A near-perfect prediction of temperatures from knowing solar and oceanic variations only

Eat your hearts out, Warmists!

Expanding upon the last post, the "sunspot integral" (accumulated departure in sunspots v. the monthly mean of 41.2 for the observational period of sunspots 1610-2009) shows good correlation with the temperature record. Excellent correlation (R2=.96!) with temperature is obtained by adding to the sunspot integral the most significant ocean oscillations (the PDO-Pacific Decadal Oscillation + AMO- Atlantic Multidecadal Oscillation*3). Various other combinations and permutations of these factors compared to the temperature record have been posted at: 1 2 3 4 5 6, although I have not located others with a correlation coefficient of this magnitude.

Contrast the R2 of .96 from this simple model (near a perfect correlation coefficient (R2) of 1) vs. the poor correlation (R2=.44) of CO2 levels vs. temperature.

SOURCE (See the original for links and graphics)


  1. An R**2 value of .44 is acceptable only for a social science type.

  2. Social science regressions are not even that strong usually


All comments containing Chinese characters will not be published as I do not understand them