Yes, you can see something interesting there. Note that the headline says that the satellite is warming faster than the surface, but actually they both show 0.4 deg between 1995 and 2015.It doesn't matter, you are being fed a line by people that know it doesn't matter. This has been done to death by climate sceptics. There are tens of thousands of temperature measuring stations across the world, you actually need only around 80 to do a global temperature reconstruction as long as you get the spatial weighting correct. It doesn't actually matter which 80 you choose , they can be good quality stations, bad quality stations, ones that haven't moved, only ones airports (which are quite carefully controlled) If you are interested have a look at the BEST work Berkeley Earth Surface Temperature, the sceptics, Jeffid and romanm. They all tried their own reconstructions accounting for any biases, I think they all showed a higher warming trend than the established analyses.
Here is one - satellite lower troposhere Vs surface temperature based on satellite measured temperatures
View attachment 55095
It's A Match: Satellite and Ground Measurements Agree on Warming
The consensus gives confidence to satellite estimates of temperature rise in remote areas with few weather stationswww.scientificamerican.com
What's more interesting is where the satellite gets its heat from because it's above the atmosphere. How much heat comes directly from the sun, how much of the sun's heat is reflected from Earth to the satellite, and how much comes to the satellite as radiation of the Earth's heat? I'm not an expert, but my instict tells me that the most heat would come directly from the sun, in which case that graph is of the sun's temperature, not the Earth, and, of course, if the sun is giving out more heat, then it seems logical that the Earth would be hotter in proportion and the two graphs would match exactly, which they do. Does that sound right?
Last edited: