The abstract seems to be a weasle worded cop-out suggesting that the previous models failed to account for something and they now accounted for it, it being "internal variability" (whatever that is) that will ameliorate climate change (what we are seeing isn't as bad as the original "end of the world" scenarios predicted, but which still allows them to retain some of the "the end is nigh!" spirit.
On the other hand, the Reuters journalist has cherry picked the alarming bits to declare:
"Global warming is forecast to set in with a vengeance after 2009, with at least half of the five following years expected to be hotter than 1998, the warmest year on record, scientists reported on Thursday."
Which seems to me to be egging the pudding a great deal.
Apart from the Reuter's person being seemingly oblivious to the recent changes in the order of alleged record hottest years (Number of the Month for August) the Met Office claims mean that they have set themselves up, as pointed out elsewhere by people who better understand these things than I, to be a first 'real' test case of the skill of a GW model. At least in so far being able to measure the outline forecast against the next ten years rather than some variable point in the middle to distant future.
So will this induce panic among the eco-controllers and Gore followers? Is that why the forecast is reported the way it is - to give impetus to the enforcement of draconian legislation as quickly as possible before the true situation is slightly revealed which may put at risk their strategy for economic control? ("Slightly revealed" on the basis that the result will be spun and in any case the results of one assessment are hardly enough to confirm absolute decisions about accuracy either way.)