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

Projections of CMIP5 Models: Will They Ever Come Together?

Reference
Christensen, J.H. and Boberg, F. 2012. Temperature dependent climate projection deficiencies in CMIP5 models. Geophysical Research Letters 39: 10.1029/2012GL053650.
In an intriguing paper published in Geophysical Research Letters, Christensen and Boberg (2012) describe how they compared monthly mean temperatures projected by 34 global climate models that were included in phase 5 of the Coupled Model Intercomparison Project (CMIP5) with observations from the University of East Anglia's Climatic Research Unit for 26 different regions covering all major land areas of the world for the period 1961-2000, which feat was accomplished employing quantile-quantile (q-q) diagrams. This work revealed the existence of what they call "a warm period positive temperature dependent bias" for "many of the models with many of the chosen climate regions," the magnitude of which temperature dependence varied considerably among the models.

Analyzing the role of this difference as "a contributing factor for some models to project stronger regional warming than others," the two scientists found that "models with a positive temperature dependent bias tend to have a large projected temperature change," and that "these tendencies increase with increasing global warming level." In addition, they say that this situation "appears to be linked with the ability of models to capture complex feedbacks accurately," noting in particular that land-surface/atmosphere interactions are treated differently and with different degrees of realism among the various models they investigated, and that "soil moisture-temperature feedbacks are relevant for temperature extremes in a large fraction of the globe," consistent with their findings.

The unfortunate upshot of these observations, in the words of Christensen and Boberg, is that "accepting model spread as a way to portray uncertainty of the projection estimate may result in an overestimation of the projected warming and at the same time indicate little model agreement on the mean value." And they say that "a non-negligible part" of this sad state of affairs "is due to model deficiencies," which have yet to be overcome to the point that all model results come together and converge on a common projection.

So, will it ever happen? Will the convergence finally occur?

Archived 30 April 2013