Seeker Blog

…Not all climatologists will fall prey to these temptations; many or most will modify their models, will see that mankind is not in as much trouble as originally thought, and move on to do work on, for example, the Indian monsoon.

Thanks to Roger Pielke, Jr. I recently discovered a new science blog by consultant and professor of statistics William Briggs. The particular post linked by Roger was this one, which examines the appropriate responses to the divergence between observed temperatures and model-predicted temperatures. As prof. Briggs put it “certainly the IPCC “scenarios” are too high, and they have been for more than five years”.

That last statement doesn’t look right — at least if Dr. Briggs is discussing the 1995, 2001, 2007 IPCC projections [the 1990 temperature projections were too high; the 1995 projections were much lower to agree with observations since 1990 — revising the models to incorporate aerosols; the 2001, 2007 projections moved slightly up]. Not to sidetrack you from Briggs’ useful discussion — but look at Roger Pielke’s Verification of IPCC Temperature Forecasts 1990, 1995, 2001, and 2007 from 14 Jan, 2008. I’ve included a thumbnail at left pointing to Roger’s two summary graphics. More on Roger’s conclusions at the end of this post. Back to Briggs:

The short answer, I will disappoint many of you by saying, is no. Like I wrote before, climatologists are generally nice people genuinely struggling with understanding the immense complexities of the oceanic-atmospheric (and space!) system. It might be that many of them are misleading themselves by custom tailoring models to show them what they expect (or desire?) to see, but this has not reached a level where it is done with intent. Most mistakes that are made are honest ones. And it is also true that much has been learned while examining climate models. Still, while scientists are in general noble creatures, there does exists the possibility of them sliding into the abyss.

So suppose, if you are able, that significant man-made climate change is false; further, that it cannot happen, and that all changes to the climate system are due to external forcings, such as those caused by changes in solar output. Just suppose all this is true for the sake of argument.

Now put yourself in the place of a climatologist, one of the many hundreds, in fact, who was involved with the IPCC and so shared in that great validator, the Nobel Peace Prize*. You have spent a career devoted to showing that mankind, through various forms of naughtiness, has significantly influenced the climate, and has caused temperatures to grow out of control. Your team, at a major university, has built and contributed to various global climate models. Graduate students have worked on these models. Team members have traveled the world and lectured on their results. Many, many papers were written about their output, and so forth.

But something has gone wrong. The actual temperature, predicted to go up and up, has not cooperated and has instead stayed the same and even has gone down. What do to? Let’s take a “What would a scientist do” quiz and find out.

Your model has predicted that temperatures will go up because CO2 has, but unfortunately temperatures have gone down. Do you:

1. Abandon the model and seek a new career
2. Discover where the model went wrong; publish results admitting why and how you were wrong
3. Sit and wait: after all, the temperature is bound to increase sooner or later, hence validating your model
4. Believe that the model cannot be wrong, else so many people wouldn’t believe it, and so posit some new source that is “holding back” warming, and only if that new source weren’t there, your model would be perfect.

The correct answer, it should go without saying, is (2), though (1) is not a horrible option for the shy, but it is really only open to beginning graduate students or professors reaching emeritus status. And if you do go for (2), as you should, option (1) naturally follows from it. (I must remind you here that significant man-made global influence is an impossibility by assumption.)

Would anybody opt for (3)? Certainly, because it’s the easiest thing to do, though not as many as you would think will go this route mostly because it would be too difficult to answer critics with a “Just wait and see!”

The slide begins with choosing (4). Nobody would, or should, abandon a well-developed model because an observation or two is not consonant with that model. Some time has to pass for enough failed predictions to mount up. How much time? That’s always difficult to tell. If the best climate models over-predict global temperature for a year, this is not cause for concern. For two years, no big deal. Even three to five years would not cause undue suspicion. But more than that, then something has gone wrong.

That is the state of the art today: climate models regularly over-predict temperatures; certainly the IPCC “scenarios” are too high, and they have been for more than five years. No climate scientist yet has gone to the quiz and opted for answers (1) or (2); several, of course, have opted for (3), saying five to ten years isn’t enough and that “more time” is needed. Nobody, that I know of, has said how much more time.

Has anybody gone for answer (4)? Yes. Already we are seeing papers — peer-reviewed, to be sure — that posit sources that are “masking” the true warming. So far, these papers are concentrate on aerosols, which are particles, caused by mankind naturally, that can, through various mechanisms, block incoming solar radiation and lead to cooling. Aerosol cooling only gets you so far, however, because aerosols are heavy, short-lived particles whose effects are actually easy to measure. So if models continue to over-predict, even after accounting for aerosols, some other source that “masks true warming” will have to be found.

Bob Park, physicist and resident curmudgeon at the American Physical Society, writes regularly on pseudo-science, and has identified “The Seven Warning Signs of Bogus Science.” Not all of these signs now apply to climatology, but number [3], “The scientific effect involved is always at the very limit of detection” is most relevant.

Since we haven’t detected the predicted warming, it must be masked or otherwise held up by something. Aerosols are one source, but an inadequate one, so another is needed. What will this source be? Of course, we cannot know for certain, but I can guess, though I blush when I do so: I predict it will be statistics…

I appreciated the link to Parks’ The Seven Warning Signs of Bogus Science — you’ll likely wish to archive that one for reference!

The connection between Briggs’ critique and Roger’s IPCC reconciliation is nicely explained in the conclusion of Roger’s first post:

What to take from this exercise?

1. The IPCC does not make forecast verification an easy task. The IPCC does not clearly identify what exactly it is predicting nor the variables that can be used to verify those predictions. Like so much else in climate science this leaves evaluations of predictions subject to much ambiguity, cherrypicking, and seeing what one wants to see.

2. The IPCC actually has a pretty good track record in its predictions, especially after it dramatically reduced its 1990 prediction. This record is clouded by an appearance of post-hoc curve fitting. In each of 1995, 2001, and 2007 the changes to the IPCC predictions had the net result of improving predictive performance with observations that had already been made. This is a bit like predicting today’s weather at 6PM.

3. Because the IPCC clears the slate every 5-7 years with a new assessment report, it is guarantees that its most recent predictions can never be rigorously verified, because, as climate scientists will tell you, 5-7 years is far too short to say anything about climate predictions. Consequently, the IPCC should not predict and then move on, but pay close attention to its past predictions and examine why the succeed or fail. As new reports are issued the IPCC should go to great lengths to place its new predictions on an apples-to-apples basis with earlier predictions. The SAR did a nice job of this, more recent reports have not. A good example of how not to update predictions is the predictions of sea level rise between the TAR and AR4 which are not at all apples-to-apples.

4. Finally, and I repeat myself, the IPCC should issue predictions for the future, not the recent past.

Roger has posted a few updates based on very recent [and minor] corrections to some of the data. The changes are only of interest to specialists, but can be found with this search.