THIS POST IS A CRITICAL EVALUATION OF THE SKEPTICAL HYPOTHESIS THAT THE CORRELATION BETWEEN CO2 AND TEMPERATURE SEEN IN THE DATA CAN BE INTERPRETED AS REVERSE CAUSATION WHERE SURFACE TEMPERATURE DRIVES ATMOSPHERIC CO2 CONCENTRATION.

HERE WE PROPOSE THAT THE DIRECTION OF THE CAUSATION CAN BE TESTED WITH CORRELATION ANALYSIS OF TEMPERATURE AGAINST CO2 CONCENTRATION AND THE LOG OF CO2 CONCENTRATION. IF THE LINEAR CORRELATION IS STRONGER THAN THE LOGARITHMIC CORRELATION IT WOULD SUPPORT THE REVERSE CAUSATION HYPOTHESIS. IF THE LOGARITHMIC CORRELATION IS STRONGER THAN THE LINEAR CORRELATION, IT WOULD FAVOR THE CLIMATE SCIENCE CLAIM TO A CAUSATION OF TEMPERATURE BY CO2.

THE TWO TRENDS IN CO2 AND LN(CO2) ARE SHOWN GRAPHICALLY BELOW.

THE CORRELATIONS AND DETRENDED CORRELATIONS ARE SHOWN BELOW

	CORR	DETCOR
CO2	0.9531	0.4252
LN(CO2)	0.9532	0.4588

THE REQUIRED GREATER LINEAR CORRELATION IS NOT FOUND IN THE DATA. THE CORRELATION IS ALMOST THE SAME BUT WITH A SLIGHT EDGE FOR THE LOGARITHMIC CORRELATION.

THE STRONGER TEST IS THE DETRENDED CORRELATION AND THERE WE SEE A SIGNIFICANTLY HIGHER DETRENDED CORRELATION IN THE LOGARITHMIC DATA THAN IN THE LINEAR DATA.

THESE DATA DO NOT PROVE THE CLIMATE SCIENCE CASE FOR THE GREENHOUSE EFFECT BUT THEY DO NOT SUPPORT THE SKEPTICAL HYPOTHESIS THAT THE RELATIONSHIP BETWEEN CO2 AND TEMPERATURE DERIVES FROM A REVERSE CAUSATION WHERE TEMPERATURE DRIVES ATMOSPHERIC CO2.

POSTSCRIPT: A REVIEWER SUGGESTED THAT THIS TEST SHOULD ALSO BE CARRIED OUT WITH SST. THAT WORK IS PLANNED AND THIS POST WILL BE UPDATED WITH THAT DATA WITHIN 24 HOURS.

THE HADCRUT4SST DATA SHOW AS FOLLOWS:

	CORR	DETCOR
CO2	0.9503	0.3097
LN(CO2)	0.9511	0.3908

THE SST RESULTS VERIFY THE FINDINGS FOR THE GMST (GLOBAL MEAN SURFACE TEMPERATURE) RESULTS.

THE ANALYSIS DOES NOT SUPPORT THE SKEPTICAL HYPOTHESIS THAT THE RELATIONSHIP BETWEEN ATMOSPHERIC CO2 AND SURFACE TEMPERATURE IS A REVERSE CAUSATION WHERE WARMER SURFACE TEMPERATURES MOVE MORE CO2 FROM THE SURFACE TO THE ATMOSPHERE.