Dr. William Gray, Professor Emeritus of Colorado State University, writes,
"There are many flaws in the global climate models. But the largest flaw is a result of the climate model’s inability to realistically deal with the small horizontal scale (and model unresolvable) changes brought about by the globe’s thousands of individual deep cumulonimbus (Cb) cloud elements. An increase in the totality of these deep Cb convective units adds drying to the upper troposphere. This is in contrast to the assumptions implicit in the General Climate Model (GCM) simulations which increase upper tropospheric water-vapor as a result of enhanced rainfall and Cb convection associated with rising levels of CO2." The upper-level drying leads to extra infrared radiation loss to space, resulting in a negative water vapor feedback.
Climate scientist Bob Tisdale published an open letter to three U.S. Senators. He asked "Why are taxpayers funding climate model-based research when those models are not simulating Earth’s climate?" and other related questions. Tisdale notes that modes of natural variability of the oceans "are now being blamed for suppressing global surface warming." This natural variability, much of which may ultimately be solar-caused, enhanced warming from the mid-1970 to the turn of the century. Climate models were aligned to that warming trend, falsely attributing almost all the warming to human causes.
High estimates of climate sensitivity to double CO2 determined from climate models rely on "highly negative aerosol forcing" to offset high CO2 forcing estimates so the forecasts approximately match the 20th century warming. A new study by Bjorn Stevens shows the aerosol forcing is much less negative than previously estimated. Climate scientist Nicholas Lewis has calculated new lower climate sensitivity estimates utilizing the new aerosol estimates. His best estimate equilibrium climate sensitivity is only 1.45 C with a likely range of 1.2 to 1.8 C, which is much less that the AR5 range estimate of 1.5 to 4.5 C.
Dr. Ross McKitrick, University of Guelph, explains that the "pause" in global warming reveals that there are flaws in the climate models. The models say there should be an increase but there is no increase in global temperature. “Social costs of carbon” are based on faulty climate models, not reality.
Dr. Clive Best reviews the IPCC fifth assessment report (AR5) attribution statement, which is based on comparing climate models to the global temperature record before this century. The statement "depends critically on them [models] being able to simulate both anthropogenic forcing and natural variability. Chapter 9 of the IPCC report shows that models are tuned so as to describe past observations. The model control runs exhibit long-term drifts. Any ‘natural’ trends generated by models in the temperature data are assumed to be an artifact and simply removed. The procedure assumes there in no natural climate change despite the recovery from the little ice age. However, the Atlantic Multidecadel Oscillation alone added 0.2 C warming to the 1950 - 2010 trend. The result is that the AR5 attribution statement is useless.