Apparently this study was published in Nature magazine which I can't seem to find; have you read the study itself or just this copy and paste article drawn from the Press Room at Columbia University?
For reasons best known to the loudest of the GW crowd all charts now seem to start with 1970, a well-known year of low temperatures having dropped steadily from 1940, just as the above graph.
I didn't read the actual article. I read the article I posted from TerraDaily. It's in the May 15 issue of Nature; I can see it but I can't access it (have to pay).
As for the starting date, it seems generally agreed that climate variability combined with sulfur emissions to lower temperatures a bit in the 60s and 70s. There also seems to be general agreement that the warming beginning in the 1970s is mostly attributable to anthropogenic cause. For example of why I say that: