Michaels revisits and updates Meltdown: The Predictable Distortion of Global Warming by Scientists, Politicians, and the Media. The authors explain why you never get the straight scoop about global warming. They refer to Robert Rosenthal "file drawer problem." For any given research area you get only to see the 5% of the studies that support the current view, and you don't get to see the 95% that do not support it. They also mention Edward Wegman in "Controversy in Global Warming" who indicated that any paper promoting the global warming theory will be reviewed very leniently by peers meanwhile one debunking it will be reviewed most critically. To improve the quality of peer review, they propose that scientific articles submitted for review be released on the Internet so that any scientist could comment on it. Instead, such articles are reviewed by familiar and supportive peers.
Near the end of the book, the authors study the global warming bias by reviewing 116 papers published in Science and Nature in 2006. The authors indicate that the scientific community thought so far that existing models did not underestimate global warming. This suggests going forward there would be a 50/50 chance that new findings indicate that global warming is either better or worst than we thought. Instead, the authors uncovered only 10 papers (less than 9% of the 116) that suggested that global warming was moderating. All others suggested global warming was worst than thought.
The authors also refer to Thomas Kuhn's The Structure of Scientific Revolutions first published in 1962. Kuhn states that scientific research supports contemporary paradigms. And, scientific work tries either to explain anomalies or to show that anomalous data are wrong. The authors suggests Kuhn's framework perfectly anticipated the behavior of the scientific community in their supporting the anthropogenic global warming paradigm. First, this community found a weak argument (CFCs) to explain out the cooling of temperatures during the mid of the 20th century. And second, it revised the data numerous times in an attempt to entirely get rid off the embarrassing mid century cooling all together.
Now, you can't even trust the data. Temperature data series have been adjusted 6 times in just the past few years. They were to factor the urban island effect and the related effect of agriculture, deforestation, and zoning changes. They all lead to artificially raising recent temperatures. So, adjusting the time series should have lowered recent temperature levels. Instead, they lowered earlier temperatures. As a result, instead of the adjustments showing a reduction in global warming, they show an acceleration. Thus, you get more warming from the same data series!
The ones who don't go along with the global warming paradigm pay dearly for it. A bunch of State climatologists (Delaware, Virginia where Michaels the co-author got fired, Oregon, Washington) have either been fired or censored by Government officials for disclosing data and analysis that does not support the global warming paradigm.
The ones who promote this global warming paradigm are often deified. Al Gore received a Nobel Prize for his work and an Oscar award for An Inconvenient Truth. Meanwhile, the authors state that all Al Gore did was developing an apocalyptic vision disconnected from the science. Al Gore projected mean expected temperature increase of 6 degree Celsius only matched by the IPCC very worst case scenario. He also projected sea level rise of 20 to 40 feet vs 8.5 to 18.5 inches for IPCC most likely outcome.
The authors also debunk numerous other exaggerations from Al Gore and followers. Long term temperature records indicate that Greenland had been warmer for several millennia than currently, and it did not shed its ice. The ice cover in Antarctica is extremely stable. Its ongoing minimal ice loss translates into a sea level rise of only 1 inch per century. The retreat of glacier on Kilimanjaro are related to a drop in moisture that far precedes the rise in CO2 concentration during the 20th century. Looking at multi-century record, fire frequency for the last 500 years has been the result of natural ocean climate cycles, and not global warming. Heat-related mortality declined as cities get warmer (graph pg. 179). Heat related mortality rates decreased by 75% for 28 American cities between the 1960s and 1990s (graph pg. 182). They also confirm that impact of CO2 on temperature rise is logarithmic. This greatly moderates the gravity of CO2 as a greenhouse gas. Also, contrary to global warming advocates wheat and corn yields have risen very rapidly in tandem with temperature increases since 1950. They also refer to Steve McIntyre debunking the hockey stick increase in temperature by simply looking at a long term average in temperature vs just averages over the past 78 years (graph pg. 218). For more on this specific issue read the second chapter in Shattered Consensus: The True State of Global Warming.
Models are still really poor at factoring cloudiness, rainfall, humidity levels. Additionally, they all project CO2 concentration that accelerates far beyond current trend way into the future. As a result, those models are highly inaccurate and exaggerate temperature increase. But, they still support the global warming paradigm. The IPCC mindset is that ten different models can't be all wrong; But, if they suffer from the same flaws and bias, they can [be all wrong].
Government policy response has so far been ineffective. Such is the case of the U.S. subsidies supporting the production of ethanol that will actually increase greenhouse emission, is highly energy inefficient, and cause substantial food inflation due to displacing a substantial portion of U.S. corn production. Meanwhile, the Kyoto Protocol has achieved little. All the member countries have failed their CO2 reduction targets by wild margins. In the end, the Kyoto Protocol just allowed the member countries to blame the non member ones. But, when you look at actual carbon emission performance over the relevant time period the difference between the two groups is ambivalent.