Glaciers in West Antarctica have thinned and accelerated in the last few decades. A new paper provides some of the first evidence that this is due to human activities.
by Eric Steig
It’s been some time since I wrote anything for RealClimate. In the interim there’s been a lot of important new work in the area of my primary research interest – Antarctica. Much of it is aimed at addressing the central question in Antarctic glaciology: How much ice is going to be lost from the West Antarctic ice sheet, and how soon? There’s been a nearly continuous stream of evidence supporting the view that the West Antarctic Ice Sheet is in serious trouble – perhaps already undergoing the beginning of “collapse”, which John Mercer presaged more than four decades ago.
Yet showing that the ice sheet has changed doesn’t really address the question about what will happen in the future. To do that, we also need to answer another one: How much of the ice loss that has already happened is a response to anthropogenic climate change? A new paper in Nature Geoscience this week is one of the first to attempt an answer, and that is what has inspired me to get back to RealClimate blogging. Full disclosure: I’m a co-author on the paper.
In this post, I’d like to provide a bit of context for our new paper, and to emphasize some points about our findings that are generally going to be lost in popular accounts of our work.
The key finding is that we now have evidence that the increasing loss of ice from the West Antarctic Ice Sheet is a result of human activities — rising greenhouse gas concentrations in particular. Now, some may be surprised to learn that this wasn’t already known. But the argument that humans are responsible has rested largely on the grounds that there must be a connection. After all, why should melting have increased only in the late 20th century, precisely when the impacts of anthropogenic climate change were becoming more and more apparent? It seems an unlikely coincidence.
As Richard Alley* put it:
It has been hard to imagine that the ice sat around happily for millennia and then decided to retreat naturally just as humans started perturbing the system, but the evidence for forcing by natural variability was strong.
To be sure, there have been studies suggesting a discernible anthropogenic impact on Antarctic surface temperature, particularly on the Antarctic Peninsula. And it’s known that the depletion of stratospheric ozone and the rise in greenhouse gases has caused the circumpolar winds to increase in strength. But there has been little direct evidence that what’s happening to the ice sheet itself can be attributed to human-induced climate changes. Consequently, there has been no paper published that makes a strong claim about this. Indeed, a formal solicitation of expert views in 2013 showed that opinion was pretty much evenly divided on whether observed changes to the Antarctic ice sheet were simply part of the natural variability of the climate/ice-sheet system. In stark contrast, agreement among those same experts was (and is) unanimous that Greenland is melting because of anthropogenic global warming.
Before getting into what is new in our paper, it’s worth starting with a bit of background on West Antarctica, and a review of the evidence for the role of natural variability. Since not everyone will want to read the play-by-play, I’ve put most of that in a separate post, here. I hope you’ll read it.
In short, glacier melt in West Antarctica has increased because more Circumpolar Deep Water (which is relatively warm) is getting from the ocean surrounding Antarctica onto the Antarctic continental shelf and reaching the floating ice shelves of the large outlet glaciers that drain the West Antarctic ice sheet into the ocean. As shown by Thoma et al. (2008) in a seminal modeling study in Geophysical Research Letters**, how much Circumpolar Deep Water (CDW) gets onto the continental shelf is strongly influenced by the strength and direction of the winds at the shelf edge. Essentially, stronger westerlies (or simply weaker easterlies) tend to cause more CDW inflow, and hence, more glacial melt.
Because of the important role played by the winds, many have assumed that there must be a link between the melting glaciers and the ozone hole. But the greatest control on wind variability along the coast of West Antarctica is the state of the tropics. Just as El Niño event causes widespread climate anomalies in the Northern Hemisphere — such as increased rainfall in southern California — it also causes changes in the West Antarctic. Indeed, the Amundsen Sea, where the largest West Antarctic glaciers are, is one of the areas on the planet that is most strongly dependent on the El Niño-Southern Oscillation (ENSO) (e.g. Lachlan-Cope and Connolley, 2006). In 2012, we published a paper showing that changes in the winds in this region in the last few decades, which correspond well with variations in the glaciers, are very well explained by changes in ENSO, and very poorly by changes in ozone. We also noted that because big ENSO events had occurred in the past, it was quite plausible that wind conditions not that different than those of today had also occurred in the past. Indeed, we have very good evidence from ice cores that climate conditions in West Antarctica in the 1940s were not very different than those in the 1990s.
It is clear from this work, and much other recent research, that ENSO plays a dominant role in determining the climate conditions in West Antarctica that are relevant to the ice sheet. And since there is little evidence for a long-term anthropogenic change in ENSO, this implies that natural variability in Amundsen Sea winds (driven by natural variability in ENSO) may be the primary driver of observed ice-sheet change in West Antarctica in the last few decades. This is what Richard Alley is referring when he says that the evidence for forcing by natural variability was strong, and it throws a lot of cold water (no pun intended) on the purported link with human activities. But that’s not very satisfying. It doesn’t answer the question of why glacier retreat is occuring now. This is where our new paper comes in.
The new work is led by Paul Holland of the British Antarctic Survey (BAS), with help from Tom Bracegirdle, Adrian Jenkins (also of BAS), Pierre Dutrieux (now at LDEO) and myself. What we argue, in brief, is that although ENSO does indeed dominate the wind variability in the Amundsen Sea on timescales from interannual to multi-decadal, there is also a longer-term trend in the winds, on which the ENSO-related variability is superimposed.
The graph below (Figure 1) summarizes the key finding. What is shown are the winds in the key sector of the Amundsen Sea, centered on ~71°S and ~108°W, with observations in blue, and model results in black and gray. The model results are from an ensemble of simulations, referred to as the “tropical pacemaker” or “PACE” runs, of the CESM climate model. Details are given in Schneider and Deser (2017). Briefly, what has been done is to adjust an otherwise free-running climate model (forced by greenhouse gas emissions) so that it follows the actual history of sea surface temperature in the tropics, but is otherwise left unconstrained by data. We use these experiments as an estimate of how winds have varied over the last century in the Amundsen Sea, a) given what we know happened in the tropics and b) given what the climate model’s physics dictates about how conditions in the tropics affect the Amundsen Sea. Critically, there is nothing done to make the model match observations outside the tropics. Yet the results are in superb agreement with the observed Amundsen Sea winds. While we can never know exactly what happened prior to the advent of satellite observations in the late 1970s, the PACE ensemble provides a set of histories that is plausible, and compatible with modern data. This is probably the best current estimate of how winds have in fact varied in this region.
What Figure 1 suggests is that the winds in this region have varied between easterly and westerly from decade to decade, throughout the 20th century. This is the natural variability associated with ENSO, and is no surprise. But in addition, there is a long-term trend. When averaged over several decades, the winds can be seen to have shifted from mean easterly in the 1920s through 1980s, to mean westerly thereafter.
The trend in the winds is small, and easily lost within the variability of individual model ensemble members, but it is robust (it occurs in all the ensemble members) and statistically significant. Moreover, we know its cause (at least in the model experiments): radiative forcing. Although these experiments also include radiative forcing changes resulting from the ozone hole, it’s clear that the trend in the winds begins well before ozone depletion begins in 1970s. Thus, the key forcing is greenhouse gases.
These results show that variations in the winds that have occurred at the same time as we have been observing the glaciers retreat (i.e., since the 1970’s) are largely attributable to ENSO, as we had thought. But at the same time, the prevalance of strong westerlies in the Amundsen Sea has gradually increased throughout the 20th century. That is, although anomalous westerlies tend to occur most often during an El Niño, the long-term underlying trend means that the likelihood of strong westerlies in any given year is increasing, regardless of whether there is an El Niño or not. Thus, the radiatively-forced change (the trend) accentuates the effect of the natural variabilty (ENSO). As we wrote in the paper, the recent wind anomalies of the last few decades “…reflect Pacific variability that is not at all unusual…. However, when superimposed on the anthropogenic trend, this variability produces periods of absolute westerly winds that are sufficiently anomalous to account for much of the current ice loss.”
Now, a couple of caveats:
First of all, our finding are “simply” the result of looking at climate model simulations. We don’t know exactly what happened in the Amundsen Sea in the last century. On the other hand, Figure 1 looks very much like the data from ice cores from West Antarctica: variability that can be related to ENSO, superimposed upon a long-term trend. (See e.g. Schneider and Steig, 2008 and Steig et al. 2013 for details.)
Second, we are assuming that the Amundsen Sea shelf-edge winds are indeed the most relevant aspect of the system to consider. Again, this is based on the body of work showing that the inflow of CDW onto the Amundsen Sea continental shelf is strongy controlled by these winds. But the physics linking wind variability and CDW inflow is complex, and not everyone agrees with our view on this. Indeed, it is most certainly an oversimplification. Furthermore, as many authors has emphasized, there are complex feedbacks and internal ice-sheet and glacier dynamics involved, and it’s not as if there is a one-to-one relationship between changing winds and glacier retreat. For an excellent discussion of this, see the paper by Christianson et al. (2018).
Third, even without the first two caveats, we are far from proving that the ongoing ice loss from Antarctica can be attributed to human-induced climate change. The challenge here is that the natural component of the wind variabilty is so large that actually detecting (with direct observations) the trend inferred from the model results is not likely to be possible for some time. As we say in the paper, “Decadal internal variability therefore dominates ice-sheet and ocean variability during the modern observational era (since 1979), and will continue to dominate observations for decades to come.” We are not likely to find the smoking gun any time soon.
That all said, our findings are supported by other experiments. It is not only CESM, which is the main focus of the paper, that shows a long-term trend in the winds. In fact, most climate models (i.e., “CMIP5” — see details in our paper) show the same thing. Also, we find that the better the agreement between a given model and observations, the stronger the trend. (Note that the wind speeds shown in the figure above are not anomalies. These are the actual modeled and observed wind speeds. As it happens, CESM has unusally low bias in comparison with observations.)
Finally, our findings provide an important opportunity to glimpse into the future. We examined additional results with CESM, from the so-called “Large Ensemble” (LENS) and “Medium Ensemble” (MENS) set of experiments. These are identical to those of the PACE set-up but without the constraint to follow the observed tropical sea surface temperature. The results are illustrated in Figure 2, below.
The ensemble mean trend in the LENS experiments is nearly identical to that of the PACE experiments, which further demonstrates that the trend is not part of the natural variability. Comparison between the LENS experiments, which uses the “business-as-usual”*** RCP 8.5 IPCC scenario for the future, and MENS, which uses RCP 4.5, shows that reducing greenhouse gases reduces the future trends.
This is a big deal! Although we humans have evidently caused a long-term increase in westerly winds along the Amundsen Sea coast (which is bad for the West Antarctic ice sheet), the future is not yet written (which is an opportunity). Lowering greenhouse gases to a more modest rate of increase might be enough to prevent further changes in those winds.
Of course, many glaciologists believe we have already passed the point of no return for West Antarctica. I personally think the jury is still out on that. But that’s a discussion for another time.
*The quote from Richard Alley is from a National Geographic article about our paper.
**Not all the most important papers are published in Nature or Science.
***Some people think calling RCP 8.5 “business as usual” is misleading. Hence the quotes.
Eric,
My questions can be seen in the Bore Hole “too hot to handle” category:
https://www.realclimate.org/index.php/archives/2004/12/the-bore-hole/comment-page-39/#comment-743376
Thanks!
[Response: Not too hot. Too foolish. If you wanted to know answers to these questions, you’d read the literature. But, please, don’t let my expertise and that of my coauthors get in the way of your confidence!
To be fair you did ask one sensible question, which was about snowfall increases. Yes, there has been an increase in snowfall over parts of Antarctica, which partly offsets the sea level contribution. But critically, it does not fully offset it. The most up to date calculation suggets about 80 GT/year of growth, vs. 220 GT/year of loss. You can do the math from there. –Eric]
It is remarkable how you were able to spin John Mercer 1976 Nature paper as a success story. So, the stability of Antarctic West sheet is measured by wind speed nowadays? Not by temperature record, satellite altimetry, or GPS position sensors embedded in the ice?
[Response: This particular paper makes no claims about the “stability of the ice sheet”. That’s a different subject, which we and others have published on extensively, using, yes, ocean and air temperature, satellite altimetry, and GPS positioning, among other things. If you are interested in more detail on the relationship between the and ice dynamics, I suggest reading the background material, extensively referenced both in my blog post, and in the paper. –Eric]
Damn it! Just read #46 After writing this post. Oh, well. Won’t be my first foray into the Bore Hole.
[Response: You’re forgiven. –Eric]
But quickly, is Eric Lundberg not conflating science and scientific method, in specific?
Though this is a science blog, it’s a damned blog, right? As such, it has a reason for being accessible to non-scientists, yes? That is, there is an intent to engage the pubic and spread knowledge, no? I have long suggested the colloquial over the pedantic for this reason. Too much pedantism and it becomes quite difficult for laypersons to penetrate the density of… stuff… and some important %, I assume, may be dissuaded.
It’s all science to me!
A. Lowering greenhouse gases to a more modest rate of increase might be enough to prevent further changes in those winds.
But the lower emissions pathway still had an increasing trend, so this is a strange onclusion. Also, if Antarctica has already been melting for decades, it makes little sense we’d expect it to stop by increasing atmospheric C more slowly.
It seems clear your work actually shows reductions of atmospheric C are necessary to try to stabilize Antarctica.
B. I am astonished I did not know ENSO affects AntSI. In 2015 I developed the hypothesis there was a measurable effect of EN’s on ASIE and posted it here at RC: https://www.realclimate.org/index.php/archives/2015/08/unforced-variations-aug-2015/comment-page-5/#comment-635199
Here’s a better listing of the data.
El Ninos Strength ASIE New Low (NL) Near New Low (NNL)
51-52 Moderate NL 53 NNL 54
52-53 Weak NL 53, 55 NNL 54
53-54 Weak NL 55
57-58 Strong NL 57, 58, 60
58-59 Weak NL 60
63-64 Moderate
65-66 Strong NNL 67?
68-69 Moderate NL 71
69-70 Weak NL 71
72-73 Strong NNL 73?,74?
76-77 Weak NL 77
77-78 Weak NL 79 NNL 80, 81
79-80 Weak NNL 80, 81
82-83 Very Strong NL 84, 85
86-87 Moderate
87-88 Strong NL 90
91-92 Strong NNL 91, 93
94-95 Moderate NL 95 NNL 97 (3rd)
97-98 Very Strong
02-03 Moderate
04-05 Weak NL 5, 7 NNL 6 (3rd)?
06-07 Weak NL 7 NNL 08, 09
09-10 Moderate NL 12 NNL 10, 11
14-15 Weak NL 16
15-16 Very Strong NL 16
18-19 Weak NL 19? k NNL 19?
https://ggweather.com/enso/oni.htm
Near New Lows
EN’s New Lows Combined
25 15 15
0.60 0.60 1.20
[Response: The relationship between ENSO and Antarctic sea ice has been known for decades. A classic paper on this is Yuan and Martinson, 2001.]
Re #54 [Response: The relationship between ENSO and Antarctic sea ice has been known for decades. A classic paper on this is Yuan and Martinson, 2001.]
Well, right. But a relationship with the Arctic has been considered to not exist. Yet, there is at least a positive trend of EN’s and low extents (and one wonders if there isn’t a clearer relationship with area and volume). We also know heat and moisture from the Pacific does have some effect on the ASI from recent studies.
Maybe it’s time to revisit the EN/ASI relationship?
It would seem to me that the kinetic energy of the wind (= 1/2*m*V^2) is the power source for the Eckmann pump delivering CDW to the melt zone, and that would imply a quadratic (or cubic since m per unit time is increasing as well?) increase in ice melt as time goes on. This would mean that the nearer we are to the tipping point, the faster we will approach it. Is this built into the models as an emergent property from the physics? Might there be some surprisingly surprising surprises in store?
[Response: The answer would perhaps be yes is this were a very simple system, with simple geometry. It’s none of those things. The physics is “built in” in the sense that climate models do the correct ocean physics, but a fully-coupled climate-ocean-ice model at the necessary resolution (< a few km in the ocean) doesn't exist. Our work in this paper looks only at the atmosphere -- the ocean, at 1° resolution -- is not well enough resolved to capture the kinds of important details you are referring to. Stay tuned for a few years.. --Eric]
The question is; if Mawson was able to do this below in 1911 to 1914, then why did this that I will furnish the link for in another post happen in 2013?
1911 to 1914 Australasian Antarctic Expedition (AAE)
Explorers during this time were focused on being the first to reach the South Geographic Pole. However, Mawson’s passion for advancing scientific knowledge inspired him to develop an Australian-led Antarctic research expedition.
Embarking on the Australasian Antarctic Expedition(AAE) with Mawson, John King Davis captained the Aurora with a crew, 31 expeditioners and materials for living huts, and wireless masts to establish the first radio communications in Antarctica. Five men set up a base at Macquarie Island, while the remainder sailed on the Antarctic continent to establish two bases: the Main Base at Commonwealth Bay led by Mawson, and the Western Base at Queen Mary Land led by Frank Wild.
http://www.antarctica.gov.au/about-antarctica/history/people/douglas-mawson
A photoprint of Commonwealth Bay, taken during the Australasian Antarctic Expedition in 1912.
https://en.wikipedia.org/wiki/Commonwealth_Bay#/media/File:F._Bickerton_looking_out_over_seas_near_Commonwealth_Bay.jpg
Sir DOUGLAS MAWSON’S second expedition on SCOTT’S Discovery to Antarctic waters south of the Indian Ocean and Australia is by this time already near the coast which he skirted and explored in the Summer of 1929-30. He identified Enderby and Kemp Lands, first seen by British explorers a hundred years before.
https://www.nytimes.com/1930/12/28/archives/mawsons-expedition.html
I would like an explanation of what caused these conditions in 2013, 100 years after Douglas Mawson being able to reach the shore of Antarctic in a sailing ship?
“A Russian vessel is stranded in ice off the coast of Antarctica with 74 people onboard, including the scientific team recreating explorer Douglas Mawson’s Australasian Antarctic Expedition from a century ago.”
[…]
Had the ship carrying the trio of explorers in 1912, the Aurora, gotten icebound the same way the M.V. Akademik Shokalskiy did, there would have been no rescue option and certain death.
[…]
One hundred years after Mawson’s journey, we still don’t know much about the Antarctic.
[…]
As may be expected, global warming might play a role in this, he suggests, particularly with respect to melted ice in the East Antarctic.
[…]Had the ship carrying the trio of explorers in 1912, the Aurora, gotten icebound the same way the M.V. Akademik Shokalskiy did, there would have been no rescue option and certain death.”
http://news.nationalgeographic.com/news/2013/12/131226-russian-ship-stuck-ice-mawson-trek-antarctica/?rptregcta=reg_free_np&rptregcampaign=20131016_rw_membership_r1p_intl_ot_w#
You state this in your article; “The key finding is that we now have evidence that the increasing loss of ice from the West Antarctic Ice Sheet is a result of human activities — rising greenhouse gas concentrations in particular. Now, some may be surprised to learn that this wasn’t already known. But the argument that humans are responsible has rested largely on the grounds that there must be a connection.” How much ice is being lost today from the West Antarctic Ice Sheet when the temperature at the South Pole, Antarctica is -58 °C?
Weather in South Pole, Antarctica
Now
-58 °C
Snow flurries. Scattered clouds.
Feels Like: -78 °C
Forecast: -52 / -55 °C
Wind: 20 km/h ↑ from Northeast
Location: Amundsen-Scott South Pole Station
Current Time: 30 Aug 2019, 23:17:59
Latest Report: 30 Aug 2019, 18:00
https://www.timeanddate.com/weather/antarctica/south-pole
[Response: Temperature at the South Pole has about as much relevance to melting ice in West Antarctica as temperature in New York City has to do with whether the swimming is pleasant in Hood River, Oregon. The ice in West Antarctica — a couple thousand miles from the Pole — is melting from below. This has nothing to do with air temperature.–Eric]
Eric, you say wind velocities in Fig 1 are actual, not anomalies. 1 m.s-1 is about 4 km.h-1. Seems a bit sluggish for latitudes well south of the roaring forties. .???
[Response: This spot is pretty much on the zero-wind velocity line, between westerlies and coastal easterlies. It’s really about the wind-stress curl, not the wind speed per se. The westerly speed goes with the curl, in the simple case where the easterlies stay constant. –Eric]
#57-59, John Swallow–
John, there were lots of problems with ships and pack ice in the ‘old days’, too. For instance, the loss of the Antarctic, back in 1903:
https://en.wikipedia.org/wiki/Antarctic_(ship)
Pack ice is highly variable, both in space and time, so just a couple of data points–such as the Mawson expedition and its thwarted recreation–don’t tell you much. An Arctic example of this variability is the contrast between the first crossing of the Northwest passage by the St. Roch, which took from 1940-42, and the second, which due to favorable ice conditions was accomplished in a single season in 1944. (This was the first time that had been done, and I expect it was a considerable surprise to Captain Larsen and his crew of Mounties.)
https://en.wikipedia.org/wiki/St._Roch_(ship)
https://en.wikipedia.org/wiki/Henry_Larsen_(explorer)#1944:_east_to_west
If you’re interested in climate change and its effects on the ice, you need more comprehensive measurements to be able to say what the larger ice trends are.
So, for example, you can go to the NSIDC page and see what the ice has been doing by a number of metrics. The link below, I think, defaults to monthly means for each year in the record, for the most recent month. So right now, it’s showing me a time series graph of the mean July extent for every year back to 1979, when satellite observations began:
https://nsidc.org/data/seaice_index/
You can see on the monthly graph that there had been a small (but, it is known, statistically significant) increasing trend for some decades. But in the last few years, that trend has abruptly reversed, and we’ve seen a bit of a crash in extent. It remains to be seen how long-lived this crash will be, and the causes of both the increase and the crash are, I believe, being actively investigated. (There are hypotheses about the increase, but I’m not aware that there is yet a robust scientific consensus.)
It’s also important to be clear what is meant when one says “ice”. What you and I have been talking about is *sea ice*–the ice floating upon the ocean’s surface and forming a semi-mobile “ice pack”. In the Antarctic, the sea ice is mostly a seasonal phenomenon, expanding in the Austral winter–ie., at this time of year–and melting back to the shore in the “warm season” of January and February. That’s why sailing ships were able to discover Antarctica, way back almost to the days following the Napoleonic wars:
https://en.wikipedia.org/wiki/History_of_Antarctica
(Note the Austral spring and summer dates involved.)
But the big story for climate in the Antarctic isn’t the sea ice, as in the Arctic, but rather the permanent ice sheet that covers most of the continental mass itself. That ice sheet–and its dependent marine-terminating glaciers–are what the current post is about. So, while ultimately warm enough conditions will melt both the sea ice and the ice sheet, the two are very different beasts, with different short-term trajectories and characteristics, and we need to keep clear on that distinction.
I never use Wikipedia as a sources of information on any matter such as the climate for this reason.
“Good research and citing your sources
Articles written out of thin air may be better than nothing, but they are hard to verify, which is an important part of building a trusted reference work. Please research with the best sources available and cite them properly. Doing this, along with not copying text, will help avoid any possibility of plagiarism. We welcome good short articles, called “stubs”, that can serve as launching pads from which others can take off – stubs can be relatively short, a few sentences, but should provide some useful information. If you do not have enough material to write a good stub, you probably should not create an article. At the end of a stub, you should include a “stub template” like this: {{stub}}. (Other Wikipedians will appreciate it if you use a more specific stub template, like {{art-stub}}. See the list of stub types for a list of all specific stub templates.) Stubs help track articles that need expansion”
http://en.wikipedia.org/wiki/Wikipedia:Your_first_article
#62, John Swallow–
In other words, you’ve found an excuse not to consider information you find inconvenient.
Please try not to be so ridiculous.
John Swallow @62 says “I never use Wikipedia as a sources of information on any matter such as the climate for this reason….”
Yet John actually quoted wikipedia himself several times above, along with the NY Times and both for the climate issue. So his assertion doesn’t make much sense.
John Swallow:
AFAICT, this very ‘John Swallow’ has a history of swivel-eyed AGW denial on the Internutz, e.g.:
-throwing ‘alarmist’ around, on Greg Laden’s blog in 2011;
-ineptly but scornfully reviewing a book that supports the climate-science consensus, on goodreads.com in 2013.
He’s manifestly DK-afflicted (at the least), thus presenting him with verifiable counter-arguments may not be productive. Should we explore his implicit warrants and values instead? Followups to the new UV thread, if you please.
nigel, #62–
Indeed he did. Yet another way in which he should try to be less ridiculous.
A little reminiscent of our beloved non-president’s serial amazement at Cat 5 hurricanes:
https://www.cbs.com/shows/the-late-show-with-stephen-colbert/video/Uxh4FvmXbKw_RsbMaQVXS0Q5SR_C52Mc/trump-s-hurricane-tweets-not-overburdened-with-accuracy/
Eric: I respectfully submit that you reasoning appears to be flawed. The problem begins with Alley’s perspective, which reeks of confirmation bias.
“It has been hard to imagine that the ice sat around happily for millennia and then decided to retreat naturally just as humans started perturbing the system, but the evidence for forcing by natural variability was strong.”
There are many examples of change that is regional rather than global. A dramatic retreat of ice in one particular location, Glacier Bay, took place mostly during the end of the LIA. The retreat of glaciers in many, but not all, locations began with the end of the LIA (and is continuing during the last half-century of rapid GHG-mediated warming). Inherently unstable systems such as ice in contact with water holding back glaciers will undergo dramatic change whenever gradual change crosses a stability threshold. IIRC, the floating ice shelves on the Antarctic Peninsula have broken off before AGW. Cause clearly CAN be regional and chaotic, rather than global. If you believe the end of the LIA was due to a change in natural forcing (rather than unforced variability), the global retreat of glaciers began before anthropogenic forcing. When everything that is happening is attributed to anthropogenic forcing, we are suffering from confirmation bias. Alternative hypotheses deserve consideration.
Suppose you are correct that westerly winds are dramatically more effective at melting ice under parts of the WAIS and that the westerly winds have increased from 40% of the time to 50% of the time this century. One might expect that 40% of the time for MANY MILLENNIA westerly winds been been undermining parts of the WAIS. If the percentage of westerly winds has risen to 50% of the time for the past half-century, is the cause of an impeding collapse the past half-century of modestly more rapid undermining or past millennia of undermining?
If you could prove that WAIS was completely stable during millennia of exposure to 40% westerly winds, then a 10% increase in westerly wind could be responsible for looming instability.
The more serious problem is that the increase in westerly wind does not correlate well with the increase in anthropogenic forcing. According to AR5, anthropogenic forcing increased roughly linearly about 0.05 W/m2/decade from PI to 1970 and roughly linearly at 0.4 W/m2/decade from 1970 to present. In your Figure 1, most of the increase in westerly wind appears to have developed before 1970. The percentage of westerly wind arguably has been constant since 1970. Whatever teleconnections exist between the Equatorial Pacific and the West Antarctic exist, it isn’t obvious why you are assuming that these teleconnections are transmitting FORCED change rather than UNFORCED variability. AFAIK, the biggest source of variability in the Equatorial Pacific is ENSO (unforced), not AGW (forced).
My apologies if this amateur analysis is flawed due to absence of some critical knowledge.
[Response:We aren’t assuming anything about the forced vs. unforced variability. This is a modeling study so we know what the forced variablilty is — it’s anthropogenic. You are absolutely right that the largest source of variability is the unforced part, related to ENSO. Hence our statement that natural variability is going to dominate the wind variability in the Amundsen Sea for decades to come, just as it has for the past few decades (my previous work!).–Eric]
Eric,
I’m noting this here because you are obviously engaged with this thread. It’s a personal communication so I’m quite happy with it being boreholed if you decide that that’s best.
You lamented about the excessive time that would be needed to adequately police the comments and the excessive resources it would take to automate the system.
I find it amazing that someone as bright and caring as you can’t see the obvious:
You guys have an incredible resource available that you ignore. This site’s commenters spend tons of time without compensation here, specifically because they want to help and find it fun.
Obviously, all you guys need to do is allow for “citizen moderation”. Pick several volunteers to take over moderation duties. I’m sure a creative structure would result, perhaps with a two-level path to the borehole, perhaps with comments separated between the few that are on topic and productive that appear right after a post and then the scrum. You know, like “above the fold page one newspaper articles” vs page 19 burial. You guys would only have to read the cream of the crop and could add value with inline comments instead of (poorly) acting like preschool teachers.
Seriously, you guys are smart enough to solve a problem instead of throwing up your hands and declaring that solutions are impossible.
Use the free resources you have available. Free programming, free moderation. Free whatever you want. Because your current model sucks.
“The problem begins with Alley’s perspective, which reeks of confirmation bias.”
I don’t see it: to me, that simply sounds like someone observing a very obvious correlation, and saying that it deserves investigation–particularly since it explicitly acknowledges evidence to the contrary.
@Frank, #67:
No evidence for globally coherent warm and cold periods over the preindustrial Common Era
https://www.nature.com/articles/s41586-019-1401-2
Regarding Al Bundy’s recommendation at 68.
I know he has an important day job, but Ray Ladbury would be the perfect moderator. Put him in charge of the Bore Hole and the Crank Shaft and general moderation and this whole place would pick up. He could create new environs like maybe Neoliberal Reveries or MAGA Safe Space or the Innumerical Void (reserved for Weaktor) – and if the post doesn’t deserve banishment to an outer hell, he can just heap a little scorn on the poster.
You might even want to consider paying him.
I know I should get boreholed for this, but hey, you only live once.
@O, #70 reminds me that Neukom et al showed that there was “No evidence for globally coherent warm and cold periods over the preindustrial Common Era.”
I would agree that there is no evidence for for globally coherent periods 2 degC COOLER OR WARMER THAN THE LONG-TERM AVERAGE over the preindustrial Common Era. Now we can start debating how much lower the limit should be than 2 degC, and I personally do think that limit should be lower. The abstract to the Neukom’s paper merely says:
“In particular, we find that the coldest epoch of the last millennium—the putative Little Ice Age—is most likely to have experienced the coldest temperatures during the fifteenth century in the central and eastern Pacific Ocean, during the seventeenth century in northwestern Europe and southeastern North America, and during the mid-nineteenth century over most of the remaining regions.”
This doesn’t tell us how much lower average GLOBAL temperature was at any time from the fifteenth through the nineteenth century; it just says that the minimum temperature occurred in different places at different times.
After seeing all of the “spaghetti graphs”, I’m perhaps unreasonably skeptical of our ability to quantify pre-industrial warming and cooling. Von Storch and others showed that more noise added to pseudoproxy data, the more variability is suppressed in reconstructions. Large volcanic eruptions have caused forced cooling in both the instrumental and pre-instrument period. Esper showed that TRW did a horrible job of reconstructing those perturbations. Then we have about 0.45 degC of warming from 1930-1945, perhaps 0.3 degC of which was unforced). As far as I know, that was effectively “global” warming. Does that show up in the proxy record?
On the other hand, I’ll challenge anyone who says that the 0.9 degC of warming over the past half century is precedented. This is one of those cases where the absence of statistically significant evidence for warm or cool periods is not “evidence of absence” of warm or cool periods”, because our limited ability to define and detect them. (Just my opinion.)
@#69 Kevin McKinney. The normal scientific method is to construct a hypothesis and perform experiments or discover evidence that might refutes that hypothesis. Alley’s approach seems to be look for evidence that rising CO2 is responsible for everything. There is an Alley AGU lecture (ca 2010) showing how every major temperature change in the last 500 million years can be explained by changes in CO2. Very persuasive, I listened twice. Since then, I’ve noticed that recent estimates of CO2 levels during the PETM really are not high enough to be explained by CO2; you need to cherry-pick the reports with the highest values and apply an ECS inconsistent with EBMs.
Then I read an AGU call for papers for a special session on the Miocene: “The Miocene (23 to 5.3 mya) is a crucial, dynamical interval in Earth’s history that provides unparalleled insights into the functioning of greenhouse climates. At times during the Miocene, Antarctic ice volume was half modern, the Arctic Ocean was ice-free in winter, and extratropical temperatures nearly as warm as in the Eocene. This is an enigma, because the continental configurations and ocean circulation were much closer to modern than in the Paleogene, and atmospheric pCO2 was in the 300-600 ppm range. Taken at face value, this implies either a system highly sensitive to greenhouse gas forcing or the presence of still unexplained forcings and feedbacks.” Alley admitted that the Miocene had been a problem, but there was some new information that showed CO2 could be responsible.
When one starts by assuming one knows the cause of some phenomena, one often doesn’t treat that cause as a hypothesis. One forgets that the scientific method should be to perform experiments designed to reject that hypothesis. Or study all the evidence that is available and decide whether each piece is consistent or inconsistent with that hypothesis. It is human nature to not closely scrutinize evidence that supposed our ideas and ignore or not retain evidence that disagrees. All humans are very susceptible to such confirmation bias, but as scientists we are expected to recognize and struggle against that problem.
jb,
While I appreciate the vote of confidence, I fear you ignore three things:
1) I am only one of many voices on here
2) I am not a climate scientist.
3) As they say of John Adams in the musical “1776,” I’m obnoxious and disliked.
Perhaps a rating system might be more appropriate–thumbs up/thumbs down or point-based. Of course, the problem with such a system is that it can be gamed, so some sort of registration system would be needed.
Going to chime in on the volunteer moderator suggestion/request by AB: Yes please. Ray Ladbury would be fine with me as moderator if he has time and interest, but we probably need two or three folks to share the duties.
Mike
Eric (response): The reason that we refer to the *rate* of increase is simply that we were comparing two model scenarios: RCP 8.5 and RCP 4.5. In both cases, CO2 is still increasing, but the rate is lower in RCP 4.5. And in that scenario, the winds no longer increase much, if at all.
AB: So, suppose that, as a cutter, you open a vein. The doctor suggests that you stop cutting or at least reduce the rate at which you further open the vein.
That would only be relevant if your vein wasn’t already opened enough to eventually kill you.
So, you said that you aren’t convinced that WAIS is toast. What leads you to the conclusion that the current bleeding can go on forever without fatal consequences?
So because Pacific variability “superimposed on the anthropogenic trend” move in the same direction, this is ‘ evidence that this is due to human activities’? Man, this is some poor science. If I did this kind of sloppy research as a biomedical basic scientist, I’d get no funding, my postdocs would starve and my lab be shut down.
First of all, there is absolutely no evidence that this is due to human activities proof current temperature increases are anthropogenic. After all, we are just coming out of an ice age, so rising temps are to be expected, at least by sane people.
Second, if those two superimposed curves move in the same direction doesn’t mean that one of them is the cause for the other. Most likely, both move in the same direction because of a third factor, e.g. sun activity, low cloud cover etc. The latter has recently been shown to completely explain global warming.
A 77: there is absolutely no evidence that this is due to human activities proof current temperature increases are anthropogenic.
BPL: Just because you’re not aware of the evidence doesn’t mean there’s no evidence. There is a mountain of evidence that the current warming is anthropogenic.
A: After all, we are just coming out of an ice age, so rising temps are to be expected, at least by sane people.
BPL: “[C]oming out of an ice age” isn’t a physical process. If you’re talking about Milankovic cycles, we passed the peak of the interglacial 6,000 years ago and were cooling since then, until we hit the industrial revolution.
A: Second, if those two superimposed curves move in the same direction doesn’t mean that one of them is the cause for the other. Most likely, both move in the same direction because of a third factor, e.g. sun activity, low cloud cover etc. The latter has recently been shown to completely explain global warming.
BPL: Who showed that, where, and when? [CITATION NEEDED]
Anonymous,
Evidently you haven’t heard of conservation of energy. I’d suggest you Google it. Temperature doesn’t rise without an energy source.
thanks eric
#78, BPL–
Plus one.
And I’d add snarkily that I hope this is not a characteristic sample of ‘anonymous’s’ “proofing” skills:
‘Cause if it is, he might well “get no funding, [have his] postdocs… starve and [his] lab be shut down.”
” If I did this kind of sloppy research as a biomedical basic scientist, I’d get no funding, my postdocs would starve and my lab be shut down.”
Yeah, yeah, yeah. Anonymous, doesn’t know diddly, and can’t write a decent sentence while bragging. His claims MUST be true.
Oh, by the way, Thinkaholics Anonymous (Congrats on that 20 year pin), the post is about a model and how it performed against reality. So the scientists who did the work know precisely what changed and no, it wasn’t sunlight. Of course, instead of blathering about the sun you could have Googled solar iridescence and looked at a graph. But I understand. That would have lost you your shot at a 25 year pin.
“Briefly, what has been done is to adjust an otherwise free-running climate model (forced by greenhouse gas emissions) so that it follows the actual history of sea surface temperature in the tropics, but is otherwise left unconstrained by data. We use these experiments as an estimate of how winds have varied over the last century in the Amundsen Sea, a) given what we know happened in the tropics and b) given what the climate model’s physics dictates about how conditions in the tropics affect the Amundsen Sea. Critically, there is nothing done to make the model match observations outside the tropics. Yet the results are in superb agreement with the observed Amundsen Sea winds.”
I am not a scientist just an interested observer. Can anyone point me to any articles anywhere that look at more than singular factors that may or may not affect changes in climate or in the case of this article the changes in the wind patterns in the Antarctic. I would think geological events, undersea volcanos, earthquakes (a slight shift in the Earth’s axis from the Indonesian earthquake of 2004) etc., could have an effect on climate patterns. Since climate has continually changed from the formation of the planet to the current inter-glacial period, I would think that many factor contribute to shifts. I think too many people confuse climate with weather.
continued from 82
So, once they confirmed that the model’s output with regard to Amundsen Sea winds matched observations they ran the model for periods where there were no observations of Amundsen Sea winds so as to derive said winds.
From there the indirectly derived winds can be used (with the appropriate caveat) to inform other research. This advance is one that allows the advancement of other understandings, which makes it valuable beyond its scope.
Nigel,
I find it amazing that those with the ability to make incremental improvements fervently believe that incremental improvement is the way to go even when someone who changes paradigms before breakfast is standing right in front of them.
The world is changing. Unless one is one in a thousand, AI and robots will soon make one useless (from a capitalist perspective).
Wow, can see why he Anonymous didnt want to use a real name.
AB, a few thoughts on moderation:
1. First recognize boards go through sequences of calm, chaos, assholery by the core group. That latter is very much a constant. So, does any given period of civility/non-civility matter?
(Ask the core questions first.)
2. The worst thing for any board (on-line forum) is when the moderator and a core group become “institutionalized,” i.e., they dominate the board and are hostile to outliers/outside-the-boxers (relative to the core group.)
3. This board was an example of #2 very, very much over the last three years, and still is. It doesn’t seem so bc I, Thomas, and other outliers are posting very little. When I do, I am largely ignored – even when I simply post links to important science. Literally, over the las three years, my science-only posts get virtually no comments while people complaining about my “style” or non-scientific thinking – according to them – get threads that can go on for months.
That, AB, is when you know a board is calcified.
4. 1 – 3 are not complaints, they are observations setting up my suggestions:
A. Boards go through phases. So be it.
B. Because of personal differences, moderators must be of an eclectic mindset, or a very narrow mindset, and nothing in between. Choose to be institutionalized (very, very easy, and very 1994) or open (very, very difficult and more democratic.)
C. If A, just get your core group (aka the Peanut Gallery in this case) to be moderators and drive any interesting dissention away. A calm, focused, less-than-creative board will result,
D. If B, then select an eclectic group to be moderators. A more quarrelsome, more creative board will result.
E. With good moderators, either C or D can succeed. What is good moderation?
E.1. Have clear rules as to what gets moderated.
E.2. Have moderators collaborate on any moderation that is due to style, language, rudeness, etc. Clear, direct insults should always be moderated. Denial, too.
E.3. Decide what gets snipped, what gets deleted, what gets Boreholed.
E.4. Do not set moderators up as gods that cannot be questioned. Totalitarianism breeds distrust.
F. Modration slows a board down and screws with continuity. If you have a group with time on their hands, then set up a site where you copy and past problematic posts, the moderators suggestion, and discuss. Ideally, a 1-day max acceptance time is the result.
F.1. It could be set up as a passive system: Here’s the original, here’s my moderation.
F.2. Any comments, suggestions or objections must be made within X hours.
G. if you don’t do F, then you have to deal with moderators having to moderate themselves as a group, which can work, but is slow and messy.
All that said, this would be the “Least Change for Maximum Effect” (permaculture principle) change I would make: Make the Borehole a moderated space for utter nonsense, OT posts and denialist posts. Let those that want to waste their time do it there.
Comments are now closed. The conversation is starting to go off the rails.
No, winds are not influenced by undersea volcanos and slight tilts of the earth’s axis don’t occur on decade timescales.
–Eric Steig