*************************EXCERPT********************************************
Marc77 says:
First, it is not enough to find potential problems with satellites. You have to explain why the satellites have measured what they have measured and you have to prove that it is not incompatible with your hypothesis.
Second, here is the graph above detrended by 1/3 degree Celsius for 1/3 of a century:
http://www.woodfortrees.org/plot/uah/detrend:0.333/plot/uah/detrend:0.333/mean:13
So, removing a trend of 1 degree Celsius per century seems to remove all of the perceived trend. Is it possible that a trend of 1/2 a Celsius per century comes from the cool Pinatubo in the first half and the 1998 super El-Nino in the second half? If you take the sine function from 0 to 2pi radians. Since it starts with the positive phase then the negative phase, you would find a negative trend even if this function does not have a trend in the long run. It would be interesting if someone knew a way to remove the artificial trend that comes from particular events.
**********************************EXCERPT**************************************
Peter Hartley says:
Given this little bit of history
http://www.worldclimatereport.com/wp-images/Michaels_Knappenberger_Nature96.pdf
followed up here
http://www.worldclimatereport.com/index.php/2010/02/26/quick-response-to-ben-santers-comments-at-realclimate/
one would think that Ben Santer would be a little more reticent to keep on bringing up the issue of detecting a “human fingerprint” in climate data.