FAIL (the browser should render some flash content, not this).

Modeling Temperature, Sea Level Pressure and Precipitation: CMIP5 vs. CMIP3

Bhend, J. and Whetton, P. 2013. Consistency of simulated and observed regional changes in temperature, sea level pressure and precipitation. Climatic Change 118: 799-810.
Introducing their study, Bhend and Whetton (2013) write that "over the past decade, demand for spatially explicit climate change information for impact and adaptation assessment has been steadily increasing." But they note that "a comprehensive assessment of climate model performance at the grid box scale in simulating recent change ... is not available at present." And, therefore, they set about to try to fill that void.

More specifically, the pair of researchers compared seasonal temperature, sea level pressure (SLP) and precipitation data for the most recent 50-year period common to both observations and simulations, with the temperature observations coming from the Goddard Institute for Space Studies (Hansen et al., 2010) with 1200-km smoothing, the SLP observations coming from the gridded HadSLP2 data set (Allan and Ansell, 2006) aggregated to 5° x 5° blocks, and the precipitation observations coming from the Global Precipitation Climatology Centre (Beck et al., 2005) on a 2.5° x 2.5° grid basis.

In discussing their findings, the two Australian researchers report that with respect to temperature, "significant inconsistencies can be found in the majority of CMIP3 and CMIP5 models in the Indian Ocean and Indonesia, the Arctic, and north-western Africa and south-western Europe in boreal summer (JJA), and central Asia in DJF where models underestimate the observed warming." In addition, they say that the models "do not reproduce the regional cooling or lack of warming over parts of the southern Ocean and western Atlantic and the north-eastern and south-eastern Pacific."

With respect to sea level pressure, they report that "the majority of the models significantly underestimate the magnitude of the observed decrease in SLP in parts of the high latitudes in the respective winter months," and that "additionally, most of the models underestimate the magnitude of the observed increase over Africa and tropical South America in DJF, and a smaller fraction of models also in the tropical Atlantic and the eastern Indian Ocean."

Finally, with respect to precipitation, the real-world data indicate it to be "strongly variable in space and large in magnitude in some regions," while "the simulated changes are considerably weaker but generally consistent with the observed change, except in boreal spring," when there are "some coherent areas of inconsistencies shared across models."

Most importantly of all, however, Bhend and Whetton say they "find no improvement from CMIP3 to CMIP5 with respect to consistency of simulated local trends per degree warming in near-surface temperature, SLP, and precipitation with the observed change." Or as they also more bluntly put it, "recent model development has not significantly altered our understanding and description of long-term regional change in these variables."

Clearly, progress in climate modeling of this nature over the past several years has essentially been no progress at all.

Additional References
Allan, R. and Ansell, T. 2006. A new globally complete monthly historical gridded mean sea level pressure dataset (HadSLP2): 1850-2004. Journal of Climate 19: 5816-5842.

Beck, C., Grieser, J. and Rudolf, B. 2005. A new monthly precipitation climatology for the global land areas for the period 1951 to 2000. Climate Status Report 2004, German Weather Service, pp. 181-190.

Hansen, J., Ruedy, R., Sato, M. and Lo, K. 2010. Global surface temperature change. Reviews of Geophysics 48: 10.1029/2010RG000345.

Archived 24 September 2013