In his new position at the Washington Post, Chris Mooney has a great new piece about the scientific consensus represented by the IPCC. Using a new study, and older ones, Mooney lays out the peer-reviewed case for why and how the IPCC sometimes underplays the risks of climate change.
The IPCC is conservative for a number of reasons, but Mooney narrows in on sea level rise as an example. In describing the worst case scenario for sea level rise, Mooney shows how the IPCC focuses on what will be the "likely" range. This ignores the 17% chance that it could be much worse, meaning the actual worst case scenario is even worse than the high end of the given range, which is 2.69 feet. Further, the 2.69 figure is JUST the mean increase expected between 2081-2100. The actual amount, not the one reported on by media, of sea level rise by 2100 is 3.22 feet. Continuing, the IPCC's estimate only takes into account a type of model run, and not expert opinion. One study (.pdf) taking into account expert analysis of different model projections shows an even higher worst case of 3.93 feet!
So while deniers will rant and rave about the alarmist IPCC, the peer-reviewed literature actually suggests the opposite: that if anything, the risks of climate change could be greater than what the IPCC overtly describes.