Lorenz et al 2006: “Tropical Cooling”

Despite the protestations of the Team, it seems to be a consensus of other paleoclimatologists that the Holocene Optimum was warmer at high southern latitudes. For now, I’ll take this as read, although it’s well worth canvassing the literature. Some citations are at ukweatherworld.

The next line of Team attack on the Holocene Optimum is the argument of Lorenz et al 2006 (building on earlier articles Rimbu et al 2004 ) – and you’ll undoubtedly hear more on this line of reasoning – that the Holocene Optimum was a period of warming in the extratropics and cooling in the tropics.

So the question becomes: how do they know that there was cooling in the tropics in the Holocene Optimum?

Their proxy is alkenone SST estimates collated in the following location: a site which unfortunately is password protected. Boy, this use of password protection for paleoclimate data is annoying. I’ve written to the corresponding author requesting access to the data and have received no reply so far. Some of the data may be available in non-password protected locations, but I haven’t been able to collate most of it.

Lorenz et al 2006 states the following:

The alkenone-derived SST records show diverging linear trends over the last 7000 years (Figures 3 and 4a). In general, the extratropics cooled while the tropics experienced a warming or no substantial temperature changes from 7 kyr B.P. to the present. The magnitude varied between 0.62C and 4.41C per 7 kyr for the cooling and between +0.19C and +1.47C per 7 kyr for the warming (Figure 3).

Lorenz et al Figure 3, supposedly demonstrating this, is shown below. It shows trends for 20 cores from the Holocene Optimum to the present. The first two rows show sites with decreasing trends; the bottom row shows sites with decreasing trends and the middle two rows show sites with slightly increasing trends. The rhetorical effect of this arrangement is to convey the impression that extratropical sites have cooled since the Holocene Optimum, while tropical sites have warmed in accordance with the claims of the article.

lorent76.jpgLorenz Figure 3. Original Caption. Examples of alkenone-derived SST records (circles) and their linear trends (solid lines) evaluated from 20 cores, whose positions are shown on the map [Kim and Schneider, 2004; see also Kim et al., 2004, Table 1]. The magnitude of SST change over the last 7000 years (calendar age) is indicated in the upper right corner of each panel.

As pointed out by Gator below, the middle rows are not necessarily tropical. But, in any event, it is these 8 sites (cores 9-16) that are relied upon both for the claim that the Holocene Optimum was regional and to support the claim that it there was “generally” cooling in the tropics. [this paragraph edited]

Lorenz Cores 9-16
Here’s the location map for the above 20 cores – pay particular attention to the locations of cores 9-16, which are the ones used to argue that the Holocene Optimum was regional.

lorent75.jpg
Lorenz Figure 3 top panel. Location map of 20 cores shown in first figure.

A first obvious point: core 9 (in the rhetorical middle group) is outside the tropics (off California). Three tropical cores (8, 19 and 20) with increasing “trends” are shown in the rhetorical “extratropical” groups.

Second, 5 of the 7 Lorenz et al tropical cores with negative “trend” are readily seen as coming from upwelling regions. In 20th century discussions, increasing proportion of (coldwater) G Bulloides in Arabian Sea sediments was interpreted by Moberg and Juckes as evidence of 20th century warming – indeed, this proxy was one of the key Moberg proxies – see elsewhere on CA.

  • Core #10 (M35006-5) comes from the upwelling Cariaco area. (David Black reported a high-res Caraico G Bulloides series discussed elsewhere at CA).

    Core #11 (ODP-1078C) is from the upwelling Benguela area. Note that core #19 (GeoB 1023-5), which is very close, is said to come from a non-upwelling area and shows a different “trend”.

    Core #12 (GeoB5844-2), is from the Red Sea – my guess is that it is extremely difficult to segregate salinity and temperature issues here.

    Cores 13-15 (TY93-905, 74KL, SO94-39KG/56KA) come from the Arabian Sea, a strong upwelling region that we’ve discussed frequently, primarily in the context of the use of Arabian Sea G. Bulloides by Moberg et al and Juckes et al. as key proxies for NH temperature increases(!) G Bulloides is a coldwater foraminifera and is a proxy for wind strength (upwelling). Moberg and Juckes argued that increased levels of G Bulloides (i.e. cooler water in the upwelling regions). Shouldn’t the Team make up their mind?

    Core 16 (17940-2) comes from the China Sea and shows increases since the LGM.

  • Perhaps the most remarkable aspect of this compilation is the complete lack of representation of the Pacific Warm Pool, where Stott et al 2004 (using Mg/Ca) shows declining SST since the Holocene Optimum. This lack of representation from an obviously essential region is presumably because of the restriction of their proxies to alkenone SST, which eliminates Stott et al 2004 from consideration (it used Mg/Ca and dO18). But how reasonable is it for a study like this to ignore the Pacific Warm Pool?

    Another class of information not considered here is from tropical glaciers. Puruogangri, Dasuopu and arguably Quelccaya were formed following the termination of the Holocene Optimum. Yet none of this evidence is considered in assessing the claim of a “cool” tropical Holocene Optimum.

    So the principal evidence from Lorenz et al is the following:

    1. Alkenone SST from many high-latitude northern sites show a significant Holocene Optimum.
    2. Alkenone SST from a number of high-latitude southern sites show a significant Holocene Optimum.
    3. Alkenone SST from a number of upwelling sites in the tropics shows a temperature increase since the Holocene Optimum. However, increased upwelling has been used elsewhere as an indicator of NH warmth (Moberg et al, Juckes et al). Lorenz et al adopts an inconsistent interpretation – cooler water in upwelling regions supposedly shows tropical cooling. Seems like sucking and blowing.

    Update: AR4 stated:

    Extratropical centennial-resolution records therefore provide evidence for local multi-centennial periods warmer than the last decades by up to several degrees in the early to mid-Holocene. These local warm periods were very likely not globally synchronous and occurred at times when there is evidence that some areas of the tropical oceans were cooler than today (Figure 6.9) (Lorenz et al., 2006).

    Also Fig 6.9 showed cool tropical oceans from 11K to 5K BP:

    Figure 6.9. Timing and intensity of maximum temperature deviation from pre-industrial levels, as a function of latitude (vertical axis) and time (horizontal axis, in thousands of years before present ). Temperatures above pre-industrial levels by 0.5°C to 2°C appear in orange (above 2°C in red). Temperatures below pre-industrial levels by 0.5°C to 2°C appear in blue. References for data sets are: Barents Sea (Duplessy et al., 2001), Greenland (Johnsen et al., 2001), Europe (Davis et al., 2003), northwest and northeast America (MacDonald et al., 2000; Kaufman et al., 2004), China (He et al., 2004), tropical oceans (Rimbu et al., 2004; Stott et al., 2004; Lorentz et al., 2006), north Atlantic (Marchal et al., 2002; Kim et al., 2004), Tasmania (Xia et al., 2001), East Antarctica (Masson et al., 2000), southern Africa (Holmgren et al., 2003) and New Zealand (Williams et al., 2004).

    [Note – above note edited a little for flow on Apr 6, 2009 – see Wayback Machine if you’re worried about this.}

    References:
    Lorenz, SJ, Kim, J-H, Rimbu, N, Schneider, R, Lohmann, G (2006) Orbitally driven insolation forcing on Holocene climate trends: evidence from alkenone data and climate modeling. Paleoceanography, 21: PA1002 url

    Kim, J-H, Rimbu, N, Lorenz, SJ, Lohmann, G, Nam, S-I, Schouten, S, Rühlemann, C, Schneider, R (2004) North Pacific and North Atlantic sea-surface temperature variability during the Holocene. Quaternary Science Reviews, 23 (20-22), 2141-2154

    Rimbu et al Clim Dyn 2004 url

    24 Comments

    1. Che Gonzales
      Posted Jan 2, 2007 at 2:47 PM | Permalink

      The data possesses extra special mojo–it is too powerful for just anyone to handle. It is to protect your crops and your goats. Be grateful that others know your limitations, even if you do not. :-)

    2. Tom Ault
      Posted Jan 2, 2007 at 3:05 PM | Permalink

      First of all, aren’t “tropical” and “extra-tropical” defined by the lattitude of a site? That is, aren’t all sites above a certain lattitude (ignoring north/south) “extra-tropical” while all sites below “tropical?” If so, then clearly some of the sites are misclassified.

      Something else to notice: the magnitudes of the temperature changes for the “tropical” sites are much smaller than the magnitudes for the “extra-tropical” sites. The former range from +0.19 to +1.47 C while the latter range from -0.62 to -4.41. Therefore, even if the tropical region did warm between Holocene Optimum and now, it doesn’t necessarily follow that global mean temperatures weren’t significantly higher than they are now.

    3. jae
      Posted Jan 2, 2007 at 3:12 PM | Permalink

      Boy Oh Boy, what artful science!

    4. MJW
      Posted Jan 2, 2007 at 3:20 PM | Permalink

      If it’s not too difficult, I’d like to see a version of the above map with the dots that indicate the site locations colored-coded according to the slope of the line.

    5. richardT
      Posted Jan 2, 2007 at 4:07 PM | Permalink

      #2
      Don’t forget that the half the earth is within 30N and 30S. Less than 15% of the earth’s surface is at latitudes greater than 60 degrees. This means that only small tropical temperature changes can balance large high-latitude temperature changes.
      The selection of cores from tropical upwelling zones is unfortunate, as these may not be representative of the whole ocean, but probably inevitable, as these are areas likely to high enough sedimentation rates.

      Where does it specify in the paper which cores are “tropical”? The colour coding in figure 3 seems to refer to the trend, not the location.

    6. nanny_govt_sucks
      Posted Jan 2, 2007 at 11:36 PM | Permalink

      Yes, what’s with all the different Y-axis scales of the 20 cores? I don’t think the slopes would be all that impressive with a standard scale.

    7. Jos Verhulst
      Posted Jan 3, 2007 at 12:06 PM | Permalink

      I’ve written to the corresponding author requesting access to the data and have received no reply so far. Some of the data may be available in non-password protected locations, but I haven’t been able to collate most of it“.

      Here’s a plea for ‘executable articles‘: an interesting idea launched by Marc Liberman

    8. 2dogs
      Posted Jan 3, 2007 at 12:06 PM | Permalink

      How strong is the effect of polar amplification? Could the effect of a, say, 2.5C global mean temperature increase be negilible in the tropics?

    9. Ken Fritsch
      Posted Jan 3, 2007 at 1:08 PM | Permalink

      Second, core #8 located offshore Mauritania is shown as an extra-tropical site, while core #9 located offshore northwest USA is shown as “tropical”. What is the rationale for this?

      I came away with the impression that the authors were listing the cores (and location of them) by temperature trend (positive versus negative) and not necessarily assigning them to extra-tropical or tropical sites. I thought what this showed, and perhaps unintentionally, was that there were several exceptions to the general claim that extra-tropics cooled from the Holocene optimum while the tropics stayed nearly the same or warmed slightly.

      On reading the paper quickly once, I also noted that, while underlying themes ran through the paper indicting that regional and seasonal differences existed during the HO and that general, i.e. very general, trends from climate modeling allowed some first look substantiation of these differences, no attempts were made at statistically analyzing the data or even simply directly comparing it or averaging it over the globe. I have a real difficult time taking these underlying themes seriously without the missing analyses. Or perhaps I am missing something here? It was if the authors presented the information in color coded picto graphs to be interpreted more by imagination than any hard analysis.

      The SST reconstructions were stated in the article to be limited to alkenone proxies because of the need for “consistency”. The use of this simple criterion would seem to me to make the study suspect for cherry picking if non-alkenone proxies that went against the trends were not included in the study or at least mentioned.

    10. Brooks Hurd
      Posted Jan 3, 2007 at 4:18 PM | Permalink

      Steve,

      core #9 located offshore northwest USA is shown as “tropical”.

      If #9 were off the east coast at that latitude, I might believe that it were tropical. I lived more than 8 years in the DC area and the Summer temperatures were tropical even in the 1980s. They are claiming that it is tropical in the Pacific NW. These authors clearly need to get out from behind their desks and travel to some of these sites.

      I live a good distance south of #9 and I can assure you that California’s Central Coast is far from tropical. Off the NAMER west coast the Pacific is cold.

      We drove through Pismo Beach last week and saw several people in swimming trunks coming out of the water. We all agreed that they were not locals. No one around here goes into the Pacific without a wet suit.

    11. Brooks Hurd
      Posted Jan 3, 2007 at 4:20 PM | Permalink

      Note: I meant “off the NAMER west coast north of here the Pacific is cold.”

    12. Gator
      Posted Jan 5, 2007 at 3:34 AM | Permalink

      Steve McIntyre has made a huge argument around what sites were defined as “tropical” or “extra-tropical.” A quick read of the article in question makes it clear that the sites are not defined as tropical or extra-tropical by the slope in the graphs shown (as Steve McIntyre says above.) This is a tempest in a teapot.

      Is this a deliberate misrepresentation by Steve McIntyre? Or just a mistake? We’ll know when he posts a correction to his discussion.

    13. Steve McIntyre
      Posted Jan 5, 2007 at 9:28 AM | Permalink

      Gator, my main concern is exactly what evidence exists for cooling in the tropical Holocene Optimum. I’m reading this article in the context of how it’s been used in the IPCC 4AR Second Draft, where they rely on it as follows:

      paleoclimatic records of the Holocene provide no conclusive evidence for globally synchronous warm periods, especially because the temperature trends appear distinct in the low versus mid- and high-latitudes during the Holocene (Lorentz et al, 2006).

      So my concern is exactly what evidence exists for such distinct trends. I haven’t finished with this series because I’m posting on many different issues right now – but I plan to talk about the Warm Pool which appears to have a Holocene Optimum, as well as other evidence on whether there was cooling in the tropical Holocene Optimum. In examining the data, there appear some sites which appear to show a warming throughout the Holocene. Not a HS, but a fairly sustained warming throughout the Holocene. Also interesting,

      So the main issue is how IPCC uses Lorentz et al 2006 – and they use it to argue that there is tropical cooling. I suppose that they were encouraged in this by the following statement in Lorentz et al:

      The alkenone-derived SST records show diverging linear trends over the last 7000 years (Figures 3 and 4a). In general, the extratropics cooled while the tropics experienced a warming or no substantial temperature changes from 7 kyr B.P. to the present. The magnitude varied between 0.62C and 4.41C per 7 kyr for the cooling and between +0.19C and +1.47C per 7 kyr for the warming (Figure 3).

    14. Ken Fritsch
      Posted Jan 5, 2007 at 12:28 PM | Permalink

      I find the writing and terminology of the paper very “soft” in analyses and conclusions — something that would make it more useful for the IPCC to reference in its tendency for generalized and less than clear cut language in spelling out uncertainties. Steve M’s reference to this paper weighs little or not at all on the interpretation of the authors’ classification of tropics and extratropics as the paper (after Steve M pointing to it) is there for all to read.

      I would image that the Holocene Optimum temperature reconstructions have presented a stumbling block for making clearer cut statements about late 20th century representing unprecedented warming and by doing a computer model simulation the issue of seasonal variation can be put into the mix (in less than precise terms in my view) and cast doubt (in a less than a compelling manner) on the average global temperature being warmer than today. The alkenone proxies are to be relegated to a measure of seasonal temperature that, while being higher than today, is averaged out by lower temperatures in seasons (as computer runs indicate) when the alkenone proxy was temperature insensitive or dormant if you will.

      The climate model simulations are also used to very generally imply that an averaging effect of all seasons and global regions will show that the temperature of the Holocene Optimum could be lower than the late 20th century — even though a straight forward look at the proxy data would indicate otherwise. More proxy information and particularly those less susceptible to seasonal sensitivity will be interesting to see.

      By the way this article noted that the model had to run for hundreds of years under the Holocene Optimum conditions in order to reach “equilibrium”. I saw that same reference made in a previously discussed paper on modeling where the first 700 years were discarded because the model had not equilibrated. The papers tend to indicate that at least part of this “equilibration” is not an artifact of the model run but deals with a relatively slow climate equilibration. In contrast the Hansen Scenarios, and I believe it was Scenario C, shows an almost instantaneous effect of lowered GHG levels. What is the estimated equilibration time for a given natural or anthropogenic change?

    15. Gator
      Posted Jan 6, 2007 at 12:47 PM | Permalink

      Re:#13 Steve McI — your comments don’t change the fact that your first two points in your article were flatly wrong. Was this a mistake? Or an attempt to make “the Team” look stupid? All it did was make you look stupid as it appears you simply did not read the article you chose to criticize.

      It would not surprise me that one would find varying patterns of temperature change that do not strictly track lattitude. There are those pesky continents and things messing up the smooth flow of the oceans and atmosphere.

    16. Steve McIntyre
      Posted Jan 6, 2007 at 1:16 PM | Permalink

      Gator: you say that the first two points in my note were flat wrong. I’m trying to assess new territory for me and certainly don’t claim to be as familiar with the data as with millennial reconstructions. I’m quite prepared to amend the points but I still don’t see exactly what I’ve got wrong. My first point was:

      Despite the protestations of the Team, it seems to be a consensus of other paleoclimatologists that the Holocene Optimum was warmer at high southern latitudes. For now, I’ll take this as read, although it’s well worth canvassing the literature. Some citations are at ukweatherworld.

      I base this comment on a statement in IPCC 4AR Second Draft which states the following:

      Other early warm periods are identified in the equatorial west Pacific (Stott et al 2004), China (He et al., 2004), New Zealand (Williams et al, 2004), south Africa (Holmgren et al, 2003) and Antarctica (Masson et al., 2000). At high southern latitudes, the early warm period cannot be explained by local summer insolation changes (see Box 6.1), suggesting that large-scale reorganisation of latitudinal heat transport may have been responsible.

      My statement was based on this statement from IPCC WG1 which seems to concede Holocene Optimum warmth at high southern latitudes.

      My second point was:

      The next line of Team attack on the Holocene Optimum is the argument of Lorenz et al 2006 (building on earlier articles Rimbu et al 2004 ) – and you’ll undoubtedly hear more on this line of reasoning – that the Holocene Optimum was a period of warming in the extratropics and cooling in the tropics.

      Again, IPCC 4AR Second Draft says:

      In contrast, tropical temperature reconstructions, only available from marine records, show that tropical Atlantic, Pacific, Indian Ocean SSTs exhibit a progressive warming from the beginning of the current interglacial onwards (Rimbu et al., 2004; Stott et al., 2004), possibly a reflection of annual mean insolation change (Figure 6.5). When considering the periods of largest temperature changes (Figure 6.9), paleoclimatic records of the Holocene provide no conclusive evidence for globally synchronous warm periods, especially because the temperature trends appear distinct in the low versus mid- and high-latitudes during the Holocene (Lorentz et al, 2006).

      I don’t see that my representation here is incorrect either.

      So I don’t see that these points, which I interpret as being the first two points, are incorrect. Perhaps you’re thinking of different points. If so, let me know and I’ll consider things some more.

    17. Ken Fritsch
      Posted Jan 7, 2007 at 1:24 PM | Permalink

      Steve M the only disagreement that I think Gator could be referring to and to which RichardT in comment #5 of this thread referred (see below) is that the authors appear to have presented data by temperature trend and not by tropic or extratropic location.

      Where does it specify in the paper which cores are “tropical”? The colour coding in figure 3 seems to refer to the trend, not the location.

      It, I think, all derives from your statement below. I agree that the authors appear to be using temperature trend and not location — even though the trends are not totally consistent with locations. I think that there are many interesting points to be made about this article, so if this relatively unimportant point, to me anyway, can be resolved perhaps the discussion can continue.

      A small point first: core 11 with decreasing SST is shown as “tropical” while core 19 with increasing SST is shown as extra-tropical. Yet they seem to be very close together – why the difference?

    18. Steve McIntyre
      Posted Jan 7, 2007 at 4:00 PM | Permalink

      #17. Ken and Gator, I’ve re-stated this post to focus more clearly on the reliance of Lorenz et al on upwelling regions in the tropics. As Gator implicitly points out, the arrangement of sites in Lorenz et al Figure 3 was a rhetorical device which conveyed an impression of differences between the tropics and the extratropics and thereby seemingly support the claim in the paper of differences between the tropics and extratropics. As Gator observed, Lorenz et al did not actually say that the sites were tropical and extratropical and so this arrangement in Figure 3 was merely rhetorical rather than according to their actual latitudes.

    19. Ken Fritsch
      Posted Jan 7, 2007 at 4:44 PM | Permalink

      As Gator observed, Lorenz et al did not actually say that the sites were tropical and extratropical and so this arrangement in Figure 3 was merely rhetorical rather than according to their actual latitudes.

      That was my impression also and in line with my earlier stated impressions that the article tends, in my view, to be very fuzzy in making points. I asked myself why the authors did not otherwise summarize all the trends in the tropics and extratropics and compare them with the seasonal computer simulations. That is why I am interested in seeing other proxy data from this period that is not as seasonal as the alkenone proxy is claimed to be and something that might test the computer simulations results for seasonal variations. The Lorenz paper seems to me to be constructed as a very quotable source for the Team and others — without having to deal with the issues with much specific detail or with regard to countervailing evidence. That is of course my opinion and I await further information and details to clarify it.

    20. J Edwards
      Posted Jan 8, 2007 at 10:18 AM | Permalink

      In the “Team vs Stott” thread, Andre posted a link to a thread on UK weather world, which has a number of links to various papers relating to the Holocene Optimum. One of those papers (posted by Andre – presumably the same one) is the following paper:

      Pronounced occurrence of long-chain alkenones and dinosterol in a 25,000-year lipid molecular fossil record from Lake Titicaca, South America
      Kevin M. Theissen1, David A. Zinniker1, J. Michael Moldowan1, Robert B. Dunbar1 and Harold D. Rowe2

      Presumably Lorenz did not include this site beacuse it wasn’t an “ocean” site.

      From the abstract:

      Using these criteria, the U37K unsaturation indices suggest relatively warmer temperatures in the mid-Holocene. In contrast to previous speculation, lipid analysis provides little evidence of a greatly increased presence of aquatic plants during the mid-Holocene. Instead, it appears that a few algal species were dominant in the lake. Based on the dramatic rise in abundances of LCAs and dinosterol during the early to mid-Holocene, we suspect that the algal producers of these compounds rose in response to a combination of physical and chemical changes in the lake. These include temperature, salinity, and alkalinity changes that occurred as lake level dropped sharply during a multi-millennial drought affecting the Central Andean Altiplano.

    21. Ken Fritsch
      Posted Jan 15, 2007 at 9:35 PM | Permalink

      I continue to think the Lorenz (2006) article is less than clear on the message that it seems to want to convey. It does very little detailed summarization of the data or the conclusions that it appears to only vaguely draw from the data.

      From the alkenone proxy data in Lorenz (2006), I calculated average tropic and extratropic temperature changes to present and came up with: tropics (30N to 30S) change from 7000 years BP (the Holocene Optimum) to present = 0.23 degrees C and for the extratropics (all other areas of the globe) this change = -1.45 degrees C. Since the tropics and extra tropics have nearly equal areas we can say that from the alkenone data alone that 7000 years BP was globally about 0.61 degrees C warmer than present.

      For the ensemble of computer simulations, the Lorenz paper presents, not data for the change from Holocene Optimum to present, but color coded global maps showing the estimated temperatures changes. I eye-balled these maps to arrive at the following temperature changes: For the maps showing changes for annual temperatures I estimated changes of +0.1 degree C for the tropics and -0.7 degrees C for the extra tropics (most of this change occurring in the northern hemisphere). So without stating this conclusion directly the computer simulations showed that the Holocene optimum was approximately 0.3 degrees C warmer than present.

      Now the authors of Lorenz (2006) do not appear to be content with this as a conclusion. They go on to attempt to show that computer model climate simulations can provide seasonal change resolution for the Holocene Optimum period to present that the alkenone results cannot. It appears to me that the interest is not so much quantitative at this point but qualitative in that they want to direct attention to a point that the seasonal variations might be of such a magnitude that alkenone proxy indications of a warmer Holocene than present might be reversed with more precise and seasonal resolved data.

      The Lorenz paper shows two color coded maps of Holocene to present temperature changes from computer simulations one for the Dec-Jan-Feb months (DJF) and the other for the June-July-Aug months (JJA). The global change for DJF shows – by eye ball – a change of 0.14 degree C and for JJA it shows a change -0.6 degrees C.

      Lorenz et al. (2006) give a less than confident observation of phytoplankton growth and the resulting relationship of the alkenone proxy relationship to seasonal temperatures:

      [39] Although the seasonal cycle of SST in the tropics is small, phytoplankton production is not constant throughout the year. It is reasonable that a change of seasonal insolation on the order of 10% is able to impact marine biological productivity. If the alkenone
      production is thought to be highest during the month with the warmest water temperature in the mixed layer, then the resemblance of reconstructed trends with the simulated trends of local summer can be taken as an indication that the time of maximum production may have changed with the insolation signal.

      Lorenz’s case is less than clear when from the link we have:

      Like their land-based relatives, phytoplankton require sunlight, water, and nutrients for growth. Because sunlight is most abundant at and near the sea surface, phytoplankton remain at or near the surface.

      ..The atmosphere is a rich source of carbon dioxide, as millions of tons of this gas settle into the ocean every year. However, phytoplankton still require other nutrients, such as iron, to survive. When surface waters are cold, deeper depths are allowed to upwell, bringing these essential nutrients toward the surface where the phytoplankton may use them. However, when surface waters are warm (as during an El Niño), they do not allow the colder, deeper currents to upwell and effectively block the flow of life-sustaining nutrients. As phytoplankton starve, so too do the fish and mammals that depend upon them for food. Even in ideal conditions an individual phytoplankton only lives for about a day or two. When it dies, it sinks to the bottom. Consequently, over geological time, the ocean has become the primary storage sink for atmospheric carbon dioxide. About 90 percent of the world’s total carbon content has settled to the bottom of the ocean, primarily in the form of dead biomass.

      Also from this link we can understand better what is not clear from Lorenz et al. (2006).

      In the tropics, there is not much in the way of seasonality, and it’s generally pretty warm. That means that the surface waters can get pretty warm, and there’s not a lot of wind to mix that warm water down very deep. Because the density of water depends on its temperature, the warming at the surface can cause the surface layer to become less dense than the water underneath it (which is cooler), and it begins to “float” on top of the cool water. Because these two layers are no longer of the same density, there is an interface between the two layers (the thermocline), which acts as a barrier to mixing, and to the movement of things like nutrients. So, the surface waters, once they become depleted of nutrients, can only be resupplied by whatever recycling of nutrients can happen in the surface layer, and by mixing of nutrient-rich deep water can happen across the thermocline (nutrients in deep water are also regenerated, but all you need to know is that deep water is generally high in nutrients). This doesn’t amount to very much, and it results in phytoplankton growing slowly, and not a lot of growth overall; because the tropics are warm all the time this pattern persists over the entire year. Therefore, phytoplankton in the tropics are generally nutrient limited all the time.

      In temperate latitudes, there is stong seasonality. In winter, it gets good and cold, and there’s lots of storms, which means lots of wind-driven mixing. So, in higher latitudes, the thermocline is seasonal, and does not occur in winter. This means that there is lots of mixing to very great depths (winter mixing in the Labrador and Norwegian seas reach depths greater than 2000m!). So, the deep, nutrient-rich water is mixed up to the surface. There isn’t a lot of phytoplankton growth (or biomass) in the winter, because they become light limited, because they are mixed to such depths. In the spring, when it starts to warm up, a thermocline develops, again separating the phytoplankton from the deep water, but this time they are in the presence of abundant nutrients, and in the surface water where there’s lots of light. When this happen, the phytoplankton begin to grow like crazy in a phenomenon called the spring bloom. Phytoplankton biomass can become extremely high during these spring blooms, which is why overall phytoplankton biomass is very high in temperate latitiudes. Of course, the phytoplankton generally use up all those nutrients eventually, and phytoplankton biomass is smaller in the summer once the phytoplankton become nutrient-limited. There is also often a smaller fall bloom as well, caused when winter mixing starts to break down the thermocline and nutrient-rich deep water beings to be mixed upward.

      The above links make the phytoplankton growth considerably more complicated than the Lorenz paper seems to in looking for supporting evidence for their conclusions. The following links: http://www.whoi.edu/science/GG/paleoseminar/pdf/rosell04.pdf and http://www.mit.edu/~jsachs/Sachs_G3.pdf flat out state that alkenone proxies calibrate with the annual temperatures and not seasonal ones.

      Global sedimentary values of U37K0 have a higher correlation factor with annual SST at 0-m water depth than for any other depth and season [e.g., Mu¨ller et al., 1998]. Therefore, SST estimates from U37K0 can be interpreted as reflecting annual mean values at the ocean surface.

      Global core top sediment calibrations of the alkenone paleothermometer [Miller et al., 1998] are most consistent with mean annual SSTs at 0-M water depth. This is enigmatic since much of the ocean is characterized by seasonal or episodic maxima in haptophyte production [Brown and Yoder, 1994].

    22. David Smith
      Posted Aug 17, 2008 at 11:19 AM | Permalink

      Here is the recent SST anomaly map for the eastern subtropical North Pacific:

      I’ve circled a dark region where SST are perhaps -5C below normal, due no doubt to strong local upwelling. (The other dark areas are land.)

      I also noticed the ECMWF subsurface anomaly cross-section for 140W (the smaller circle on the SST map is at 140W):

      This shows another upwelling pocket of cold water, with anomalies of 5C or greater just 30 meters below the surface. This pocket has been rising for several months.

      I don’t know how unusual this is. I’ve observed a few other such cold pockets in recent years but they tend to be small and disappear quickly. That may prove to be true of these, too – we’ll see. But, if they are more widespread and longer-lasting, I wonder if they’ll affect the trade wind strength and cloud cover, which in turn may affect ENSO behavior. Something to watch.

    23. Chris
      Posted Aug 17, 2008 at 11:27 AM | Permalink

      #22 David – thanks for the SST maps, I wonder if there is any connection with/sheds any light on an issue I was raising in my post on the Sea Ice thread less than an hour ago….?

      http://www.climateaudit.org/?p=3336#comment-289660

    24. Chris
      Posted Aug 17, 2008 at 1:32 PM | Permalink

      #22 David: just to mention, my recent comments #656 on the sea ice thread might be of interest to you here (the comments followed on from my earlier comments/links at the end of #341 on the same thread)

    2 Trackbacks

    1. [...] Interestingly, I had already wondered in Janurary 2007 about the potential impact of coastal upwelling sites on IPCC’s analysis of the Holocene Optimum and had assessed the sites in Lorenz et al 2006 against exactly this standard. See here . [...]

    2. [...] high latitudes were warm.   Really? The only evidence comes from eight seabed cores taken from areas of ocean upwelling, which show slight warming rather than marked cooling since 7,000 years ago[viii]. But more cool [...]

    Follow

    Get every new post delivered to your Inbox.

    Join 3,329 other followers

    %d bloggers like this: