Runaway Temperatures, or Runaway Numerical Modeling in the Natural Sciences?
“One of the messages that I like to convey after these five years of working in the topic is that I believe that in the general world community there has been a breaking of the balance between the three key aspects of any research in the natural sciences, which is observation, theory – by theory I mean physical, chemical, biological theory based on principles – and numerical modeling. I believe the three are extremely important. They all should be interacting, but they should be balanced.
My view is that in the last 20 years there has been much too much on the numerical side and absolutely not enough on observation, and I would suggest that observation is the key thing that should be supported in the coming decade.
So I begin with a critical look at some global and regional temperature data. I think this is either from […] or Fred Singer – I stole this slide from. It’s just to remind people what the recording of temp at a station looks like. On the left I don’t know which station but from 1850 to the year 2000 you have all the temperature data and you see that actually the trend you are going to try to extract is the mean on the right. When we talk about climate change in terms of temperature, what we are looking at is the figure on the right, which is always expanded on the vertical axis in such a way that you forget where it’s extracted from. It doesn’t mean that it is not interesting or not meaningful. It means one you should remember how difficult it is to extract this small signal, and two, it doesn’t have an error bar. Someone asked the question about uncertainties. Another one of my messages is that I believe uncertainties in many cases have been enormously underestimated, and we should always bear them in mind.” (To be cont’d.)