Sea Temperature Time Delay

Over on WUWT I made a comment:

That article is about some interesting patterns in Sea Surface Temperatures (mostly which oceans are warming and which are cooling). The comments here will make more sense if you have read the article there, and looked at the ocean temperature charts in that article.

What I think I noticed is that some temperatures were time lagged relative to others, and that shifts start in The Southern Ocean and arrive last at the North Atlantic and the Arctic Ocean. Similarly, time for cold water to get into the Indian Ocean is time lagged as it crosses the Pacific to get there.

This posting is just to preserve that comment where I can find it, and remind me to do a bit more “Digging” to see if I can prove the thought, or disprove it.

With that, the comment:

E.M.Smith says:
April 30, 2012 at 12:18 pm

At the conference in Chicago a couple of years back, there was a presentation on the heat flow patterns in the Pacific. It demonstrated a, roughly IIRC, 18 year time lag from a cooling or warming spike into the center of the Pacific, and when that band of temperatures eventually reached Alaska.

Given that, I would expect that it will be a while before the full impact of cold is felt in the Arctic. ( One might presume a similar pattern for the Atlantic, but with different time constants due to different size and currents).

In 1998 or so we had our peak. A bit later a cold dagger of south polar water ran up the coast of Chile and out into the center of the Pacific. At the time, I noted that there was now a timer running before the Global Warming rant about heating in the Arctic and Canada would run into a coldening North Pacific.

IMHO, these graphs show that initial cold plunge in the Southern Ocean, then the “fall off a cliff” in the North Pacific about 2008, 2011. There’s also an interesting “Spike and plunge” in 1998 / 2000 that is often seen in stock charts. (It’s a natural artifact of many systems. When riding a bike, for example, to turn right, the handlebars are first deflected just a tiny bit left first, that then leans the bike and you enter a right hand turn, the handlebars are then adjusted right to balance the turn. Learning to do that counter intuitive ‘right then left’ is why so many kids fall over when learning to ride a bike… It is taught as a deliberate awareness in motorcycle classes as a 1000 lb superbike does not respond as much to body lean and the handlebars become much more important to understand…) To me that “pop and drop” is the signature event of a major reversal.

Then we get the 18 year time lag to full effect. Call it 2016. Mark your calendars…

BTW, I can only wonder if we did a little study of the Indian Ocean if we might find a longer time period for the Pacific / Southern ocean cold to make it to that ocean. Once that happens, the whole system is in dramatic cooling.

As that cold spike headed up the Chilean coast, it had to suck other water down to the Southern Ocean to replace it. That water would have come from oceans fronting on the Southern Ocean, which would then have pulled other water into those oceans. IFF that water came from the Indian Ocean and / or the South Atlantic, those oceans would have gained replacement water from warmer areas more north (or East in the case of the Indian ocean – from the mid-Pacific…)

IF that has some truth to it, one would expect the Indian and South Atlantic oceans to follow the cooling of the Pacific with some years of time lag. So I’m putting down a “Watch This Space” marker. By 2020 both of those oceans ought to be showing significant cooling trends.

That’s the point where AGW as a thesis is “deep sixed”… as it will be obvious to everyone we’re in a dramatic coldening process of multi decadal length.

This speculation would benefit from looking at an actual map of ocean currents, where they come from and where the go to.

Makes it look like the North Indian ocean is a bit isolated with a circulating warm current, so slow to cool. South Atlantic looks to be fed from the Southern Ocean with the North Atlantic a bit isolated (so it ought to lag, too – just NOW getting a warm spike like that seen in the Pacific and Southern back in 1998; so I’d guess about a 10 to 12 year lag).

So a ‘first blush’ look at that (simplified and maybe just surface currents) map would seem to offer some confirmation of the idea. One would also speculate that Australia and the Pacific Islands ought to show the cold turn very rapidly. Wonder if it’s been cold Down Under lately… Any tendency to cold on the West Coast would imply a cooling Indian Ocean ‘on the way’ as the Southern Ocean current has to pass by there first.

Last on the list looks to be Eastern USA / Southern Europe. (Gee, hasn’t it been cold on the West Coast of the USA and warm on the East Coast… I think maybe this thesis ‘has legs’… certainly enough for a ‘dig here!’…)

It would be interesting to plot AMO vs PDO and European / African thermometers vs West Coast USA / Australia / Chile thermometers and see if there is a decade or so ‘offset’ between peaks and valleys. (Another “dig here!”)…

Simple to check, interesting if it plays out…

I’d also note that there was a follow up that prompted this bit of detail about the reference to an 18 year ocean lag in the Pacific:

E.M.Smith says:
May 2, 2012 at 1:46 am

@Bob Tisdale:

The time delay in the Pacific to which I referred was a lag in propagation of surface water temperatures. I think it was due to the time it took for bands of water to move from the central Pacific (where they arrive from along the coast of Chile / Peru) to work their way all the way up to Alaska.

My notes from here:


Gary Sharp?

We then got a bonus of a video clip that I think was presented by Gary Sharp. It showed the heat / cold cycling of water in the pacific over decades as El Niño comes and goes. How to put a movie into words? Not well… But you see the warm and cold moving and swirling and you start to see patterns, one is that it drifts north over time.

The Punch Lines being that that heat reaches the Arctic going past Alaska about 18 years after generation in the Pacific. So the warming in 2008 melting ice comes from a 1990 hot Pacific. None of the models allow for that time lag and “If you don’t have that in your model, your model is broken”. (as a pretty good paraphrase).

Hope that helps…

I think this picture of the Thermohaline deep circulation might also help:

Thermohaline Circulation

Thermohaline Circulation


When added to:

Surface Ocean Currents

Surface Ocean Currents

It’s pretty clear the water originates around the south polar current, then works it’s way up to points where it submerges and then returns as deep water currents. Add in various time delays and I think this idea ‘has legs’. I do note that the Thermohaline Circulation has some bits looking like they go in the opposite direction of some of the surface currents on the other map, so I think there is an elevation detail that needs working out too. Perhaps someone already has a combined ocean currents model somewhere. Hmmm….

I’m also wondering if these tie to the cycles of PDO / AMO as time lags make different areas colder and warmer and if the making of “global averages” hides the important bit of information about leading / lagging areas and reduces the precision with which “Regime Change” in the long duration weather cycles can be observed. If, for example, one could show that what happens in Chile shows up in Japan 5 years later and in both Nome and Jakarta 10 years later, that could be quite interesting.

Subscribe to feed

About E.M.Smith

A technical managerial sort interested in things from Stonehenge to computer science. My present "hot buttons' are the mythology of Climate Change and ancient metrology; but things change...
This entry was posted in AGW Science and Background and tagged , , , , . Bookmark the permalink.

20 Responses to Sea Temperature Time Delay

  1. LB says:

    I’ve been thinking recently about how climate reacts to changing in forcing. Is there any way of estimating the reaction?

    Well, we do have a daily change in forcing. Temperate varies from day to night. Temperate reacts very quickly to that change.

    Now look at the annual change. Here again you see a change that will be faster than a year.

    Why would there be an 18 year lag?

    [ Reply: Because the cold water from the south pole first shows up as a cold current shooting out into the middle of the Pacific. It then takes a very long time for that water to migrate north, and during that time it is very very slowly being warmed by incoming sunlight. While the AIR reacts very quickly to daily and seasonal changes, the WATER is of much higher heat capacity per unit volume and has a huge mass to heat down to the limit of penetration of UV. Instead of 1 gram / liter it’s 1000 grams / liter, and with specific heat of 1 Calorie / Gram too. Basically it’s all about air heating / cooling vs water heating / cooling. But frankly, why doesn’t matter quite so much as the fact that it has been observed to happen. -E.M.Smith ]

  2. adolfogiurfa says:

    Enlightening article today at NIA:
    And the peculiar properties of water:
    Prof.Giorgo Piccardi´s “Structure of Water … Subtle Properties”:

    Click to access piccardi2.pdf

  3. adolfogiurfa says:

    Not forgetting the influence of the GMF :

  4. David says:

    Humm, even the surface currents appear to be moving in opposite directions, the antarctic suppolar, vs the circumpolar. I read somewhere, but can not now find it, on how these currents have changed over the years. Also. in T-5 I had some questions on some alarmist type concern over “missing” bottom water. Here is the link to that article, although I do not know if it is of help here.

  5. David says:

    The deep water formation appears to have to drop right through the antarctic subpolar current moving in the opposite direction.

  6. vukcevic says:

    Absorbed heat between two ‘tropic’ parallels is distributed by ocean currents pole-wards. In the southern hemisphere the circumpolar current is the principal component.

  7. David says:

    The Weddel current, apparently is part of the antarctic suppolar, but actually moves in a circular clockwise direction, with, from what I have found so far, little understanding of details of its interaction with the circumpolar current. Here is a more detailed illustration. According to the link concerning bottom water the flow of this cold salty water is fifty times the volume of the mississippi river, so that is a lot of cold water, formed by melting sea ice, and the super cold winds blowing off the continent, and over the sea. I guess I need to see the prominent direction of said winds in all seasons and changes in all currents over the last few decades (-;–JMuGM2gWlv5HlCQ&usg=AFQjCNHXnmFpNjrIkjpe-Lz7PYb_OkxHLw

  8. David says:

    Antarctica has “katabatic winds”. These are caused as a result of the height and shape of the continent. Antarctica is a high (average 2300m) dome-shaped continent. Air cools and then starts to fall (the opposite of hot air rising) as it does so it flows down the slope from inland towards the coast, travelling many hundreds or thousands of miles driven largely by gravity.

    As these katabatic winds reach the coast and flow over the surface of the sea, they produce a west-flowing ocean current known as the East Wind Drift. The rotation of the earth causes this wind flowing offshore from the land to produce an ocean current at a right-angle to the wind direction.

    Not very helpfull so far, as the currents in some locations are circular, however the air is clearly very cold and, and, as it is induced by cold air flowing down from the high points, probably stronger in winter then summer, therfore the increase in winter ice is likely to affect the location of these wind formed currents, moving them to lower latitudes and possibly creating greater interaction with other ocean currents.

  9. wayne says:

    Glad to see you raising the time delay again. It amazes me such long lags tend to be totally ignored as if every movement of decadal energy upon this globe were instantaneous and the slow deep ocean currents is one place that magnifies and stretches out this lag effect. Has anyone performed the correlations with the +/- eighteen year in the Pacific? Seems I commented to you month’s ago but in that case we were speaking of currents around the cape and upward into the Atlantic to the arctic. I’d still be curious what that lag is from the cape, or better the Indian Ocean, to the arctic.

    To me it all boils down to the solar grand maximum we went through in the 80-90’s and this very lag you are speaking of here. It’s great you keep bring this up E.M.

  10. Sandy McClintock says:

    Perhaps someone has can explain…
    In Al Gore’s film there was the famous graph of CO2 and temperature moving up and down ‘together’ over millions of years. Then it was pointed out that there was about 800 years lag between temperature changes and CO2 changes based on ice-cores.
    We see CO2 going up currently, so one might expect temperatures to have been going up 800 years ago. Would the Medieval Warm Period be the driver of the CO2 rise we see now?
    To what extent is the current CO2 rise due to man vs an 800 year lag?
    (yes I know your time lag is a bit shorter than Al’s ;) )

  11. E.M.Smith says:


    Thanks for pitching in! Not sure what it was you want to ‘condense’, the links or th several postings. No way for an individual to combine comments, though a moderator can. If it’s the links, “they are what they are” most of the time. There’s the tiny URL site where you can map one to another, but I’ve never used it.

    It does look like you posted a link with the image search in it. You could, instead, after the search: Close the top image box, find the image in the page, right click on it and chose “open image” and then get this shorter image link without the page in the background (but you loose the information about the page from which it came):

    though you could then say “From:” and give the page link:


    which shows up when you close the top image from the google search…

    Basically, take the two parts the google search link gives you (the page and the image) and fish out the actual link for each separately by closing the foreground image “popup”, getting the page link, and scrolling through the page to find the image and then rt-click open it to get the link for just the image.

    @Sandy McClintock:

    The CO2 cycle Algore was showing was for ice age glacial / interglacial cycling; so may not be pertinent to processes inside an interglacial. I just don’t know. The speculation I’ve seen is an 800 year lag for dissolved CO2 in the ocean; or less commonly, for CO2 in soils and from plant decay. (Trapped under retreating ice, being exposed and then decaying).

    As for us, now: I’d expect a lot of it comes just from our having cut down the global forests in large part. Elsewhere (the “Got wood?” posting) I showed how a bamboo stand (or fast growing forest) completely depletes the air column over it of CO2. Cutting down whole continents of forest is going to have an impact… But the Warmers ignore that… The quantity of carbon “sequestered” in soils is also quite large. Tilling reduces it as it becomes exposed to oxidation (loss of ’tilth’ happens). That, too, is ignored. Then there is the whole speculative tie between warming cycles and reduced volcanism / cooling cycles and increased volcanism. So was the world more “volcanic” during the LIA? Could a load of that CO2 from deep ocean volcanic action (that is the bulk of the global volcanoes…) take a few hundred years to outgas from the ocean? Maybe… I suspect nobody really knows how the whole carbon cycle works and what the individual time lags might be.

  12. R. de Haan says:

    “I have no idea why Aliens became a hot topic on a channel that’s supposed to be about history”

    Since the UN appointed an Ambassador for aliens in 2010

  13. Pascvaks says:

    Re: “Llanquihue Glaciation, Southern Andes”
    ( (Map (see via link below) showing the extent of the Patagonian Ice Sheet in the Strait of Magellan area during the last glacial period. Selected modern settlements are shown with yellow dots.))
    “The Llanquihue glaciation takes its name from Llanquihue Lake in southern Chile which is a fan-shaped piedmont glacial lake. On the lake’s western shores there are large moraine systems of which the innermost belong to the last glacial period. Llanquihue Lake’s varves are a node point in southern Chile’s varve geochronology. During the last glacial maximum the Patagonian Ice Sheet extended over the Andes from about 35°S to Tierra del Fuego at 55°S. The western part appears to have been very active, with wet basal conditions, while the eastern part was cold based. Palsas seems to have developed at least in the unglaciated parts of Isla Grande de Tierra del Fuego. The area west of Llanquihue Lake was ice-free during the LGM, and had sparsely distributed vegetation dominated by Nothofagus. Valdivian temperate rainforest was reduced to scattered remnants in the western side of the Andes.[35]”

    It would seem that during a glacial period the “circuit breaker” is in the Southern Hemisphere in the vicinity of Tierra del Fuego. As the Antarctic cools the surface Antarctic Circumpolar Current must expand until it is totally disrupted when it runs into the Western Coast of S.America, this would also put a ead to Atlantic and S.Africian branches of the Antarctic Circumpolar. Gradual but effective?

  14. p.g.sharrow says:

    @Pascvaks; You may be right, but I think changing the Bering Sea to the Bering plains may have a greater effect then the reduction of surface exposure at Cape Horn. Warm water enters the Arctic from the North Sea and leaves through evaporation and cold waters through the Bering Straits. Just a thought. pg

  15. E.M.Smith says:

    Gives a pretty good view of how the circulation happens.

    The map in this page:

    shows cold water going IN to the Arctic through the Bering Strait (and warm from the Gulf Stream) with combined cold coming out near Eastern Canada…

    If you look at the map on this page, you’ll see how water moves through the Arctic Ocean. Cold, relatively fresh water comes into the Arctic Ocean from the Pacific Ocean through the Bering Strait. This water meets more fresh water from rivers and is swept into the Beaufort gyre where winds force the water into clockwise rotation. When winds slack off and the gyre weakens, fresh water leaks out of the gyre and into the North Atlantic Ocean (follow the blue lines on the illustration toward the bottom of the map). Of course, water can go both ways, and water does come into the Arctic Ocean from the North Atlantic (red lines on the map). This water is warmer and relatively salty. Because of its increased salinity, it is denser and sinks below Arctic waters.

    That’s why the cold spike into the center of the Pacific takes 18 years to get up to the Arctic. You’ve got a lot of water in the way and a fairly small strait! (Yes, some sinks in the cold too).

  16. Wayne Job says:

    Well done E.M. I have been saying for a while we are running out of warm water, it has been a while since the sun was pumping hard. My understanding was that the cycle average was 16 years and that is about over. However I would defer to your superior knowledge and accept the 18 year time frame. Maybe the time table is ahead of schedule in the southern hemisphere as the weather in Australia has gone into a cooler and wetter cycle so a La Nina dominant period is our immediate future. The portends of cooling seem more dominant than the those of global warming.

  17. E.M.Smith says:

    @Wayne Job:

    Thanks! Take a look at the video clip. It’s pretty clear that the Indian Ocean tends to be in an isolated gyre. That would explain the blip up on 1998 solar peak, but lack of a drop on the cold surge out of the Southern Ocean.

    The North Atlantic gets fed by a current that heads up from South Africa, so will likely be quite late in the process. (Thus the Eastern USA and Western Europe staying warm a bit longer than the rest). It’s South America and Australia that are interesting in that they look to get the first surge of cold water out of the Southern Ocean. California and Alaska get the cold after a 15-18 year lag, so we got cold in the last couple of years.

    BTW, that 18 years is how long it takes for a cold spike into the center of the pacific (from water running up the Chile / Peru coast and upwelling from the deep off Ecuador) to reach the Bering Strait. It gets to other places proportionately sooner. So 9 years at 1/2 way from the equator to the Bering Strait. I really wish I could find that video on line as it is a “visual thing” where you can see the water forming bands that migrate north over time.

    You can see a hint of that in the video just above. The central pacific goes West, then just north of it a return current goes east, then north of that… etc. The first bands are strong, then harder to see pattern above that, but the simple fact is that the water slops back and forth a couple of times on the way north.

    Other oceans are a bit less ‘structured’ in bands, IMHO. More based on gyres and on things like The Gulf Stream, so will have other time delays. In particular, I find the way that the South Atlantic has a current feeding up to the North Atlantic that seems to originate in a current carried by gyres out of the Indian Ocean and around the horn of Africa, ought to imply they are last on the temperature change timeline. (Oddly, the AMO swap is about a decade after the PDO swap and Hadley chose a baseline a decade later than GISS chose… right at the cold bottom for both, given their respective places in the timeline.)

  18. F. Guimaraes says:

    “… To me that “pop and drop” is the signature event of a major reversal…”
    that’s very interesting because I’ve got the same impression from the evolution of solar cycles and temperatures, for example, (despite the fact that I’m not a great fan of C14 dating)


    which show the deeper solar minima being preceded, more or less closely, by strong maxima.
    Both graphs show the strongest local maxima and the lowest minima with a steady downward trend, basically since the very early times of the Holocene.
    We have just come from ~ 70 years of strong solar cycles, from C18 to C23, suggesting a subsequent period of strong minimum, which seems to have already started.
    If my previous statement about “downward trend” makes sense then the present minimum has a good chance to reach the lowest point of this Interglacial, or even be the beginning of the new glacial period-new Ice Age.

Comments are closed.