The only thing we can be sure of about the future is that it will be absolutely fantastic. So if what I say now seems to you to be very reasonable, then I have failed completely. Only if what I tell you appears absolutely unbelievable, have we any chance of visualizing the future as it really will happen.

- Arthur C. Clarke (h/t Brin)

Saturday, February 16, 2008

The First Big Surprise

A season or a few seasons colder or warmer than the trendline fall in the awkward gap between climate and weather. People who think about such things in terms of the particular peaks and troughs of the present and the near future tend to call themselves climatologists, but I think
they are doing something very different. Bill Gray is perhaps the best known of this type; he had a good heuristic for predicting hurricanes for a while. As the climate system slides or lurches into new configurations that approach will increasingly fail.

It's that slipping vs lurching that is on my mind this morning. Until whoever is funding the Heartland Institute and such mercifully pulls the plug on the denialist conspiracy, or until we actually encounter a year warmer than 1998, we will be hearing the "world is actually cooling since 1998" noise among the various distractors. (And after that we may well come around to hearing something like "the world has actually been cooling since 2011" for awhile...)

But what if the world actually cools for a bit?

Many places in the northern hemisphere are experiencing a cold winter. Not extraordinarily cold, sure, but cold enough that younger people accustomed to our recent half-winters might find it very striking. Does this event require explanation?

I'm concerned that it will fall under the "natural variability" radar for real climatologists (people interested in global dynamics and statistics), the tea-leaf-readers will simply conjure up some "oscillation" to account for it. Now the existence of oscillatory energy at multi-year time scales (mediated by ocean dynamics and possibly sea ice) is real enough, and the need for dynamic explanations real enough.

That all said, the possibility that an event is neither part of a gradual trend nor of a background oscillation can't be excluded a priori.

Now we understand the system well enough that we don't anticipate dramatic lurches in it without some mechanistic explanation. Is there a mechanistic explanation here?

Well, did anything unusual happen in the last few months? Yes, there was an abrupt, huge and unanticipated retreat of sea ice in the boreal summer. What's more, there was a comparably abrupt and huge recovery since. (I've seen a curve that shows it above the baseline mean for the first time in years; can't turn it up just now.)

We don't really have a mechanism for the rapid retreat, but consider the rapid rebound. What does that mean? Well as anyone who has lived in an icy climate can attest, ice melts faster in the presence of salt; i.e., it refreezes faster in its absence. The abrupt melting last summer must have freshened the Arctic allowing for an unusually effective freeze following. Does this portend a full recovery, or an increased seasonal cycle under greenhouse forcing, or something in between? Moderate that I am, I go for somewhere in between; some of the fresh water will have been exported into the Atlantic, and the forcing has not gone away.

Aha, but this brings us back to the original "unpleasant surprises in the greenhouse". Freshwater pulses into the Atlantic are associated by many paleoclimate theorists with severe cold anomalies. Can this mechanism account for severe winters in the north?

Note, this is mere speculation as of now. I would wait to see the scenario repeated before I would even endeavor to build a publication-worthy model of this. Please don't quote me without noting that I am speculating on my blog, not publishing.

If this makes sense it would be doubly unfortunate; I am sure even the most climate-aware people in the midwest are finding themselves wishing for "more global warming about now". Both the sense of urgency and the sense of confidence could be weakened by what in the grand scheme of things would amount to a delay in the temperature signal while extra heat is devoted to melting ice which causes a noticeable cooling signal.

In a larger sense, nature resists change. When you dig a deep hole, mud rushes to fill it in. Things that happen rapidly tend to kick off countervailing processes that are slower. Also, systems systems ring and wobble when you hit them. That's another source of increasing variability.

Neither of these principles operate magically. There are always physical mechanisms at work, and we need to identify them. The fact is, though, that the harder we perturb the system the weaker our ability to predict its responses will be. Consequently, surprises lay ahead.

Abrupt changes in biological systems notwithstanding, the abrupt retreat and advance of Arctic sea ice of the past year remains without known precedent and is perhaps the first unanticipated physical phenomenon of the greenhouse world. Whether my speculation about the relationship to the cold winter holds water or not, we are already into the realm of the unanticipated.

R. Pielke Jr. asks at what point the theory on which we operate might be falsified. I have to say I have a hard time answering. If this means falsifying the theory of the greenhouse effect, it's baffling. The question is sort of like what it would take to abandon the idea of gravity. It is pretty much incomprehensible to me how the theory of radiative transfer might be falsified without taking the whole of science down with it.

We are changing the radiative properties of the earth. In the long run we expect tremendous warming. In the meantime we expect a warming trend and maybe a bumpy ride. The appearance of bumps in the temperature record will surely be taken as evidence that we have nothing to worry about by irresponsible people grasping at far weaker straws than that, but rationally we should take no solace in the bumps.

5 comments:

Steve Bloom said...

Per this Cryosphere Today graph, the winter ice level will be higher than last year's record low, but in any case not terribly out of the line with the overall very slightly declining trend. It seems very unlikely to reach the 1979-2000 mean since the growth season will be over very soon. Last summer I recall various modelers saying they expect this sort of recovery to occur for some years even if the summer trend continues downward.

But these maps don't show thickness, and apparently the ice in the Arctic basin is behaving just like last year only more so: Most of the remaining thick multi-year ice (which is more resistant to melting) is streaming out now, which is setting things up for a new record. In another article a few weeks ago a scientist described the ice off of Banks Island (pretty high up in the Arctic) as behaving like "Styrofoam in a bathtub." See also this interesting and apparently unusual behavior (although I don't know how extensive it was otherwise).

BTW, Michael, in the second link there's a reference to what sounds like a model prediction that warming will force the AO into a more and more positive pattern (important because that's how the old ice gets flushed out), but with observationalists disputing that. Can you shed any light?

Anonymous said...

Michael,

you might want to look at the data of salinity of the Atlantic ocean and at the strength of the Gulf stream. But, I have to say, that the winter in Europe was very mild, not quite what one would expect from weakening of the gulf stream :-)

Yes, there was cold in parts of China and there was snow in Baghdad for the first time in modern history - but linking warming trend and weather anomalies is always difficult...

Alexander Ac

David Duff said...

Do I detect a Tobis oscillation? Such fun!

Anonymous said...

Here in Southeast Michigan we have had our share of cold and windy days. We've also had unseasonable warmth. This winter has been one of tremedous oscillation, with several snow dumps (5-12 inches) followed by complete melting. It's very rare for a foot of snow to melt in January.

Hank Roberts said...

Ward's been making the point that we don't expect the ocean circulation to halt, nowadays -- we expect it to get rearranged. His worst case ("Under a Green Sky") is that the area where surface water sinks moves southward, and so instead of cold oxygenated water sinking, the deep ocean gets fed a slug of warm, low-oxygen water, and Earth does anaerobic exercises for a while, to the delight of the bacteria generally.

I have no idea whether this is even being watched for (the ARGO system, maybe? or of course the US Navy, who certainly will have the data about areas of rising and sinking water, and temperatures, from submarine work but likely won't be disclosing it.

I wonder if there's a set of instruments the climatologists could specify, that the Navy could be convinced to carry on each working submarine --- even if the Navy won't promptly publish what they find, collecting the data would be invaluable. And trends could be published.

What would the scientists wish the submarines were detecting, every hour or every minute? (Which of such a list are they already tracking?)

pH, temperature, salinity/density, oxygen, methane, CO2, other dissolved gases? Tracer gases, there is a list of them from C14 to chlorofluorocarbons to other rare but highly detectable stuff that's entirely manmade. Coal plant output like mercury, thorium, uranium, would probably already be tracked through the atmosphere so the atmospheric scientists would know where it's falling out into the ocean, and the Navy could find out where it's going afterward.

I realize the ARGO and other systems ought to be doing this kind of thing eventually.

But -- nuclear submarines are out there now every day. The Navy's already doing climate modeling, quite high powered.

Is it possible to come up with a list of what the scientists know could be measured by the Navy on cruises?