Study: abrupt shifts occurred in the ancient European climate

From the University of Helsinki and the “no SUV’s needed” department comes this study which suggests big cold snaps occurred right in the middle of the warm Eemian period. My only concern is perhaps they over-rely on climate models. For reference (and not part of the study) here’s the Eemian graph in context. Data sources listed int he graph. – Anthony

Slowdown of North Atlantic circulation rocked the climate of ancient northern Europe

Major abrupt shifts occurred in the climate of ancient northern Europe, according to a new study from University of Helsinki, Finland. The research reports that sudden cold spells, lasting hundreds of years, took place in the middle of the warm Eemian climate period, about 120 thousand years ago. These cold intervals saw a fall in temperature of a few degrees, and replacement of forests by tundra, at the study site in northern Finland. The Eemian, which took place before the last Ice Age, had a climate generally warmer than present. This has made the Eemian important for climate scientists assessing the modern climate warming.

According to the researchers, the sudden shifts of Eemian climate are connected to disturbances in North Atlantic circulation which happened during that time. Today, the warm oceanic currents of the North Atlantic maintain a relatively temperate climate in Europe. The future development of this oceanic circulation has been hard to predict, however, and possible disturbances have not been ruled out.

“These results strongly suggest that the North Atlantic circulation is sensitive to disturbance, with major effects in Northern Europe”, says University of Helsinki postdoctoral researcher Sakari Salonen who headed the study.

“Our results show that the oceanic circulation was in fact perturbed the last time climate was warmer than today. This is where our study significantly adds to the prediction of future climate done based on numerical models”, Salonen says.

The research was conducted by a multinational research group lead from the University of Helsinki. The team utilized intelligent computer algorithms based on machine learning.

“These new numerical methods significantly improve our understanding of abrupt changes of past climate”, notes professor Miska Luoto, also a team member from University of Helsinki

These results are based on a geological deposits studied in Sokli, Northern Finland. At this site, thick geological layers have been preserved in a deep depression, allowing them to survive through the following Ice Age.

“The Sokli site is unique in the northern parts of the world, which has made the site invaluable in the study of past, long-term climate change”, says Karin Helmens, a researcher from Stockholm University and a long-time coordinator of the studies at Sokli.


The study was funded by an Academy of Finland programme on long-term environmental changes and the Swedish Nuclear Fuel and Waste Management Company (SKB). The results were published in Nature Communication.


Salonen JS, Helmens KF, Brendryen J, Kuosmanen N, Väliranta M, , Goring S, Korpela M, Kylander M, Philip A, Renssen H, Luoto M (2018) Abrupt high-latitude climate events and decoupled seasonal trends during the Eemian. Nature Communications 9:2851.

0 0 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
July 23, 2018 12:13 pm

… yet, here we are, right at the best time we could be here
and people are falling for some s c a m telling them how bad it is

Wiliam Haas
July 23, 2018 12:24 pm

Apparently what they are telling us is that the climate cycling that we observe today was also observed during the Eemian. This is not new. Based on the paleoclimate record and the work done with models one can conclude that the climate change we are experiencing today is caused by the sun and the oceans over which mankind has no control. Despite all the hype, there is no real evidence that CO2 has any effect on climate and plenty of scientific rational to support the idea that the climate sensitivity of CO2 is zero. It is all a matter of science.

Greg Goodman
Reply to  Wiliam Haas
July 23, 2018 12:37 pm

Yeah, well I’d find it a lot more convincing if they did not screw around with an x-axis that suddenly changes scale by a factor of 40x.

Reply to  Greg Goodman
July 23, 2018 1:40 pm

If you’re referring to the chart above then that was not part of the original paper. Figure 4 in the paper has a uniformly scaled x axis.

Reply to  Greg Goodman
July 23, 2018 2:18 pm

If the x-axis weren’t compressed, you’d be complaining that the chart didn’t fit your scren.

Reply to  jorgekafkazar
July 23, 2018 2:59 pm


So just make it smaller.

Reply to  Greg Goodman
July 23, 2018 8:23 pm

You are correct that the way the chart changes scale on the x-axis is confusing at first. But I think they are trying to show the additional detail available in the modern time, while also showing the oldest period in context, without blowing out the size of the chart. Logarithmic scales on charts are an old, old practice & this is basically a variation of that.

July 23, 2018 12:26 pm

It doesn’t sound from that summary that they used climate models. They used “artificial intelligence” methods to teach computers to give the expected results. Surprise, they got the expected results.

Greg Goodman
Reply to  Anonymoose
July 23, 2018 12:38 pm

that sounds like we used AI to create out climate model, rather than them not using one.

Curious George
July 23, 2018 12:38 pm

Abrupt shifts still occur in Phoenix, AZ: Today’s max 116F, min 87F.

July 23, 2018 12:42 pm

Apparently supernovae explosions could be one of the reasons for the global climatic changes.
Scientist from the National Research Nuclear University (MEPhI) (Moscow, Russia), Institute of Physics (Trondheim, Norway), the Astronomy Department of the University of Geneva (Switzerland), Paris Diderot University (France), and The Sorbonne Observatory (Paris, France) suggest that ” a supernova explosion occurred about two or three million years ago within 220-450 light years of the Solar System ……”
comment image
“If the event we have described happens again, the Earth will be in for global climatic changes and much higher radiation levels. If, however, a supernova explodes within a 10 times shorter distance, the consequences for biota on our planet will be disastrous …. Dmitry Semikoz and his US and European colleagues have already shown that cosmic ray “attacks” on Earth resulting from supernova explosions within 300-600 light years approximately 3.2 and 8.7 million years ago should have wreaked havoc with the biological clocks of animals and accelerated biological evolution on the planet due to the dramatic surge in the number of mutations caused by radiation.”
If my memory serves me right this is the period when large herbivores were start disappearing and territorial spreading out of humans was occurring.

Reply to  vukcevic
July 23, 2018 1:55 pm

So whatever we do we’re toast.

Tom Abbott
Reply to  Robertvd
July 24, 2018 1:57 pm

The Earth might be toast, but the human race can survive in artificial habitats in space.

Rich Davis
Reply to  vukcevic
July 23, 2018 2:00 pm

Maybe I misread it, but it sounded to me like they aren’t claiming global climatic effects, but rather a surge in the rate of mutations driving creation of new species.

Reply to  Rich Davis
July 23, 2018 2:19 pm

“If the event we have described happens again, the Earth will be in for global climatic changes …”
it’s in the first paragraph below the illustration

Reply to  vukcevic
July 23, 2018 3:04 pm


I don’t think Rich is entirely wrong as the article goes onto describe said mutations if the event was ten times closer.

Reply to  HotScot
July 24, 2018 3:50 am

it said for 220+ LY distance
“If the event we have described happens again, the Earth will be in for global climatic changes and much higher radiation levels. ”

and for 22+ LY distance (10 times shorter than the above)
“If, however, a supernova explodes within a 10 times shorter distance, the consequences for biota on our planet will be disastrous
two statements referring to two entirely different type of events dependable on distance as measured in LY (light years)

Reply to  vukcevic
July 24, 2018 12:27 pm


Fair comment. My ignorance is revealed once again.

phil salmon
Reply to  vukcevic
July 24, 2018 12:18 pm

There has also been speculation that the end-Ordovician mass extinction could have been linked to irradiation from a nearby supernova. Not conclusive though.

July 23, 2018 12:48 pm

The team utilized intelligent computer algorithms based on machine learning.

Don’t know whether to laugh or cry. I hope this sentence was the product of a middle school report and not the work of a professional journalist/writer. The cluelessness about intelligence, algorithms, and computers is astonishing.

Reply to  Gary
July 23, 2018 4:34 pm

Bingo. The “intelligent computer” and/or (worse) “intelligent computer algorithms” are tells. Pathetic.

Smart Rock
Reply to  Gary
July 23, 2018 8:15 pm

Gary, you need to be able to distinguish between a news release written by someone who’s not a scientist, and the abstract and/or the full paper. The silly wording is from the news release. We see a lot of these and they make mediocre science look like bad science. They are designed to attract attention in an information-saturated world, not to provide information to scientifically literate skeptics like most of us.

Bruce Cobb
July 23, 2018 12:55 pm

Ah yes, the old “warming may lead to cooling via disruption of the North Atlantic current” gambit.

Pompous Git
Reply to  AndyHce
July 24, 2018 9:52 am

Thanks Andy. I enjoyed very much reading that at the time. I still have one small quibble. Richard Seager wrote that the activities of the Serpentine Swimming Club would be unthinkable on the opposite side of the Atlantic where temperatures are typically below -10C. I have friends who swim in the Antarctic…
comment image?custom=true&w=168&h=94&stc=true&jpg444=true&jpgq=90&sp=68&sigh=2eviGa0phxYdE7lZBDoVWw1NT8U

NZ Willy
July 23, 2018 1:03 pm

It looks to me like their conclusion that the North Atlantic current was responsible is just sheer speculation without any grounding in evidence whatsoever.

Joel O'Bryan
July 23, 2018 1:26 pm

“The team utilized intelligent computer algorithms based on machine learning.”

Those algorithms do nothing more than their training/learning process provides, that is an efficient mult-parameter/multi-output interpolation process. It is the human capacity for extrapolation and abstraction that allows us to imagine all sorts of future boogeymen. And to imagine going to the Moon and creating the technology to do it.

Reply to  Joel O'Bryan
July 23, 2018 3:07 pm


Oh come on. Give them a break.

They can’t keep calling them computer models, so now they’re AI models. Seemples.


July 23, 2018 1:43 pm

The x-axis changes scale in the middle. It goes from 5,000 years per division to 200,000 years per division (one of the divisions is 180,000 years).

I am reminded of today’s xkcd cartoon.

The problem with the graph is that nothing happened for a very long time and then something happened quickly. You want to show the context and the detail in the same figure. In drafting you use line breaks or magnified views to achieve the same end.

There’s nothing wrong with changing the scale of the x-axis, but it does set a trap for the unwary (which is the point of the xkcd cartoon).

Reply to  commieBob
July 23, 2018 2:16 pm

The demarcation in the middle of the graph is perfectly clear.
“it does set a trap for the unwary”
Good Lord, commieBob. How unwary do you have to be.
“Touring the Grand Canyon with your nose stuck in you cellphone and walking right off the cliff” unwary? Darwinian selection in action, not “survival of the fittest” but rather “death to the stupidest” unwary?

Good cartoon, BTW. I never thought of making a graph like that, could come in handy.
XKCD is always good.

commieBob, this is too good, I just saw it, my 15 min edit is still open.
American tourist falls off a cliff taking a selfie. Darwin Award contestant.

Reply to  TonyL
July 23, 2018 3:13 pm
Reply to  HotScot
July 23, 2018 3:29 pm

I always liked the Don Quixote cartoon.

Pop Piasa
Reply to  HotScot
July 24, 2018 3:43 pm

(This might work, might not… Trying for the halibut.)
comment image
(Gitterduuun! Smell me, WordPress!)

Reply to  TonyL
July 23, 2018 4:42 pm

Thanks Tony. I spent the latter part of my career working with students. They all have pieces of paper saying that they all have the requisite math skills. Somehow a few always managed to surprise me with the level of their unpreparedness.

I didn’t witness this one myself, it happened to one of my buddies. A student was having trouble with a lab. It turned out that he couldn’t solve for c given a = bc. How did he get out of high school?

Reply to  TonyL
July 23, 2018 6:02 pm

It seems this is becoming a commonplace method of perishing. Australian and British persons too;

Reply to  commieBob
July 23, 2018 2:33 pm

It’s a kind of a Manniac trick, choosing the x-axis jump point where the two parts of the graph ‘join’ at the same temperature level, creating an impression of continuity.

Reply to  commieBob
July 23, 2018 3:09 pm


Aw Poo!

I just spent an hour asking a whole lot of daft questions below, and you guys answered most of them before I posted.

July 23, 2018 1:58 pm

now, this is a bit of surprise
hmm …. maybe, but most likely maybe not.

July 23, 2018 2:08 pm

More wins reported today by the GWPF.

Finally, even the dummies out there are starting to “see though” all the phony green non$en$e.

1) Carbon Tax May Bring Down Canadian PM Justin Trudeau
National Post, 21 July 2018

2) Gallup Poll: Americans No Longer Regard Global Warming As A Main Problem
Breitbart, 22 July 2018

3) A Green Republican’s Carbon Tax Would Hike Household Energy Bills by $688
Americans For Tax Reform, 21 July 2018

4) Britain’s Pension Fund Association Rejects Climate Change Law For Investment Decisions
The Actuary, 20 July 2018

5) Are Britain’s Feed-In Tariffs For Renewables Dead?
GWPF Energy, 18 July 2018

6) King Coal To Rule For 20 More Years, Says Australia’s Energy Market Operator
The Australian, 18 July 2018

7) Told You So: UK Recycling System ‘Open To Fraud And Error’, National Watchdog Warns
The Sunday Telegraph, 22 July 2018

July 23, 2018 3:17 pm


I liked 4) the best. All these freaky deaky green types complaining about fossil fuels, either ignoring or oblivious to the fact that most pension funds invest in fossil fuels because it’s a safe bet.

They will all retire on fat pensions thanks to coal.

Ooo, the irony.

July 23, 2018 4:51 pm

1) makes me happy, but he was supposedly out next election as it was. The more pile on to that, the better.

Reply to  Daelyn75
July 23, 2018 6:29 pm

Justin has nice hair, and women like him. He is also a congenital idiot, so stupid he does not even realize that he is the front man for the Liberal crime syndicate of Quebec.

Justin is the most credible argument ever against giving women and imbeciles the vote! 🙂

July 23, 2018 2:20 pm

Hmmm, As per the graph above, for approximately last the 10,000 yrs the Earth seems to be in a stable “climate”…Before that, chaos..So they are complaining about ???

Reply to  Marcus
July 23, 2018 2:25 pm

They should be complaining about the state of the science given that questionable graph.

Joel O'Bryan
Reply to  Marcus
July 23, 2018 2:34 pm

They are Ehrlich-Malthusians.

So they are complaining that Earth’s human population will continue to increase in a slightly warmer, more bio-productive planet. We will consume more resources. Increasing pressures on ecosystems, especially in poorer 3rd World Countries with high birthrates, like Bangladesh, sub-Sahara Africa, Afghanistan, etc where it is likely to lead to regional collapses when the inevitable cold climate downturns hit and other black-swan events.

The watermelons are eagerly awaiting the collapse of humanity.

July 23, 2018 2:57 pm

I’m lousy at reading graphs. No idea why, perhaps it’s graphical dyslexia or something. However, I’ll have a go as this one seems straightforward enough, but I’m hoping for confirmation of what I think I see with the rest of you on here. So please bear with me, most of the questions (all of the questions) will be simplistic Janet and John stuff.

Nor are my questions much to do with the subject matter, just the graph. Sorry.

The ‘0’ (Zero) designated horizontal line on the X axis is presumably meaningless other than to demonstrate the temperature anomaly, not the actual temperate, but the difference between temperatures over the last one million years (1,00 x 1,000).

The black line (Liseckie, Raymo and Hansen), the dark blue line (Antarctic ice cores?) and light blue line (Greenland ice cores?), and most of the red line (Marcott)are presumably paleo (and I understand that to be observations of temperatures taken from fossils, tree rings etc.) observations.

The first thing that strikes me is that it appears the scale of the period ‘0’ (Zero) on the Y axis, is presumably the present time (far right hand side) to 20 thousand years ago, but it’s not the same as the scale from 20,000 years ago to 1,000,000, which give a distorted impression to an uninformed observer (me)

And I appreciate the graph isn’t designed for the uninformed observer and that this type of representation may be normal convention in scientific circles. It just makes it difficult for me to visualise. Wouldn’t the peaks and troughs from 20 K to 1000 K be far less dramatic and convey a different impression were they to the same scale as the rest?

So in the last one million years, there has been temperature variations as much as 15°F (around 9.4°C) many quite dramatic, although the 20 K+ scale exaggerates that impression. And we reached the depths of an ice age around 20 K years ago (LGM – Lower Grand Minimum? – No idea) and then as the planet begins to emerge from it, the Greenland and Antarctic data disagree with each other on two occasions by perhaps 5°F. That would suggest to me either two cataclysmic events (one labelled YD – even less of an idea of that one) which affected one hemisphere but not the other, or problems with data.

Then we get to the Holocene Optimum, perhaps ‘Optimum’ because it’s mankind’s optimum survival period? And the red line which looks to me for all the world like the state of my putter after missing the match winning shot, dislocating my knee on it, before hurling it into the blue rough. Many years ago!

The red line is mostly comprised of paleo data as I don’t think mankind had the desire, or the communications to be concerned with global temperature measurements until around 200 years ago. Possibly revealing more of my ignorance here but it appears accurate (?) temperature measurements didn’t begin until around the head of the putter (far right) when, of course there’s a marked increase in temperatures, which has been blamed on man increasing atmospheric CO2.

Which seems to me, at the least, a wee bit daft, and at worst, cavalier science. I’m not sure how accurate paleo temperature measurement are, but I’m pretty sure they cant be within a couple of °F even over the shorter 20 K period of this graph. Humans just aint that smart yet.

That being the case, there must be a margin of error applied to any measurement of that nature. Surely that should be represented on a graph of this nature, prominently.

So then I started to wonder about atmospheric CO2 over this period and I dug out my old faithful Geological time-scale of temperature Vs CO2. A graph even I can understand. I also believe it to be universally accepted as a fair representation over the life of the planet.

I hope it shows up.

For the last 2,000,000 years or so, it seems CO2 has been bumping along the bottom of meaningful life’s existence window, with a slight uptick today, geologically speaking. And what I see is that during that period, whilst temperatures have been fluctuating by what appears to be 2°C, CO2 is absolutely flat despite the claim of it driving temperatures.

Meanwhile, back to the putter head. How accurate is the observed date going back 200 years? I mean, ocean temperatures were probably taken by a cabin boy chucking a bucket over the side of a ship and the water temperature taken with a not easy to read thermometer. These were undoubtedly, predominantly along regular shipping rout’s, so the rest of the oceans were largely ignored. Even until Argo buoys were introduced more recently, ships were tasked with the job of sea surface temperature data.

Then there’s land temperatures, where the value of Stevenson screens were, I suspect, simply for local data collection. And whilst the standard of construction was understood, really, how good was the standard of maintenance? We know the horror stories Anthony uncovered, and that’s within the last 20 years. What were they like 100 years ago?

And who interrogated and recorded the data, on a miserable, wet, freezing, or stinking hot day? Probably the office tea boy, because it had no real global importance, so a few days when no one took the data, or some days when the very tall scientist read the thermometer instead of the very short tea boy, didn’t really matter.

So, back to the graph, which seems to convey no allowance for these variables. Or is it scientific convention to ignore them, or perhaps every variable is considered and allowed for, which I believe is called homogenising. Which seems a dishonest way to present data with as many wild variables as climate change exhibits. Why not just present the variables? Admit the failings, plot the average which would be no different to to homogenised data, and highlight the margin of error.

Am I being completely unreasonable?

So where do I go from here? Is my interpretation wrong (entirely likely) Is the data dodgy i.e. are paleo observations as inaccurate as I suggest, is historic recorded data accurate to within 0.1°F, are cabin and tea boys reliable? Why has CO2 remained flat-lined for over a million years while global temperatures have fluctuated wildly? And can I have my putter back please.

Reply to  HotScot
July 23, 2018 3:39 pm

A little help with the acronyms:
LGM: Little Green Men, typically Martians. (Also Last Glacial Maximum)
YD: You Denier, obviously. (Also Younger Dryas, search it – lots of info out there)

Reply to  TonyL
July 23, 2018 5:06 pm


“(Also Last Glacial Maximum)”

Aw come on, I was close.

“(Also Younger Dryas, search it – lots of info out there)”

You try putting YD into Google. I got “The yard (abbreviation: yd) is an English unit of length”.


Apr 27
COMIC1☆13新刊「し-39ab OrangeMaru」にて配布します。




[The mods are unclear whether any of the above is misspelled. .mod]

Holly Birtwistle
July 23, 2018 3:21 pm

I guess it’s some aspect of human nature, I find astounding but not surprising, is that a ‘study’ which provides an educated guess about abrupt climate changes 120,000 years ago – endows the researchers with the hubris to declare ” future disturbances of this circulation are hard to predict – and possible disturbances can’t be ruled out”. The NA circulation has been changing and oscillating for millions of years, and ‘ they can’t rule out it could happen again.’ Their knowledge of this process is almost nil, yet….

July 23, 2018 5:02 pm

Slightly off topic —

Ice core data is often used to show that CO2 levels react to global warming some 500 to 800 years later. 800 years ago the global temperature were quite warm, so the rise in current CO2 levels should be expected.
If no, why?

July 23, 2018 5:18 pm

“Abrupt shifts”? Really? And what could have caused that? Volcanic eruptions, perhaps?

July 23, 2018 5:19 pm

Or was it just Mother Gaia taking a hissy fit?? Mmmm, which could it be?

Patrick MJD
July 23, 2018 7:25 pm

Cyclones are being driven further south. Read this bit of “nifty” Australian research…

Krudd Gillard of the Commondebt of Australia
July 23, 2018 7:29 pm

“The team utilized intelligent computer algorithms based on machine learning.”

Open the pod bay doors, Hal.

Richard M
July 23, 2018 7:33 pm

I suppose this one depression in Finland could be combined with the one tree in Yamal to tell us everything we ever needed to know about historic climate. eye::roll

Loren Wilson
July 23, 2018 8:24 pm

You mean to tell me that it was warmer in the past and we didn’t all die?

Stephen Richards
July 24, 2018 1:18 am

“These results strongly suggest that the North Atlantic circulation is sensitive to disturbance,

And yet it has been stable for millenia

July 24, 2018 3:34 am

Flat out contradiction between two climate research articles in the same week.

Here we are told by the University if Helsinki (probably correctly) that AMOC slowdowns in the Eemian caused episodes of cooling.

While just a few days ago this article says the exact opposite – that the current AMOC slowdown will do the opposite – cause more warming:

Yes it’s a complex system, with multiple mutually interacting processes. The Gulf Stream that warms NH high latitudes is inseparably part of the same process that downwells cold dense deep water and cools the deep ocean.

Can we acknowledge chaotic complexity? Or do we just pick the part we like.

July 24, 2018 7:51 am

What the heck does it mean that the North Atlantic Circulation are subject to disturbance? What caused the disturbance? Over what time scale? Decades, centuries, millennia? Does their model even suggest what the cause was? Just because it was warmer then than now does not mean they are demonstrating that that era has any relationship to today’s circumstances.

July 24, 2018 9:11 am

Sorry mate…………..I LOST INTEREST AT :
……………………………………” My only concern is perhaps they over-rely on climate models”.

July 24, 2018 2:20 pm

Look at that!? The previous interglacial had extremely sharp temperature increases, followed almost immediately by deep temperature plummets.
comment image

A result absolutely dependent upon program code that invokes Gulf Stream stoppages, every time certain temperature boundaries are reached. All as written by alarmist disciples.

It is amazing how they program in such steep temperature trends, independent of CO₂ levels…

“p-hacking” is an excellent description for such specifically tailored programmed responses resulting in preferred “findings”.

Considering that their “programmed” imitation of paleo temperatures does not resemble current temperature reconstructions.
comment image

Reply to  ATheoK
July 25, 2018 7:32 am

Bleeding-edge mother of all climatic reference works.

Rodney R Chilton
July 27, 2018 3:51 am

Oh here we go again, when in doubt re: climate shifts, blame the North Atlantic!

[? .mod]

%d bloggers like this: