The cycle has, or is getting ready to start now. Bond et al. 1997 Last one was dated 1410 years ago.
[Response: There is no evidence of this “cycle’s” existence in the Holocene (last 11,000 years). What Bond showed is that there are quasi-cycles that Bond argued have the same cause as the ones during the glacial period. Personally, I’m very skeptical about this, but it is just a matter of conjecture on both parts. Unfortunately Gerard Bond died last year so he can’t defend himself. Certainly he had a more nuanced view than “the cycle is starting now”! –eric
How confident is the number 1470? What’s the source of that figure, and the error range on that? I was looking at the GISP data, and the periodic behavior is undeniably present. But using GISP temperature from 12000 yBP to 50000 yBP, I get a “period” of 1461 +/- 4 yr. A minor point, admittedly — but I’m curious.
[Response:Stefan wrote a very nice paper on this some years ago, showing that the +/- is remarkably good, much better than one expects from "noise". However, there are serious questions about this, because this nearly pure "cycle" only shows up in GISP2. It is not demosntrably so precise in any other record. It is not a minor point, because we don't expect such perfect cycles in sloppy geophysical systems like oceans. We'll say more on this soon, I'm sure.--eric]
[Response: I’m currently in Nairobi so a bit slow to respond, but the paper Eric mentioned is here. Indeed, more on this will come out some time, I just haven’t had the time to write it up yet. -stefan
Has any correlation turned up between mud/sediment cores and the ice cores? I wonder, given the recent news about the various ways plankton actively affect the oceans, including churning the upper 100 meters, if any of the cycles could reflect big changes in which species predominate over time.
Grin. To me those lines were reminiscent of the patterns in ecosystems of predator-prey-parasite interaction.
Could we actually be in the beginning of a new D/O event?
There is a warm sea surface anomaly that has been persistent in the area between Greenland & Newfoundland for the past few years (even in winter) that may be part of the cause for loss of ice on the margins in the south of Greenland.
Is this a coincidence or the start of a trend related to a new D/O event?
[Response: This is a cute idea but there is absolutely no evidence for it. Let's not translate climate from ice cores -- more than 12,000 years ago -- direclty to the present, shall we. The recent cooling over Antarctica is well understood and most certainly has nothing to do with ocean circulation. See. e.g. here. Note that it is essentially meaningless to talk about 'trends' that are only 3 years long, because the natural year-to-year variabilty, especially in Antartica, is simply too great.--eric]
[Response: My answer to that is a clear “no”. There is no DO events in the Holocene – that’s an observational fact, if you look at the ice core data – the last one in my view is the end of the Younger Dryas. Since then we are in a permanently warm state, which already is like a permanent DO event. Convection is in the Nordic Seas in the Holocene, so if convection starting in the Nordic Seas is the mechanism of a DO event as our theory has it, then you can’t get a DO event any more. -stefan
Very interesting. About your conclusion: What could be the physical link between a small variation in solar cycles (causing no major climate change of the “first kind”) and a treshold-crossing in oceanic circulation? That is: how such a solar variation could imply a second kind change without noticeable effects on first kind change?
I think I found the article you’re referring to (Rahmstorf 2003, GRL 30, 1510). Stefan concludes that the periodic phenomenon responsible for D/O events has an internal “clock error” <= 7% of the period, or just about 100 yr. The original estimate of 1470 yr. for the period apparently comes from Grootes et al. (1993, Nature, 366, 552). I haven’t got a copy, but if anybody does and wants to report the results of their period analysis, please do!
Of course the underlying clock “period” can be nailed down more precisely than the clock stability. My analysis based on the CLEANEST Fourier spectrum (Foster 1995, Astron. J. 109, 1889) indicates a period of 1461 +/- 4 yr. However, the error calculation is predicated on a null hypothesis which we know to be false; under such circumstances a better estimate of the error range is the FWHM in a Fourier transform. This gives an estimated period of 1461 +/- 27 yr. So, the underlying “driving force” has a very precisely determined period, but it may exhibit random “cycle-to-cycle” fluctuations contributing to some of the irregularity in the timing of D/O events in the GISP2 data.
BTW, Stefan’s analysis of sources of the irregularity in event timings was, in my opinion, very nicely done. It reminds me of “O-C” analysis as is regularly applied to variable stars in astronomy, but Stefan treats the sources of variation correctly (as is, sadly, too often not the case in astronomy); my compliments.
I don’t know that that I would feel comfortable saying there is no evidence of warming entering the area. There does appear to be a warming in the NA at the Labrador Sea interface.
Looking at the Rossby Wave effect on the Jet Stream it appears to be affecting the Bermuda High whaich has moved from the normal 60 Deg. W to around 30 Deg. W which would be sufficient for an extension of warmer tropical waters moving towards this region and appears to be causing the higher temperature subtropical SSTs to pool in the this area.
(Based on the 2005 data there had appeared to be a notherly wind reducing the ice pack in the Bering Straits and now we appear to have the opposite effect this Fall in both the Norwegian Sea and Labrador Sea areas.)
When looking at the ice pack data this Fall the extension of the Arctic ice pack towards the Bering Strait seems greater and the icepack build up aroung the islands at the western edge of the Labrador sea is not as great.
To be certain I still need to get a link to insure that there is no chance of a Gulf Stream Meander or ring from setting up shop near by as well. In the meantime, here are a few links that seem to support this possibility:
[Response: See Stefan's repsonse following mine, above. Neither of us are saying there is no evidence of warming in the North Atlantic. We're saying that jumping to the conclusion that we're seeing the beginning of a DO event is pure conjecture, and contradicts pretty much everything we know about such events in the past. --eric]
I know about a physical phenomenon called “stochastic resonance”: Some noisy systems may greatly amplify weak signals, expecially if thresholds are involved.
May this be the case? Some unknown natural forcing exists with a period of 1470 yrs, and the climate system is subject to stochastic resonance during glacials. During interglacials, it is not as sensitive as in glacials, or thresholds are shifted to other values, which are no more critical (I know all this is very generic, but I’m not phisicist or a climatologist, sorry!).
A good understanding of this may be very important: does global warming keep us out of danger (of D-O cycles: I’m not suggesting that global warming is good!), or it makes the whole system more sensitive? We don’t know, of course. But I’d like to know if studies are ongoing on this topic.
This was a great post. As usual, but this time the argument is among my favourites!
[Response: Well, good idea! Our papers on DO events triggered by stochastic resonance are here and here. -stefan
[Response: On the other hand please see our paper showing where the stochastic resonance idea may be wrong. Don't get me wrong here; I think the papers Stefan cites are important, and largely correct, but the statistics supporting "stochastic resonance" as a good model are simply not convincing, at least to me.-- eric]
To reiterate, according to the above article there are two types of warming, one of heat distribution, and the other of total heat accumulation.
The comment I would like to make is that an abrupt climate change that affects global heat distribution would also affect regional deforestation, devegetation, precipitation, and desertification. In other words, abrupt climate change would increase rapid climate change.
For instance, changes in ocean convection over the north Atlantic would affect the jet stream, which would have a dramatic effect upon regional precipitation, probably leading to acute drought in the northern hemisphere, which will weaken and push the northern jet stream toward the Arctic.
The point I am trying to make is “when it is claimed that DO events represent a much larger and more rapid climate change than anthropogenic global warming,” perhaps DO events do cause rapid regional climate change larger and more rapid than anthropogenic global warming generally.
In other words, a DO event (brought on this time by anthropogenic global warming) should be seen as larger and more rapid climate change than anthropogenic global warming.
I think that temperature extremes brought on by global warming lend themselves more to rapid climate change than higher average temperature. The extremes are more a matter of regional heat distribution than global heat accumulation.
Great article RC, just goes to show how complicated the climate is and why this site is necessary for all of us interested quasi scientific types. Fantastic stuff, better than buying any book (and I have bought a few of em).
As impressive as it is, how sure are we that our growing understanding of the climate isn’t on too small a time-scale? Having read the recent NYT article about evidence on a paleontological time scale makes me think that we might get close to underastanding climate until panama falls into the sea, or some more likely, but equally momentous thing happens.
[Response:That NYT article you quote is extremely misleading, and in some places simply wrong. We're putting up a post on that today. -eric]
I suspect that the possible warmer temper intrusion above the NA into the Arctic appears related to the surface fetch and pooling related to the anti-cyclonic zone near 30 Deg. W (“Bermuda High”). Now if you back up and go look at Greenland Sea – Norwegian Sea interface you can see a clear intrusion between Svalbard Is. and the east coast of Greenland here:
Whether this is a abnormally high intrusion of warmer temperatures remains an opportunity for additional study. As to whether this is due to winds or GS also remains. (I suspect it is purely due to the force behind the GS as the increased Arctic Easterlies should push most of the warmth westward. However, with the “Bermuda High” east and slightly north it may be sufficient to drive the Gulf Stream with enough force to intrude deeper into the Arctic.)
The important point is that this really is not news, anyone with a PC and the interest should be able to see this data. The tough part would be to define how or why this is occurring. So far this seems to be eluding us all… (Please note that during most of the years in which there appeared to be significant support for GW, that the Bermuda High appeared in it’s “Classic” position of approximately 60 Deg. W and 30-35 Deg. N during the peak of summer, and at times would retrograde towards 25 Deg, N and 75 Deg. W.) Is it possible the Rossby Wave effect on the Jet Stream is the driving force behind the NAO and PDO? If so what drives the deviation of the Jet Stream?
In addition to Alley’s Two Mile Time Machine, another excellent book that describes the Greenland ice cores and what they might mean in terms of abrupt climate change is
Climate Crash: Abrupt Climate Change and What It Means for Our Future by John D. Cox. Cox is a science journalist, albeit a good one, and the book is written at about the Scientific American level. Perfect for someone with a technical background coming to this topic for the first time. I’ve read it twice in succession because I find it so helpful for understanding and supplying context for the articles here in RealClimate.
If the theory is that DO events are driven by some “regularly varying external factor” why would these events not be seen during the Holocene? And if this is externally driven shouldn’t the driver of DO events still be in play today?
[Response: In our model they only occur under glacial conditions, not under Holocene conditions. The reason is that the Holocene has a different pattern of ocean circulation with vigorous convection in the Greenland-Norwegian Seas. It's already in something like a permanent DO event state, the basin is well-flushed, and small variations in freshwater input, which accumulate under glacial conditions, cannot accumulate. Hence we don't get events there from the same forcing that triggers events in the glacial, when the circulation is close to a threshold so a small forcing can trigger big changes. -stefan]
At the risk of being one of the non-experts confounding first and second kinds of climate change, I assume it is possible both could happen at once. Using the bathtub analogy, we could be running more water into the tub at the same time we are sloshing the water around.
Is there any evidence that something like this might be involved in the transitions between glacial and interglacial periods? In other words, there is general warming occurring at the same time ocean circulation changes to accelerate the general warming. Or, a general cooling accompanied by a change in flow in the opposite direction?
[Response: Absolutely. During the transition from glacial to interglacial climate, the evidence is that the MOC became more vigorous. So in Greenland it got warmer both because of higher CO2, more sunlight at high latitudes during summer, AND because of increased poleward heat flow. It is still debatable just how much the increased poleward heat flow really increased, but it almost certainly did change.--eric]
I brought up GW to my anthro students, and one claimed she read in SCIENCE recently that the Antarctic is cooling, thus disproving GW. The other claimed her father in Utah (name??) is the leading GHG scientist, and he says GW is not happening.
Are they tapped into some reality that most others are missing?
[Response:[Some alternate reality perhaps, but not this planet. Antarctica cooled in the 1980s to 2000. But it was warming from the 1960s to the 1980s. If you look at the overall trends 1957-2005, for most of the continent, it didn't change at all (on average). Oh, and the Antarctic Peninsula is the fastest warming place on the planet, and has been for some 50 years or more. None of this either proves nor disproves anything about global warming. We've had several posts on this, beginning with the very first week of RealClimate. Have your students read these articles here and here then come back to you. As for the famous Utah professor, um.. not anyone we know of. — eric
Comment by Lynn Vincentnathan — 9 Nov 2006 @ 9:59 PM
I would like to elaborate on my previous posting (i.e. abrupt climate change vs. rapid climate change, and heat distribution vs. total heat accumulation).
Dr Lovelock is known for a unifying theory of earth’s eco-systems. In other words, the gestalt of earth’s eco-systems. He came away from a visit to the Hadley Climate Centre with a belief that earth’s eco-system is going to crash sooner than was conventionally believed by other scientists.
This is directly relevant to abrupt climate change, because heat distribution unbalancing eco-systems means earth will soon catch a “fever” (I apologize for the anthropomorphic imagery). In other words, the total heat accumulation won’t abruptly rise (which is impossible per the above article), but abrupt climate change will affect heat distribution, which will unbalance eco-systems, which will led to rapid global warming (i.e. rapid heat accumulation).
When it is written that within a decade or two abrupt climate change will occur and the earth will switch from the mild climate of the last ten thousand years (the Holocene), to a hotter dryer climate that has been responsible for mass extinctions in the past, what is meant is not that total heat accumulation will instantaneously jump.
Instead, what is meant is that heat distribution will change, consequencially leading to a higher rate of total heat accumulation. Abrupt climate change shouldn’t be dismissed as regional, and not be implicitly tied to an abrupt rise in total heat accumulation, but instead should be seen as a big step toward an increased rate of total heat accumulation (i.e. a fever).
Let me go farther: the problem with many global climate models is they don’t properly evaluate events of imbalanced heat distribution. Under the right circumstances glaciers melt abruptly, regional forests and soil become carbon emitters abruptly, and oceanic/atmospheric currents shift abruptly, leading to rapid global warming.
In my opinion, these events only work one way (i.e. to make global warming more rapid). In other words, we may warm up quicker than consensis scientific opinion predicts, but we are going to cool down much slower after the “fever” peaks.
Changes in the atmosphere, ie; increased moisture levels at higher elevations, are contributing in driving the climate change, so I believe the ocean models need to include atmospheric data and how those combine to drive wind currents.
Additionally, the solar energy that is now being absorbed rather than reflected in the Artic must be a major contributor to weather patterns.
The composition of the troposphere has been altered through manmade contributions, such as: exhaust from combustion engines accumulating over a more than a century, and the destruction of forests (which act as sinks and replenish oxygen, as well as keep surface temps cooler..)
The comparison of ice cores and postulating effects of gases and temperatures appears intriguing, but one must admit that what is occurring NOW is a first in a Global lifetime.
Just explain how anything in the past history of the Earth can compare to the changes that have occurred in the last century…
[Response: You are right on your first point - the models we use of course are coupled climate models, not ocean models, so they account also for changes in the atmosphere. Also, we do not use the past as an analogue, comparing the present with ice core data to draw conclusions. Rather, we build a physical understanding of past changes, and assume that the laws of physics stay the same. -stefan]
Comment by Joseph O'Sullivan — 10 Nov 2006 @ 6:33 PM
Good post stefan. Just curious, when you say:
It’s already in something like a permanent DO event state, the basin is well-flushed, and small variations in freshwater input, which accumulate under glacial conditions, cannot accumulate.
Can you clarify this? I am under the impression that freshwater only affects the thermohaline belt if a massive amount of water is released at once. For instance, a giant lake on top of a glacier in canada bursts free and changes the salinity of the water. You seem to imply there are other ways freshwater can affect the circulation.
[Response: You're talking of cold events that are a response to massive freshwater influx to the Atlantic and a subsequent shut-down of deep water formation. That holds for Heinrich events (surges of continental ice into the Atlantic) or for meltwater influx (e.g., the 8k event). Now this post was on DO events, and these are warm events. There are no signs of massive forcing of these events, and my feeling (based on our model runs) is that they don't require that, because the ocean was very close to a threshold, so some subtle changes to the freshwater budget were all it took to trigger DO events. -stefan] Changes in Antarctica are much smaller and more gradual, as it is far from the centre of action and the vast reservoir of ocean around it acts as a heat store
Is it just the sheer amount of water that protects Antarctica? Or does the circumpolar circulation act as an “water door” much like commercial “air doors” operate?
I vaguely rember a Carl Winsch talk where he noted that if you take two signals with 365 and 365.25 day periods, the resulting beat frequency would have a 1460year period! – someone analysing ice cores mucking up the number of days in a year :)!
“To prevent this drift between the calendar year and the astronomical (seasonal) year, we add one extra day every four years. Thus, over the four year period, we have 1461 days, not 1460, for an average of 365.25 days per year. …. This innovation was imposed in the year 709 AUC (ab urbe condita, after the founding of the city), when Julius Caesar regulated the calendar. …”
But if we really want to apply this theory, maybe we should use the difference between the “calendar year” (365.2425d in the Gregorian calendar) and the actual tropical year (365.2422d). This gives us a “beat period” of 1,217,475 years.
Which of course proves that Nostradamus was right. ;)
I am not a scientist by occupation, so you can stop reading here if you are only interested in comments by scientists.
What strikes me in most discussions regarding rapid climate change is that everyone seems to be looking for “causes” and “effects”. If X happens, the Y and Z will follow. However, from what I have read about chaotic systems it would seem that it is futile to search for simple causes and effects. A chaotic system simply moves in a trajectory in its phase space and it is impossible (impossible not because of lack of data, but because it is – well – impossible) to forecast precisely where it will be at a given moment in the future. If we forecast that the system will be at point A and it in fact turns up at point B, we can not conclude that something specific has “happened” to the system. Well, we can do it, by all means, but it will be a quite arbitrary conclusion. In the case of the Earth’s climate, there are obviously billions of things “happening” all the time, and it is the sum of all these events that brings the system to a certain state at a certain point in time. We will effectively never find out what the human impact on the climate system actually is – whether we are heating the planet or cooling it or having no impact whatsoever. The only sensible thing to do, therefore, is to minimize our ecological footprint. Then, when disaster finally strikes, it will at least not be our fault!
Comment by Anders Lundqvist — 14 Nov 2006 @ 8:07 AM
“the European Project for Ice Coring in Antarctica (EPICA) established a precise link between climate records from Greenland and Antarctica using data on global changes in methane concentrations derived from trapped air bubbles in the ice.”
while your graph notes oxygen.
I’m curious about how one gets temperature from methane or oxygen concentration.
[Response: The methane in the bubbles in the ice is used to line up the records on the same time scale (since changes in methane are recorded in both cores at pretty much the same time). The proxy used for temperature is the isotopic composition of the ice (the ratio of 18O to 16O in the H2O (ice)) which is measured independently. The linking of the time scales through methane then lets you see the leads and lags of the system and in particular demonstrates that there is a partial see saw in the temperatures at both poles. This would be impossible to derive without the matching because the other ways of dating the cores this far back have enough uncertainty that you wouldn't be able to tell whether the South was leading or following or behaving exactly synchronously. - gavin]
Why do you characterize the climate system as chaotic? It may have some chaotic behaviours on some time scales, but I think it might be best described as broadly deterministic with some chaotic aspects.
Re ” We will effectively never find out what the human impact on the climate system actually is – whether we are heating the planet or cooling it or having no impact whatsoever.”
This is just plain wrong. We know what effect humans are having. We may not understand all the details yet, but there’s no question that a) the world is warming, and b) human activity is most of the cause. You can’t just babble about chaos theory and assume it somehow overturns thousands of empirical observations. It doesn’t.
Comment by Joseph O'Sullivan — 22 Nov 2006 @ 12:43 PM
“BEYOND THE YOUNGER DRYAS
Collapse as Adaptation to Abrupt Climate Change in Ancient West Asia and the Eastern Mediterranean
Although Holocene climate events are relatively minor on a glacial/interglacial perspective, the small Holocene changes in the polar vortex and atmospheric storminess documented by O’Brien et al. (1995) would probably cause widespread disruption to human society if they were to occur in the future (Keigwin and Boyle 2000:1343).”
There is evidence to suggest that strengthening and more ‘stable’ polar vortices can cause rapid climate change.
The â��increased storminessâ�� noted by Dr. Weiss may be the â��amazingâ�� snowstorm that froze Dr. Thompsonâ��s little Andean plants during the 5200 YBP event.
â��In 2002, Thompsons team made a surprising find along the margin of the Quelccaya ice cap a remarkably preserved wetland plant that had been remarkably preserved under the ice. Later testing yielded viable DNA from the plant and dated it back 5,200 years ago.
This is a soft-bodied plant, he said. It had to be captured by a very large snowfall at the time, a snowfall and climate change that began very abruptly fast enough to capture a plant but not kill it. That is astounding.â�� http://researchnews.osu.edu/archive/quelcoro.htm
Thus, I believe it possible that there is a 1470-year solar cycle (perhaps 7 of the provable 210-year cycles â��overlaidâ�� over some other cycle) that can force rapid change.
“Borehole temperatures in the ice sheets spanning the last 6000 years show Antarctica repeatedly warming when Greenland cooled, and vice versa… The phenomena has been called the polar see-saw… Attempts to account for it have included the hypothesis of a south-flowing warm ocean current with a built in time lag… There is (however) no significant delay in the Anarctica climate anomaly…
While mechanisms involving ocean currents might help to intensify or reverse the effect of climate changes, they are to slow to explain the almost instantaneous operation of the Antarctic climate anomaly…”
Svensmark’s explanation for the polar see-saw is GCR effects on cloud formation. He provides calculatons and data to support the hypothesis that an increase in low level clouds above 75 degree South, can produce warming. This is opposite to lower latitudes, where an increase in low level clouds, results in cooling.
“In polar regions the clouds can have a warming effect if their re-radiation of long-wave radiation downward dominates over the loss of short wave solar energy blocked by the clouds.”