Peer-Reviewed Climate Sensitivity Paper ‘Wrong & Lacks Credibility’

  • Date: 09/01/16
  • Nicholas Lewis, Climate Audit

Marvel et al.’s claim to have shown that TCR and ECS estimated from recent observations will be biased low is wrong. Their study lacks credibility.

Appraising Marvel et al.: Implications of forcing efficacies for climate sensitivity estimates

A guest article by Nicholas Lewis

Note: This is a long article: a summary is available here.


In a recent paper[1], NASA scientists led by Kate Marvel and Gavin Schmidt derive the global mean surface temperature (GMST) response of the GISS-E2-R climate model to different types of forcing. They do this by simulations over the historical period (1850–2005) driven by individual forcings, and by all forcings together, the latter referred to as the ‘Historical’ simulation.

They assert that their results imply that estimates of the transient climate response (TCR) and equilibrium climate sensitivity (ECS) derived from recent observations are biased low.

Marvel et al. use the GISS-E2-R historical period simulation responses to revise estimates of the transient climate response (TCR) and equilibrium climate sensitivity (ECS) from three observationally-based studies: Otto et al. 2013, Lewis and Curry 2014 and Shindell 2014. Their revisions give figures that are substantially higher than in the original studies. Remarkably, the Marvel et al. reworked observational estimates for TCR and ECS are, taking the averages for the three studies, substantially higher than the equivalent figures for the GISS-E2-R model itself, despite the model exhibiting faster warming than the real climate system. Not only is the GMST increase simulated by GISS-E2-R is higher than that observed, but the ocean heat uptake rate is well above the observed level.[2] No explanation is given for this surprising result.

The press release for the paper quotes Kate Marvel as follows:

‘Take sulfate aerosols, which are created from burning fossil fuels and contribute to atmospheric cooling,’ she said. ‘They are more or less confined to the northern hemisphere, where most of us live and emit pollution. There’s more land in the northern hemisphere, and land reacts quicker than the ocean does to these atmospheric changes.’

and continues by saying:

‘Because earlier studies do not account for what amounts to a net cooling effect for parts of the northern hemisphere, predictions for TCR and ECS have been lower than they should be.’

However, this is not true when the effective radiative forcing (ERF) measure of aerosol forcing – preferred by IPCC AR5 and used in the observational studies Marvel et al. criticises – is employed. When calculated correctly using Marvel et al.’s data, bases and assumptions, aerosol ERF had a transient efficacy of 0.97 – almost the same as the 0.95 for GHG forcing and 1.00 for CO2 forcing. This result is in line with the findings in Hansen 2005.This implies that aerosol forcing has had almost the same effect on GMST since 1850, relative to its ERF, as did CO2 and GHG forcing. Its concentration in the northern hemisphere did not lead to a greater cooling effect globally since 1850.

Studies like Marvel et al. can be valuable in showing the effects of differing forcing agents in climate models, which – if similar across climate models – may provide a guide to their effects in the real climate system. Unfortunately, I believe that the Marvel et al. results are substantially inaccurate and misleading. Its conclusions are therefore unfounded. But, as with any single-model study, even were its results unimpeachable they would reflect the behaviour of the particular model involved, which may be very different from that of other models and, more importantly, from that of the real climate system. […]

Full post

Recent Popular Articles

We use cookies to help give you the best experience on our website. By continuing without changing your cookie settings, we assume you agree to this. Please read our privacy policy to find out more.