This is an interesting
Nature article about atmospheric temperature change controversies.
Warming is occurring even faster in the Earth's atmosphere than on its surface, according to a new look at old data. The finding could bolster the case for global warming by eliminating a primary sticking point in the debate.
...
"This might actually change the picture if it is accepted by the community," and could have important implications for public policy concerning greenhouse gas emissions, says atmospheric scientist John M. Wallace of the University of Washington in Seattle. But the community is a rather small and polarized group, and consensus does not appear to be forthcoming.
...
The one thing both camps agree on is that the problem of troposphere warming is a very important point in the global warming debate, and despite its complexity, the satellite data is the best way to tackle the question. But we may still be a long way from the answer.
Old data is valuable even though it's sometimes shaky since we can't travel back in time to run the data collection experiments again with better equipment. To make use of this data a lot of analysis and theory is required which inevitably spawns scientific disputes. Sometimes these disputes are difficult to resolve since no one has strong empirical arguments.