Earlier in this series we reported on a paper by Nic Lewis and Judith Curry which brought some key climate data series up to date and showed that the energy balance method of computing ECS yielded an estimate of about 1.6° C. But afterwards the IPCC released new data on aerosol forcing, and some new compilations of surface temperature data came out that filled in missing Arctic records, conveniently raising the global trend and wiping out the “hiatus” of all things, so the same authors had a look to see if their earlier ECS estimate needed revision. Yes, they found, causing it to soar to... 1.8° C, still too low for greenhouse gas emissions to be a problem.
The importance of getting the aerosol forcing effect right is that aerosols have a cooling effect while greenhouse gases have a warming effect (that’s the theory, at any rate). The two together have to add up to the net warming observed. Over the 20th century there has been a small amount of overall warming. Suppose aerosols have a very strong cooling effect: then greenhouse gases have to have a very strong warming effect, which implies a high ECS value. But suppose aerosols have very little cooling effect. Then greenhouse gases can only have a small warming effect otherwise the models would have way too much warming to match the 20th century record.
So whenever the IPCC puts out new aerosol forcing data, people like Nic Lewis, who know what to look for, check it to see how that affects the ECS estimate. Over the past decade the revisions have always led to a reduction in the estimated aerosol cooling effect. On its own this would mean ECS should go down. But at the same time the IPCC embraced a new data set of surface temperatures which tried to infill the high Arctic where there is a lot of missing data. Since the Arctic is warming faster than other places, by adding in its trend it pulled the global average warming rate up. Also there were new ocean heat content data series available from the Argo float network so Lewis and Curry took that into effect as well.
They found the overall effect was slightly positive, raising ECS from 1.6C to 1.8C. That’s still right at the low end of the traditional IPCC range and too low for greenhouse gases to cause any real world problems. It also implies that most climate models are wrong, or as Lewis and Curry put it:
“The implications of our results are that high best estimates of [ECS] derived from a majority of CMIP5 climate models are inconsistent with observed warming during the historical period (confidence level 95%).”
So real world estimates are inconsistent with most climate models, i.e. models predict too much warming. Normally that would cause the scientists to fix their models. But not in the upside-down world of climate science, where they simply ignored the finding and kept on publishing alarmist projections based on models with artificially high climate sensitivity. Follow the science, we say, which means we should ignore the projections from hot climate models.