It is the trends that are being compared, and moreover trends
in backcasts using common forcing inputs, so use of ensemble means is certainly appropriate.
Does this suggest excessive tuning
in the backcast well its some evidence for it.
Not exact matches
However
in the first decade which GISS forecasts, rather than
backcasts, this opposite trend
in signals emerges.
I just give you data that shows, on the
backcast, the models, surprisingly, hit all the decadal trends since 1950
in a signal it was supposedly not tuned to match.
Here is the comparison of the
backcasts of global temperature from the Canadian model
in comparison with the observations.
# 308 When I asked Gavin Schmidt about this at RC he denied the role that tuning plays
in getting the
backcasts fit so well.
I'm having a hard time putting
in to words my meaning, but it's different than hind /
backcasting.
Smoothing the
backcasts first, to make the errors comparable to the ones they are criticising, results
in errors of + / — 0.3 dC.
In this chart, the red is the «
backcasting» of temperature history using climate models, the black line is the highly smoothed actuals, while the blue is a guess from the models as to what temperatures would have looked like without manmade forcings, particularly CO2.
There are numerous models that show a relationship, but the models omit more elements of atmospheric complexity than they include (and exclude any solar forcing with the exception of irradiance
in some models), they can not
backcast past ca 1900, they can not forecast next month, and they do generate greatly dissimilar forecasts when fed with the same assumptions of future conditions.