Tuesday, July 03, 2018

Hansen 1988 Retrouvé

Celebrating the 30th anniversary of the 1988 publication of Global climate changes as forecast by Goddard Institute for Space Studies three-dimensional model Hansen, J., I. Fung, A. Lacis, D. Rind, S. Lebedeff, R. Ruedy, G. Russell, and P. Stone Eli along with others (like Gavin Schmidt, Nick Stokes and others) Eli went back and took a look at what had been discussed in these parts as well as those.  There are a lot of very good people who contributed to that paper, so it henceforth hereabouts shall be known as HFLRLRRS.

One of the most important things is that the current discussion focuses on the predictions of forcing as well as outcomes.  The first is more or less ecomonics, the second physics.

That being the case, HFLRLRRS were pretty good on both.  Back in 1988 Hansen described the choice of scenarios
For the future, it is difficult to predict reliably how trace gases will continue to change. In face, it would be useful to know the climatic consequences of althernative scenerios. So we have considered three scenarios for future trace gas growth, shown on the next viewgraph. 
Scenerio A assumes the CO2 emissions will grow 1.5 percent per year and that CFC emissions will grow 3 percent per year. Scenerio B assumes constant future emissions. If populations increase, Scenerio B requires emissions per capita to decrease. 
Scenerio C has drastic cuts in emissions by the year 2000, with CFC emissions eliminated entirely and other trace gas emissions reduced to a level where they just balance their sinks. 
These scenarios are designed specifically to cover a very broad range of cases. If I were forced to choose one of these as most plausible, I would say Scenario B. My guess is that the world is now probably following a course that will take it somewhere between A and B
In 2006, Eli pointed out that the CO2 prediction was eerily accurate

It shows that many details in emission scenarios are unimportant, or rather that false estimates in one direction are most likely going to be cancelled by false estimates in the other for a different forcing agent, and that on average the scenarios should be useful for larger periods.
Which is what happened using the illustration from Real Climate

A neat thing in HFLRLRRS which has not been commented on much were the color coded maps of temperature anomalies by decade in the future which can be compared to measurements. Eli finds this much more interesting and informative about the early GISS model than the time one dimensional variation of averate global temperature change.  Comparing the result for the 2010s with measurements between 2014-2016 show that HFLRLRRS captured the broad picture but was not perfect

Arctic amplification is clear in the model and 30 years later in the observation.  Otoh, the Antarctic in the model is clearly too warm as is the Arctic south of Greenland. The third pole (the Tibetan Highland is too warm also. It would probably be better to map out more of the current decade wrt the El Nino - Southern Oscillation, 2016 being a monster El Nino, but the period also includes about a year of La Nina.
Another, not remarked much upon thing is that  HFLRLRRS calculated the temperature change compared to the interannual variabilty of their 100 year control run.  The change is everywhere (with the exception of the Indian Ocean off Perth) positive compared to the null of the control.  What would be interesting (to Eli) to compute would be the model interannual variability for the 2010s compared to that of the control run.  Indeed were Eli in charge, CIMP runs would include such a metric.


Everett F Sargent said...

I think that Hansen88 needs a forcing correction per ESRL AGGI ...

CO2 is close, between A and B
N2O is below A and B and above C (say ~ halfway between A and C)
CH4 is below C in its entirety
CFC11 is close to C but slightly below C for future projection
CFC12 is very close to C
Some groups include O2 (ozone) which gives a CO2e somewhere north of 500 ppmv today.

JC@CE has a post by McKitrick and Christy and they suggest not making the forcing adjustment ...

"Second, applying a post-hoc bias correction to the forcing ignores the fact that converting GHG increases into forcing is an essential part of the modeling."

Which actually supports making a forcing correction, use the forcing that actually happened, not Hansen's higher than actual forcing estimates. As the models are quasi-linear to begin with in the 1st place.

jb said...

What would be interesting (to Eli) to compute would be the model interannual variability for the 2010s compared to that of the control run.

You might find this interesting: Yettella, et al, 2018, "An Ensemble Covariance Framework for Quantifying Forced Climate Variability and Its Time of Emergence".

Everett F Sargent said...

Cross posting this from moyhu (Nick Stokes) ...

I've also been given the impression, by Gavin, several posts now, that the actual forcing to date are lower than the CMIP5 forcing as Gavin shows in this figure ...


In 2006, Eli pointed out that the CO2 prediction was eerily accurate

Nice job- Eli, but why nobody calls liinear extrapolations eeiry when they beat apocalyptic predictions on accuracy is a semiotic puzzlement.

Mike said...

How do Hansen et al's projections compare with projections made by others around the same time? Did any of the folks sometimes described as deniers make any public projections or forecasts?