At the EDS Climate Change and Business conference in Auckland this October, Prof. Tim Naish of the VUW Antarctic Research Centre revealed that current supercomputer modelling, with improved cloud physics, seem to show that the climate is more sensitive to GHG than previously thought.

He estimated warming of 6-7 C by 2100 on RCP8.5 (Business as Usual), rather than 4C, but said it won’t be official until the next IPCC report comes out.

His presentation is here:

NB: While we’re not strictly on the RCP8.5 pathway, due to mitigation efforts underway, it appears that we’ve got a lot less time and carbon budget than previously thought, regardless of trajectory.

Here is an article on this disturbing development.

Given the importance of this news, I asked Dr. Jim Salinger to comment on the new modelling results; his edited comments, and those of other climate scientists he canvassed, follow.

Jim Salinger

The jury is still out – this is the feedback I am receiving from my modelling colleagues:

“On the models, I would be, for the time being at least, very suspicious of the new CMIP6 results. The most likely reason for the increased sensitivity is a markedly bigger decrease in projected low/medium level cloudiness than in CMIP5 models or a marked increase in cirrus clouds, or both.”

Peter Barrett

The key question for planners will be how much risk to accept for decision-making on strategies for the future. As a geologist, my bet is on the high end with increasing climate sensitivity with warmth on account of feedbacks.

Martin Manning

… summaries of the issues can be found at:


It seems that so far the two French modelling groups are the only ones to have said these are the final results for the WMO / CMIP6 model intercomparison project, but groups in the UK and USA are coming up with similar results and CMIP6 is running about 12 months behind schedule to finalise its results for use in the next IPCC assessment report.

The deadline for that report cannot be changed, but if the IPCC still has any control of its reporting process then it should probably now say there will be a follow up special report on climate projections about two years later.

That is because it is going to take at least that long to try and resolve differences between different ways of estimating the temperature response to increasing GHGs.

At the moment those using paleoclimatic data and those using recent observations of temperature over the last 100 years are staying with the lower estimates for climate sensitivity. So at this stage it is still a question about the reliability of the models and whether future temperature changes are going to be like a simple extrapolation of past changes.

The bottom line is that the experts are still not sure whether the new models are right. That has been the case for model estimates of sea level rise for more than ten years now.

I think we are just going to have to get used to recognising that uncertainty ranges for the future are increasing rather than decreasing.