On whether Arctic Sea Ice, once lost, could recover
Mike MacCracken
tipping point with respect to Arctic sea ice--that is, on whether, once
melted, Arctic ice might stay melted and not be recoverable if radiative
forcing was brought back down to, say, the present level or below. The Ewing
and Donn ice age hypothesis [see Science, 15 June 1956, Volume 123, Number
3207 for first of three articles] is one place where this is suggested--they
postulate that the Arctic would stay open for some thousands of years even
after glaciers had started to form around the Arctic.
It seems to me that the recent article in Science [Svend Funder, et al.,
2011: "A 10,000-Year Record of Arctic Ocean Sea-Ice Variability--View from
the Beach" Science 333, 747-750] now provides a pretty clear historical
analog indicating that a reduction in the energy balance for the Arctic
would indeed lead to a return of Arctic sea ice. Basically, the article
provides shoreline evidence that the Arctic was quite open roughly 6000 to
8500 years ago and since then has become more ice covered. The article's
analysis focuses on how winds and circulation may have played a role in this
and then in its last rather long paragraph talks about the strengths and
weaknesses of results from model simulations of this period in representing
what the reconstructions show.
What is very surprising about the article is that it does not discuss the
significance of the changes in the orbital parameters at this time. If one
goes to the "Insolation" article on wikipedia, however, one can find a plot
of the time history of peak summer radiation at 65 N (I'd prefer to have the
summer integral and for the Arctic area, but this will do to make the
point). The caption to the figure in wikipedia comments: "Past and future of
daily average insolation at top of the atmosphere on the day of the summer
solstice, at 65 N latitude. The green curve is with eccentricity e
hypothetically set to 0. The red curve uses the actual (predicted) value of
e. Blue dot is current conditions, at 2 ky A.D."
What happens to the solar radiation over the Holocene was commented on by
Robert Tulip at
http://www.bautforum.com/showthread.php/109120-Insolation-and-Global-Warming
saying: "[The] Wikipedia chart for insolation shows that solar watts per
square meter on the summer solstice at 65N latitude has decreased over the
last ten thousand years by about ten percent from 530 to 480 w/m2, and will
increase to about 490 w/m2 over the next few thousand years. As explained at
Insolation Application to Milankovitch cycles, this cycle is a product of
the interaction between precession of the equinox, orbital ellipticity and
obliquity."
While there are lots of complications about how changes in the solar flux
distribution over season and latitude might affect the global climate
[remember that Milankovitch cycling does not change the annual integral of
top of the atmosphere solar radiation--the cycles only redistribute it by
season and latitude--and then variations in albedo have their influence as
well], the decrease in solar radiation shown by the plot gives a sense of
the likely relationship--namely, with 10% more solar radiation, one would
have a lot less sea ice, and it appears that there is apparently not some
large hysteresis as the sea ice extent expanded as the level of solar
radiation was coming down to its present level.
I would thus suggest that, consistent with Caldeira and Wood and with
simulations that we have undertaken since then and are in review for
publication, it would seem that, while there are uncertainties on timing and
rates, if the amount of energy going into the Arctic can be reduced, the sea
ice will return and there is not some very large hysteresis that would make
that impossible.
It thus seems to me, as I have urged elsewhere, that research (and
development) determining the potential for reducing solar radiation into the
Arctic as a way of limiting climate change in that region and beyond is
justified by the amount of change occurring and, given the rates of decrease
in Arctic sea ice, the mass of the Greenland Ice Sheet, and the extent of
permafrost, research and development (covering disciplines from physical to
ecological to social to political science) should be proceeding far and fast
enough so that governments can consider possible deployment of such an
approach on a decadal time scale. And as to how it should be organized, I
would favor both an integrated effort focused on addressing this particular
objective (and maybe part of an broader risk reduction focused
effort--focused efforts should have clear objectives) and a widely
distributed effort for gaining general knowledge and critiquing the focused
effort--and then also an assessment effort to provide the public and
decision makers an independent view for use in their decision-making
efforts.
If such a focused effort is not made, especially given the slow pace of
international negotiations, it would seem we will need to just give up on
having a cold Arctic, and that will mean very significant environmental and
social impacts for the region, for the weather of much of the Northern
Hemisphere mid-latitude areas, and for coastal areas around the world as the
pace and long-term duration of sea level rise kicks in.
Best, Mike MacCracken
Andrew Lockley
Mike
Thanks for informing the debate. However, in 'tipping elements in the earths climate system" lenton notes the vulnerability of Greenland and also of cryosphere methane.
The fact that sea ice extent appears not to be a tipping element should therefore not be seen as a significant reassurance on climate stability as regards the arctic in general.
During the recent banff geoengineering summer school, I noted David Keiths apparent confidence that SRM geoengineering has no practical upper bound of capability, with this recalled statement (paraphrased) sticking in my mind: "you could get a snowball earth if you wanted it." I am interested to know whether this is widely agreed. Or, to put it another way, is geoengineering itself a potential tipping point, with its capacity to stabilize climate coming to an abrupt end at some scale or temporal limit?
The answer to this question potentially informs substantially the debate on the possible futures for the arctic region.
A
On 8 Aug 2011 18:09, "Mike MacCracken" <mmac...@comcast.net> wrote:
Mike MacCracken
On whether there is an upper limit to geoengineering, a few comments. Clearly, if you invested enough in a satellite deflector at the L1 point, you could cut solar radiation enough to get a snowball Earth. And maybe one could with mesosphere particles David has talked about. Whether even theoretically one could get a snowball Earth with stratospheric sulfate seems to me not at all clear for the more sulfur you put into the stratosphere, the larger the particles become and so the lower their lifetime. With very great loadings of soot, the nuclear winter studies did show that if you absorbed all solar radiation in the stratosphere (so above virtually all of the greenhouse gases), then the surface could get pretty cold (sort of isothermal below the stratospheric cloud).
On whether the Earth would stay as a snow-covered Earth with present solar radiation if once gotten there by some sort of geoengineering, I am pretty dubious (based on my early simulations of the situation back during my dissertation research). I think most of the modeling studies that have suggested a snowball Earth would stay that way have not had a diurnal cycle, and the low thermal transmissivity of snow means that it will melt in the noonday equatorial Sun even with a reasonably high albedo (say 80%). And once it melts and refreezes at night, it is ice, and that has a lower albedo (say 50-60%), so I am dubious of model simulations with 24-hour average solar radiation and that keep a high albedo.
Now, ice does have a higher thermal conductivity, so it has the potential to transmit heat into it and then back out to radiate away at night, so an ice-covered Earth becomes a more interesting question—but doing so with 60% albedo or so is again pretty hard to sustain. I’d also note that if one were to freeze the oceans so no new snow (and then ice), it would all likely get dirty (from volcanic ash, meteoric dust, etc.) and the albedo goes down—and once the albedo is down a bit, it is really hard to sustain the full cover, so one spot starts melting, and then albedo feedback would keep it going.
Now, can one engineer the climate enough so elevated land even at low latitudes would have at least seasonal snow cover? Quite probably so, given the Little Ice Age, but that is different than a truly snow-covered Earth.
As for me, I am much less worried about risk of engineering ourselves into a snow- or ice-covered Earth as compared to not doing engineering and ending up very warm.
Mike
John Latham
We do not yet know whether we'll be able to satisfactorily resolve all
remaining technological and other questions relating to the viability
of Marine Cloud Brightening (MCB).
We can say, however, that if this SRM technique functions as is assumed
in our GCM computations, the application of MCB in a 2xCO2 environment
could roughly restore the sea-ice coverage to current values. Please
see Rasch et al 2009 (attached) for more details.
Cheers, John. lat...@ucar.edu
John Latham
Address: P.O. Box 3000,MMM,NCAR,Boulder,CO 80307-3000
Email: lat...@ucar.edu or john.l...@manchester.ac.uk
Tel: (US-Work) 303-497-8182 or (US-Home) 303-444-2429
or (US-Cell) 303-882-0724 or (UK) 01928-730-002
http://www.mmm.ucar.edu/people/latham
________________________________________
From: geoengi...@googlegroups.com [geoengi...@googlegroups.com] on behalf of Andrew Lockley [andrew....@gmail.com]
Sent: Monday, August 08, 2011 7:08 PM
To: geoengineering
Subject: Re: [geo] On whether Arctic Sea Ice, once lost, could recover
Mike
A
On 8 Aug 2011 18:09, "Mike MacCracken" <mmac...@comcast.net<mailto:mmac...@comcast.net>> wrote:
--
You received this message because you are subscribed to the Google Groups "geoengineering" group.
To post to this group, send email to geoengi...@googlegroups.com.
To unsubscribe from this group, send email to geoengineerin...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/geoengineering?hl=en.