After a bit more investigation, motivated by comments on my earlier post on Jim Hansen’s 1988 predictions (thanks all), it turns out that Jim Hansen has two sets of scenarios labeled A, B, and C. This is confusing to say the least. The conclusions of the earlier post remain unchanged, however, the analysis below may help to explain some things.


The first set of scenarios A, B and C are from his 1988 paper and the details can be found in the appendix to that paper. The second set of scenarios named A, B, and C was apparently generated for his 1998 paper in order to rescale emissions projections based on the lack of realism in the 1988 projections. The earlier post of mine therefore is actually a comparison of the 1998 scenarios with what has transpired since, and in that case it is true that Scenario C (let’s call it C98) is clearly the most comparable to how things have evolved since 1998.

For his 1988 paper here then from its appendix are the relevant annual growth rates for each scenario relevant to the time period 1990-2005, which I am now labeling with the year affixed for clarity:

Scenario A88

CO2 1.5%
N2O 0.2% 1990-1999, 0.4% >2000
CH4 1.5%

Scenario B88

CO2 1% 1990-1999, 0.5% >2000
N2O 2.5% 1990-1999, 1.5% >2000
CH4 1.0% 1990-1999, 0.5% >2000

Scenario C88

CO2 1.5 ppm 1990-2000, fixed at 368 ppm after 2000
N2O 0.0%
CH4 0.5% year 1990-2000, fixed at 1916 ppm after 2000

How do these scenarios compare? From the earlier post here again are the actuals:

CH4 0.46%
N2O 0.95%
CO2 -0.52% (Annex 1 Countries 1990-2003)
CO2 >0.45% (Non-Annex 1 Countries 1990-2003) (note typo caught in earlier post)
CO2 1.5 PPM (1990-2005)

How to evaluate these scenarios?

From 1990-2000 Scenario C88 was right on target with respect to CH4 and CO2, and off with respect to N2O (The description of Scenario C88 actually doesn’t mention N2O in the appendix describing the growth rates, so I assumed that it was zero).

Scenarios A88 and B88 were off by factors of 3 and 2 respectively with respect to CO2 and CH4. With respect to N2O Scenario A88 was by a factor of 0.2 too small and B88 was a factor of 2.5 too large. These are large errors, much larger than those documented with respect to the 1998 projections, as would be expected.

So, there are several conclusions to draw from this exercise.

1. With respect to Jim Hansen’s 1988 predictions, his Scenario C88 was the most accurate with respect to emissions 1990-2000. That scenario froze time in 2000, meaning that going forward there are two evolving scenarios which both have dramatically overestimated emissions. The lower of the two is thus “more accurate” than the other. Neither is particularly accurate or realistic. Any conclusion that Hansen’s 1988 prediction got things right, necessarily must conclude that it got things right for the wrong reasons.

2. With respect to Hansen’s 1998 predictions C98 has been the most accurate.

3. In two sets of predictions compared with experience, Jim Hansen’s predictions of emissions have proved to be overly aggressive both times with respect to rates of emissions with his lower estimate proving most accurate.

4. When discussing Hansen’s scenarios make sure that you differentiate between 1988 and 1998 versions.

More comments welcomed!