‘A better path’ toward projecting, planning for rising seas on a warmer Earth
From Princeton University, by Morgan Kelly, Office of Communications
More useful projections of sea level are possible despite substantial uncertainty about the future behavior of massive ice sheets, according to Princeton University researchers.
In two recent papers in the journals Nature Climate Change and the Proceedings of the National Academy of Sciences (PNAS), the researchers present a probabilistic assessment of the Antarctic contribution to 21st-century sea-level change. Their methodology folds observed changes and models of different complexity into unified projections that can be updated with new information. This approach provides a consistent means to integrate the potential contribution of both continental ice sheets — Greenland and Antarctica — into sea-level rise projections.
“No single ice sheet model or methodology for projections provides sufficient information for good policy and planning decisions,” explained lead author Christopher Little, a postdoctoral research associate in the Program in Science, Technology and Environmental Policy in Princeton’s Woodrow Wilson School of Public and International Affairs.
“Furthermore, there are fundamental limitations in the observational data available on and near ice sheets,” Little said. “Projections of their response to 21st century climate changes are thus very poorly constrained. There’s unlikely to be a single answer in the near future: a better objective is a comprehensive, transparent baseline that can be improved over time.”
The Princeton approach provides a more informative projection of future sea levels that politicians and coastal planners can use to prepare for higher seas, said Little, who works in the group of co-author Michael Oppenheimer, Princeton’s Albert G. Milbank Professor of Geosciences and International Affairs. Little and Oppenheimer worked on both papers with Nathan Urban, a former postdoctoral researcher with Oppenheimer now at the Los Alamos National Laboratory.
In two recent papers, Princeton University researchers present a probabilistic assessment of the Antarctic contribution to 21st-century sea-level change. Their methodology provides a consistent means to integrate the potential contribution of continental ice sheets such as Greenland and Antarctica into sea-level rise projections. In existing projections, the contribution of Antarctica to future sea-level rise is almost entirely derived from locations where present-day mass loss is concentrated (area 15, above). This is despite evidence that future discharge in other drainage basins — which comprise more than 96 percent of the ice sheet’s area — remains uncertain. (Image courtesy of Christopher Little)
Little explains the findings of both papers as follows:
“Gauging the future rate of sea-level rise is critical for climate-change policy and coastal-planning efforts. One crucial component is the melting of polar ice sheets.
“During the past 20 years, the Greenland and Antarctic ice sheets have lost an increasing amount of ice and now contribute roughly one-third of the rate of global mean sea-level rise. However, the standard tools used to project these ice sheets’ contribution to future sea levels are limited by inadequate process understanding and sparse data. Ice sheets interact with the ocean on small spatial scales, and their motion is strongly governed by poorly understood properties of the ice as well as the sediment hidden several miles beneath it. Sea-level rise projections should reflect these uncertainties.
“Recently, several groups have used alternative techniques to forecast maximum possible sea levels — known as upper bounds — that do not explicitly model ice dynamics. Upper bound estimates by the year 2100 projected using these techniques are up to 6 feet (three times higher than future sea level estimates from the Intergovernmental Panel on Climate Change (IPCC)). However, the physical basis underlying these projections and their likelihood of occurrence remain unclear.
“In our group, we think we can more consistently assess disparate sources of information. In two recent papers, we introduce a novel framework for projecting the mass balance of the Antarctic ice sheet that allows for the conversion of current and future uncertainties of ice-sheet dynamics into probability distributions that may be supplemented by expert judgments. The power of this framework arises from its ability to improve and compare projections in a transparent manner.
“Like watersheds on land, ice sheets discharge precipitation that falls over a wide drainage basin through relatively narrow outlets. Although ice flow is linked across basins, each basin may remain relatively independent over time periods less than a century. The framework described in these two papers projects mass balance separately for each drainage basin, while allowing for correlated trends driven by underlying physical processes occurring at larger spatial scales.
“The first paper, published in PNAS, introduces this ‘basin-by-basin’ framework and reveals that, even with limited information, a comprehensive probabilistic approach can provide insight that is missing from previous projections. We performed sensitivity analyses by changing the set of assumptions applied to each basin. For each set of assumptions, Monte Carlo simulations [computer algorithms based on random sampling] were used to generate 30,000 to 50,000 scenarios of mass changes originating from each basin and the continent as a whole.
“In previous scenario-based projections, the contribution of Antarctica to future sea-level rise is almost entirely derived from locations where present-day mass loss is concentrated. This is despite evidence that future discharge in other drainage basins — which comprise more than 96 percent of the ice sheet’s area — remains uncertain.
“By incorporating the entire ice sheet, the PNAS study demonstrated that uncertainty in ice discharge outside regions where scientists ‘expect’ ice loss might result in additional sea-level rise that must be considered in projections. In addition, we quantitatively show that the likelihood of upper bounds must be taken into account when assessing their magnitude and appropriate uncertainty reduction efforts.
“The second paper, published in Nature Climate Change, extended the framework to include Bayesian updating, which allows prior assumptions to be updated as new data are collected. We combined model-based basin-level projections with data-based extrapolations and previously reported continental-scale observations to forecast the Antarctic contribution to sea-level change.
“The paper projected a 95th percentile ice-mass loss equal to a 13-centimeter (5.1-inch) increase in sea level by 2100; other estimates provide upper bounds reaching up to 60 centimeters (roughly 23.5 inches), but with no quantification of probability. This paper suggests that most earlier projections either overestimated Antarctica’s possible contribution to sea-level rise; implied physical changes inconsistent with underlying methodological assumptions; or, assume an extremely low risk tolerance.
“Future work on this framework includes further addressing inconsistencies in different methodologies, which will continue to refine the range of upper-bound sea-level projections. Our group also intends to include the solid earth and gravitational response that modulates sea-level changes at the local level, allowing the generation of a global map of the local probability distribution of sea-level rise.”
The paper, “Probabilistic framework for assessing the ice sheet contribution to sea level change,” was published February 26 in the Proceedings of the National Academy of Sciences. The article, “Upper bounds on twenty-first-century Antarctic ice loss assessed using a probabilistic framework,” was published online March 17 in Nature Climate Change. Both papers were funded by the Princeton Environmental Institute‘s Carbon Mitigation Initiative, and Princeton University’s Program in Science, Technology and Environmental Policy.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
“Probabilistic” or problamatic? more along the lines that Hoser points out. This is what politicians point to as “peer reviewed science” when explaining what their decisions are based on. Former Congressman Bart Gordon was the hand-picked successor to Al Gore’s seat in The House. Using his leadership position on the Science and Technology Committe he funneled billions to dodgy studies like this. His response on the release of the Climategate emails “we need more research” With the gig up he then reversed his decision to run for re-election and is now a lobbist working on science and energy issues. This study should have appeared in a political science mag.
ICE MELT ALERT..!
Its spring (although you wouldn’t know it here in the UK) and the Arctic sea ice is (at last) beginning to melt..!
Man the lifeboats..!
Maybe I read it too fast, but are they actually predicting a sea level rise from 13-60 centimeters, attributable to Antarctic ice melt, at a time when the Antarctic is increasing ice mass and sea ice area to record levels? Do the models account for the observational increases in actual ice? So what good are models that don’t begin with real world observations?
They get an immediate “F” for inaccuracy!
““During the past 20 years, the Greenland and Antarctic ice sheets have lost an increasing amount of ice and now contribute roughly one-third of the rate of global mean sea-level rise.”. All data I’ve seen says Antarctica ice sheets have shown a net increase over the past 20 years and the ice mass is still increasing!
Two simple ways to predict sea level rise:
1) Linear projection over then next 20 years based on the average annual increase over the past 100-120 years.
2) K*(^GL + ^AA) = ^SL. (tough to do equations on basic iPad)
K = conversion constant (if you don’t know, guess!)
^GL = change in Green Land ice content
^AA = change in Antarctica ice content
Can I get the copious funding now for my brilliant work?
Bill
So they declared Uncertainty to be non-existent with the power of SCIENCE !
Just go to show that if you get enough PhD in one room they can declare the sun to be blue and the politicians would run on it.
The planet is cooling and Antarctica leads the way!
My fellow Dutchmen have completely gone off their trolley. The ice is expanding because it’s melting?
But you wouldn’t want to feed the many that lap up such crap like manna from heaven.
Proving once again that there is a clean end on a turd.
GIGO
“Their methodology folds observed changes and models of different complexity into unified projections that can be updated with new information. This approach provides a consistent means to integrate the potential contribution of both continental ice sheets — Greenland and Antarctica — into sea-level rise projections.”
This sounds like the IPCC’s ensemble of models. You know, this may seem like a good idea on casual inspection, but from my perspective, making a big honkin framework that mixes the less bad with the worse isn’t the answer; it just makes it harder to figure out exactly where the problem is when the
predictionsprojections are found to be totally off base. They can sayall they want, if there’s any support for the idea that this framework will allow projections to be compared and improved in a transparent manner, they’ve failed to explain it here.
Not buying it.
I’m not so upset with this. They admit it is very uncertain, introduce a second paper that will allow corrections to be made based on changes in data and are only predicting 5 inch increase in 100 years. Not the 9 feet that some fools are.
If and when sea levels are measured to not be increasing by much at all, then this 5 inches can be adjusted downwards (or upwards) as the data indicate. If there is any cooling at all, this will cause contraction of oceans and an even smaller sea level rise. Don’t be so gloomy. They’re not all incompetent or liars.
The majority of non-scientists sadly do not understand science at all. The majority of “scientists” are button pushers, paper shufflers, and/or downright toadies. That leaves a VANISHINGLY SMALL percentage of the population that understands, and is performing, actual scientific work. These papers are unfortunately the product of “scientists” that fall into the button-pusher, paper-shuffler, toady categories, as is the case with over 95% of all climate-related studies at this time.
Nice headline, Anthony 🙂
“There is something fascinating about science. One gets such wholesale returns of conjecture out of such a trifling investment of fact. ”
Mark Twain
——-
Just for the fun of it….
“The difference between the right word and the almost right word is the difference between lightning and a lightning bug.”
Mark Twain
“The man who doesn’t read good books has no advantage over the man who can’t read them.”
Mark Twain
This is what is at the core of the problem with climate science. They go into a study with assumptions already built in. Result = models diverge from observations = fail. Then they try again using another model run.
Recently scientists have been reported to be baffled by the 16 year temperature standstill. Why? Assumptions.
Ay least the 30 meters by 2100 Sea-Level rise projections placed into play by the dimmer bulbs seem to have drowned (into the rising seas???)!!!
Unfortunately, Ian, in climate science today you have to consider every possible way you might be being deceived. The scientific bond of trust has largely been broken as far as I am concerned.
Consider the sentence from the article;
Now, imagine: If more snow/condensation formed during some months, then more might possibly melt during other months, right? Or one location increase while another location decreased. Hence an “increasing amount of ice loss” could easily occur when total amounts were increasing. (Also note they don’t even use the word melt. It could be glacial flow rate).
Also imagine: If the rate of global sea level rise decreased, and Antarctic ice loss remained the same, then the fraction it contributed would increase. But it would be an increasing fraction of a decreasing amount.
I’m not saying either is the case, but these are the pathetic little tricks which get sprung on readers both in science and general-media. The two hypothetical examples above would not be lies, but would be unethical deceptions I am no longer be surprised to find in this area of politicized science.
A political-environmentalist camel has been invited into the climate-science tent, but there are still other camels defecating on the exterior. What is the purpose of the tent now?
“During the past 20 years, the Greenland and Antarctic ice sheets have lost an increasing amount of ice” is offered with no proof; it is part of the (failed) hypothesis.
I managed to download the Tide Gauge database from the Permanent Mean Sea Level Service. Anybody who has tried to do this before knows that it comes down in over 2000 individual files and one would need two grad students, fortran programming and much help from the PMSL database managers over several months in order to track how the Tide Gauges are changing over time.
http://www.psmsl.org/data/obtaining/complete.php
I took one of the archive files and kept sorting and consolidating it down until all that was left was the Tide Gauge measurements by Year. There are 30,966 individual annual measurements going back to 1807 and as many as 400 individual Tide Gauge measurements in some years.
So this isn’t really tracking each Gauge over time, but more the average of all measurements in each year. It will not be 100% accurate since Gauges moving into and then out of the database which will change the baseline a small amount over time, but it is what it is.
So, here are all 30,966 Tide Gauge measurements by year going back to 1807. (note there are some groups that are rising or declining at a high rate due to glacial isostatic adjustment which is still continuing today).
http://s17.postimg.org/si6ly8q27/Sea_Level_Measurements_PMSL.png
On average, the Tide Gauge sea level is increasing at 0.28 mms/year since 1807. (Note that GPS indicates the Land is rising by an weighted average of 0.3 mms/year so the actual volume/sea level might be rising at 0.6 mms/year on average over the period.
To test the trends over time, I’ve Zoomed into the 1930 to 2009 period. Generally, from 1930 to 1980, the trend was similar to the previous rise and, from 1980 to 2009, the Tide Gauge sea level increase went up to 1.41 mms/year.
http://s2.postimg.org/xcp9tsz6x/Sea_Level_Measurements_PMSL_1930_1980_2009.png
I don’t really see a 60 year AMO cycle in there or any further acceleration, but there is certainly an El Nino and a volcano impact which should be taken into account. So, the Satellites, from 1993, start right at the low stand provided by the Pinatubo eruption.
Here is the PMSL Tide Gauge database going back to 1960 (1.13 mms/year) against the other sea level reconstructions which have been done by Church2011 (1.9 mms/year), Jarajeva2006 (1.6 mms/year), Domingues2008 (1.5 mms/year) and the Satellite Altimetry (now up to 3.2 mms/year).
http://s8.postimg.org/9ysbkpw51/All_Sea_Level_Measurements_1960_2013.png
At 1.41 mms/year in reality since 1980, sea level is rising very slowly and there is no reason to consider it to be dangerous at these rates.
Regardless of your stance on the overall issue here, I think a lot of commenters fail to understand some basic issues. First, there is a difference between sea ice area and land ice mass. The article refers to the latter, but many commenters point to the former as evidence that ice mass is increasing. I’m not trying to support the authors’ conclusions here, but discussing sea ice area is completely irrelevant the points they are making.
Also, a model is inherently almost always “wrong”. As the article states, their model is a “probabilistic assessment” that considers the likelihood of various potential outcomes. The probability that a specific one of these outcomes will occur is very small. These kinds of models are used for all kinds of things, including local weather forecasts. Forecasts more than 24 hours out are very rarely 100% precise, but they are still helpful for planning purposes. This doesn’t mean that I think the model in the paper is correct, but to attack the concept of a model is just poor reasoning.
I’m not even close to an AGW supporter, but nor am I a blind critic. It seems to me that many here are guilty of the same logical fallacies that you accuse the warmist crowd of. I was hoping to find a reasonable community for logical discussion here, but so far I sense a serious lack of credibility. Even the title of this article, with the phrase “uncertainty be damned”, applies an unnecessarily negative bias toward an integral part of any projection.
AR 5 is essentially focused on the dangerous sea level change. That is what Jouzel in France is pushing hard, that is what is promoted, models, ad hoc satellite “corrections” especially since Envisat was suicided.
This is their last chance at fearmongering…
Typical climate change denier gobbledygook, but it begs the question: “Has anyone tried to accurately measure the rate of change – if there is one – in the volume (rate of change!) of the total Antarctic ice cap?”
I made no reference to sea ice, Brian, or (don’t get me started) models. Not everyone addresses all the points that could be addressed. If you want to find a site where everybody is always correct and deferential then you are going to be searching for a long time.
What I sense most from contributions here is a serious lack of gullibility, not credibility.
Brian says:
April 5, 2013 at 8:02 am
…
Also, a model is inherently almost always “wrong”. As the article states, their model is a “probabilistic assessment” that considers the likelihood of various potential outcomes. The probability that a specific one of these outcomes will occur is very small. These kinds of models are used for all kinds of things, including local weather forecasts. Forecasts more than 24 hours out are very rarely 100% precise, but they are still helpful for planning purposes. This doesn’t mean that I think the model in the paper is correct, but to attack the concept of a model is just poor reasoning.
…
—–
You’ve got to put the distaste around here for models in context. Models have been used as justification for extremely painful and disruptive policy changes, and have overstated warming spectacularly. It’s not unreasonable for people to view the subsequent application of further models to climate problems with a certain amount of prejudice in my view. Particularly when we read lead author Christopher Little, a postdoctoral research associate in the Program in Science, Technology and Environmental Policy; this seems to imply a continued drive to associate policy with model results. Models can be great, just take the necessary decades to properly test and validate your results before using them as a basis for policy, that’s all I ask.
Michael Hart,
I understand that many people here know what they’re talking about. I was referring to the slew of comments about Antarctic sea ice growth on nearly every article that has anything to do with ice. And those who attack various accepted scientific procedures because they don’t agree with the results.
I’m not looking for a site where “everybody is always correct”, but one with two-sided dialogue involving educated people who are simply trying to understand climate. Sadly, I don’t believe that such a site exists. The fact that there are very few dissenting opinions (from the poster’s intended point) on this site indicates either widespread gullibility or a severe case of confirmation bias. I think there are at least as many commenters who want to be spoon-fed material to demonize AGW proponents as there are commenters who want to learn and discuss. And the posters cater to those people, which makes sense, but credibility is still lost because of it. I am fully aware that all pro-AGW sites suffer from the same symptoms as well.
I think it would help if posters, like yourself, who know their stuff would call out incorrect comments, even if they support your conclusions. I think it would make us all a bit smarter and less susceptible to preconceived biases.
Additionally, who cares about the actual sea level? What matters is the relationship of the land level to the sea level and whether we can adapt IF the water level is going up with relation to the land level.
http://tidesandcurrents.noaa.gov/sltrends/trends/8518750.png
http://www.ecy.wa.gov/climatechange/images/hightides/20100218_fig1.gif
http://www.environment-agency.gov.uk/static/images/Research/Figure-5-climate-change(1).jpg
What part of the increase in these graphs is melt water?
And if you were going to PREDICT how these so called curves might continue?
In addition, what part of these increase are C02?
Yeah, kinda hard to tell.
Chuck L says:
April 5, 2013 at 5:02 am
“These papers and the people who write them are like cockroaches. As soon as a paper is squashed and debunked, there is a lot of scurrying around in the dark by “researchers” and a new one appears.”
Note that the faculty mentor and co-author is “Michael Oppenheimer, “Princeton’s Albert G. Milbank Professor of Geosciences and International Affairs.” He is all propaganda, all the time. He published the scare article claiming that global warming would cause a flood of people across the Mexican border into the US.