Here's the trend for the last ten years - it's flat: http://www.woodfortrees.org/plot/hadcrut4gl/last:120/plot/hadcrut4gl/last:120/trend http://www.woodfortree

Here's the trend for the last ten years - it's flat: http://www.woodfortrees.org/plot/hadcrut4gl/last:120/plot/hadcrut4gl/last:120/trend http://www.woodfortrees.org/plot/hadcrut4gl/last:120/plot/ha... The trend has been flat enough for long enough to falsify all the model predictions that assumed CO2 was the primary driver and nothing else really mattered. Those predictions looked GREAT in the late 1990s but don't work now, so the hunt is on for new explanatory factors. Everybody's got their favorites. The main thing to consider with regard to new years being dubbed "the hottest on record" is selection bias: When a new year looks like it MIGHT be "the hottest on record", that makes for big scary newspaper headlines. Then when it (often) turns out that it actually WASN'T the hottest on record, that's not considered newsworthy. Nor is it newsworthy when the following year is only the 6th or 10th or 12th hottest. If you only see headlines about it when there's been a record or somebody has predicted that there WILL be a record, due to availability bias you are left with the impression that records are happening constantly. (And even when the whole year isn't the hottest, the same dynamic applies to hottest summer, hottest winter, hottest in the northern hemisphere, etc.) Another thing to consider is that although the current trend is essentially flat (and has been for over a dozen years), it's flat at a relatively high level. Since the temperature plateaued but hasn't yet significantly decreased, people can still claim it's a hot decade or invent silly "X of the last Y years" constructions to make it seem like the warming trend hasn't stopped.