The climate model we employ has a global mean surface air equilibrium sensitivity of 4.2 C for doubled CO2. Other recent GCMs yield equilibrium sensitivities of 2.5-5.5 C.....
Forecast temperature trends for time scales of a few decades or less are not very sensitive to the model's equilibrium climate sensitivity (reference provided). Therefore climate sensitivity would have to be much smaller than 4.2 C, say 1.5 to 2 C, in order for us to modify our conclusions significantly.Eli pointed out that
We are getting to the point, twenty years on, where the high estimate of climate sensitivity is making itself felt. OTOH, the 1988 paper estimated the forcings slightly on the low side. The result was a pretty good prediction. Definitely in the class of useful models.We are also at the point where Scenario C, the one where CO2 went to zilch in 2000, would, according to the 1988 model, plateau, but one would expect that this far on the 1988 predictions would begin to look a bit too high because of the large climate sensitivity.
So in the spirit of the thing, allow Eli to plot the 1988 model data, let's throw in the GISSTemp temperature anomalies, as was done in the 1988 paper, basing the anomaly on the 1951-1980 period.
UPDATE: As any thinking bunny might expect, for this Nick Stokes has a better gimmick, but he should work on the background. Eli understands that Pat and Chip could loan him their Photoshop.
Ah, is it Lint again, already? Careful; it'll clog your pipes and set your hair on fire.
ReplyDeleteSo, when we compare the early predictions by Hansen to the GISS temperature data adjusted by Hansen we get a good match.
ReplyDeleteHardy Cross
Waldo observes:
ReplyDeleteWhen I caclulate linear trends for the scenarios and data since 1979, I get:
HansenA: 3.15 K/century
HansenB: 2.83 K/century
HansenC: 1.97 K/century
GISSTEMP: 1.57 K/century
HadCRUT3: 1.47 K/century
UAH LT: 1.35 K/century
RSS LT: 1.34 K/century
Had SST: 1.25 K/century
RSS MT: 0.79 K/century
UAH MT: 0.46 K/century
For the MSU era, all observations appear to be warming a rate less than scenario C.
Hansen originally used a met stations only index (now GISS Ts) in his comparison plot. There's a case for continuing to use it on that basis.
ReplyDeleteIn any case, I made a gadget here for trying out different datasets.
ReplyDeleteAnonymous said...
So, when we compare the early predictions by Hansen to the GISS temperature data adjusted by Hansen we get a good match.
Hardy Cross
Hardy,
If you download the GHCN raw data used (but not owned, controlled, or manipulated) by Hansen/GISS and run it through even a crude hand-rolled temperature anomaly gridding/averaging program, you will get results surprisingly close to the results published by NASA/GISS.
No manipulation, homogenation, or adjustments required -- just run the raw data through what is basically a glorified averaging procedure and you get Hansen's results.
Problem: Pinatubo
ReplyDeleteAnswer 1: Compare obs scrubbed of non-modeled events. Foster and Rahmstorf 2011.
Answer 2: Rerun GISS ModelII ca 1988 with 1988-2011 forcings.
Prefect anticipation!
ReplyDeleteAnd thanks for the kind words. I did rather like the '80s grunge on the background plot.
Dear Eli, would you now be so kind as to predict tonight's Euromillions lottery numbers with similar accuracy. Promise to send some of the winnings your way. Thanks in advance.
ReplyDeleteRon,
ReplyDeleteBecause it fell in the first half of the MSU era,
Pinatubo and El Chichon actually amplify temperature trends
above what they would otherwise have been ( because they
cooled the early part of the record and then stopped ).
More El Ninos in the '90s and more La Ninas in the '00s
provides a better excuse.
But I am amused at the new affection towards Foster and Rahmstorf.
Volcanoes, ENSOs, and solar fluctuations are part of reality.
They are part of the past, and they will be part of the future.
In contrast, Foster and Rahmstorf are imaginations of what might have been.
Waldo
Nick reminds me that Scenario B & C include a volcano in 1995 and 2005 and 2015. My only fondness for F&R is that I thought (mistakenly) that it might decrease the gap between real world forcings and scenario forcings. We might be able to salvage the approach though, take scenario forcings, subtract real world forcings, and use the difference to 'correct' the obs to better test the skill of the model.
ReplyDeleteI still prefer option B. Check GISS modelIIv1988 out of cvs, feed it the real world forcings from 1988-2011 and test against real world observations. If I can just find the cvs repository ... :D
Waldo,
ReplyDeleteYou seem to be rather unclear on the purpose of a scientific model--namely to explain as many observations with as few causes as possible. That F&R manage to show consistent warming when taking into account 3 known and well validated forcings is actually quite significant. In searching only for evidence that confirms your prejudice, you miss that.