A recent study that is getting a lot of press suggests that the massive ice sheets of Antarctica are on average growing rather than shrinking, and thus, not contributing to sea level rise. (The authors of the study warn that this will reverse in the near future with global warming.) However, there is reason to believe that these conclusions are incorrect.
Antarctica is the sleeping giant of climate change. Human activity, mainly the release of greenhouse gasses from burning fossil fuels, has been changing the climate rather dramatically for the last few decades, and the consequences of this change are mostly negative. Failed agricultural systems have led to failed states and regional political instability. Dramatic changes in weather patterns, including droughts in Australia and California, a series of unprecedented tropical storms over the last several years, major flooding (if anyone from Texas is reading this, nice to know you have internet access in your tree), all have a global warming contribution, because weather is climate and climate is changed and changing. But sea level rise, while mostly a thing of the future rather than the present, may have the biggest effect of all, at least on land. As we warm the planet, the polar ice sheets will contribute much of their ice to the sea, and based on what we know of the past, direct measurements over the last 20 years or so, and from models of the medium term future, this could mean an increase in sea level of several meters. The best available science currently suggests that by the end of the century average sea levels could be about a meter higher than they are now. It would not be unreasonable to regard that as a conservative estimate.
I’d like to take a moment and point out an important aspect of the sea which people, especially those that don’t live on the sea, forget. The average altitude of the sea at a particular point along the shore is not the part you have to worry about. Well, that is important, but it is not the part that bites. Consider the cobra snake. A cape cobra can strike at a distance well over half its own body length. So if you are standing ten feet away from a fifteen foot long cobra, the snake might seem a safe distance away, but you are actually within its striking range. One could say that the sea has two overlapping but distinct distances at which it strikes. One is the normal storm range. If you raise the sea along a beach in Cape Cod by six inches, nothing interesting happens most days. But the dozen or so medium size storms that will occur over a year (especially in winter when the storms come in from the Atlantic) will convert that foot of elevation into several horizontal feet of beach erosion, in a very short amount of time. The second is what happens when more serious storms, like tropical cyclones or their extratropical spawn, come along. New York City was built and reinforced from the sea, over time, mainly when the Atlantic was about a foot lower than it is now. A couple of years ago, when Super Storm Sandy came along, the storm gathered up that extra foot of sea level and turned it into an extra large storm surge sufficient to flood the subway system in lower Manhattan. Long before the sea in that area rises another three feet, there will be the occasional storm surge that will be even more severe.
Since a large percentage of the world’s population, a large percentage of the world’s agricultural activity, and an even larger percentage (probably) of the world’s real estate value will become subject to flooding, sometimes severe, and eventually be replaced by the rising sea over the next century and beyond, sea level rise is a very important phenomenon.
You have probably already heard about the study, “Mass gains of the Antarctic ice sheet exceed losses” by H. Jay Zwally and others (see citation and abstract below), that came out a couple of weeks ago telling us that the contribution to sea level rise by the Antarctic is currently zero or negative. Or at least, that is how many press outlets are reporting the story.
There are two problems with this study that you need to know about. First, the study examines a data set that ends in 2008. The second problem is that there are indicators that the study is simply wrong, even though it likely has significant merits.
The last decade of research on Antarctica have shown, in many studies using a variety of techniques, that Antarctica is contributing to sea level rise. They have also shown that the rate of melting in Antarctic is probably increasing. Even more importantly, they have indicated that certain areas of Antarctic are current in a state of instability, suggesting that the rate of contribution of the southern continent’s ice mass to sea level rise may increase abruptly in the near future.
The fact that the study being reported uses older data could explain why it conflicts with everything else the science is telling us. Michael Mann, quoted in The Guardian, notes, “…the claims are based on seven-year-old data, and so cannot address the finding that Antarctic ice loss has accelerated in more recent years.” To this I’ll add that it is somewhat annoying that those reporting the story, including, oddly, the authors of the study, are using forms of the word “current” to describe the result. These results are old, out dated, and while potentially valuable, a data set ending in 2008, when speaking of a rapidly changing system, is not current.
Sou at HotWhopper has a nice graphic showing estimates of Antarctic ice melt before and after 2008, strongly indicating the problem with using a study from older date to understand current conditions.
Average global sea level is a measurable verifiable established fact, and the contribution of major ice sheets to this has been measured and found to be important. If the study is correct, and Antarctica was not contributing to sea level rise during that period prior to 2008, then something is terribly wrong. There is simply not enough wiggle room in the other sources of sea level rise to account for the missing volume of water. One could argue that a beautiful hypothesis (positive mass balance in Antarctic ice) has been killed by an ugly fact (actual observed sea level rise). But Zwally’s study does not present a mere hypothesis, but rather, is based on detailed observations incorporated into a set of carefully done calculations.
So, perhaps the observations are wrong. There may be two reasons the observations (and the calculations derived from them) are wrong. One is simply that the satellite data they use are inherently less accurate than needed. The measurements are of a very small change over time over a very large area. If the satellite method is just a little off, this could cause a problem. (By the way, the data end in 2008 because the instrumentation on the satellite stopped working then.) This study’s main contribution may, in the end, to be to point out a problem with the instrumentation prior to that time. This doesn’t seem that likely for the simple reason that the whole point of putting fancy instruments in a bird is to get super accurate information.
The second possible reason seems more likely. Part of the process of determining that Antarctica has a positive mass balance (more ice over time rather than less) involves assumptions (and some measurements) about the response of the bedrock underneath the very thick ice sheets. If that is wrong, then that is a problem.
Since the sea level has in fact been going up, and there is no easy way to account for that than a certain contribution to Antarctica, and all the other science shows an increasingly melting Antarctic, and the study uses older data, then I’m afraid I have bad news. Sea level is still going up, Antarctica is still contributing to it, and the amount of this contribution is still, as the science has been suggesting for several years no, only going to increase.
The following resources will be of interest to anyone following this story.
The original paper is here.
Zwally, H. Jay, 2; Li, Jun; Robbins, John W.; Saba, Jack L.; Yi, Donghui; Brenner, Anita C. 2015. Mass gains of the Antarctic ice sheet exceed losses. Journal of Glaciology, International Glaciological Society.
Mass changes of the Antarctic ice sheet impact sea-level rise as climate changes, but recent rates have been uncertain. Ice, Cloud and land Elevation Satellite (ICESat) data (2003–08) show mass gains from snow accumulation exceeded discharge losses by 82 ± 25 Gt a–1, reducing global sea-level rise by 0.23 mm a–1. European Remote-sensing Satellite (ERS) data (1992–2001) give a similar gain of 112 ± 61 Gt a–1. Gains of 136 Gt a–1 in East Antarctica (EA) and 72 Gt a–1 in four drainage systems (WA2) in West Antarctic (WA) exceed losses of 97 Gt a–1 from three coastal drainage systems (WA1) and 29 Gt a–1 from the Antarctic Peninsula (AP). EA dynamic thickening of 147 Gt a–1 is a continuing response to increased accumulation (>50%) since the early Holocene. Recent accumulation loss of 11 Gt a–1 in EA indicates thickening is not from contemporaneous snowfall increases. Similarly, the WA2 gain is mainly (60 Gt a–1) dynamic thickening. In WA1 and the AP, increased losses of 66 ± 16 Gt a–1 from increased dynamic thinning from accelerating glaciers are 50% offset by greater WA snowfall. The decadal increase in dynamic thinning in WA1 and the AP is approximately one-third of the long-term dynamic thickening in EA and WA2, which should buffer additional dynamic thinning for decades.
Could also be a homogenization problem between the ERS and ICESat satellites, as we saw (and continue to see) in the hoopla over the tropospheric hot spot. Either way, as you note, there are tight constraints on the SLR budget and ice mass balance contributions, so the preponderance of thought casts doubt on Zwally et al's findings.
I suspect we haven't heard the last word on this study.
Don't those who are crowing over this new study realise that it represents very bad news for global sea level in the long term?
Taking the study at face value: if Antarctica contributed little or nothing to the long term trend in sea level rise (SLR) pre 2008—which was 3mm pa and is still rising now—then, given we know from other studies that Antarctic ice loss is today accelerating, we can surely expect to see a major ramping up of the rate of SLR as recent Antarctic ice loss starts to kick in. This means that we're in for way over the, very conservative, foot of sea level rise by 2100 predicted by the ICPP's RCP8.5. In fact, given that the Antarctica ice sheet is the largest mass of ice we have on the planet, we could easily be into metres of sea level rise by 2100, at an incredible cost to society.
The past time I looked at the Antarctic ice mass data, the ratio between loss of ice, staying the same, and ice gain was 17:40:1. Has that changed?
#1: I wonder how robust the new study's conclusions are. More funding would greatly help.
I read the paper. Argh. There are data that contradict the conclusions, which makes me wonder why the paper was published. Did the journal editor write a note along with publishing the paper? The measure of global average sea level rise has shown an increase in rate--- where did all of that water come from if not Antarctic?
Great post, Greg.
Your third ref. in the Resources section goes to the Colorado Sea Level page, not to Al Jaz.
John, I think you are basically correct. I did not address that in my post because it is an argument that draws from wrongness of a study rather than veracity of a study, so it may get too convoluted. But yes, if the Antarctic is suddenly taken out of the equation, then other causes have greater meaning. But, they probably don't.
Desertphile, this is a head scratcher from the peer review aspect. At the very least the reviewers/editors should have insisted that the time frame be clarified, and the snow/not snow contradictions be explained better (or just explained).
Then again, if Zwally et al have erred, let us also not discount the possibility that various death rays are being tested in Antarctic lairs of supervillains, causing heat disruptions and melting the ice from underneath. Basal melt? Try laser melt, folks.
The GRACE data is quite clear. Mass loss exceeds mass gain.
Trying to infer the mass of a thick ice sheet from altimeter data requires estimating the density of the ice very precisely. The density of ice is a strong function of temperature near the melting point. As ice sheets melt, they do not only heat up from the top and bottom, they can heat up in the middle, due to ice flowing down (because liquid water is denser than ice) and freezing, depositing its heat of fusion where it freezes.
The temperature profile of the ice sheet is not known with sufficient precision to estimate the density of the ice accurately. At the melting point (at atmospheric pressure), the density of water goes from ~0.93 to 1.0. At higher pressures the density change is larger, from ~0.95 to 1.11 (at ~230 K).
As ice warms up, it expands. If the wrong density is used, the mass of ice in the bulk of the ice sheet will be miscalculated.
I quickly looked at the paper. They assume that the density distribution of the ice sheet is in "steady state". It is well known that the Antarctic ice sheet altitude is subject to short term fluctuations due to water flows at the base.
The GRACE data is much more reliable and is a direct measure of the mass of ice.
Another excellent article, Greg. You make a small goof in the third paragraph: You posit six inches of sea-level rise. The next sentence is "But the dozen or so medium size storms that will occur over a year (especially in winter when the storms come in from the Atlantic) will convert that foot of elevation into several horizontal feet of beach erosion, in a very short amount of time." You probably meant to write "that half a foot of elevation."
Looking at the paper again, and citations.... I am still missing something. At least two of the citations refute the paper's conclusions.... I assume I am just too ignorant to understand why those papers were cited. ERS and ICESat data contradict GRACE data , as different ice densities are involved--- did the writers just assume an increase in altitude (snow) equals hard, dense ice?
Not that it's important for me to understand. That's not my job.
I'm so tired of the mental gymnastics that global alarmists go through when the evidence goes against their proclamations, which are almost exclusively based on computer models that can be tweaked to support any conclusion that the programmer desires. Global temps have leveled off since 1998. The Antarctic is no melting but increasing. As the NASA study points out, since 1992 this has been decreasing sea levels, undercutting one of the basic apocalyptic global alarmist Book of Revelations visions.
michael goodwin: I’m so tired of the mental gymnastics that global alarmists go through when the evidence ....
I have no idea what those are. If you have an irrational fear, why share it with anonymous strangers on the Internet? There are health care providers trained to help you with mental and emotional issues.
I’m so tired of the lies and propaganda of those in denial as they cherry pick their way through the evidence, accepting with open arms every new study they think* supports their denial cause and rejecting as 'unscientific' (or worse) any that says something they don't want to hear.
[*Michael Goodwin @14 thinks the new NASA Antarctica study supports his agenda. How wrong can he be? NASA Scientist Warned Deniers Would Distort His Antarctic Ice Study -- That's Exactly What They Did.]
John Russell: "... accepting with open arms every new study they think* supports their denial cause and ...."
Denialists also love to insist computer models are utterly worthless, always wrong, and unreliable... until they discover a computer model that they believe supports their politics.
This recent ( April 2015) clip :
seems to indicate we're in probably soon going to be in a lot more trouble from melting Antarctican glaciers - including Eats as well as West Antarctica - such as Totten than the "present" (till 2008 in this papers case right?) trends seem to suggest. Reckon its well worth a watch as is the rest of that series.
Okay, so I Googled "global alarmists" to see what that/they is/are. This is what I found:
The first two sentences: "NASA is having to eat its own words. They’ve come up with satellite data that shows the arctic ice is growing, not shrinking."
Heh. I don't recall Arctic ice being mentioned in the study. No where is the phrase "global alarmists" defined.
Questions for michael goodwin:
During the last interglacial (Eemian ~130 - 115ka BP) mean sea level peaked at >6m above present values. The Eemian was, on average, 1 - 2C warmer than the present.
1/ Why was sea level >6m higher than the present?
2/ Where did the water come from?
I’m so tired of the mental gymnastics that climate change deniers go through when the evidence goes against their proclamations, which are almost exclusively based on self-centered wishful thinking that can be tweaked to support any conclusion that the denier desires. Global temps have not leveled off since 1940. The Arctic is melting not increasing. As the NASA study points out, since 1992 this has been increasing sea levels, undercutting one of the basic global conspiracy alarmist Book of Revulsions versions.
I find it peculiar that two authors in the Zwally et al 2015 paper, Zwally and Donghui Yi, were also contributing authors in the Shepherd et al 2012 paper - http://www.ess.uci.edu/.../science-2012-shepherd-1183-9.pdf - which, combining an ensemble of satellite altimetry, interferometry, and gravimetry data sets, showed that Antarctica, between 1992 and 2011, was undergoing negative mass ice balance and contributing to SLR, while Zwally et al 2015 showed a positive mass balance from Antarctica and reduced sea level from 1992 to 2008. The Shepherd et al paper was published in 2012, but included later data than the Zwally et al 2015 paper. Why did Zwally and Yi sign on the the Shepherd et al paper including data up to 2011, and then later submit a paper contradicting the Shepherd et al conclusions, but whose data extended only to 2008?
The starting date for both papers was 1992. The Shepherd et al paper included European Remote-Sensing (ERS-1 and ERS-2), which spanned from May 1992 to September 2010. The Zwally et al 2015 paper also included those data, but only included data from 1992 up to 2001. The Shepherd et al 2012 paper also included Envisat data, the Zwally et al paper did not. Both studies included ICESat data from 2003 to 2008. The Shepherd et al study also included GRACE data. So the Shepherd ensemble of data sets was definitely more inclusive, both in terms of data and the time interval covered.
The Shepherd et al study found that Antarctica as a whole, lost 1350 ± 1010 Gt of ice, whereas Zwally et al show an average gain of 82 ± 25 Gt per year from ICESat data (2003-2008), and an average gain of 112 ± 61 Gt per year from ERS data. The Shepherd et al interpretation of the data show an annual contribution to SLR from Antarctica from 1992 to 2011, but the Zwally et al interpretation of the data, which is a subset of the Shepherd data set ensemble, show an annual reduction in sea level from Antarctica from 1992 to 2008.
Michael Goodwin: I’m so tired of the mental gymnastics that global alarmists go through when the evidence goes against their proclamations, which are almost exclusively based on computer models...
You're tired? I'd love to grab you by the scruff of the neck, drag you to a blackboard, and force you to write 100 times: "THE CONSENSUS ON GLOBAL WARMING IS NOT BASED ON COMPUTER MODELS."
That still wouldn't equal the number of times I've heard this drivel from Denialists.
sdm: That might be mostly meaningless quirky history, but there could be some interesting factors at play. There are a lot of reasons a particular paper might be published at a certain time, with certain authors, with certain data. Sometimes to close out a grant you need to submit a promised peer review publication. This looks to me like a paper that might have been hanging around a while.
A key fact is that the basic honest (one hopes and strives for) of the scientific endeavor is sometimes going to come out with odd sequences of events like this. They had data, they had an analysis, they had results. More than once. the idea that results from within an overlapping team can contradict one another is more likely when the results are close, as is the case here. You don't "sign on" to a paper because it fits or does not fit other findings. Etc.
Dear Greg. I wish to thank you for your most welcome post. It is very informative. I'm writing a book about todays worrying world topics. Antarctica plays a vital role in it. I think your article is very clarifying. I would appreciate it very much if you would give me permission to publish your article under mentioning your authorship and Blog. I hope to hear from you. It would mean a lot for me. Respectfully Yours, Gerrit Bogaers, Laren NH, Netherlands, Friday 25 March 2016, 19.17 Dutch Time PM
Gerrit I send you an email. Cheers,