Global sea surface temperature and model projections, with Bob Tisdale

A couple of days ago at WUWT Bob Tisdale posted a whole heap of charts of sea surface temperature and compared them to CMIP5 models (archived here). He was doing his usual thing of complaining the climate models don't model weather. He wrote:

The climate models used by the Intergovernmental Panel on Climate Change (IPCC) are not simulating climate as it exists on Earth. 

What he was really complaining about was that:

The multi-model mean of climate models used by the Intergovernmental Panel on Climate Change (IPCC) are not simulating climate as it exists on Earth. year to year variations in the weather as recorded by the older version of Reynold's OI v2 data, when looking at some sections of the ocean. 

Well, they aren't meant to do that. Climate models are for, you guessed it, climate not weather. The CMIP5 models are used for climate projections as well as learning more about particular aspects of climate. They are pretty good when it comes to global projections. Not as good (but not that bad) when looking at large areas like entire oceans. They are not intended to be used for weather forecasts. (Bob wants them to time ENSO events at the same time as they happen. That won't happen. There are other models specially developed for localised projections looking ahead a few months.)

There were some charts that were noticeable by their absence, so I figured I'd fill in the gaps in his article.