Honestly, I didn't know there was any real controversy about whether the tropospheric temperature was rising or not. I'm sure I've seen some papers questioning trends, etc, but a few stories I've seen in the popular press [e.g.,Economist.com] lately have made it seem like the skeptics actually had a case.
That link to the Economist article (no byline) explains this controversy and how it might be getting resolved. Basically, numerical models of the atmosphere predict a large warming in the free troposphere (about 1 km to 10 km), associated with increased carbon dioxide. The observations of tropospheric temperature are known to be kind of flaky, but not bad. Those observations, from some satellites and weather balloons, show no warming comensurate with surface temperatures. What does that mean?
Well, let me tell you this. It does NOT mean that we don't understand global warming or the atmosphere. We can see warming at the surface, with reliable thermometers and long records. There's no question. That warming can not be independent from the free troposphere. I've recently seen some skeptical web sites saying the warming is concentrated over a few years in the early 70s, right before satellites go up, and that might explain why the satellite records show no trend. Wrong.
In the "current" (I have to verify that) issue of Science, three papers discuss how the instrumental record (of weather balloon data and satellite data) could be wrong. In two of them, when corrections are applied, the free troposphere is actually warming. All these findings will have to be verified by other people who work with observations, and this might even warrant re-running our reanalysis models with the correctly-calibrated data, but it seems that if this turns out to be true, we will have a much better picture of global warming so far. I doubt that it will improve our projections, since simulations were already doing the right thing.
More on this later.