2011-08-22 08:02:23Lessons from Past Climate Predictions: IPCC FAR
Dana Nuccitelli

Next entry in the lessons from past predictions series.  Figure 4 is the money shot.  Really damn impressive projections if you ask me.

Lessons from Past Climate Predictions: IPCC FAR

2011-08-22 12:56:27
Sarah Green

Hi Dana,

I'd suggest extreme caution with your sentence:

we can adjust the IPCC projections to approximately reflect the observed atmospheric GHG changes since 1990...

"adjusting" is a red flag that obscures rationality. Perhaps say "using the original 1990 model but known GHG data over the last 20 years..."

There are actually several intemingled projections here: (i) GHG emissions, (ii) radiative forcing (including aerosols), (iii) T response. The first is most dependent on human activities, and this hardest to project.  Anyhow, I'd point out that that's the main thing that was mis-projected (but not by much- it's quite remarkable, actually). Do you have a sense of how much of the lower than projected GHG emissions are due to the collapse of the soviet union and the more recent global economic downturn? 

The dates should be made more obvious to keep emphasizing throughout these are 20-year old projections. e.g.:

the projected BAU GHG radiative forcing in 2010 was...

Could be:

the radiative forcing projected by scientists in 1990 for conditions in 2010 was...



2011-08-22 13:24:27
Dana Nuccitelli

Thanks Sarah, good comments.  I made a few changes.

2011-08-23 06:06:19I'm gonna go with a Judith Curry "wow" here
John Cook

Figure 4 is indeed the money shot and needs to go in our graphics page. But first, echoin Sarah's comment, I'd take "adjusted" out of the title. Perhaps call it something like "IPCC 1990 Projections with observed GHG emissions" (both in the title and legend).

This is powerful and useful stuff, Dana. I see this culminating in a very powerful presentation being a sequence of physics based predictions (FAR, Broecker, Hansen, TAR, etc) compared to a series of climastrology predictions, with a title like "Physics vs Climastrology".

2011-08-23 06:55:37adjusted
Dana Nuccitelli

Thanks John.  I've thought a bit about how to compare the good vs. bad predictions at some point in the future.  Maybe just a post showing the various graphics one after the other.

But for now I'll modify the title, and I'll send you the Excel file tonight so you can add it to the graphics page.  Actually I think I'll delete the 'low' and 'high' too, because the way it's drawn is deceptive, since technically they should start at the same temperature in 1990.  They don't because the IPCC started the projections way back in 1850 and then ran them forward.  So I guess another option would be to include the hindcast in the comparison.   I'm not sure if I've got all the digitized data from 1850 though.

I'll play with it a little tonight, but the best option might be to just show the 2.5°C 'best' projection starting in 1990, which is basically the 'money shot' sans the low and high estimates.

2011-08-23 07:47:23
Mark Richardson

Have you or are you going to discuss the bits where current understanding falls down, and why?



E.g. climate models are pretty poor at putting precipitation in the right place and time. Or at least, old ones have been (dunno about the latest ones).

But they appear to be doing a good job of getting global temperatures about right, and we expect them to do a much better job because it's a different problem*!




*sure, water cycle is very important to the heat flow balance, but Soden & Held (2005?) covered why this probably doesn't matter too much pretty nicely.

2011-08-23 08:30:29
Dana Nuccitelli

Well, I've been focusing on temperatures, because it's the most common type of prediction (even a few 'skeptics' stuck their necks out with temperature change predictions), and easy to evaluate.  I can add a note that climate models aren't as good at projecting precipitation changes, but I wouldn't want to go into too much detail on this, since the focus is mainly on temps.

But if anyone wants to add to the series with some other types of predictions - precipitation, sea ice, glacier retreat, or what have you - that would be great.  Any climate predictions fall within the scope of the 'Lessons' series.

2011-08-23 14:04:27
Dana Nuccitelli

Post updated.  I didn't add anything about precipitation because it didn't really fit.

2011-08-23 19:06:42
Rob Painting

Typo under figure 5 - "Now we see that had the iPCC FAR" = IPCC

And again in next paragraph - "iPCC "best" model equilibrium sensitivity of 2.5°C may be somewhat too low."

Quite 'jammy' how the projections, over the last decade, match the observations so well.

Looks good.

2011-08-24 02:14:06
Dana Nuccitelli

Thanks Rob.  You would think if there were a letter I wouldn't have a problem capitalizing, it would be "I"!

2011-08-24 02:19:58
Alex C


>>>The FAR greenhouse gas (GHG) radiative forcing and CO2-equivalent for the scenarios are shown in Figure 1.

>>>In the B, C, D scenarios, the  projected 2010 forcing was nearly 3 W/m2.  The actual GHG radiative forcing was approximately 2.8 W/m2, so to this point, we're actually closer to the IPCC FAR's lower emissions scenarios. 

A question, was the overestimation of radiative forcing due to a given GHG concentration due to overestimation of other GHG concentrations as well, or of a primitive (for lack of a better word off the cuff) understanding of the relationship between concentration of CO2 and forcing?  For example, was a different coefficient factor used instead of Myhre et al's 5.35 (which I recognize is from a study eight years after FAR was published)?  I don't think it necessary to discuss this, just a bit curious.  I also note that this would have a somewhat canceling effect with the lower estimated equilibrium climate sensitivity.  You later discuss possible factors not taken into account with the model that could imply a larger ECS - the overestimation of radiative forcing would be another factor, and while you say it's not a pertinent matter how well they projected *concentration* (and it's not), it is somewhat important how they projected radiative forcing, as those projections are not only based off of a guess to future CO2 levels but also what the physical relationship was best estimated at the time.


"The actual GHG radiative forcing was approximately 2.8 W/m2"

I'm getting 2.6-2.7 for the greenhouse gases, up to 3 if you include the ozone.  Long lived GHGs only, I get a total of 2.64 W/m^2.  You sure, or are you taking into account the new concentrations even after AR4 was published?

Looks good otherwise.

2011-08-24 03:53:24
Dana Nuccitelli

Good question Alex, I'll have to go back and look, but I think they overestimated the increases of other GHGs as well.  I recall reading that until recent years, methane didn't increase as much as climate scientists expected, for example.  But I'll see if I can find their CO2 forcing formula.

The 2.8 I estimated as an update from the IPCC AR4, since which time the CO2 forcing has increased about 0.17 W/m2, if I recall correctly.  Or maybe it was 0.11.  Anyway I'm also just eyeballing the FAR forcing graph, so none of this is terribly precise.