Guest post by Verity Jones
Man is not the primary cause of change in the Arctic says book by Russian scientists
Forget the orthodox view of Arctic climate change – this book has a very different message. (h/t to WUWT commenter Enneagram)
Published last year, this is a synthesis of work by the Russian Arctic and Antarctic Research Institute (AARI). It sets out the data and experience of scientists over 85 years, drawing together much already published in the area. For a book that is billed under a climate change heading, this is actually more an antidote to the hype usually associated with warming in the Arctic. A few pages of each chapter are available on-line and even that is well worth reading; no doubt even better in its entirety.
The Preface sets the tone of the book very clearly – “.…scientists have predicted a significant decrease in sea-ice extent in the Arctic and even its complete disappearance in the summertime by the end of the 21st century. This monograph presents results of studies of climatic system changes in the Arctic, focused on ice cover, that do not justify such extreme conclusions.” “Many studies and international projects, such as the Arctic Climate Impact Assessment (ACIA), attribute the air temperature increase during the last quarter of the 20th century exclusively to accumulation of greenhouse gases in the atmosphere. However these studies typically do not account for natural hydrometeorological fluctuations whose effects on multiyear variability, as this monograph shows, can far exceed the anthropogenic impact on climate.”
The book begins by examining the major effects of the Polar Ice caps and their overall stability on Earth’s climate – affecting albedo, and regulating the heat flux from the sea to atmosphere. Climate variations are discussed and the WMO’s “30 year average” definition of climate is not considered applicable in the Arctic because fluctuations in the polar climate are so large.
Chapter 2 looks at what is known about changes in sea ice in the 20th century. The Russian data sets probably hold the most extensive information available for the first half of the century due to interest in the Northern Sea Route in the 1930s. In addition, measurements of ice thickness also go back to the middle of the 1930s when they were taken regularly for coast-bound ice at many of the Polar stations.
It is particularly interesting what they say about Arctic air temperatures (Chapter 4). “Periodic cooling and warming events are evident in air temperature fluctuations in the Arctic during the 20th century, similar to changes in ice cover.” A cool period at the beginning of the 20th century was followed by what is commonly referred to as the “Arctic Warming Period” in the 1920s-1940s. Relative cooling was widespread between the late 1950s to late 1970s, followed by the current warming period peaking in recent years. Gridded average temperature anomalies for 70°-85°N produce a curve that fits a polynomial trend to the sixth power and the cycle periodicity is 50-60 years (Figure 4.1). Other indicators in Arctic and Antarctic support this cycle and show its global nature. On the subject of polar amplification, whereby weather and climate variability increase with latitude, a number of models and explanations are discussed. None of these involve CO2.
The authors point out there is an abundance of hypotheses as to the possible causes of climate and ice variation and climate change (a ‘long-term’ phenomenon) but these lack detailed long-term data. They state “where data do exist, we should prefer data to computer models”; they believe model projections of future ice area fluctuations are unreliable. Actually, they have some deliciously scathing remarks about climate models.
“The models neglect natural fluctuations because they have no means of incorporating them, and put the entire blame for climate changes since the 19th century on human activity.”
On possible future changes they predict that “..in the 21st century, oscillatory (rather than unidirectional) ice extent changes will continue to dominate Arctic seas.” A new ice maximum in 2030-2035 is predicted (Figure 6.1) and this will have major implications for shipping in the region.
From the results of spectral analyses, they conclude that there are 50-60 year cycles and less prevalent ones at 20 years, 8-12 years and 2-3 years. These are closely related to variations in general atmospheric circulation. In the longer term the decreasing trend of ice extent may be a segment of a 200 year cyclic variation responsible for the Medieval Warm Period and Little Ice Age. Much of the discussion about solar effects is behind the paywall for the book, however there are some strong conclusions about solar effects on Arctic climate. Despite the small variation in Total Solar irradiance (TSI) through solar cycles, solar activity may have a greater effect on high latitudes because of interaction with the Earth’s magnetic field. Solar system “dissymmetry” (barycentre) influences are also mentioned as closely corresponding to the 60 year cycles.
The authors conclude that the simulation by the general circulation models does not appear to reflect the cyclic features in Arctic ice extent and climate, and, if their cyclic interpretations of climate variation are correct, ice cover will continue to fluctuate as there is little connection with the anthropogenic burning of fossil fuels.
Climate Change in Eurasian Arctic Shelf Seas: Centennial Ice Cover Observations. Authors: Ivan E. Frolov, Zalmann M. Gudkovich, Valery P. Karklin, Evgeny G. Kovalev, and Vasily M. Smolyanitsky. Published by Springer/Praxis (2009) ISBN 9783540858744
============================
Verity is one of WUWT’s moderators and contributors. She also has her own website at Digging in The Clay. Be sure to visit it and bookmark it – Anthony
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Hmm, chances of anything from this book appearing in future IPCC publications?
Anyone dropped a copy into Tammy, see how “Open Minded” he is about this?
Rabe says:
October 16, 2010 at 3:25 am
Hmmm – so if you measured say light intensity outside, and made a high order polynomial fit to it, and found an oscillation, this would show that night and day were an illusion and an artefact of naive maths?
Hey Tamino (Grant Foster) – here’s Arctic data more than 6 years back.
There is tons of ice research buried in the Russian archives that has never seen the light of day in western circles, being published or written in Cyrillic. In the late 60s I was spending a lot of time on snow literature, and by far the most detailed and voluminous research was Russian, even on snow alone. The Russians have had a pragmatic and strategic interest in ice and snow for decades. Nobody knows more about north polar ice aand snow then the Russians. Nobody.
Maybe these guys could explain how HMS Investigator managed to sail to its final resting place in 1853, as Grant Foster can’t/won’t, and the models of history produced by the Hockey Team certainly don’t.
Maybe Grant Foster and The Hockey Team would prefer the alien spaceship theory to accepting that their science does not permit any other explanation
Excellent synopsis, Verity.
So, so far I can see the ENSO 2-3yr cycle, the 12 year would be close enough to the sunspot cycle, the 20 year would be close enough to the 18.6 year lunar cycle, the 200 year possibly the deVries (PDF) cycle which lines up with Sporer, Maunder and Dalton minimums, and whose next one is due about now.
The sixty year cycle is the most fascinating, because everywhere I see temp reconstructions, it’s there. Would this be Jupiter and Saturn lining up with Earth? Looking here, they don’t seem to really line up in a way that would have an effect on Earth at the important dates.
If we can find the underlying cause of that cycle, I think we’ll go a long way to figuring out our “climate”. No CO2 involved.
richard telford says:
October 16, 2010 at 2:07 am
“The models neglect natural fluctuations because they have no means of incorporating them”
This is only true for simple deterministic models. It is not true for the more complex GCMs used in IPCC AR4. Unforced (natural) variability in the models arises from the stochastic nature of the models. For example, I have a 600 year unforced model run that contains many cycles in model climate.
Its good to know that stochastic models are exhibiting cycles – its just that they are not the right cycles.
There is very little chance of a model being useful as a predictive tool for reality as the chaotic interacting systems that make up the climate are not all known and many of their inputs are not understood or unknown. Even relatively obvious and high impact issues such as the effects of clouds are not yet understood or quantified despite the entire AGW hypothesis relying on hyrdologic cycle feedbacks. A chaotic climate model would need to have very accurate starting values for all the variables that affect climate miss one input or get the timing, periodicity or scaling wrong for an input in a chaotic system and it will rapidly deviate from the real world it is trying to model.
All the models have proven is that climate system cycles can appear out of a chaotic system.
phlogiston: …naive maths?
Umm, …yep. My experience is that most natural phenomena map either (partly) to a polynomial order of at most 3 or are cyclic (for which one would use fourier analysis) or are chaotic (for which one would better give up) or the sum of some of those.
Guess my preference on this topic. 😉
It seems cyclic … I wonder why.
richard telford says:
October 16, 2010 at 2:07 am
“This is only true for simple deterministic models. It is not true for the more complex GCMs used in IPCC AR4. Unforced (natural) variability in the models arises from the stochastic nature of the models. For example, I have a 600 year unforced model run that contains many cycles in model climate.”
What code are you running, Richard? What equations are being solved? What boundary conditions, initial conditions? Is there complete documentation someplace?
There is a geopolitical fight brewing over parts of the Arctic which has little to do with science, of course. Canada is enforcing licensing of foreign vessels in waters it claims, Norway is moving forward with oil exploration, Russia is moving forward with updating its nuclear-powered icebreaker fleet and Russian, American and Canadian claims overlap. The US icebreaker fleet is out of commission currently, from what I’ve read. Part of the cover story uses the idea that AGW will open up the Arctic to shipping, but frozen or not, they want to rape the resources asap.
Rabe says: (October 16, 2010 at 5:34 am )
“My experience is that most natural phenomena map either (partly) to a polynomial order of at most 3 or are cyclic (for which one would use fourier analysis) or are chaotic (for which one would better give up) or the sum of some of those.”
Don’t just read my summary. They do use wavelet analyses (check out pages 18-20 which you can read in the preview on Google Books, but not on the Springer site) and this is mentioned in the conclusions as how they arrived at the 50-60 year cycles and less prevalent ones at 20 years, 8-12 years and 2-3 years.
Re Paulhan 0505am and the 60-year cycle, Scafetta has a 2010 paper which also uses a 60-year (and 20-year, and quadratic) cycle which is tentatively attributed to Jupiter-Saturn modulation of solar processes, in an unspecified manner.
Thus does astronomy turn, in some sense, into astrology, i.e. an influence on humans by the planets, via very tenuous connections.
Rich.
richard telford says:
October 16, 2010 at 2:56 am
Juraj V. says:
October 16, 2010 at 1:41 am
The models do not capture the AMO-like variation at all
Not so. See http://www.nature.com/ngeo/journal/v3/n10/full/ngeo955.html
Not sure how this proves your case without admitting more error. The paper talks about the primary drivers of the Atlantic sea temperatures being volcanic activity and external solar forcings. Not sure how the models take this into considerations since temperature series’s have been omitting the waves and troughs of climate for awhile now due to “massaging of the data.”
Dr. Hansen years ago started smoothing the multidecadel data, and the reasoning is correct when trying to figure out a climate picture you would want to smooth out 60 year cycles, but when you are modeling using GCM philosophy of boundary condition, this smoothing introduces error that can not be removed. But that error is a different story….the fact remains that GCM’s are built based on the multidecadel data being smoothed to a fraction of its original, so any assumptions including effects of carbon dioxide on sea temperatures will be especially wrong.
As I have stated previously, building models like this makes them incredibly unreliable outside of 60 year windows where they have a shot at being close. And attempting to predict temperatures in any specific area of the world….good luck with that. And so we go on with the argument about boundary conditions versus initial state, but boundary condition systems have never been found to provide reliable results for small portions of the system. That is just not what they are designed for. They are a big picture system that sees the system as a whole, but might get entire areas wrong. That is if the input is correct and other assumptions are correct too..
Doug @ur momisugly 12:42 AM
Yes, it is my impression that the intransigence of the BRICs at Copenchaos was from strong doubt as to the anthropogenic effect on climate. It seems that they were playing along with the charade as a strategy to shakedown the developed West over its imagined carbon guilt. The strategy failed, but their chagrin over its failure was masked by their outburst of outrage over the backroom deal promoted by the thug mentality of one Western leader in particular. Whoa, did he play his unintended and hamstrung part.
===================
I disagree. Al Gore knows more than the Ruskies. He said that the Arctic could be free of ice by 2013. /sarc off
http://nobelprize.org/nobel_prizes/peace/laureates/2007/gore-lecture_en.html
Here’s the source for Al Gore’s belief. All these fools are soon going to end up with egg on their faces.
http://news.bbc.co.uk/2/hi/7139797.stm
Ian W says: October 16, 2010 at 5:16 am
A chaotic climate model would need to have very accurate starting values for all the variables that affect climate miss one input or get the timing, periodicity or scaling wrong for an input in a chaotic system and it will rapidly deviate from the real world it is trying to model.
It does not have to be a linear model to be useful. Tsonis et all have been using neural net analogues of the oceanic currents and are getting fairly good descriptions of nature. Unfortunately the idea has not been attractive to the rest of climate modelers. I believe it is the way modeling should go.
One will not get exact numbers, but will get attractors and track trajectories, i.e. ranges in the parameter space. The GCMs think they are simulating this with their large number of runs, on their linear models, but that is nonsense.
richard telford says:
October 16, 2010 at 2:07 am
“The models neglect natural fluctuations because they have no means of incorporating them”
“This is only true for simple deterministic models. It is not true for the more complex GCMs used in IPCC AR4. Unforced (natural) variability in the models arises from the stochastic nature of the models. For example, I have a 600 year unforced model run that contains many cycles in model climate.”
I see that you have reasonably well-confirmed physical hypotheses about natural regularities which underly the natural fluctuations and that you have programmed them into a GCM so that you can learn more about what they imply. That’s nice, but the really valuable part of your work is in the physical hypotheses that you have used to explain and predict these natural regularities. Give us those. We are not interested in the models.
Paul Coppin – A relative who was , for many years, a CPO on several US Navy subs,
said:”The Russians know more about cold and ice,than George Hamilton knows “Toasted”..” For those who aren’t familiar George was famous for his tanning habits.
later made commercials for a Taco Chip co…
Do not discount Russians. They still are space and seafarers, the US is Navel Gazing now…
maelstrom
Russia is also building floating nuclear power plants for use in the Arctic. Oil and gas exploration being one reason.
O/T
I was shocked to find that my local education board is part of 10:10. (10:10 is the group that produced the children snuff video. While my local education board has a major influence on what my children are being taught.)
I am not even in the UK.
I recommend readers check their local community education committees and schools by performing a search here
http://www.1010global.org/allcountries/education/learn
I recommend writing to your local school board if you are concerned.
richard telford says:
October 16, 2010 at 2:07 am
Yes, you can easily write a simple program to simulate lottery tickets, but that doesn’t guarantee the winning ticket will be generated, nor does that mean that your model run will hindcast all the winning tickets. On the other hand, you could ‘adjust’ the previous winning numbers as proof of the model accuracy and sell the model program for a tidy sum. The model will look genuinely accurate until someone audits the ‘winning numbers’ and finds they were tampered with.
I wonder if this will be included in the next IPPC report?
I think not.
It will be the “Great Control Knob” instead!
I can see it; The Prime Ministers and the Presidents sitting there in their offices.
Playing with the “Great Control Knob”.
Maybe they will get a special suitcase for it so they can carry it around at all times?
Nice to see someone doing real science (gather data, analysis, etc.).
Models are nice and all, but mostly they just inform our ignorance. Where the model goes off the rails is where you have something to learn.
(I ran a supercomputer center doing modeling for many years. You can use the model, but must TEST it against reality, not the other way around. We would get 1 in 10 or so ‘model runs’ that would give faulty plastic die issues (weld lines, voids). That was with ONE well characterized fluid in a precisely constrained system designed by experts for the purpose of being easy to model / likely to succeed. And a 10 hour run time.)
FWIW, at a talk about Neural Nets a guy doing DOD research on tank identification told a story about one of their “successes”. Got a 100% accurate identification of Russian vs US tanks. They started cutting the pictures down to see what part was the “key”. Gun? nope. Turret? Nope. Treads? NOPE. Surface PAINT? NOOOPE.
In desperation, they put in a picture of a tree NEAR the tanks. 100% accurate. It could even identify pictures of Russian tanks without the tank…
Seems the Neural Net had “learned” to identify high speed spy film ASA / grain vs high quality low speed low ASA low grain posed US tank photos.
Use, don’t trust, and always verify models and neural nets…
Verity, didn’t you find the same or very similar 60 year cycle in one of your postings about the GHCN data? I remember seeing a graph rather like that one up top, but I’ve forgotten what the article name was…
REPLY: and there’s the 60 year cycle Scaffeta was talking about too. See:
http://wattsupwiththat.com/2010/10/13/scafetta-on-60-year-climate-oscillations/
-Anthony