In a recent post, I linked to a 2015 blog post from Jerry Taylor. I was concerned that his own chart showed the opposite of what he claimed. Specifically, it seemed that the climate models that were published in the 2007 IPCC report had overpredicted actual warming from 2007 forward. And since you can calibrate the models (by adjusting parameters such as the reflectivity of aerosols in the atmosphere), the fact that the models “match” the observations before 2007 is not so reassuring.
Of course, because of the discrepancy between the models and surface average temperature, the difference between modeled and observed ocean-only readings is large, and post -1998 only matches the models during the recent El Nino (too bad this reply section will not accept an illustration).
There’s the further problem that Taylor tries to sweep away: The satellite/radiosonde comparisons with the IPCC model average show a huge error in the vertical in the tropics. Given that the vertical stratification is what determines tropical precipitation, that means the modeled rainfall is systematically wrong. Given that the presence of surface water dramatically alters the partitioning of incoming radiation (less sensible heating of wet surfaces), that means the daily thermal regime is also mis-specified, which will further screw up the rainfall etc…
At any rate, as shown by Hourdin et al. in the latest Bulletin of The American Meteorological Society, the models are tuned to match the 20th century surface history often with physically unrealistic adjustments, possibly a cause of the huge vertical error.
Because Pat said he wished he could post an image, I emailed him and offered him the option. He took me up. So below is the chart he wanted to post, along with his further commentary:
Many things to note. Even in the land-only, the CMIP model mean tends to be too warm post 1998. You will also see that by showing the real datapoints that the fit looks much less fortuitous than in Jerry’s post. And in the other 70%–the ocean surface—every post-1998 datapoint is below the model average, and it only gets close in the recent El Nino. The consequence of getting the ocean surface warming rate wrong is that means the flux of water vapor is being overestimated (it may be tuned in retrospective mode) for the near term and in the future. The calculated water vapor feeedbacks have to assume the forecast is correct, which it most clearly is not.
That leads to a further speculation: If the water vapor flux is wrong in the models (too high) that means that the models will consistently overpredict temperature, so I would surmise that the only way they can match is if they are tuned. I’m sure you have seen Hourdin et al. in BAMS. And that just scratches the surface, so to speak!
Pat then added in a final email: “[D]on’t forget to re-emphasize that the models were tuned to match the past which is why they fit El Chichon and Pinatubo.”