Decisions. Author Harold Doiron argues that TCS – essentially, the projected increase in global temperature by 2100 – is a more relevant measure for policymakers than ECS. Using CO2 concentration data and surface temperature data from 1850 to 2012, Doiron estimates that the upper bound value for TCS is 1.6°C, and that the upper bound value for ECS is 2.9°C — lower than both the AR4 “best estimate” and the average of AR5 model estimates.

It can’t be said too often. Climate change is not ‘worse than we thought.’ It’s better than they told us.