A Current Comparison of Models and Observations

As you probably heard by now, GISTEMP measured June as the warmest June on record. Also, they have updated their methodology -- you can read more here. Basically, they updated their land and sea surface numbers to the new versions put out by NOAA in the recent Karl et al paper in Science. (Reminder to people who don't like new data -- the Karl et al changes are smaller than the recent UAH changes to their lower troposphere numbers.)

With that, here's the model-observation comparison Gavin Schmitt of NASA GISS tweeted the other day (though I'm not sure why he wouldn't use CMIP5):

So here are model predictions of temperatures from the early 2000s (grey shading is mainly weather uncertainty): pic.twitter.com/vgnMM5iT0I

— Gavin Schmidt (@ClimateOfGavin) July 16, 2015

Which comes with the usual caveat:

(Note that these climate models are not initialized to observations, and cannot predict phase of El Niño etc. ahead of time).

— Gavin Schmidt (@ClimateOfGavin) July 16, 2015