Background-Greenland collapse
Andrew Lockley
https://mobile.reuters.com/article/amp/idUSKCN25A2X3
Douglas MacMartin
What is not correct in the media report is this sentence: “This process, however, would take decades.” Well, I guess arguably that’s true, it’s just it would take a LOT of decades. Melt rate is currently of order 1-2mm/yr equivalent SLR, so to get the 6m from melting all of Greenland would take a few thousand years. Obviously it can speed up a lot, but “hey, it’s losing mass” does not remotely imply “therefore we only have a few decades before we lose our coastal cities”. So no, you can’t use this study to claim that geoengineering is required to keep our coastal cities. The problem with relying on mitigation+CDR is time-scale, but this study doesn’t prove that our response time-scale needs to be faster than what CDR can (at least hypothetically) provide.
d
--
You received this message because you are subscribed to the Google Groups "geoengineering" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
geoengineerin...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/geoengineering/CAJ3C-07%2BJc0dmp26W2_H08Rsrs1V_sjEs1pkG6xuZZ75OcTa%2Bw%40mail.gmail.com.
Andrew Lockley
Stephen Salter
Hi All
The attached note has
estimates for ice loss numbers for Arctic ocean ice, Greenland,
Antarctic and for deep ocean temperature rise. Greenland is not
the biggest. Some of you may have seen this before and values
for input assumptions have been made. Suggestions for changes
to the input assumptions would be welcome.
It would be exciting to see if the buoyancy of melt water below 3000 metres of ice could give enough lubrication for a thick chunk to slide into the sea and make a spectacular tsunami but I expect that there would be objections from the environmental community against practical tests.
It would be useful to
compare estimates for low probability of dangerous events with
our choice to fly on an aircraft with the same crash
probability.
Stephen
To view this discussion on the web visit https://groups.google.com/d/msgid/geoengineering/CH2PR04MB6936220B9B7203A2CB55EA4B8F410%40CH2PR04MB6936.namprd04.prod.outlook.com.
Douglas MacMartin
True that, whether one uses lots of CDR and/or SRM, it is possible that neither of those options would arrest mass loss without cooling beyond preindustrial, and that over the ensuing millennium we would eventually lose our coastal cities no matter what we do, but that we would delay that if we start solar geo sooner. But (a) we don’t know and I think your second statement is correct, that we don’t have the modeling capability for that, (b) I’m more concerned about what we do in the next 50 years, myself, so questions of whether the cities are ultimately un-savable on timescales of centuries I don’t think matters for near-term decisions (c) personally I’m a lot more worried about Antarctic than Greenland, (d) nothing you said justifies your original claim (insofar as solar geo doesn’t guarantee keeping coastal cities over long timescales either). I agree that starting solar geo today would result in less SLR by 2100 than not starting it, if emissions stayed the same, but the only thing we know in that sentence is the word “less”.
Your original claim was also based on the premise of throwing out all other options. Yes, generally in any situation in life, if you refuse to consider all of the options but one, then yes, you have only one option. That’s true, but only trivially. Yes, if you rule out large-scale CDR to tackle historical emissions, then it is unequivocally true that SRM would be the only way to reduce temperatures beyond the current level. That’s not exactly a profound observation…
Jasmin S. A. Link
"The more advanced the process, the more momentum it has, and the
harder it is to stop or reverse." The same can count for social
dynamics. Self-reinforcing processes with the tendency towards a
lock-in are path-dependent processes. That is, why it is not easy
to just change carbon intensive behavior towards low carbon
emission behavior. Many social, economical, and technical
processes have the production of carbon emissions as side effects
which accumulate during the intensifying self-reinforcing
processes (cp. Figure 4 p. 63 in
https://ediss.sub.uni-hamburg.de/volltexte/2020/10431/pdf/Dissertation.pdf
).
But what about the side effects of SRM that would accumulate over time while deployment in an intensity of "driving the earth into a glacial period"?
It is necessary to decouple self-reinforcing processes from producing carbon emissions as a side effect. And try to potentially include the reduction of carbon emissions as side effects of new self-reinforcing processes that show the dynamic of "the more advanced the process, the more momentum it has, and the harder it is to stop or reverse".
If you would plan the potential deployment of SRM (especially on
a large-scale) you would in the same way have to consider the
potential side effects beforehand to assure to decouple a
deployment of SRM from these potential side effects. Otherwise you
would risk the accumulation in acidification of the oceans,
self-reinforcing the reduction in biodiversity, loss in coral
reefs, accumulating in social tensions (just to name a few of the
interconnected potential side effects, cp. Figure 3.1.,p.71 in the
transdisciplinary network of potential side effects in:
https://www.iass-potsdam.de/sites/default/files/2018-06/EuTRACE_report_digital_second_edition.pdf
).
Furthermore, with SRM you address the global mean temperature. If
this would work at all, this would not imply the globally even
reduction of temperature, regional effects may differ a lot,
potentially increasing also the climate related risks for coastal
cities, which you have mentioned as your main concern to safe.
Best,
Jasmin
To view this discussion on the web visit https://groups.google.com/d/msgid/geoengineering/CAJ3C-04pLr9SO5mJQXfUQUXxwUorOSNT1EgdO2AWLVv9Ba4h%2Bg%40mail.gmail.com.
-- Dr. Jasmin S. A. Link Research Group Climate Change and Security Center for Earth System Research and Sustainability University of Hamburg eMail: jasmi...@uni-hamburg.de www.clisec-hamburg.de www.cen.uni-hamburg.de The CEN combines expertise from the natural and social sciences and unites nine institutes and research centers at Universität Hamburg.
Michael MacCracken
Hi Doug et al.--I'm a bit late to this particular conversation, but I am astonished by the suggestion that Greenland can only cause such a small potential rate of rise in sea level. There was just a kerfuffle with the IPCC authors on their draft projections of rates (see attached letter). While surface melt rate may be relatively slow as often calculated, it is not the main loss of mass process--ice stream flow is very likely the major loss rate once it gets going and the calculations that are done in most models do not include this term, nor do they include the effects of ice shelf thinning that is going on. From the peak of the last interglacial to 8 ka, sea level rose at an average rate of a meter per century while global average temperature rose at an average rate of a degree C per 2000 years, and the CO2 concentration was less than 300 ppm. The documentary "Chasing Ice" shows how fast ice can disappear, and not just in the ice stream calving that is the most amazing aspect of that film. And paleo evidence also makes very clear that ice sheets go away much faster than they build up.
And the question is not so much when the cities will be under water as when it will become inevitable that they will be under water--given that consideration and the paleo sensitivity being something like 15-20 meters per degree C warming (and this is not just me saying this, but see Eric Rignot talk to the NAS last year--see https://vimeo.com/332486918 ).
Based on this sensitivity, we're already past the point where it would be good to have climate intervention underway if we want to avoid significant and early risk to our cities with a very high likelihood (and this is the criterion that is often used in building infrastructure--avoiding 1 in 100 year events or even rarer ones--consider the Dutch for their levees--1 in 10,000 year storms).
Mike MacCracken
To view this discussion on the web visit https://groups.google.com/d/msgid/geoengineering/CH2PR04MB6936220B9B7203A2CB55EA4B8F410%40CH2PR04MB6936.namprd04.prod.outlook.com.
Ronal Larson
On Aug 15, 2020, at 6:32 AM, Jasmin S. A. Link <jasmi...@uni-hamburg.de> wrote:"The more advanced the process, the more momentum it has, and the harder it is to stop or reverse." The same can count for social dynamics. Self-reinforcing processes with the tendency towards a lock-in are path-dependent processes. That is, why it is not easy to just change carbon intensive behavior towards low carbon emission behavior. Many social, economical, and technical processes have the production of carbon emissions as side effects which accumulate during the intensifying self-reinforcing processes (cp. Figure 4 p. 63 in https://ediss.sub.uni-hamburg.de/volltexte/2020/10431/pdf/Dissertation.pdf ).
But what about the side effects of SRM that would accumulate over time while deployment in an intensity of "driving the earth into a glacial period"?
It is necessary to decouple self-reinforcing processes from producing carbon emissions as a side effect. And try to potentially include the reduction of carbon emissions as side effects of new self-reinforcing processes that show the dynamic of "the more advanced the process, the more momentum it has, and the harder it is to stop or reverse".
If you would plan the potential deployment of SRM (especially on a large-scale) you would in the same way have to consider the potential side effects beforehand to assure to decouple a deployment of SRM from these potential side effects. Otherwise you would risk the accumulation in acidification of the oceans, self-reinforcing the reduction in biodiversity, loss in coral reefs, accumulating in social tensions (just to name a few of the interconnected potential side effects, cp. Figure 3.1.,p.71 in the transdisciplinary network of potential side effects in: https://www.iass-potsdam.de/sites/default/files/2018-06/EuTRACE_report_digital_second_edition.pdf ).
Furthermore, with SRM you address the global mean temperature. If this would work at all, this would not imply the globally even reduction of temperature, regional effects may differ a lot, potentially increasing also the climate related risks for coastal cities, which you have mentioned as your main concern to safe.
To view this discussion on the web visit https://groups.google.com/d/msgid/geoengineering/915645b9-6707-1546-4a9b-61cc8d46f032%40uni-hamburg.de.
Douglas MacMartin
Hi Mike,
- I don’t know (and I’m not sure anyone really does) how much Greenland ice sheet mass loss can accelerate, but agree that it won’t stay at the current rate.
- Mainly my point was that the media link Andrew sent was silly by implying that the 6m could happen in decades (even if it didn’t technically say that), and the paper doesn’t make any claims about how much the loss rate will accelerate… and to support the claim that geoengineering is the *ONLY* way to avoid catastrophic sea level rise requires evidence that we need to intervene now rather than over the rest of the century. I do agree that there’s good reason to suspect that Andrew’s claim may be true, but it is certainly not supported by the paper he was referring to, and I don’t think we can prove that the claim is true.
- Paleo evidence makes it clear that staying at even the current CO2 levels for millennia would be catastrophic. It doesn’t do a great job of constraining how rapidly we need to change, e.g. if CDR over the rest of the century would also be an adequate alternative to SRM. (And paleo evidence also shows that it is possible to get ~5m of SLR in a single century, but that’s coming out of the last glacial when there was a lot more available to be melted, and that doesn’t say that it is possible to get anything close to that this century.)
- Arguably this is simply quibbling over whether we can prove we’re past the point where even aggressive CDR would work, or whether there’s simply a risk that we’re past that point, in which case arguing over whether that is 20% or 50% or 10% might not matter for policy. I do agree that we are gambling with the climate, and with odds that no-one would accept in any other circumstance.
- Personally, given how little research has been done, I don’t think there’s strong justification for saying that the long-term climate consequences of waiting another 10-15 years for research (and to develop governance capacity) will be so bad that we should go ahead and deploy something now without doing the research (though if we did deploy something now, I’d worry more about the societal response than the physical issues). But, just like the US didn’t use the first few months of this year to prepare for covid when it knew it was coming, it would be truly awful to not do the research now, leaving us in the same boat yet another decade later.
doug
From: Michael MacCracken <mmac...@comcast.net>
Sent: Saturday, August 15, 2020 12:54 PM
To: Douglas MacMartin <dgm...@cornell.edu>; andrew....@gmail.com; geoengineering <geoengi...@googlegroups.com>
Subject: Re: [geo] Background-Greenland collapse
Hi Doug et al.--I'm a bit late to this particular conversation, but I am astonished by the suggestion that Greenland can only cause such a small potential rate of rise in sea level. There was just a kerfuffle with the IPCC authors on their draft projections of rates (see attached letter). While surface melt rate may be relatively slow as often calculated, it is not the main loss of mass process--ice stream flow is very likely the major loss rate once it gets going and the calculations that are done in most models do not include this term, nor do they include the effects of ice shelf thinning that is going on. From the peak of the last interglacial to 8 ka, sea level rose at an average rate of a meter per century while global average temperature rose at an average rate of a degree C per 2000 years, and the CO2 concentration was less than 300 ppm. The documentary "Chasing Ice" shows how fast ice can disappear, and not just in the ice stream calving that is the most amazing aspect of that film. And paleo evidence also makes very clear that ice sheets go away much faster than they build up.
Douglas MacMartin
Thanks Jasmin,
Agree that all the side effects of SRM need to be considered, and need to evaluate options holistically.
That said, SRM does not cause ocean acidification; CO2 causes ocean acidification. So one should never list ocean acidification as a side effect. (It is true that SRM doesn’t solve ocean acidification, but it also doesn’t solve car accidents… and no-one lists that as a reason not to consider SRM.) Implicit in listing acidification in any discussion of SRM is an assumption that somehow we’re required to choose between reducing CO2 or using SRM, in much the same way that we have to choose whether to drive safely or wear a seat belt, but we’re not allowed to do both.
doug
From: geoengi...@googlegroups.com <geoengi...@googlegroups.com>
On Behalf Of Jasmin S. A. Link
Sent: Saturday, August 15, 2020 8:32 AM
To: geoengi...@googlegroups.com
Subject: Re: [geo] Background-Greenland collapse
"The more advanced the process, the more momentum it has, and the harder it is to stop or reverse." The same can count for social dynamics. Self-reinforcing processes with the tendency towards a lock-in are path-dependent processes. That is, why it is not easy to just change carbon intensive behavior towards low carbon emission behavior. Many social, economical, and technical processes have the production of carbon emissions as side effects which accumulate during the intensifying self-reinforcing processes (cp. Figure 4 p. 63 in https://ediss.sub.uni-hamburg.de/volltexte/2020/10431/pdf/Dissertation.pdf ).
To view this discussion on the web visit https://groups.google.com/d/msgid/geoengineering/915645b9-6707-1546-4a9b-61cc8d46f032%40uni-hamburg.de.
Russell Seitz
Kevin Lister
To view this discussion on the web visit https://groups.google.com/d/msgid/geoengineering/CH2PR04MB69360BBA550F9276C54C33218F410%40CH2PR04MB6936.namprd04.prod.outlook.com.
Jasmin S. A. Link
Thank you Doug,
> If you would plan the potential deployment of SRM
(especially on a large-scale) you would in the same way have to
consider the potential side effects beforehand to assure to
decouple a deployment of SRM from these potential side effects.
Otherwise you would risk the accumulation in acidification of the
oceans, self-reinforcing the reduction in biodiversity, loss in
coral reefs, accumulating in social tensions (just to name a few
of the interconnected potential side effects, cp. Figure 3.1.,p.71
in the transdisciplinary network of potential side effects in:
https://www.iass-potsdam.de/sites/default/files/2018-06/EuTRACE_report_digital_second_edition.pdf
). <
Regarding ocean acidification and acid rain: Yes, putting up
mirrors would not directly cause ocean acidification but SAI might
cause acid rain (depending on the aerosol used). And no, it still
is related via an indirect effect in terms of social behavior: If
people comprehend SRM in a way as that there is some sort of
technical compensation happening for their carbon emissions, they
are likely to rather not reduce their own carbon emissions, but
instead increase their own carbon emissions. This is marked in
Figure 3.1. as "less necessity for direct emission reduction?"
connected to a "rise of CO2 emissions", which causes multiple
feedbacks such as the necessity to further increase the SAI
deployment. I think, there have been more recent simulation
studies at MIT on this kind of behavior that support the relevance
of this potential indirect effect.
In my argument on the potential accumulation in acidification of
the oceans, I had this indirect effect of the SRM deployment on
the rather increased carbon emission behavior in mind: Officially
applying a method to reduce global warming which might be
understood as "fixing the problem with engineering" might rather
reduce than increase individual mitigation efforts. (Of course, we
know that the problem is not really fixed, but try to explain that
to usual consumers who would know at some point that their
government would regularly spend large sums of money on SRM and
who might feel more comfortable to stick and increase their former
behavior than to really change it, due to path dependence).
Best,
Jasmin
To view this discussion on the web visit https://groups.google.com/d/msgid/geoengineering/CH2PR04MB69362DA78244A77013989D198F410%40CH2PR04MB6936.namprd04.prod.outlook.com.
Jasmin S. A. Link
Dear Ron,
Thank you very much, for your interest an for emphasizing the relevance of my thesis in the context of SRM and CDR.
Please, feel free to add me to the CDR list as well.
The detailed potential side effects networks I have published only for SAI (Figure 3.1.) and BECCS (Figure 3.2. in the same EuTRACE final report) so far. I have made one for afforestion, but that has not been as requested and intriguing for discussion yet.
I think, some of the potential indirect side effects of SRM - such as a potential reduction in individual mitigation efforts - might also apply to a large-scale CDR. Depending on the method of CDR and its public presence (like artificial trees), this might also be true for even relatively small applications of CDR. Thus, it might be important to estimate in advance, if the contruction of one artifical tree rather decreases carbon emissions, by capturing them, or in total rather increases carbon emissions, by triggering passengers to feel free to emit more carbon.
Best regards,
Jasmin
Douglas MacMartin
Thanks – I agree completely that moral hazard is a serious risk, perhaps the biggest risk. (But I also think it is important to be more explicit about one’s assumptions.)
There’ve been a few studies trying to look at moral hazard, with inconclusive results even on the sign of the effect – very hard to make predictions about, but, of course, very clear that it is a serious possibility.
(And re acid rain, that’s not significant in terms of ocean acidification.)
doug
Andrew Lockley
To view this discussion on the web visit https://groups.google.com/d/msgid/geoengineering/CH2PR04MB69368468A10ABDA3D77B25938F5E0%40CH2PR04MB6936.namprd04.prod.outlook.com.
Jasmin S. A. Link
Yes. I am not sure, whether moral hazard or even morale hazard, as Andrew has pointed out, exactly name what I try to describe. Moral hazard or morale hazard seem to me both have at least a deliberate decision, at least concious, perhaps even rational or optimizing in some sense, of action making in common. Regarding emission behavior, people usually do not deliberately choose to emit CO2 directly or indirectly. Usally, the emissions are just a side-effect of their consumption, their way of living. Thus, a side-effect of their routines. For many people it is even difficult to reflect on how much carbon they emit via which choice of consumption. Consequently, an individual attempt to reduce carbon emissions is a great effort and usually a large step out of the comfort zone for a single person. Calling it moral hazard, if they "just relax" and fall back a bit in their routines, or enjoy intensifying them, sounds a bit hard. This kind of behavioral decision-making processes may be far from what you might expect as rational decision-making.
In my earlier referred thesis (Link 2018) I coin the term
"path-dependent behavior", which is the result of the influence of
existing path-dependent processes and is rather a following
behavior, following routines/standards/institutions, the masses,
neighbors etc. It is an efficient way of decision-making,
resulting from a short-cut in the brain (applying the
least-effort-principle from social psychology). But the resulting
action may be a different one than the result of a utility
function would be. Like following the advice of your computer
expert (Herbert Simon has also described the phenomenon of
personal experts) does not mean that you have optimized the
variety of choices yourself. But if the choice you have made that
way is producing more carbon emissions along the way than a
different one would have, that you might have not even thought
about when only relying on your personal expert: Can you be
accused for moral(e) hazard?
Best regards,
Jasmin
To view this discussion on the web visit https://groups.google.com/d/msgid/geoengineering/CH2PR04MB69368468A10ABDA3D77B25938F5E0%40CH2PR04MB6936.namprd04.prod.outlook.com.
Andrew Lockley
To view this discussion on the web visit https://groups.google.com/d/msgid/geoengineering/196ff2cc-1636-8df8-3a7e-9d4f06dc46d9%40uni-hamburg.de.
Michael MacCracken
Hi Doug--While the thread has headed off onto issues of moral hazard, etc., getting back to the situation with Greenland, here is the latest paper on this: https://scitechdaily.com/climate-scientists-sound-the-alarm-warming-greenland-ice-sheet-passes-point-of-no-return/ And I think Eric Rignot would suggest some of the ice streams for Antarctic ice sheet are looking less than stable.
I would also note that the last deglaciation melted back about
two-thirds of the maximum ice--we have one-third left and it is
now starting to go. So, saying the past amplified rate was due to
having lots more ice around needs to be more carefully
stated--there was more at mid-latitudes, but now mi-latitude
weather is being pushed to higher latitudes (consider Siberia this
summer), so I see no reason for reassurance about the stability of
the ice.
Douglas MacMartin
Certainly the term moral hazard as used in other literature doesn’t imply any conscious thought process as a requirement. (Examples might be better termed risk compensation; someone is more likely to drive close to a cyclist who is wearing a helmet, for example.)
Perhaps better to simply use the term mitigation deterrence and not make any implied judgment.
To view this discussion on the web visit https://groups.google.com/d/msgid/geoengineering/CAJ3C-04yYV8GfBhiL5mYi6mKNbxAN_hGEqmpc%3D3k3250EtyaVQ%40mail.gmail.com.
Ronal Larson
On Aug 15, 2020, at 5:54 PM, Jasmin S. A. Link <jasmi...@uni-hamburg.de> wrote:Dear Ron,
Thank you very much, for your interest an for emphasizing the relevance of my thesis in the context of SRM and CDR.
Please, feel free to add me to the CDR list as well.
The detailed potential side effects networks I have published only for SAI (Figure 3.1.) and BECCS (Figure 3.2. in the same EuTRACE final report) so far. I have made one for afforestion, but that has not been as requested and intriguing for discussion yet.
I think, some of the potential indirect side effects of SRM - such as a potential reduction in individual mitigation efforts - might also apply to a large-scale CDR. Depending on the method of CDR and its public presence (like artificial trees), this might also be true for even relatively small applications of CDR. Thus, it might be important to estimate in advance, if the contruction of one artifical tree rather decreases carbon emissions, by capturing them, or in total rather increases carbon emissions, by triggering passengers to feel free to emit more carbon.
Best regards,
Jasmin
Am 15.08.2020 um 19:47 schrieb Ronal Larson:
Dr. Link and list: