hrothgar, on 2011-December-01, 11:34, said:
Three simple questions
1. Is it a good idea to forecast long term cycles based on five years worth of data?
2. What happens to your regression lines if you extend your data set to include earlier data?
3. What should we conclude about someone who gets some thing this basic this wrong:
(a) They are sufficiently stupid/incompetent that we can safely discount anything that they say
(b) They are deliberately misrepresenting data
There are more significant questions than that - for example, it is not at all clear to me, and I do theoretical physics for a living- that increased radiative heating will warm the oceans. There are three counterflows to imagine
(1) Warming the ocean by radiative forcing may lead to a much warmer surface layer without much affecting the layer below, this might lead to much greater evaporation.
(2) Warmer surface temperatures might break up the convection currents which bring cold water to the surface, resulting again in much greater evaporation and lower overall temperatures.
(3) Much of the energy might bleed into the atmosphere resulting in higher average wind speeds - even a small amount of wind can lead to much greater evaporative cooling.
If any of these effects seem implausible, I refer you to the Mpemba effect. Essentially, hot water can sometimes freeze faster the cold. The most convincing experiment yet done suggests that the reason is that the momentum of the convection current keeps mixing it more and leading to more efficient cooling than is possible for the cold water.
Finally, it is not clear to me what depths these measurements were taken - as I understand it the argo buoys measure a compete temperature profile from about 1km down. Also don't forget that ocean currents vary the temperature on 20 year cycles by dredging up cold water from teh bottom (reduced cooling greater average temp, lower surface temp), and by putting it back.
The physics is theoretical, but the fun is real. - Sheldon Cooper