"hottest summer on record!", or whatever -- was it just in the U.S. or world wide?, and whose record and how long is that record I wonder. Something on the order of one ten millionth of the Earths history, I suppose.. We have of course proxy records for temperature going back a very long way. They are very unreliable and inaccurate yet they are not so unreliable and inaccurate for us to be certain that today's temperature variation lies well within the the natural variation of Earth's temperatures. Ninety percent of the species that ever lived are extinct and man had nothing to do with it, and so far it certainly looks like man has nothing to do with climate change either. The hottest summer on record! Don't make me laugh. Let's wait and see what the winter is like --maybe it will be the coldest on record, who knows? Then let's wait another century and see what the temperature does. By then we aught to have at least a rudimentary understanding of our planet's climate.
Current climate models predict greater increases in temperature than we have actually experienced. They do this by assuming (yes, it's a guess based on an hypothesis) slight additional warming from a few additional CO2 molecules per volume gets amplified by water vapor and clouds (water is the primary green house gas, CO2 the minor one) to increase CO2's effect -- positive feedback, in other words. Unfortunately we still don't know whether the effect of clouds is net warming or cooling. So this assumption of the models is tenuous. (The IPCC, in its reports, has acknowledged that clouds are a large source of uncertainty in climate models.)
Furthermore, the phase relationship between observed changes in temperature and CO2 concentration is inconsistent with these positive feedback models. There are detectable increases in temperature leading by many months increases in CO2, then as CO2 increases, temperature drops. It's cyclical, depending on what time esolution one uses to look at the data. The models do not correctly predict the observed phase relationship. It is as though anthro-CO2, which also varies somewhat on a short time scale, is a mere bystander to a larger naturally occurring variation in CO2, with the variation riding on slowly increasing, long-time-scale CO2. This will eventually get sorted out, because if CO2 concentration is dominated by natural events, temperature excursions being only one of them, then, if we wait long enough, we can expect to see an overall drop in CO2 against an anthro-CO2 emission that does not drop. This would be consistent with what we now know about the magnitudes of natural CO2 sourcing and sinking compared with the much smaller magnitude of anthro-CO2 emissions.
So far, any temperature anomaly attributed to CO2 is undetectable against a background of natural temperature variability. This is the primary dilemma faced by those seeking a relationship between CO2 concentration and temperature via models. All predictions of higher temperatures caused by anthro-CO2 emissions are based on models. Unfortunately, no model has yet been shown to be reliable. One has no choice, therefore, other than to conclude that predictions of catastrophic temperature excursions driven by anthro-CO2 emissions are also unreliable.
The opportunity for bad science here is staggering! And politics has made it nearly impossible for the good science to win out. But it eventually will. It always does.