Paper Reviewed
Ao, C.O., Jiang, J.H., Mannucci, A.J., Su, H., Verkhoglyadova, O., Zhai, C., Cole, J., Donner, L., Iversen, T., Morcrette, C., Rotstayn, L., Watanabe, M. and Yukimoto, S. 2015. Evaluation of CMIP5 upper troposphere and lower stratosphere geopotential height with GPS radio occultation observations. Journal of Geophysical Research: Atmospheres 120: 1678-1689.

In the words of Ao et al. (2015), the 13 scientists hailing from six different countries describe how they conducted "a detailed comparison of geopotential height fields between the Coupled Model Inter-Comparison Project phase 5 (CMIP5) models and satellite observations from GPS radio occultation (RO)," focusing on "the annual mean, seasonal cycle, and inter-annual variability of 200 hPa geopotential height in the years 2002-2008." And what did they learn by so doing?

The international research team reports that (1) in the annual mean "there was considerable spread among the models," that (2) "in the Southern Hemisphere high latitudes, the majority of the models showed a negative bias relative to observations," that (3) "in the Northern Hemisphere extratropics, the intermodal bias shows large seasonal variations," with (4) "positive bias during the summer" and (5) "mainly negative bias during the rest of the year," while in the Southern Hemisphere extratropics, (6) "the negative bias is present through all seasons and peaks in autumn," that (7) "for the seasonal cycle, the models were found to have excessive seasonal variability relative to the observations," that (8) for interannual variability "the correlation was shown to be very low in the subtropics and the extratropics," which finding (9) "likely reflects the deficiency of models in capturing baroclinic variabilities in the extratropics."

So what is the take-home message of these findings? Although it has come a very long way indeed from the time of its inception, the climate modeling enterprise still has a very long way to go before it achieves the ability to present a reasonably accurate reflection of reality in the case of geopotential height field behavior.