The Potsdam Institute has something of a reputation for being a research unit of loose morals. The letter written by one of their star scientists in the Economist suggests that their grasp of the scientific method is even looser.
Prof Anders Levermann's missive responds to the Economist's article on climate sensitivity and is pretty jaw-dropping:
SIR – The reduced warming of the past decade is brief and can be understood in terms of natural fluctuations from the El Niño phenomenon, the effects of volcanoes, the solar cycle and the uptake of heat from the oceans, which continues, in contrast to your statement. There are and will always be fluctuations in global temperature, but the underlying trend is robust, man-made and consistent with a climate sensitivity of around 3°C.
The IPCC’s range on sensitivity is supported by, but not merely based on, models. It is deeply rooted in physics. Quantum physics and thermodynamics, the same physical laws that underlie the functioning of our computers and power plants, yield a baseline climate sensitivity of about 3°C. This is based on the facts that carbon dioxide, water vapour and methane absorb infra-red; a warmer atmosphere holds more water; and ice and snow melt under warming. Any deviation from this baseline needs a reason. As long as we do not find modern physics to be fundamentally wrong, we will have to plan for a climate sensitivity of 3°C.
So, the empirical evidence suggests climate sensitivity is low. A hypothesis based on "quantum physics and thermodynamics" suggests it's higher. And Professor Levermann thinks we should accept the hypothesis until the deviation from reality is explained.
Scientific method fail.