Why climate models are failing - Climate Change: The Facts - page 14

cited goes back to one of the very large number of general circulation climate models
(GCMs) now in existence.
Given that there are—at least in theory—only two major anthropogenerated alterations
intheatmosphereusedinclimateforecasts,namelyanincreaseininfraredabsorptioninthe
lower atmosphere (from CO 2 and other greenhouse gases) and an increased backscattering
ofincomingradiation(fromanthropogeneratedaerosols),allsubsidiaryeffectsmustderive
from these two phenomena.
There is one variable in GCMs that changes all the subsidiary phenomena, such as
cloudiness, lapse-rate derived precipitation, or changing intensity and distribution of the
dynamic systems that initiate rainfall and snowfall. It is an average warming of the lower
atmosphere and a related change in the vertical lapse rate.
A peculiar phenomenon in modern climate studies is that cause and effect are not easy
to parse, and in some instances they are not even explored. In a prominent recent example,
President Obama's science advisor, John Holdren, citing the work of Francis and Vavrus 2 ,
suggested that the cold and snow of the winter of 2013-14 in the eastern United States
wascausedbyawarming-relatedmodulationofthecircumpolarvortexfueledbydeclining
Arctic sea ice. This was subsequently shown to enjoy no support in the observational data
by Barnes and Ballinger et al. 3 (Holdren nonetheless persisted with this meme in support
of President Obama's 'Climate Action Plan.')
What is truly odd is that there is no comprehensive examination in the refereed
literatureofthebehavioroftheentirecommunityofclimatemodels,whenitcomestotheir
'prime mover' output, a warming of global average surface temperature. Our interest in
this has been piqued by 'the pause' in warming, in which annual data in the scientifically
popular HadCRUT4 temperature history of Morice et al. 4 are now in their eighteenth
consecutive year without a statistically significant warming trend in global average surface
temperature.('HadCRUT ' is the name given to the combination of the data sets of two
organisations namely the Hadley Centre of the UK Met Office and the Climatic Research
Unit (CRU) at the University of East Anglia).
Additionally, over the last thirty years (1984-2013) the suite of 108 climate model runs
used in the 2013 compendium of the Intergovernmental Panel on Climate Change (IPCC)
produces an average surface warming rate of 2.6°C per century, while the observed value
was 1.7°C, a considerable difference for so long a period.
Materials and methods
We examined data since 1950 for the 108 model runs used in the Working Group I
(Science) 2013 IPCC Fifth Scientific Assessment available from KNMI Climate Explorer