The graph shows recent CO2 changes. Top are the CO2 themselves (monthly, plus 13-month running mean); bottom are change-on-12-months earlier. Data from Mauna Loa; sadly I can only find them up to end 2004. Which means I can't provide a decent answer to SB, who asked about reports that this years CO2 rise was 2.2 ppmv. However, even if it was 2.2 (can anyone find a report, or better still the data?), that looks to be nothing surprising, in view of the past. And its clearly lower than the 1%/y assumed by some of the idealised scenarios. When I find the SRES data I want to look at how the various scenarios look around now.
I find it hard to get too excited about the exact levels for any one year; what matters are the long term trends; for the future, those are clearly Upwards, but by how much depends on economics that are rather tricky to forecast. But T rise (etc) can largely be scaled to CO2 level, rather than date. James Annan (ref lost) pointed out that methane levels seem to be well below SRES trends (indeed this shows them to be currently flat; whether that will continue I don't know).
Meanwhile, JF has a post sort-of about this too, with a vigorous exchange of comments.
[Update: And John Quiggin has a post too. Pure co-incidence... he too appears to be on the it-does-make-much-difference idea, but with the advantage of understanding the economics]