Bond Cycles and the Role of The Sun in Shaping Climate

Bond cycles are defined by petrological tracers from core samples in the N Atlantic that link to the pattern of drift ice distribution. They provide a record of shifting ocean currents and winds, in particular periodic weakening of the North Atlantic current and strengthening of the Labrador current. These cycles shape what we perceive as climate change in the circum North Atlantic realm, for example the Little Ice Age and Medieval Warm Period. They leave a small mark on global average temperatures that are difficult to resolve in the proxy temperature records

Bond Cycles correlate with cosmogenic 10Be suggesting that variations in solar and terrestrial magnetic field strength somehow link to changes in atmospheric circulation and ocean currents. My favoured explanation is changes in solar spectrum that accompany changes in the magnetic field.

With tens of thousands of papers published in climate science, it is possible to pick any 50 and come up with almost whatever narrative one wants. In this post I focus on evidence from ice rafted debris (IRD) dispersed in the N Atlantic from drift ice as presented by Bond et al 2001 [1]. I like the data because it is coherent with what is known about historic climate change in the N Atlantic realm (Figure 3).

Bond Data

Glaciers entrain rocks and rock fragments from the bedrock across which they grind and when they enter the sea to become icebergs and begin to slowly melt this detritus rains down to the sea bed (see inset photo up top). This ice rafted debris (IRD) can tell us something about where the icebergs came from. If the fragments are of granite or schist then this does not tell us anything specific about the source since granite and schist is common in many bedrock areas. But if the fragments are of volcanic glass, then they can only come from Iceland in the North Atlantic realm.

Bond et al identified 3 petrological markers from sedimentary cores from three locations in the N Atlantic. 1) detrital carbonate derived from NW Greenland, 2) hematite (iron) stained grains from NE Greenland, Svalbard and Severnaya Zemlya and 3) volcanic glass from Iceland. (Figure 1). For icebergs to make the journey to  well VM29, off the coast of Ireland, then the pattern of ocean circulation and winds was clearly much different then compared with today.

Figure 1 Map from Bond et al [1] showing bore hole locations and their complex interpretation of shifting currents and atmospheric circulation pattern.

Figure 2 shows the style of cyclical petrological marker change at the various locations. The data may appear complex but to simplify things Bond et al produced an average stack shown as the lowermost panel in Figure 2. It is this average stack that I use as the background image in the charts below. Low values equate with little IRD and a warm N Atlantic and vice versa.

Figure 2 The various mineralogical indices from the various wells that are stacked to produce the average pattern in the bottom panel.

Bond et al interpret their data in terms of shifting ocean currents and atmospheric circulation. In particular a weakening of the North Atlantic Current, strengthening of polar winds and southwards movement of the Polar Front, all enabling a more southerly dispersion of drift ice compared with today.

Comparison with CO2 and GISP2 ice core

Global average temperature is the main historic control over atmospheric CO2, at least for the last 2.6 million years. There is no evidence for Bond cycles either varying with or causing CO2 to change (Figure 3). Thus we must concluded that Bond cycles are neither caused by changes in global average temperature; nor do they cause CO2 to vary. As already noted, Bond et al favoured cyclical changes in the pattern of ocean and atmosphere circulation. This may have periodically brought extreme cold winter conditions to Europe whilst perhaps bringing milder winters to other parts of The Globe.

Figure 3 Bond cycles compared with CO2 and historic climate cycles in Europe. While Bond et al counted 9 cycles (grey numbers starting at 0), I count 10 giving a mean cycle length of 1200 years. LIA = Little Ice Age; MWP = Medieval Warm Period; DA = Dark Ages; RWP = Roman Warm Period. There is no obvious connection between Bond Cycles and global CO2. Note that the way Bond et al plot their data, time is passing from right to left and warm is down and cold is up.

It may be expected that the forces causing changes in drift ice dispersion may also affect temperatures on Greenland. Figure 4 does indeed show cyclical change in temperature on the Greenland Summit (GISP2 ice core) with a similar pattern to that presented by Bond. There are offsets between peaks and troughs that can most likely be explained by different time scales and time scale calibrations.

Figure 4 Temperature data from the Greenland summit shows a similar pattern of cyclical variance to the Bond cycles although some peaks and troughs are offset.

Beryllium 10

10Be is a cosmogenic isotope formed in the atmosphere by the action of galactic cosmic rays on oxygen and nitrogen. Beryllium is a solid, alkaline earth element. Hence when gaseous O or N gets converted the Be falls out of the sky in precipitation.

There are two main factors that control the concentration of 10Be in snow and ice: 1) the precipitation rate and 2) the rate of production of 10Be in the atmosphere. In the following charts, the concentration of 10Be has been corrected for precipitation rate [2] hence the residual reflects the production rate in the atmosphere.

The combined magnetic fields of the Sun and the Earth partially shields the Earth from cosmic rays. Hence changes in the magnetic field strength, particularly of The Sun, is considered to be the main variable controlling the production rate of 10Be. Magnetic field strength also correlates with sunspot numbers, hence when we talk about low sunspot numbers, like we have right now, we are also talking about low magnetic field strength and greater exposure to cosmic rays.

Figure 5 Variations in 10Be and Bond cycles. 10Be data ref [3].

Figure 5 shows quite strong coherency between 10Be and the Bond Cycles suggesting that changes to The Sun’s magnetic field is somehow implicated in shaping the patterns of atmospheric and ocean currents. More on processes later.

Dansgaard–Oeschger cycles

Dansgaard–Oeschger (D-O) events are cyclical temperature excursions observed in the GISP2 ice core. If we look at a deeper / older section of GISP2 we see that 10Be also correlates with these Dansgaard–Oeschger events. But we also see many more events recognisable in the 10Be data than are obvious from the temperature record. Although closer inspection of the T record does show they are present, for example see 10Be event 15.

Figure 6 Temperature and 10Be profiles from deeper/older levels of the GISP2 ice core, same references as before. Chart from an earlier Energy Matters post [ref 4].

I count 20 D-O 10Be events in 23,ooo years giving a mean cycle length of 1150 years, remarkably similar to the duration of the Bond cycles. The solar cycle that actively changed climate in the N Atlantic during the Holocene also impacted climate during the last glacial but with much more punctuated outcomes.

The Finn Men

For those finding it hard to believe that icebergs once dispersed IRD off the west coast of Ireland, an additional strand of evidence comes from the infrequent sightings of Finn Men (Inuit) off the northern Islands of Scotland. One such Finn Man came ashore near Aberdeen around 1728 where he died shortly afterwards [5]. His kayak and equipment are preserved in Marischal Museum (Figure 7), adjacent to where I once studied for a PhD in isotope geochemistry.

It seems reasonable to presume that the edge of the sea ice was much closer then than today and that the Finn Men got caught on drift ice and blown southwards to the northern isles where they once again took to the sea.

Figure 7 One of several 18th century Inuit kayaks that reside in Scottish museums.

The Winter of 2010

Years ago the UK public were informed by the climate science community that our children would grow up not knowing snow. But then the winter of 2010 came along and ruined the climate science script. We all got a taste of what LIA conditions might have been like with lochs and rivers freezing over that had not frozen for very many years.

Figure 8 Not a scene from The Day After Tomorrow but the UK during the winter of 2010. Image from UK Met Office.

The observed reason for the extraordinary and prolonged cold was atmospheric circulation going into reverse with arctic winds blowing across Europe from Siberia. This took place on the back of the extreme low point of solar geomagnetic activity that took place between solar cycles 23 and 24 in 2009. Coincidence? Cycle 24 has become the weakest sunspot cycle since cycle 14 in 1906.

Figure 9 Sunspot cycle 24 is to the right. Look closely and you will see how low the number got in 2009 when the sun had no spots for many months. Image from Solar Influences Data Analysis Center.

In response to Britons freezing to death, the UK MET office published a fine paper in 2011 [6]  explaining how new data showed that changes in the spectrum of energy leaving the Sun was much larger than previously known (one of these known unknowns) and they presented a computer model that showed how change in UV could drive the observed change in atmospheric circulation and account for the abnormally cold winter of 2010.

Spectral Irradiance Monitor satellite measurements indicate that variations in solar ultraviolet irradiance may be larger than previously thought. Here we drive an ocean–atmosphere climate model with ultraviolet irradiance variations based on these observations. We find that the model responds to the solar minimum with patterns in surface pressure and temperature that resemble the negative phase of the North Atlantic or Arctic Oscillation, of similar magnitude to observations. In our model, the anomalies descend through the depth of the extratropical winter atmosphere. If the updated measurements of solar ultraviolet irradiance are correct, low solar activity, as observed during recent years, drives cold winters in northern Europe and the United States, and mild winters over southern Europe and Canada, with little direct change in globally averaged temperature. [6]

With the culmination of solar cycle 24 fast approaching in around 2021, we will shortly discover if low solar activity does indeed cause freezing cold winters in Europe.

Concluding thoughts

This post is written as a partial response to Dr Colin Summerhayes who had a guest post on Energy Matters a couple of weeks ago that stimulated much debate [7]. Dr Summerhayes narrative was based around the notion of the Little Ice Age being the culmination of global cooling, driven by orbital geometry and subsequently reversed by the activities of Man and in particular CO2 emissions.

Insolation peaked around 11,700 years ago, melting the great North American and European ice sheets and parts of West and East Antarctica. Since then, insolation has declined (Berger and Loutre, 2002). As a result, Earth’s climate cooled over the past 10,000 years (Marcott et al., 2013; see also the PAGES 2k Consortium, 2013, for the past 2000 years). The cooling trend culminated in the Little Ice Age of 1350-1850.

Orbital calculations show that we should remain in this cold condition for the next 5,000 years (Berger and Loutre, 2002). So, why are we not still in the Little Ice Age? [7]

The evidence I present here suggests that the Little Ice Age was not unique and was but one of 10 similar cold events (Bond Cycles) of the Holocene brought about not by orbital geometry and its impacts on insolation but by changes to the solar magnetic field and attendant changes in solar spectrum. The subsequent warming in the North Atlantic Realm following the Little Ice Age is but one of 10 such natural warming episodes of the Holocene brought about by changes to the North Atlantic current and atmospheric circulation patterns.

I suggest therefore that there are at least three distinct drivers of climate change 1) orbital forcing of insolation, 2) solar forcing of ocean and atmosphere circulation and 3) forcing from Manmade land use changes and emissions. In order to empirically estimate the magnitude of the latter, it is important to untangle observations of the last 150 years from the solar forced Bond / D-O cycles.


[1] Persistent Solar Influence on North Atlantic Climate During the Holocene Gerard Bond, et al. Science 294, 2130 (2001)





[6] Solar forcing of winter climate variability in the Northern Hemisphere Sarah Ineson et al; NATURE GEOSCIENCE PUBLISHED ONLINE: 9 OCTOBER 2011 | DOI: 10.1038/NGEO1282


[Inset image credit of ice bergs]

This entry was posted in Climate change and tagged , , , , , , , , . Bookmark the permalink.

114 Responses to Bond Cycles and the Role of The Sun in Shaping Climate

  1. Euan Mearns says:

    This is a reply from Polar Scientist who has had a problem commenting:

    In my previous post I was covering the past 450 million years of Earth’s climate and there was not the space to go into detail about Bond Cycles. These are real events, concentrated in the North Atlantic sector. Euan’s graphs show that they were stronger (more ice rafted debris) prior to about 5000 years ago, when there was a ‘step change’ down to lesser levels of debris. This reflects the fact that earlier in the Holocene there were still remnants around of the great northern ice sheets, and the North Atlantic was relatively cold, while later we were left with just Greenland and the North Atlantic was warmer (despite ongoing orbital cooling).

    It has long been realised (e.g. from 14C records), that the sun’s energy fluctuated throughout the Holocene. Back in the early 1980s, Minze Stuiver found that the 14C in the fossil CO2 from ice cores increased during the Maunder Minimum, confirming that solar output was weak then (Stuiver and Quay, 1980, 1981). The solar connection was confirmed when Dansgaard et al (1984) and Oeschger et al (1984) found that 10Be was also abundant during that event. Stuiver found similar increases in 14C at the times of the Wolf Sunspot Minimum (1280 to 1345 AD) and the Spörer Sunspot Minimum (1420 to 1540 AD). Analyses of 14C and 10Be by Schüssler and Schmitt (2004) found maximum values coinciding with the Oort Sunspot Minimum (1040-1080 AD), and the Dalton Sunspot Minimum (1790-1820 AD). Another sunspot minimum, the Gleissberg Minimum covered the period 1890-1910 (Steinhilber et al, 2012). Low 14C and 10Be values suggesting high sunspot activity occurred in medieval times (1100-1250 AD) and the modern era (1960-1990) Stuiver and Quay, op cit).

    By 1998, Stuiver et al had extended the variation of 14C and the solar signal back to 24 Ka. Eduard Bard et al (2000) confirmed Stuiver’s findings for solar activity for the past 1,200 years, finding radionuclide highs associated with the Maunder, Dalton, and Gleissberg sunspot minima. The radionuclide high associated with the Maunder Minimum lay within a rather long period of low solar irradiance between 1450 and 1750 AD. In between the solar minima were solar maxima representing warm periods, including one with values slightly higher than today centred on 1200 AD. Bard et al (2000) concluded that the radionuclide data supported the idea that variations in solar output contributed to the Medieval Warm Period and the Little Ice Age. Their findings were refined by Muscheler et al (2007), who identified solar maxima tied to warming, at 1100-1200 AD (the Medieval Warm Period), at 1750-1800 with a peak at 1790, and at 1960, with somewhat lesser maxima at 1370, 1550-1630, and 1850-1870. Intervening solar minima were associated with cool periods typical of the Little Ice Age.

    My analysis of the available data shows that those solar variations were superimposed on the long decline in northern hemisphere temperature imposed by the Holocene decline in orbital insolation. That decline led to the solar-driven effects of the Little Ice Age being the coldest of the past 10,000 years.

    With regard to the effects of changing sunspot cycles in recent years, it may well be the case that the natural low in sunspot activity accentuated the cooling of the European winter in 2010. But as the sunspot graph of Figure 8 shows, sunspot peaks have been in decline since 1980, and this at a time when temperatures have risen globally to the warmest year of the 20th and 21st century in 2015. So the solar output curve is moving in the opposite direction to the temperature curve and we cannot blame the sun for the warmth of 2015 or the warming since 1980. The rise in greenhouses gases is to blame.

    Stuiver, M., and Quay, P.D. (1980) Changes in atmospheric carbon-14 Attributed to a variable sun, Science 207, 11-19.

    Stuiver, M., and Quay, P.D. (1981) Atmospheric 14C changes resulting from fossil fuel CO2 release and cosmic ray flux variability, Earth Plan. Sci. Letts. 53, 349-362.

    Stuiver, M., Reimer, P. J., Bard, E., et al. (1998) Intcal 98 radiocarbon age calibration, 24000–0 cal BP, Radiocarbon 40 (3), 1041–1083.

    Dansgaard, W., Johnsen, S.J., Clausen, H.B., Dahl-Jensen, D., Gundestrup, N., Hammer, C.U., and Oeschger, H. (1984) North Atlantic climatic oscillations revealed by deep Greenland ice cores, in Climate Processes and Climate Sensitivity (eds J.E. Hansen and T. Takahashi), Geophys. Monogr. 29, Am Geophys. Un., Washington D.C., 288-298.

    Oeschger, H., Beer, J., Siegenthaler, U., Stauffer, B., Dansgaard, W., and Langway, C.C. (1984) Late glacial climate history from ice cores, in Climate Processes and Climate Sensitivity (eds J.E. Hansen and T. Takahashi), Geophys. Monogr. 29, Am Geophys. Un., Washington D.C., 299-306.

    Schüssler, M., and Schmitt, D. (2004) Theoretical models of solar magnetic variability, in Solar Variability and its Effects on Climate (eds J.M. Pap and P. Fox), Geophys. Monogr. 114, Am. Geophys. Un., 33-49.

    Steinhilber, F., Abreu, J.A., Beer, J., et al. (2012) 9,400 years of cosmic radiation and solar activity from ice cores and tree rings, Proc. Nat. Acad. Sci. 109 (16), 5967-5971.

    Bard, E., Raisbeck, G., Yiou, F., and Jouzel, J. (2000) Solar irradiance during the last 1200 years based on cosmogenic nuclides, Tellus 52B, 985-992.

    Muscheler, R., Snowball, I., Jos, F., et al. (2007) Reply to the comment by Bard et al. on “Solar activity during the last 1000 yr inferred from radionuclide records”, Quat. Sci. Revs. 26, 2301–2308.

    • Euan Mearns says:

      Colin, thanks for this thoughtful and well referenced response. The background on 14C and climate change is appreciated. I am familiar with the work of Bard et al. Bard sent me their 14C data a few years ago and I made some lovely charts that got lost in a transition from one version of MS XL to the next.

      I have mulled my reply to your response for several hours, mainly because you seem to be agreeing with most of what I have to say and I am agreeing with most of what you have to say. So where does the controversy lie? Digging around I found this:

      The sensitivity of the Earth’s climate to CO2 could be double what has been previously estimated, according to a statement issued by the Geological Society of London.

      And so I think this lies at the heart of the matter. I see climate sensitivity off the bottom end of IPCC estimate range and you see it towards the top end.

      And so to address a couple of points in your response:

      My analysis of the available data shows that those solar variations were superimposed on the long decline in northern hemisphere temperature imposed by the Holocene decline in orbital insolation. That decline led to the solar-driven effects of the Little Ice Age being the coldest of the past 10,000 years.

      The data I have from Clive Best shows that N Pole insolation has basically been stable for the last 5000 years. So I have trouble with your assertion. Perhaps you could email me a graph to support yours that I will post. And you say the LIA was the coldest for the last 10,000 years. Do you have evidence for this other than Marcott (which has a torpedo heading for it in my next post). The Bond data are actually more consistent with progressive warming, as is the LR04 benthic foram stack.

      As for solar geomagnetic field strength. I would say that cycles 21, 22, and 23 were all pretty similar from the sunspot proxy. But cycle 24 takes a step down that corresponds with what was once a pause in warming. That’s enough for now, probably more to follow.

    • Euan Mearns says:

      This is another reply from Polar Scientist

      Euan, In response to your question about the evidence for declining temperature and insolation, please see the following figures.

      Figure 1 shows the global temperature anomaly based on 30-year means. It is from PAGES 2k Consortium, 2013, Continental-scale temperature variability during the last two millennia. Nature Geoscience 6, 339–346.
      It was repeated as Figure 5 in Wanner, H. Mercolli, L., Grosjean, M., and Ritz, S.P. (2015) Holocene climate variability and change; a data-based review. Jour. Geol. Soc. Lond. V. 172, 254-263.

      Figure 2 shows declining northern hemisphere insolation and decreasing ice volume over the past 10,000 years, as calculated by André Berger. E.g. see Berger, A. and Loutre, M.F. (2002) An exceptionally long interglacial ahead? Science 297, 1287–1288.

      Figure 3 shows the decline in Greenland temperature derived by subtracting the effects of changing altitudes due to ice growth and decay. It is from Vinther, B.M., Buchardt, S.L., Clausen, H.B., et al. (2009) Holocene thinning of the Greenland ice sheet, Nature 461, 385-388.
      Vinther’s team saw that both altitude and past thinning caused by warming shaped the ∂18O record that had previously been used for the temperature reconstruction. “Contrary to the earlier interpretation of ∂18O evidence from ice cores” Vinther said, “our new temperature history reveals a pronounced Holocene climatic optimum in Greenland coinciding with maximum thinning near the GIS [Greenland Ice Sheet] margins”. Their new record of Greenland’s temperature history broadly follows the pattern of Northern Hemisphere summer insolation.

      I argue that the solar variability throughout the Holocene provided minor modifications to the overall effect of declining northern hemisphere insolation. Those modifications, which include the Medieval Warm Period and the Little Ice Age form the up and down wiggles on the curve of declining Holocene global temperature shown in Figure 1.

      What does this mean for understanding Bond Events? The climate signal is complex and has to be disaggregated. The declining northern hemisphere insolation at the beginning of the Holocene (Figure 2) leads to gradual warming of the northern hemisphere (and Greenland, Figure 3) as the enormous northern hemisphere ice caps and their associated albedo decline. The ice sheets were gone by about 7000 years ago. Which is when Holocene northern hemisphere climates were warmest at high latitudes (despite declining insolation). The northern Atlantic Ocean was colder and Bond Events were bigger while all that melting ice was around. Once the ice had gone, the northern Atlantic Ocean warmed and Bond Events declined. Paradoxically, insolation was in decline throughout this period (Figure 2), as was northern hemisphere (Greenland) temperature (Figure 3). The matter is complicated because outside influences must also be considered – in this case that Atlantic warming was driven by changes in northward transport by the Gulf Stream, which would have been more effective once the great ice sheets and their associated sea ice and melt-water had disappeared or, at least, significantly diminished. The matter is not as straightforward as you might like to think.

  2. Pingback: Bond Cycles and the Role of The Sun in Shaping Climate | Climate Collections

  3. Euan: As you know I’ve long been a proponent of the theory that ice sheet dynamics are a major control on the advance and retreat of ice sheets and that these advances and retreats themselves modify the Earth’s climate. So now I’m going to propose a variant of this theory. Bond events record surges in the NH ice sheets, like the present-day glaciers on Svalbard but at a larger scale. The surges are related to temperature, ice accumulation and a number of other factors, but lag the climatic changes by variable amounts because of ice sheet inertia. This could explain some of the miscorrelations and lags you see on your two Figures 4. (Note that your Figures after Figure 4 are misnumbered. Apologies for not picking up on this yesterday).

  4. Pingback: Bond Cycles and the Role of The Sun in Shaping Climate –

  5. Euan Mearns says:

    Before responding to Polar Scientist, I wanted to post this further example of solar forcing of climate change from stalactites in a cave in Oman. Dr Summerahyes (Colin) introduces the evidence from d14C that is analogous to 10Be since both are cosmogenic isotopes. But 14C finds its way into substances that contain C like the carbonate stalactites in Oman.

    The data are from: Strong coherence between solar variability and the monsoon in Oman between 9 and 6 kyr ago
    U. Neff*, S. J. Burns23, A. Mangini*, M. Mudelsee§, D. Fleitmann2 & A. Matter2
    * Heidelberg Academy of Sciences, Im Neuenheimer Feld 229, Heidelberg, Germany D-69120
    2 Geological Institute, University of Bern, Baltzerstrasse 1, Bern, Switzerland CH-3012
    § Institute of Meteorology, University of Leipzig, Stephanstrasse 3, Leipzig, Germany D-04103

    NATURE | VOL 411 | 17 MAY 2001 |

    In short, the data records changes to the Indian Ocean Monsoon responding to the same solar signals recorded by Bond. These solar signals change climate, if not everywhere, in very many places at once.

  6. donb says:

    VERY interesting post.
    The flux of galactic cosmic ray particles (GCR) at Earth is MUCH more dependent on solar-generated magnetic field than on Earth’s field. Solar is much stronger. The average GCR energy is 3 billion electron-volts, whereas solar wind protons (largely deflected by Earth’s field) are only 1 thousand electron volts. Coronal mass ejection protons (CME) of intermediate energy produce much of the strong electromagnetic interactions in Earth’s atmosphere.

    Polar Scientist’s comment about insolation decreasing for the past ~11 kyr because of orbital effects is partially true. However, such insolation variations alone cannot produce global cooling. This is because the whole Earth receives roughly constant insolation; it is the polar regions that vary. Likely, insolation-produced changes in the Arctic regions produce other changes such as in ocean currents, cloud albedo, land and increased ice albedo, etc., which then cause the actual global cooling.

    P.S. Like you, I am an isotope & nuclear geochemist, but not in Sm-Nd. Much of my work was in nuclear reactions in nature.
    Keep up the good posts.

  7. Euan Mearns says:

    The chart above was posted by Polar Scientist in support of the contention that global temperatures fell steadily for 1400 years before being sharply reversed allegedly by Man. We have to guess at the scale. The source is given as PAGES 2K which I believe is this:

    I’ve not read it, just looked at the pictures 😉

    So this is clearly the raw data upon which Colin’s chart is based (taken from the above link). We see that the range in temperature is a rather paltry 0.6˚C. Just eyeballing the 4 coloured lines gives a very different picture to the average displayed as dots. I think the reason for this is that all the data come together in the last 200 years and so there is no cancelling and smoothing. Before then there is a lot more variance between the 4 models that tends to cancel the structure that is evident in each model. But I am struggling to create the trend shown as dots from the 4 trends shown as lines.

    We see that this reconstruction is continental and NOT global. And its worse than that….

    Assuming this shows the regional distribution of proxies used we see that prior to 1200 AD it is based on very little data. We also see that post 1200 AD the reconstruction is dominated by N America and Asia while before 1200 AD it is dominated by Asia and the Arctic (if I am reading this correctly which I may not be).

    If I am correct here, then this temperature history is not robust.

    • clivebest says:

      The contentious element of Pages2K is of course the 20th century uptick. The proxies are mostly land based and the uptick is influenced by just a few of them showing strongly rising ‘temperatures’.The effect is most likely due to humans but may not be due to temperature at all, since proxies don’t measure temperature directly which are instead inferred through sediments or tree rings. For example the rapid development in new settlements and agriculture must affect the pollen count. Some Arctic proxies show a temperature rise while many others don’t. The strongest rise is at Igaliku in Greenland which was resettled in the early 20th century and is likely due to land use changes.

      A rise of over 2C since 1850 is also too large when compared to instrument data. Even CRUTEM4 shows a signal of no more than 1.5C since 1850.

      • Owen says:

        The hockey stick at the end is plain wrong anyway, because after a 1C rise to 1950, there was a similar drop till 1986 bringing us back to post LIA temp levels, then a rise again till 2010.

        So the hockey stick brings the whole graph into question

  8. polarscientist says:

    The PAGES 2k group data (shown in the link given by Euan, and in his latest comment, above) are represented as dots with error bars in their Figure 4b. That data was the basis for my Figure 1, which came from one of the PAGES 2k authors, Heinz Wanner, who, in effect had joined the dots (see Wanner reference in my post above). The PAGES 2k data were compared in their Figure 4 with data from four other sources (Figure 4a – see Euan’s post above), all of which were from the Northern Hemisphere, where the Medieval Warm Period was quite well developed. It was less well developed in the Southern Hemisphere, which probably explains the difference between those curves (Figure 4a) and the PAGES 2k data (Figure 4b). Interestingly, all 5 palaeodata sets show the same uptick in the last 200 years. That suggests that the data are robust. In any case, the notion that the Northern Hemisphere climate has cooled since the mid Holocene is clearly supported by the data from Greenland (Figure 3 in my earlier post, above). The Wanner and PAGES 2k figures demonstrate that this phenomenon is not merely restricted to Greenland.

    Regarding insolation as a driver for climate change it is abundantly clear from Figure 2 in my post that there is an extremely close relationship over the past 200,000 years (right hand side of graph) between northern hemisphere ice and insolation. I am puzzled by the statement by Donb that “insolation variations alone cannot produce global cooling”. After all, it is the variation in insolation with time that has produced the fluctuations of the recent Ice Age, as we know from the initial work of Croll in the 1860s and Milankovitch in the 1920s-40s and Berger in the 1970s to present. Of course, the effect of insolation is exacerbated by the growth and decline of the albedo as ice grows or melts in the polar regions, and by other factors, such as the expansion and contraction of greenhouse gases CO2 and CH4, which grow in warm periods and shrink in cool periods.

    • donb says:

      I clarify my comment above about insolation changes and glaciation. I DO believe that insolation changes in the northern hemisphere (NH) initiate glacial cycles, and I agree that the ~41 kyr and ~100 kyr orbital cycles are closely correlated (but not perfectly) with significant global temperature changes over the past 200+ kyr.

      But, when the NH receives much less insolation (as occurred ~20 kyr ago) the SH receives much greater insolation, such that the globe total insolation remains rather constant. So how does colder temperature in the NH cool the globe when the SH is being warmed? I argue that greatly cooling of the NH in a down cycle (the NH at 65N saw a ~90 wats/m^2 decrease in insolation between insolation max in the Eemian ~120 kyr ago and the great temperature drop ~20 kyr later) initiates other changes that affect the whole globe. Among these are increasing ice albedo (sea and land), exposed sea bed with changed albedo, vegetation cover changes, CO2 uptake by the cooling ocean, greater restriction on entry depths for Atlantic Ocean currents to enter the Arctic, changes in global ocean currents, and possibly significant changes in global cloud cover. IT IS THESE SECONDARY CHANGES that actually cool the whole globe.

      Accepting this, one sees that more modest changes in these secondary factors, independent of orbital insolation changes, might bring about some of the more modest regional and global temperature changes deduced for the Holocene.

      • polarscientist says:

        Donb, To answer your question about the difference between the SH and the NH, it’s worth bearing in mind the observations of Huybers and Denton We can draw on the work of Huybers and Denton (2008: Antarctic temperature at orbital timescales controlled by local summer duration, Nature Geoscience 1, 787-792.),which notes that although insolation increased during the Holocene in the Southern Hemisphere summer, it decreased during the Southern Hemisphere spring. Also, when summer insolation was high in the Southern Hemisphere, the length of summer was shorter and the length of winter was longer than average because the Earth was then at perihelion (close to the sun). These factors conspired to cool the Earth throughout the Holocene. In other words it’s not just total insolation per hemisphere input that leads to climate outputs.

        Aside from that, you are of course right, that many feedbacks come into play to alter the absolute effect of rather small changes in insolation. As the oceans warm with climbing insolation, more CO2 is emitted, contributing to warming and the evaporation of H2O vapour, which further accentuates warming. Equally, warming reduces albedo as ice melts. The changing area of the ocean as ice melts is equally important, as is the movement of the Inter-tropical Convergence Zone, which expands with warming. Then again we have to consider how ice sheets behave – their inertia slows sea level rise, and their changing weight can lead to iceberg outbreaks. These various feedbacks conspire to give us the climate we get.

        • donb says:

          POLAR SCIENTIST. Not sure I understand all of your comments in the first paragraph above, or if they are inconsistent with my suggestions.
          Currently, perihelion occurs in Jan, when the SH experiences summer and NH winter, and NH summers occur when Earth is farthest from the Sun. (Early Holocene was almost reversed.) But orbital speed varies (by Newton’s laws of motion) such that Earth moves rapidly when close to the Sun, but more slowly when distant. Thus, greater insolation rates at close approach are largely canceled by rapid movement, and lesser insolation rates at distance are largely compensated by slower movement, such that total insolation received over time is constant.
          But the last sentence of your first paragraph is also my point.

        • polarscientist says:

          DonB, I think you’ll see what I mean if you read the Huybers and Denton paper to which I refer.

    • steve says:

      Glad to see someone of your seniority back Polarscientist. The current stuff is a bit too much for a non-geologist, but could you possibly give a brief explanation on the ice core re-dating which you cited on 18.3, when you answered my query of 24.3 about the subject, saying that the bubbles in the ice cores did not move and the re-dating was done by methodology. Greg Kaan put on the paper by Parrenin 2013 and commented that it contained a lot of assumptions and suppositions. Having read it 3 times I would add assertions, lack of explanation of methodology and graphing. Various words ending with ‘ffle’ came to mind.

      Could you give some idea as to how bubbles frozen in at one date can be re-measured to age 600 or so years later? Possibly some qualified readers may not understand too.

      • polarscientist says:

        The Parrenin team (France) recognized that previous analyses used an inadequate model to estimate the difference in age between the ice and the bubbles that the ice surrounds at any given depth. They realised that within the firn (snow in the process of becoming ice) the ratio of the nitrogen isotopes in the bubbles (15N to 14N) would be enriched with depth due to gravitational settling favouring the heavier isotope, and that this enrichment depended on the thickness (i.e. depth) of the firm layer. Given the nitrogen isotopic analyses from the bubbles they could more accurately assess the age of the bubbles enabling the CO2 concentration in the bubbles to be related to ice of the same age, and hence to the temperature at the time of deposition. Quite independently, another group, led by Joel Pedro (Tasmania), used ice cores with rapid rates of accumulation to re-assess the relationship between CO2 and temperature. Using a different technique they also found that the difference in age between the CO2 and the temperature was much less than had been supposed by using the previous age-depth model for air bubbles. [see J. B. Pedro, S. O. Rasmussen, T. D. van Ommen, Clim.
        Past 8, 1213 (2012)]. Basically the earlier models used to assess the age of the bubbles back in the late 1980s was wrong. Using a more sophisticated approach and new data not available then (e.g. nitrogen isotopes), allows a reappraisal. This is how science progresses. For a review see Brook, E.J., 2013, Science 339, 1042-1043.

        • steve says:

          Thanks Dr Summerhayes and for making the relevant part of your interesting book available via the Brook E J ref. I have been reading it today, particularly from p 276 Dansgaard, where you give the original lag as between 600-1900 years.

          From the Vostock graph, the depth of the ice core to the Holocene was c 1700m and the time 130k years, giving the depth of the core per year around 13mm. The original volume measurement of CO2 is now thought to be wrong and not the age of the gas in the particular annual core, or is it over more years that the sample is taken? The Parrenin measurement is made using isotopes of Nitrogen and their proportion, measuring this from the firn, or increasingly compressed snow before the ice seals the gases in permanently and giving very reliable samples. Am I right that this firn being recalibrated may be up to 100m thick, as it compresses to reliable quality ice?

          This must be very clever stuff. Parrenin has come to simultaneous timing +- 200 years, and other follow up with a lead of 720+- 350, 460+- 360 and one in the other direction, a lag of 620+- 600 years. There has been a heroic effort to find out whether the large differences are due to other factors, such as how far the Earth was from the sun 130k years ago and whether the tilt favoured the northern or southern hemispheres, among other factors. I have made notes but may be completely wrong.

          It was pleasing to note that NASA’s Hansen of the Hockey Stick estimated response to surface temperature at 100 years and this is now thought to be 100,0000.

          I will ask for a paper copy for Christmas and recommend your book as readable and interesting for any other non-scientists, or scientists with £55 spare. Finally, p289 -‘that CO2 preceded warming- is now known to be wrong- Parrenin’. Is this right?

          Many Thanks,


          • polarscientist says:


            The quote on p 289 relates to the assumption by Kawamura et al (2007) that CO2 preceded warming. Parrenin (among others) showed that assumption to be wrong.

            Regarding the firn depth, the firn is the top 50-100m of an ice sheet, and comprises snow that is gradually being compressed to (eventually) form ice.

            Because air in pockets in the firn is continually in contact with outside air until the pockets become bubbles trapped in ice, the age of the air in the bubbles is not the same as that of the ice that surrounds them. Assumptions had to be made to determine the likely age of bubble air in ice cores. Parrenin used clever chemistry to re-determine the most likely age of the air in the bubbles, and by doing so showed that the original age assumptions were wrong. Pedro et al used a different method, but came up with a similar answer (i.e. the original estimates of the age of the air in the bubbles were wrong).

            In a way this is not surprising. Imagine – you increase insolation, which warms sea-water, and CO2 is emitted (immediately). In theory (i.e. basic physics) there should be no delay between the warming and the rise in CO2. The early Vostok analyses using the old age model for the bubbles found an unexpected lag between the temperature rise and the CO2 rise. Parrenin has now shown why that analysis was wrong. And we are back with what the new data demonstrating in fact what basic physics says should be the case.

            I hope that clarifies matters.

          • Euan Mearns says:

            But at the inceptions the CO2 time lag is of the order 8000 years while CH4 is perfectly aligned with T and the old gas age calibration. We go from interglacial to full glacial conditions with interglacial CO2 concentrations.

          • steve says:

            Thanks Polar. Bloody Honest for a change. Will buy Your book. S

          • polarscientist says:


            In response to Euan’s comment: “But at the inceptions the CO2 time lag is of the order 8000 years while CH4 is perfectly aligned with T and the old gas age calibration. We go from interglacial to full glacial conditions with interglacial CO2 concentrations.”

            My comment would be that Parrenin has shown in his paper that you are wrong. See Figure 4 in his article. CO2 and Temperature are now shown to rise and fall together in Antarctic cores. CH4 and CO2 move together at inception (c. 18,000 years ago). However, the two diverge at around 15000 years ago when northern hemisphere warming kicked off. That’s most probably because CH4 is driven by changes in wetlands, while CO2 is driven primarily by oceanic warming or cooling. CH4 then follows the northern hemisphere Younger Dryas pattern, which did not occur in the southern hemisphere. CO2 does not follow the Younger Dryas cooling. By contrast with the northern hemisphere and its Younger Dryas, Antarctica has the Antarctic Cool Reversal between 15,000 and 13,000 yrs BP when Greenland shows warming. This shows that the polar regions of the two hemispheres move in concert at some times, and in different ways at other times, depending on local, conditions and on the activity of the bipolar seesaw.

          • Euan Mearns says:

            @ Polarscientist

            Colin, in referring me to Fig 4 in The Parrenin et al paper, you are referring me to a high resolution analysis of the TERMINATION of the last glaciation where I don’t (and never have) disputed that CO2, CH4 and T are pretty well aligned. My comment about the 8000 Y time lag refers to the INCEPTION.

            I’ve had a quick look at the EPICA CO2 data and find that it seems to fall into 3 time intervals:

            1) 0-22,000 Y
            2) 393-664 KY
            3) 611-800 KY


            I’ve not plotted any data yet, but it appears that the time interval of Vostok has not been analysed. There are two time lags. 1) the 200 to 600 Y lag that everyone talks about which I’m quite happy to see adjusted with high resolution ice age dating. But the ELEPHANT in the room is the 8000 Y time lag at the inceptions. CH4 follows T down but CO2 does not. There is an 8000 Y time lag between CO2 and CH4/T at each inception in Vostok. Full glacial conditions are established at interglacial CO2 levels.

            Best Euan

          • polarscientist says:

            Hi Euan,

            Apologies, I thought you were referring to the inception of the termination. I am sending you a graph that I hope you can insert in this reply. It seems to me from this EPICA Dome C core that there is no delay.

          • Euan Mearns says:

            From Polar Scientist:

          • Euan Mearns says:

            All these charts are clickable for a large version. This is Vostok using the original GT4 timescale.

            Top chart shows T and CH4 “perfectly” aligned.
            Middle chart shows CO2 and T well aligned at the terminations but with several thousand year time lag at the inceptions.
            Bottom chart shows detail of time lag at beginning of last glaciation.

            The source of the data plotted is given on the charts. Its the original data from Petit et al. Now its not possible to get rid of the 8000 year time lag by adjusting the gas ages without totally screwing up the alignment of CH4 and T unless folks are going to use different bubble closure ages for CH4 and CO2.

            Colin, I’m also surprised that your chart shows CO2 data from Dome C (EPICA) since I can’t find any CO2 data from EPICA for the 22 to 393 KA interval. Here’s the portal.


            Could you perhaps post a link to the data that is plotted in the Luthi chart? Eyeballing it, the 0 to 400 KA interval looks as if it could be Vostok to me.

          • Javier says:


            There is a possible explanation for the slow decay of CO2 at glacial terminations,

            When temperatures go down oceans become a sink for CO2, but the biosphere becomes a source as there is a huge progressive die off of plants and animals due to the worsening climate whose carbon is progressively released to the atmosphere as CO2 due to organic decay. Only when the biosphere source becomes smaller than the sink after a few thousand years does the level of atmospheric CO2 drop towards equilibrium.

            Land becomes a lot less productive during glacial periods. Deserts expand greatly. Everything is consistent with a great mobilization of carbon from biological stores.

          • Euan Mearns says:

            On warming, oceans exhale CO2 “immediately”. On cooling, it takes time for ocean biota to grow (plankton) and to pump down the atmosphere. “Solution pump”, biological pump, ocean circulation and time lags. ± Boreal forest and all the rest.

            The slow decay is at glacial INCEPTIONS!

          • polarscientist says:

            I agree with Javier, though my response was a little more extensive.

          • polarscientist says:

            Ah, at last I see what you are getting at. Once glaciation starts, temperature drops faster than CO2 does. Yes, it does indeed. One reason is that as cooling begins, sea ice starts to re-grow, which will shrink the surface of sea-water through which CO2 can be absorbed. Another is that as temperatures start to fall, so does sea-level, which again restricts the surface area of the ocean. So even though a cooling ocean should absorb CO2, it has more difficulty doing so because of these areal changes. The growing sea ice (and snowfall on land) would also provide a growing albedo feedback, tending to push temperature down fast. But we have to consider the biology too. Cooling leads to drying and and the shrinkage of areas covered with vegetation (e.g. the sudden growth of the Sahara at about 6000 years BP as the Inter-Tropical Convergence Zone was forced south). There are fewer plants to absorb CO2. Equally, the death and decomposition of land plants with cooling would provide more CO2 to the atmosphere. Meanwhile, in the ocean, upwelling currents that supply CO2 from depth are still likely to be active or even to be enhanced as winds strengthen with cooling. It’s a source versus sink problem.

  9. clivebest says:

    Interestingly, all 5 palaeodata sets show the same uptick in the last 200 years.

    It isn’t that simply because they all use the same overlapping proxy data. I accept that there has been an increase in global temperature ‘anomalies’ since 1850, but I am not convinced that the paleoclimate data consistently matches up to the measurement data. It is all a bit too convenient in my mind, because some proxies show no change while others show large increases in the same location. In addition there are some opaque area averaging procedures, and a couple of inverted proxy data were also found by Steve McIntyre.

    The climate science community needed hard evidence confirming that the 20th century was the warmest during this interglacial. PAGES 2K apparently succeeded whereas Mann’s first paper failed to convince. I may be completely wrong here, but I have the uneasy feeling that group thinking has influenced the data analysis.

    • Euan Mearns says:

      A reply from Polar Scientist

      Clive, you are casting aspersions. The palaeoclimate science community did not NEED hard evidence confirming anything. They were and are interested in how climate changes through time, and whether or not we can identify the various climate forcing factors at different times back from the present. This is one of the most intellectually demanding questions in geology, and has been preoccupying geologists from the time of Charles Lyell (1830), and natural scientists like Buffon even back in the 18th century. Serious studies of past climate change based on ocean sediment cores began in the 1950s and were greatly expanded with the advent of the Deep Sea Drilling Project in late 1968; its descendant still operates today. From it we gained the remarkable discovery that orbital signals are reflected in the sedimentary record (Hays, Imbrie and Shackleton, 1976), a discovery as intellectually stimulating as that of plate tectonics in the late 1960s. These exercises all took place BEFORE significant attention was drawn to the possibility than man was altering the climate. There was no attempt to ‘CONFIRM’ that the 20th century was the warmest of this interglacial. Indeed, it was NOT the warmest – recent temperatures still do not reach the heights of the so-called mid Holocene climatic optimum. Later, ice cores were drilled to see to what extent they might enhance our understanding from deep sea cores of the nature of climate change in the late Pleistocene and Holocene. It is through these multifarious studies that we have come to appreciate, for example, that there was a steady cooling in the Northern Hemisphere, driven by declining orbital insolation, in the past 7,000 years, of about 0.5-0.6 deg C, following the disappearance of most northern hemisphere ice sheets (see my Figure 3, above). In the process of analysing the multitude of palaeodata, which extends into the 20th century, it became obvious that there had been a marked warming starting in about 1850. This was independently noted, for example, by the Scandinavian palaeobotanists studying the remains of trees, plants and pollen in their region, starting in the late 1880s. Scandinavian and Scottish palaeobotany, deep sea cores, and ice cores independently tell the same story for this region. Declining orbital insolation predominated through the Holocene. Once the ice had melted, there was a thermal optimum. Cooling then continued with some small solar-driven ups and downs, into the Little Ice Age. The orbital data say we should still be in the Little Ice Age, albeit in one of its well known warm phases. The geological and modern data say we have moved outside the natural envelope of the past 2000 years or so (but not yet to mid-Holocene levels). These data are from a multitude of sources and from national scientists from every country around the Arctic region. Their data are robust. There is no conspiracy, and this is not group think. This is a lot of people working to out-compete each other in top journals, and finding that they come up with similar results. You do not get science prizes for group-think. You get them for providing the best data and the best interpretation of it.

      • donb says:

        I extend my comment above to you about orbits and glaciation to the Holocene.
        For about the past 10 kyr the NH at 65N has lost about 35 w/m^2 of insolation. Earth is approaching a minimum in orbit-driven insolation, with another ~5 w/m^2 to go. As you say above, this has not (yet?) initiated much Holocene cooling, although it is a significant fraction of the ~90 w/m^2 cooling that occurred after the Eemian. So why not?
        I argue it is because NH cooling (prior to recent times) was insufficient, for whatever reason, to significantly “turn-on” the above secondary drivers that produce significant global cooling.

        • polarscientist says:

          DonB, I think I have made clear that the Holocene has cooled since the mid-Holocene climatic optimum, a point that Javier has also emphasized in his posts. I think that perhaps you are asking why did the cooling not lead to a glaciation? There are two answers. First, Andre Berger’s work (see references in my earlier post) shows that the decrease in insolation is not sufficient to have caused the next glaciation, and that it is unlikely to do so before 30-50,000 years from now. Secondly, Ruddiman and his colleagues point out that human activities conspired to start increasing CO2 (from burning forests at the start of agriculture) around 6000 years ago, and CH4 (from rice farming starting about 3000 years ago) [Ruddiman, W.F., Fuller, D.Q., Kutzbach, J.E., Tzedakis, P.C., Kaplan, J.O., Ellis, E.C., Vavrus, S.J., Roberts, C.N., Fyfe, R., He, F., Lemmen, C. and Woodbridge, J. (2015) Late Holocene climate: natural or anthropogenic? Reviews of Geophysics doi: 10.1002/2015RG000503 ] . That conspired to prevent us from falling into a glaciation. Indeed, Ganopolski argues that we now have enough CO2 in the atmosphere to postpone the next glaciation for 100,000 years [ Ganopolski, A., Winkelmann, R. and Schellnhuber, H.J. (2016) Critical insolation–CO2 relation for diagnosing past and future glacial inception. Nature 529 doi:10.1038/nature16494].

          • donb says:

            POLAR SCIENTIST. My question above was rhetorical. I agree that NH insolation decrease over the Holocene was insufficient to trigger those secondary effects needed to produce sufficient cooling to begin glaciation. As you say, orbital changes in insolation predicted over the next 50 kyr are also likely to be too small to produce either a new glacial cycle or intense warming.

          • Javier says:

            Dr. Summerhayes (polarscientist),

            To my surprise I agree with almost everything you have said in your comments to this article, and I am surprised because it is very difficult these days to read that a climatologist thinks that we are currently below Holocene hypsithermal average temperatures. And it is becoming more and more difficult to read about paleoclimatology where the record is not being reinterpreted as CO2 being the main driver of temperatures since the atmosphere formed.

            I don’t want to look too negative for bringing up only the things I disagree, but you are supporting two popular hypotheses that have absolutely no evidence behind.

            Berger’s hypothesis of a long interglacial for MIS 1 (Holocene) as analogy to MIS 11 is based on the erroneous interpretation that 65°N summer insolation is the main driver of glacial cycles, and thus as long as it is not low we are safe. This hypothesis is easily refuted by looking at the proper analogy to MIS 1, which is MIS 19.

            Unlike MIS 11, MIS 19 has an astronomical signature that is almost identical to MIS 1, as this figure from Pol et al., 2010 shows (red for MIS 1 and black for MIS 19):

            Glacial inception arrived to MIS 19 despite almost identical values of 65°N summer insolation, and will arrive to MIS 1 at the next low of the Hallstatt cycle in 2000 years. By the way, if we were living 770 Kyr ago and we were at one of those AIM (Antarctic Isotope Maxima) in the figure, we would be very worried about unprecedented global warming.

            The reasons why a long interglacial is unlikely are discussed in detail in the comments section of Clive’s recent article at:
            And illustrated by the figure:
            That clearly shows that the conditions for an interglacial are in order of importance:
            1. Rising obliquity that defines the window of opportunity (Huybers, 2007)
            2. Very low temperatures / extensive ice sheets / very low sea levels at the beginning of the window (Paillard, 1998)
            3. High 65°N summer insolation at the end of the window.

            Very high eccentricity can cause a double interglacial like in MIS 7 and MIS 15.

            Our low eccentricity and relatively high 65°N summer insolation will just make sure that there won’t be another interglacial in 30,000 years. The world will have to wait at least 70,000 years for the next interglacial.

            The other misguided hypothesis is Ruddiman’s early anthropocene, which amounts to anthropocentrism at its worst. The origin of the rise in CH4 since 7000 yr BP is clearly northern wetlands that appeared after the ice sheets melted and expanded due to increased precipitation, not rice cultivation that was anecdotical when human population was so low. There are several good arguments against Ruddiman’s hypothesis that cannot be ignored (Broecker and Clark, 2003; Broecker and Stocker, 2006; Kleinen et al., 2010; Stocker et al., 2011).

            Broecker, W. S., & Clark, E. (2003). Holocene atmospheric CO2 increase as viewed from the seafloor. Global Biogeochemical Cycles, 17(2).

            Broecker, W. S., & Stocker, T. F. (2006). The Holocene CO2 rise: anthropogenic or natural?. Eos, Transactions American Geophysical Union, 87(3), 27-27.

            Huybers, P. 2007. Glacial variability over the last two million years: An extended depth-derived agemodel, continuous obliquity pacing, and the Pleistocene progression. Quat. Sci. Rev. 26 37-55.

            Kleinen, T, et al. “Holocene carbon cycle dynamics.” Geophysical Research Letters 37.2 (2010).

            Paillard, D. 1998. The timing of Pleistocene glaciations from a simple multiple-state climate model. Nature 391 378-381.

            Pol, K. et al. 2010. New MIS 19 EPICA Dome C high resolution deuterium data: Hints for a problematic preservation of climate variability at sub-millennial scale in the “oldest ice”. Earth & Planet Sci Lett 298 95-103

            Stocker, B. D., et al. “Sensitivity of Holocene atmospheric CO 2 and the modern carbon budget to early human land use: analyses with a process-based model.” Biogeosciences 8.1 (2011): 69-88.

      • clivebest says:


        I think you misunderstand my point a little. Paleoclimate is a fascinating science that has made many insights into past climate changes especially the pleistocene ice ages. The scientists involved are professional and progress has been very impressive – so I 95% agree with everything you write above. My only concern are those few papers which purport to confirm unprecedented 20th century AGW warming, especially when they also claim that the current climate is warmer than at any time during the last 2000 years. This started with Mann et al’s hockey stick, continued with Marcott et al. and followed by the Pages 2k publication. This is such a hot potato because the political implications are huge. Therefore all such results must be robust and reproducible.

        When I re-analysed Marcott’s proxies using exactly the same algorithm for global averages as that used for station data by Hadcrut4, I got a completely different result.

        Only a few of the Marcott proxies extended into the 20th century and the way the data was binned made the crucial difference. I have not looked in detail at Pages 2K but Steve McIntyre has. His conclusion is not that there is no rise in trends during the 20th century, but that the analysis methods chosen exaggerated it.

        Steve McIntyre has now become one of the world’s expert on the use and misuse of proxy data. He found that the final uptick in the global data was dominated by Australasia and the Arctic. These in turn were caused mainly by particular proxy series. The Arctic trend is mostly robust, whereas the Australia trend is suspect.

        Some of mistakes he uncovered on the way were : Hvitarvatn (used upside down by PAGES2k), Igaliku (contaminated data used by PAGES2K) and Kepler Lake (same d18O behaviour as rejected Mt Logan) in the preparation of McKay and Kaufman 2014 (Nature Geoscience).

        The Pages2K regional temperature anomaly averages show upticks for the Arctic and Australasia regional averages.

        Yet when the PAGES2K authors made the global average, they first converted back into Standardised Units (SD) rather than simply average the regional temperature anomalies. This weights proxies with sharp changes in SD (as in the Gerkis reconstruction in Australia) thereby generating a massive uptick. This then feeds directly into the global average once it is converted back into temperature.

        You can see that the origin of the global blade lies in the Arctic and in Australia. Yet how robust is that Australia data really. Steve McIntyre writes:

        The largest problem with the Neukom-Gergis network is grounded in the data: the long ice core and tree ring series don’t have a HS shape. However, there is a very strong trend in coral d18O data after the Little Ice Age and especially in the 20th century. Splicing the two dissimilar proxy datasets results in hockey sticks even without screening. Such splicing of unlike data in the guise of “multiproxy” has been endemic in paleoclimate since Jones et al 1998 and is underdiscussed.

        I just don’t think the evidence is there yet from paleoclimate proxies for the sharp hockey stick blade which is expected by the climate science community.

        • polarscientist says:

          Clive, I don’t think that the climate science community EXPECTS there to be a sharp hockey stick blade. I think that they OBSERVE a sharp hockey stick blade. I have cited several paleoclimate papers that make this point. Javier has cited several others. These are based on good solid lake and marine sediment core and ice core data from multiple sources and in different parts of the world. Proxies can take us into the early part of the 20th century. The proxy data overlap with modern measurements made over the past 150 or so years. There is continuity between these different data sets. You suggest looking at the PAGES 2k data on the climate audit site at, where you say “You can see that the origin of the global blade lies in the Arctic and in Australia”. But that’s not true. The only site that doesn’t show and uptick is Antarctica, and I have commented in my book “Earth’s Climate Evolution” that the authors made a mistake in not separating climate signals from East and West Antarctica, which lie under contrasting climate regimes. West Antarctica does show warming. An ice core from Siple Dome in the Ross Sea area, began to warm around 6 Ka ago, in parallel with the rise in both summer and winter insolation at 60°S and the gentle global rise in CO2. That warming steepened after around 1750 AD, following the curve of rising CO2. Over the past 1,200 years the concentrations of calcium and sea salt increased at Siple Dome, telling us that the westerly winds were getting stronger there, bringing in more dust from surrounding continents and warm air from the open ocean. In contrast, in East Antarctica – represented by an ice core from Law Dome – there was a cooling trend from around 1300 AD to the present (see Mayewski, P.A., Meredith, M.P., Summerhayes, C.P., et al. (2009), State of the Antarctic and Southern Ocean climate system, Rev. Geophys. 47, RG1003, doi:10.1029/2007RG000231. 38 pp).

          • clivebest says:

            The point is simply this.

            If the climate regularly varies by 1-2C such as between the MAWP and the Little Ice Age, then 20th century warming would not seem excessive.

            Has Paleo-climatology proved Hubert Lamb wrong?

          • polarscientist says:

            The climatic range between the LIA and the MWP was about 0.6 deg C (look at Euan’s copies above of the PAGES 2k graph Figure 4a). Lamb’s key diagram regarding the MWP and LIA was based on the Central England database, and it can be taken as broadly representative of westernmost Europe, but not much else. It was not, contrary to popular assumption, a global curve. The PAGES 2k data suggests that temperatures now may have risen to about where they were in AD400. As you say, that is not excessive.

  10. I have two spreadsheets that I started to put together some years ago and have been adding to ever since. Together they contain over 100MB of data and they include (or at least used to) just about every solar reconstruction and ice core/DSDP record in existence. I’ve been through all of them, getting down in some cases to absurd levels of detail, such as plotting temperature correlation coefficients against obliquity. But never was I able to come up with anything that clearly demonstrated a link between solar actitivity and climate.

    And to the best of my knowledge neither has anyone else.

    Take, for example, your graphs. Only two of them compare a solar variable against a climate variable. Your Figure 5 shows a match between the Bond drift ice index and GISP2 10Be that’s intriguing but not compelling, and your figure 6 shows a match between GISP2 d18O and 10Be which is frankly too good to be true. Most likely it’s a precipitation-related depositional effect (if you remember the spikes line up almost exactly with spikes in Ca, Na, K, Mg etc, which makes it difficult to relate them to solar activity or temperature). So there really isn’t any good evidence there for a solar-climate relationship either. And as you note elsewhere the variables are solar and climate proxies, not direct indicators.

    And neither is there any good evidence for a link between temperature and CO2 during the Holocene, unless of course it’s backwards. The strongest evidence against the CO2-dunnit hypothesis is that temperature leads CO2, as you have pointed out on a number of occasions. Polar Scientist’s counterargument that the increased CO2 amplified the warming is in my view ad-hoc.

    But you hit the nail on the head when you said at the end of your last comment “the temperature history is not robust”. I have concluded that we don’t have enough data to make meaningful quantitative estimates of global surface air temperature much before about 1920 or of SSTs much before about 1950. In short, we really have very little idea what global temps in the Holocene did, and until we do – which may be never – you (and I) will continue to believe that solar was the dominant influence on climate in the Holocene and PS will continue to believe it was CO2. No one will succeed in changing anyone’s mind because the hard evidence necessary to do so just isn’t there.

    • Euan Mearns says:

      A reply from Polar Scientist

      A sophisticated statistical analyses of Bond’s Holocene data by a French group in 2007 (Debret, M., Bout-Roumazeilles, V., Grousset, F., et al. (2007) The origin of the 1500-year climate cycles in Holocene North Atlantic records, Clim. Past 3, 569-575.) suggested that Bond’s millennial cycles were – as Bond had first thought – most probably caused by internal oceanic oscillations. They did find evidence in Bond’s data for solar cycles, but with periodicities of 1,000 (the Eddy Cycle) and 2,500 years (the Hallstatt Cycle), not 1,500 years.

      I would like to correct your statement that “PS will continue to believe it was CO2”. I do NOT argue for a prominent role of CO2 during the Holocene. I argue that the predominate driver of Holocene climate was orbital change, modulated by solar change.

      I do not agree with you “that we don’t have enough data to make meaningful quantitative estimates of global surface air temperature much before about 1920”. There are abundant palaeoclimate studies from a variety of sources indicating the likely change in temperature over Holocene time.


    • Euan Mearns says:

      Roger, I’m with you and against you 😉 The first point is that in my post I say that Bond cycles perhaps leave a small imprint on global average temperature. My thinking here was originally guided by trying to reconcile the original Mann hockey stick with everything else we knew. I believe that Bond / D-O cycles represent changes in the pattern of ocean / atmosphere circulation. As a result the climate may change everywhere on Earth but with little change to global average temperature. I can accept the ±0.2˚C implicit in the Pages2K scenarios. I’m sitting in my open plan kitchen living area with the wood burner on, its less than 6˚C out. My thermometer reads 20˚C. It is scaled from -30 to +50˚C. If I squint up it reads 21˚C and if I squint down it reads 19˚C. So de we really believe that we can measure mean atmosphere temperature 1000 years ago with a precision of ±0.2˚C?

      So that’s where I’m with you. I think your comments on 10Be are a bit off. In one case you think the relationship is not that convincing. In the other its too good. In fact in my Figure 6 the relationship is not that good at all since there are many more 10Be cycles than there are D-O events. All I’ve done is to correct 10Be concentration for precipitation rate – what more can you do? The link between climate cycles, solar geomagnetism and cosmic rays is reinforced by d14C data which is more robust as a ratio of C/C as opposed to 10Be that is an absolute abundance. Javier’s comment I believe emphasises this point.

      So you believe and I believe that The Sun plays a more prominent role than currently bestowed by IPCC. My theory is that the truth lies in dSun, dSpectrum, dOzone, dCirculation pattern and dConvection.


  11. Javier says:


    Thank you for an interesting article. I have also looked at the issue of Bond events with certain detail, and I’d like to share my point of view.

    Bond events register increased debris transported by icebergs, and we know from ship records that icebergs were a lot more common during colder periods, like the early 20th century when the Titanic sunk.

    We also know of several very cold episodes during the Holocene, like the 8.2 Kyr event that is registered in every temperature proxy analyzed, and even shows up in the methane record from ice cores. Or the 5.2 kyr cooling event that has left numerous evidence all over the world, from Ötzi, who was frozen in the Alps only to be released a few years ago to some of the narrowest rings in the oak series from Ireland and Lancaster. These known strong cooling periods coincide with peaks in the Bond series.

    We can therefore conclude that the Bond series is a record of cold periods during the Holocene. No more, no less.

    However the Bond series doesn’t say anything about the causes of the cold periods reflected in the peaks in ice rafted debris, and a lot of confusion has come from people trying to assign the same cause to all the peaks, when we know very clearly that different factors affect temperatures.

    Gerard Bond was prejudiced, and it shows in the way he numbered the peaks in the series. One would naively assign the numbers as you did in your figure 3, although it is clear that peaks 1 to 5 are made from 2 to 4 sub-peaks. Calling number 3 a peak is already a mistake since its central date at 3 Kyr BP (before present) has a lower detrital level than about 85% of the series. However in a stretch of self deception Gerard Bond assigned only eight numbers (your figure 2), because he wanted to manipulate the average spacing of his series. 12000/8=1500 et voilà the magical spacing of the D-O series. If we count as you did 10 we only get 1200 years and if we even count double peaks as independent coolings, as they probably are, we get a cooling period for every millennia. That is the real frequency of strong cooling periods during the Holocene.

    Already the foolishness of trying to match the frequency of the D-O series to the Bond series has made it to the literature. This figure, a composite from Debret et al., 2007,
    shows clearly that prior to 7 Kyr BP the Bond series fits very well a 1000 yr filter, while afterwards it fits somewhat not so well a 1500 yr filter.

    We could leave it at that as we have already learned the fundamentals of the Bond series. It is a record of cold periods. It doesn’t tell us anything about the causes of the cold periods that are probably diverse with each peak having its own single or combination of causes. And it doesn’t have a regular periodicity but shows a clear shift of regime after the melting of the ice sheets around the Mid-Holocene transition.

    If we want to proceed further into the causes of the Bond peaks we must abandon the canon of paleoclimatology that states that solar variability is too small to cause significant changes in climatology. The available evidence shows otherwise, but scientists that dedicate their careers to study the influence of solar variability on climate change for decades have damaged their careers and reputation.

    I am going to use the following numbering of Bond peaks:
    8 – 11.2 Kyr BP
    7 – 10.3 Kyr BP
    6 – 9.3 Kyr BP
    5b – 8.3 Kyr BP
    5a – 7.3 Kyr BP
    4b – 6.2 Kyr BP
    4a – 5.3 Kyr BP
    3b – 4.5 Kyr BP
    3a – 4.0 Kyr BP
    2b – 3.2 Kyr BP
    2a – 2.8 Kyr BP
    1b – 1.5 Kyr BP
    1a – 1.2 Kyr BP
    0 – 0.4 Kyr BP

    If we accept that the detrended change in 14C during the Holocene represents mainly solar variability with little climatic contamination then we can see the strong correspondence that the 14C record displays with the Bond series. The 14C record shows certain periodicities that correspond with the Bond record. The main periodicity is a ~2500 years solar cycle that occurs during the entire Holocene. There is also a ~1000 years solar cycle that is prominent from 11.5 Kyr BP to about 7 Kyr BP and then weakens to the point of almost disappearing until it becomes very strong again at about 1.5 Kyr BP. We can see this behavior of the ~1000 yr solar cycle in the wavelet analysis of the following figure from Kern et al., 2012:
    You can also see very clearly the ~2500 years solar cycle band in that analysis.

    It is important to note that there is no ~1500 years solar cycle. The ~1500 years periodicity in climate has no solar origin. We have already enough information to understand the ~1500 years cycle, because as Sir Arthur Conan Doyle said, when the impossible is eliminated, what remains, however improbable, must be the truth.

    We can now match the 14C record to the Bond record (SGM stands for Solar Grand Minimum):

    8 – 11.2 Kyr BP – Preboreal SGM – ~1000 yr solar cycle low
    7 – 10.3 Kyr BP – Boreal 1 SGM – ~2500 and ~1000 yr solar cycles lows
    6 – 9.3 Kyr BP – Boreal 2 SGM – ~1000 yr solar cycle low
    5b – 8.3 Kyr BP – Sahel 1, 2, 3 SGM + Lake Agassiz – ~1000 yr solar cycle low and glacial lake outburst
    5a – 7.3 Kyr BP – Jericho 1, 2, 3 SGM – ~2500 and ~1000 yr solar cycles lows
    4b – 6.2 Kyr BP – Two unnamed SGM – ~1000 yr solar cycle low
    4a – 5.3 Kyr BP – Sumer 1, 2, 3 SGM – ~2500 yr solar cycle
    3b – 4.5 Kyr BP – Not solar – ~1500 year oceanic cycle low
    3a – 4.0 Kyr BP – unassigned
    2b – 3.2 Kyr BP – Not solar – ~1500 year oceanic cycle low
    2a – 2.8 Kyr BP – Homer SGM – ~2500 yr solar cycle low
    1b – 1.5 Kyr BP – Not solar – ~1500 year oceanic cycle low
    1a – 1.2 Kyr BP – Roman SGM – ~1000 yr solar cycle low
    0 – 0.4 Kyr BP – Wolf, Spører, Maunder SGM – ~2500 and ~1000 yr solar cycles and ~1500 yr oceanic cycle lows

    Nearly all big solar grand minima that had received a name coincide with peaks in the Bond series. The exceptions are Noach at 4.8 Kyr BP, Greek at 2.3 Kyr BP, and Oort at 0.9 Kyr BP. However two of the three are within he dating periods of geomagnetic excursions that can cause an increase in 14C production from cosmic rays without corresponding decrease in solar activity. The 4.8 Kyr BP increase might correspond to a geomagnetic excursion detected in China (Zhu et al., 1998), while the 2.3 Kyr BP increase could correspond to the Sterno-Etrussia geomagnetic excursion that some records date between 2.2 and 2.4 Kyr BP (Raspopov et al., 2003).

    The link between cold periods during the Holocene and periods of increased production of 14C is extraordinarily strong. Why then is not accepted by science that low solar activity can produce a global cooling of 0.3-0.5°K as required by evidence instead of the 0.1°K currently calculated by solar physicists as the upper value? The answer is because we do not know of any mechanism that could do it. The decrease in Total Solar Irradiation is too small. The decrease in UV is larger, but UV radiation is a very small part of total radiation and the change takes place at the top of the atmosphere. We do not know of any mechanism by which changes in the Sun’s magnetism or solar wind could affect climate. Changes in gravitation from the Solar system baricenter and/or changes to the Earth’s rotation are not even considered by serious scientists. And Svensmark’s theory of clouds being affected by cosmic rays has a very serious problem with changes in the Earth’s geomagnetism not affecting climate while they largely determine the amount of cosmic radiation that arrives to the planet.

    But whenever there is such a flagrant contradiction between theory and evidence, best practices recommend going with the evidence and assume that the theory is either incomplete or wrong. When Alfred Wegener argued that the evidence supported that the continents had separated from a joint position he was ridiculed by geologists for a lack of any mechanism that could cause a continental drift despite the support from paleontological evidence and continental geometry. Wegener died discredited but in the end theory was wrong and evidence, however absurd, was right.

    Despite the lack of any known amplification mechanism, the evidence is very clear that periods of very high 14C production are associated with periods of strong cooling during the Holocene. Since not all the cold periods are associated with high 14C production it is difficult to argue that the cold is causing an increase in 14C production. The most likely explanation is that only when the increase in 14C production is due to a reduction in solar activity a severe cooling takes place. We lack a proper understanding of the amplification mechanism because modern science has taken place during a period of solar stability.

    Perhaps the period of reduced solar activity that started with solar cycle 23 and that is likely to continue at least until cycle 25 ends will give us a better understanding of the real effect of solar variability on climate change. I would not be very surprised if global warming stalls until the level of average solar activity doesn’t return to its values during the second half of the 20th century.

    Debret M., et al., 2007. The origin of the 1500-year climate cycles in Holocene North-Atlantic records. Clim. Past Discuss., 3, 679–692.

    Kern, A.K. 2012. Strong evidence for the influence of solar cycles on a Late Miocene lake system revealed by biotic and abiotic proxies. Palaeo. Vol. 329–330. pp. 124–136.

    Zhu, R. X., et al., 1998. “Sedimentary record of two geomagnetic excursions within the last 15,000 years in Beijing, China.” Journal of Geophysical Research-All Series- 103: 30-323.

    Raspopov, O.M., et al., 2003. “Ezekiel’s vision: Visual evidence of Sterno‐Etrussia geomagnetic excursion?.” Eos, Transactions American Geophysical Union 84.9: 77-83.

    • Euan Mearns says:

      Javier, many thanks for this very thoughtful comment. I need to plot some charts, but not tonight. You make a couple of crucial points here:

      The link between cold periods during the Holocene and periods of increased production of 14C is extraordinarily strong. Why then is not accepted by science that low solar activity can produce a global cooling of 0.3-0.5°K as required by evidence instead of the 0.1°K currently calculated by solar physicists as the upper value?

      Climate science doggedly sticks to TSI even though it is known that changes in TSI cannot explain the observations. Climate Science is locked into radiative forcing despite the fact that convection is the dominant process.

      When Alfred Wegener argued that the evidence supported that the continents had separated from a joint position he was ridiculed by geologists for a lack of any mechanism that could cause a continental drift despite the support from paleontological evidence and continental geometry. Wegener died discredited but in the end theory was wrong and evidence, however absurd, was right.

      Earth Science (I prefer the term Geology) is empirically driven. Climatology is but a small fraction of all geology. And so we won’t work out what is really going on until it is observed. The winter of 2011 I believe gave us a taste. I believe different modes of ocean – atmosphere circulation are important and may lead to small changes in convection and water vapour that may clearly have a large effect.

      • polarscientist says:


        Clearly we need more data (well, we always need more data!) to reduce uncertainty, but just because you BELIEVE something to be the case does not make it so. You have to cite plausible (and testable) scientific reasons for thinking something may be the case. It goes without saying that ocean and atmospheric circulation ar important diistributirs of heat, but they are

      • polarscientist says:


        By the way, my comment below (at 1018) related to Javier’s analysis of Bond cycles.

        Turning to your latest post, above (0913, April 13), you cite your BELIEF that ocean and atmospheric circulation are important and may have a large effect. But in science we don’t use the word belief. You must explain in what way ocean and atmospheric circulation changes have worked to create in the Holocene the changes we have seen. Where and what is the evidence that they played a key role IN THE ABSENCE OF EXTERNAL FORCING? We do indeed have several natural oscillations in the climate system that seem to have no external driver (e.g. El Nino events, the Pacific Decadal Oscillation, the North Atlantic Oscillation). But these do not explain the larger fluctuations we see in the Holocene. There has to be some external driver to force change in atmospheric and ocean circulation of the magnitude we see in the Holocene. As Javier points out (and I agree – see my reply below) variations in solar output are likely to be important drivers, combined, of course, with the added insolation effect of declining insolation.

      • Javier says:

        Climate science doggedly sticks to TSI even though it is known that changes in TSI cannot explain the observations.

        Solar physicists look at the Sun and don’t find there the type of variability necessary to explain climate changes. We look to the Earth and find a correlating variation between cosmogenic isotopes and climate change during cold periods. It is impossible to reach a consensus with the available evidence.

        I am privileged of having a conversation at WUWT with Dr. Leif Svalgaard, one of the foremost solar physicists of our time. Over this conversation many of the arguments together with supporting bibliography are exposed. Leif is at times a little dismissive, but he has had to contend for years with the attacks of the sometimes rowdy solar crowd, so it is very kind of him that he still answers the same arguments.

        If you are interested, the conversation starts here:

        • Euan Mearns says:

          Javier, you seem to have one foot in each camp. I’m interested to know how you explain the observed warming of the last 150 years?

          • Javier says:


            I am evidence driven. For as long as the evidence indicates that the increase in 14C production that took place during the LIA was due to an increase in the arrival of galactic cosmic rays due to a weakening/change in the heliospheric magnetic field I will believe that reduced solar activity was the main cause behind the LIA, with increased volcanic activity and oceanic circulation changes as contributing factors.

            If we take a naive view to temperature reconstructions for the past 100 Kyr we see that during glacial periods there are no cooling events, just warming events (D-O events) followed by a slow recovery of cold conditions. And when we look at the past 9000 years we see the opposite. There are no warming events during the Holocene, just cooling events (8.2, 5.5, 2.8 Kyr events, LIA) followed by a slow recovery of warm conditions. This is only logical as they are opposite extreme states and variability is towards the mean.

            As Dr. Summerhayes points the temperature is set by the orbital configuration so it is a decaying line during most of the Holocene, not a straight line as most people imagine.

            So if we find ourselves in a warming period, we have to ask what happened before. If there was no previous cooling, now that would be unprecedented as there are no warming events during the Holocene. But we find ourselves just before the second strongest cooling event of the Holocene that according to all evidence (mainly glacierology) included the coldest centuries of the entire Holocene. So we are clearly in a recovery period, and the deepest the cooling, the highest the warming as we might even see some of a rebound effect before we equilibrate to present orbital target temperature.

            So imagine that you go to the kitchen and you find that the soup in a pot that has very thick metal base and walls (high inertia) is not warm enough, so you raise the fire a little bit and put the lid on. Your apprentice Sciency comes after you and starts taking measures of the soup temperature and the vapor content and temperature inside the pot. He does this each 15 minutes and after half an hour reaches the following conclusion:
            “The fire has not increased during the last half hour yet the temperature of the soup is rising, thus it must not be the fire since it is constant. However the vapor content inside the pot has increased, and also the vapor temperature, so it is clear that the reduction of the exchange between the soup surface and the kitchen is responsible for the warming of the soup.”
            The mistake is in assuming equilibrium conditions at the start of the measurements, and ignoring the system thermal inertia.

            Solar activity went back to normal around 1725 and except for two periods around 1825 and 1900 has been normal since. Temperatures have been on the rise since 1650 except for those two periods.

            Solar influence on global warming has been underestimated by scientists because we don’t know how solar variability affects climate despite the strong evidence that it correlates very well for thousands of years. CO2 influence on global warming (the lid) has been overestimated by scientists.

            Global warming is now about 350 years old, and according to solar cycles it should be running out of steam, as solar activity cannot go any higher. If CO2 has been overestimated, global warming will not get very far during the 21st century and depending on the real value of CO2 climate sensitivity we could even see a 0.1-0.2°K cooling over the next decades. Maximal temperatures should be reached between 2015 and 2090, and afterwards solar cycle evolution and orbital settings will make sure the long descent into the next glacial period is resumed.

    • polarscientist says:

      Now that is a comprehensive and credible analysis! In my book “Earth’s Climate Evolution”, I concluded that “it seems highly likely that most of the short cool periods of the Holocene represent periods of low amplitude in century-scale solar cycles. These cool periods occurred irregularly within a cycle having a broad envelope of 1-2 Ka, much as Bond had suggested in moving away from a precisely defined 1,500-year signal”.

  12. Euan Mearns says:

    This is how I see Pages2K.

    The grey lines are basically following Ljungqvist. Moberg prior to 600 AD needs to be ignored unless we are to believe that the Roman Warm Period was as cold as the Little Ice Age. The data back then seem based mainly on Greenland.

    Its not obvious to me how this gets turned into a hockey stick.

    • polarscientist says:


      As I mentioned in an earlier post, the 4 coloured lines at the top of PAGES 2k’s Figure 4a (which you re-show above with your grey curves superimposed) are for the Northern Hemisphere, not for the globe. The data in Figure 4b, on the other hand, are the GLOBAL averages. And they do show an uptick. That uptick is also abundantly apparent in all of the scientific publications referred to below by Javier (April 14, 1.34 am). As he suggests, readers of this post should acquaint themselves with Tamino’s analysis at

      Anyway, it seems that Javier and I are on the same page regarding the Holocene temperature profile. I hope from the abundance of evidence in pour support that readers will finally accept that the data are telling us that there has been a real long term cooling, which in recent times has reversed.

      • Euan Mearns says:

        Colin, with respect to 4a and 4b, thanks for keeping me right. So the structure of N hemisphere temperatures appears to be dominated by Bond cycles. Which means the S hemisphere must show quite a lot of cooling to produce the global picture shown as (b). Is there a source that shows the S Hemisphere plotted the same way as (a)?

        I have just recently looked at Epica for the first time. It shows an Epic hockey stick 😉 Its just that the temp spike is dated as 1895.

        I was also surprised to see that the Holocene warming in Antarctica pre-dates warming in Greenland by about 2000 years? Is that correct? Or have I got something screwed up here?

        • polarscientist says:

          Regarding the Southern Hemisphere Marcott showed three temperature curves – one each for the Northern Hemisphere, the Tropics, and the Southern Hemisphere. Regarding the difference between the SH and the NH at the beginning of the Holocene, that’s because in t he NH we had the Younger Dryas cold event (possibly a sort of Dansgaard-Oeschger or Heinrich Event) immediately before the Holocene, and there was no Younger Dryas event in the SH. Regarding temperatures in Antarctica, the EPICA cores are not high resolution. Those from West Antarctica show 20th century warming – see Mulvaney, R., Abram, N.J., Hindmarsh, R.C.A., et al. (2012) Recent Antarctic Peninsula warming relative to Holocene climate and ice-shelf history, Nature 489, 141-145; and Abram et al 2013, Acceleration of snow melt in an Antarctic
          Peninsula ice core during the twentieth century. Nature Geoscience. Note that this pattern does not characterize James Ross Island, which lies in the Weddell Sea side of the Peninsula and is more like East Antarctica.

          • Euan Mearns says:

            Colin, I’ve been walking my dogs and recalled the Younger Dryas while doing so. Believe me, I’m trying to take on board all this data and what you are saying but remain a long way from being convinced. For example, conceptually, I find it hard to accept that the N hemisphere gets cast into Younger Dryas full on glaciation and that this leaves no mark on the isotope record in Antarctica. I’m not questioning that it was so.

            And in the Yair paper I find it conceptually hard to accept that the temperature histories of the N and S oceans are quite different, especially since the oceans mix quickly on a time scale of the order 1000 years.


          • polarscientist says:

            The YD is merely the latest in the long line of D-O events in Greenland. Cold events in the north were preceded by warm ones in the south. The connection was through the ocean.

            You would expect the northern and southern SSTs to differ because the insolation differs (see PAGES 2k Fig 4h).

            Read my book – it’s all in there!

          • Euan Mearns says:


            Javier posted link. I guess I should say Rosenthal.

  13. Javier says:


    And you say the LIA was the coldest for the last 10,000 years. Do you have evidence for this other than Marcott (which has a torpedo heading for it in my next post). The Bond data are actually more consistent with progressive warming, as is the LR04 benthic foram stack.

    The temperature profile of the planet during the Holocene I would say it is pretty well known, and the only big point of contention is the magnitude of the temperature changes.

    The global temperature maximum was probably reached around 8000 BP, except in northern high latitudes where it was delayed until the last remnants of the ice sheets melted, and was maintained until 6000-5500 BP when the Holocene climatic optimum ends. The Holocene highstand, the highest sea level, was reached at most locations between 6500-4500 BP, since then sea level has been falling until modern rise.

    A significant cooling took place between 5500-5200 BP giving rise to the Neoglacial period, and since then temperatures have been dropping on average. The second millennium AD has been the coldest millennium in the entire Holocene. And the third millennium is likely to be even colder despite its auspicious start.

    I will offer some bibliography on this:

    Global glaciers are a good marker of Holocene cooling since 5500 BP, as they have progressively extended more.
    A) Koch & Clague 2006 meta-study of global glacier extent showing the progressive increase in glacier extent during the second half of the Holocene.
    B) Thompson et al., 1995 study of the Huascarán glacier. Ice-core of the glacier with the temperature proxy showing that glacier temperature has been dropping in the second half of the Holocene.

    Sea surface temperatures during the Holocene have been beautifully analyzed in the Pacific Ocean during the Holocene by Rosenthal et al. 2013, showing a very similar picture.

    Holocene sea levels are analyzed in Woodroffe & Horton 2005.

    Temperature proxies during the Holocene are dissected by Ljungqvist 2011 both for the Northern and Southern hemispheres. This is an important paper to estimate temperature changes during the Holocene.

    A final comment on Marcott et al 2013 reconstruction of Holocene temperatures. It is full of problems as Steve McIntyre showed, and appears politically motivated which is a sin in science, however some of the most egregious problems can be fixed as Tamino did here:
    In the seventh graph there, labelled “Differencing Method”, the black curve corresponding to published ages, is probably the best global temperature reconstruction of the Holocene available. It still has issues and could be done better, but nobody has done it yet, and global reconstructions are very rare.

    Koch, J., & Clague, J.J. 2006. Are insolation and sunspot activity the primary drivers of Holocene glacier fluctuations? PAGES News, Vol. 14 No 3 pp 20-21.

    Ljungqvist, F. C. (2011). The spatio-temporal pattern of the mid-Holocene thermal maximum. Geografie–Sborník ČGS, 116(2), 91-110.

    Rosenthal, Y., et al. 2013. “Pacific ocean heat content during the past 10,000 years.” Science 342.6158: 617-621.

    Thompson, L.G. et al. 1995. Late Glacial Stage and Holocene Tropical Ice Core Records from Huascaran, Peru. Science vol. 269, 46-50.

    Woodroffe, S. A., & Horton, B. P. (2005). Holocene sea-level changes in the Indo-Pacific. Journal of Asian Earth Sciences, 25(1), 29-43.

    • Euan Mearns says:

      Javier, thanks for all the links. I’m finding this to be highly instructive and I’m sure a lot of other readers will too. I’m also approaching information over load. The graphic is from the Yair paper:

      I am a bit colour blind and have a question about the top left panel where it seems that the green band dives down before jumping up? (that is for 30 to 90 N).

      We also see that the N and S hemispheres are behaving very differently (Green top left and blue top right). How is this explained?

      • polarscientist says:

        what is “the Yair paper”?

      • Javier says:


        Yes I understand that these are complex issues with a lot of evidence sometimes contradictory to ponder.

        That figure from Rosenthal et al., 2013 is a torpedo to several aspects of the Marcott et al., 2013 reconstruction.

        As you keenly observe, the dive in their northern hemisphere reconstruction between 1600 and 1850 is contrary to every other reconstruction and an artifact as McIntyre demonstrated. It is countered by an equally absurd and unjustified rise in the southern hemisphere reconstruction (the blue upward spike) that leaves a more palatable global reconstruction. Both northern (green at A) downward spike and southern (blue, B) upward spike take place before modern warming took off.

        This figure here illustrates the difference in temperature evolution between both poles due to the anti symmetrical nature of precession changes (ignore the instrumental curve that should not be there).

        But Antarctica is not representative of the southern hemisphere. In their paper, Rosenthal et al., 2013, are looking at tropical waters south of the Equator and they find the profile shown in panels C and D. there are several important conclusions:

        – There is no way that surface temperatures could show only a cooling of 0.6°K from the Holocene climatic optimum to the bottom of LIA as Marcott et al., 2013 claim, when tropical waters are showing a drop of 1.5-3.0°K both at surface (SST) and Intermediate (IWT) levels. If the sea surface was cooling so much and even more in the northern hemisphere, so was the land.

        – Perhaps (or not) land temperatures are higher now that during MWP, but ocean temperatures are clearly not, as figure three and Oppo et al., 2009 show.

        Peter Clark’s group at Oregon State (Marcott et al., 2013; Shakun et al., 2014) know very well how to do things, but instead they have chosen to do political science and push claims like current temperatures being well above Holocene climatic optimum that are not supported by evidence, including their own evidence. No doubt he is increasing his citation index and pushing forward his career at the expense of supporting false claims and becoming irrelevant in the future.

        Oppo, D. W., Rosenthal, Y., & Linsley, B. K. (2009). 2,000-year-long temperature and hydrology reconstructions from the Indo-Pacific warm pool. Nature, 460(7259), 1113-1116.

  14. polarscientist says:

    Owen, The rise in temperature from about 1900 to about 1945 was about 0.5 deg C, not 1.0 deg C as you claim. There was then a drop of about 0.2 deg C to 1950, after which temperatures stayed fairly flat until 1970. They then started their climb to where we found them to be in 2015, about 1.3 deg C above where they were in 1910. These widely available figures from reputable sources in NOAA, NASA and the UK Met Office disagree entirely with your conception.

    • Owen says:

      I have some 1st hand sources from late 40s and early 50s that show a 1C rise and sources from 70s that show 82% of scientists actually thought we were entering long period of global cooling. I do not know how to post pics here unfortunately. I also use Valentia temp records which is as reliable record as one can get. It speaks for itself, no correlation with CO2 emissions :

      • polarscientist says:


        You can send figures to Euan and he will post them.

        I know that during the late 60s and early 70s some scientists did think that the world might be cooling, but my sources say it was around 10%, not 82%.

        With regard to temperature data, yes there have been some adjustments. They were to compensate for things like the necessary move of a weather station from an area that had become concreted to a nearby grassy area like the one when the station was first established, or for the differences at sea between engine intake temperatures and bucket temperatures.

        Don’t forget, when evaluating whether or not there has been warming, that there are a multiplicity of other measures indicative of warming – like the melting of glaciers, the disappearance of sea ice, the advance of spring time, the rise in sea level (which is measured by two entirely different methods – satellite altimetry and coastal tide gauges). Then of course there is the fact that land temperatures show much the same increase as ocean temperatures.

        The global temperature records come from highly reliable sources – the US NOAA, the US NASA, and the UK Met Office. I doubt very much that these three different agencies are conspiring with one another to deceive the general public.

        • A C Osborn says:

          I have not often seen so much misplaced trust in one person.
          Especially for the UK Met Office who change their Climate” stories” regularly, continually contradicting themselves and each other.
          As for the 1.3 degrees C since 1910, 0.8 of that is from Adjustments.

          • Euan Mearns says:

            This a response to AC Osborne from Polar Scientist

            It is instructive to consider the case of New Zealand, where those who believed that the government agency responsible for weather records had adjusted the data to suit a concept, lost their case in court of law recently. I would like to see all such cases brought to the law courts, where an impartial body could evaluate the competing claims. I suspect that the result would be the same as it was in NZ.

            The problem with independent investigators pointing to adjustments to the temperature records is that they do not have the back-story. They usually do not know the grounds that the responsible agency had for making the adjustments. Under those circumstances it is all too easy to cry foul – and to imagine that there is some sort of conspiracy going on. But do you seriously imagine that government agencies all around the world, are all working to some grand global plan to defraud their own publics? What on earth would be the point? How could such a global-scale con work? Someone from within would be sure to blow the whistle. Anyway, surely some bright spark would have found them out by now, and clearly that did not work in NZ.

            Also, let’s not forget that a group of global warming skeptics led by Richard Muller decided to test all the data purporting to show global warming, and found to their surprise that it was indeed occurring and the data adjustments were genuine (i.e. necessary). Theirs was the Berkeley Earth Surface Temperature Project, and the results were published in the New York Times of July 28, 2012 (

            Finally, as I pointed out in my post to Owen (above), it’s not just about the temperature record. Don’t forget, when evaluating whether or not there has been warming, that there are a multiplicity of other measures indicative of significant warming – like the melting of glaciers, the disappearance of sea ice, the advance of spring time, the rise in sea level (which is measured by two entirely different methods – satellite altimetry and coastal tide gauges). Then of course there is the fact that land temperatures show much the same increase as ocean temperatures despite being measured by different means. These independent facts make it hard to dismiss the evidence that the climate has indeed warmed significantly since 1900. And if that evidence (all of it from ocean, atmosphere, ice, lakes, tree rings, stalactites, corals and plant movements etc) is correct then we have to seek for probable cause. The Sun is in decline. CO2 and other greenhouse gases are on the rise. I rest my case.

            Kind regards

          • Javier says:

            Dr. Summerhayes,

            We are clearly not talking about the same degrees professor Kelvin established when we talk about present global warming and Holocene temperatures.

            We are told global temperatures have increased above 1°K since instrumental records began. We are also told global temperatures declined 0.6°K from Holocene climatic optimum to modern average (zero anomaly).

            However we know that glaciers have retreated only to their level 5000 years ago but are still significantly larger than during Holocene climatic optimum. And we know from theory that glaciers are specially sensitive to CO2 increase because the air above glaciers is very dry and contains a lot less water vapor to compete with CO2 for radiation.

            So we probably have not retreated in temperatures to the level 5000 years ago, and we are somewhere between 5000-2000 years ago in terms of temperatures. That is what the ice thermometer says.

            And despite all the hype we lack the expected biological response. We know biology also responds quickly, within decades, to temperature changes. In fact changes in pollen subdivided the Holocene into preboreal, boreal, atlantic and subatlantic periods. The northern tree line has been receding for thousands of years, and shows few signs of a significant advance. Young trees grow in a few years if conditions are good, and seeds can travel several kilometers north every few years if global warming is taking place. Other animal species can push their ranges north even faster if conditions allow.

            Yet despite all the hype about how much the biosphere is being affected by global warming, we lack evidence that the biology has reclaimed the ground lost to cold since the Roman times, when the wine industry was prosperous in Britain.

            As I always follow the evidence, my conclusion is that if recently we have warmed above 1°K then the Holocene involves a cooling of about 3°K. This is more than what the proxies support and doesn’t compare well with the temperature changes that we believe separate glacial conditions from interglacials (~5°K). If we believe the Holocene involves a more reasonable cooling of about 1.5°K then we cannot have warmed more than 0.6°K in the instrumental era. The rest we must have added through adjustments.

            The ice and the biological species do not use thermometers, but they always get temperature changes right, and they are not trying to push any hypothesis.

            And if you think that scientists as a collective cannot get sidetracked and blinded to evidence to the point of rejecting all the evidence against a favored hypothesis and accepting flawed evidence supporting such favored hypothesis, you should set aside 10 minutes and read the following article:


          • polarscientist says:


            In focusing on the temperature record, perhaps you are missing the significance of the rise in sea level evident from salt marsh data, tide gauge data and satellite altimetry. These seem to support the notion that the world has been warming significantly since the late 1800s, with the rate of increase itself increasing in recent decades. The assumed contributions come about 1/3 each from increasing atmospheric heating, increasing melting of glaciers, and increasing loss from ice sheets.

          • A C Osborn says:

            First of all Richard Muller has at no time been AGW “Sceptic”, I suggest that youi look at some of his historical statements, ie before the Global Temp project.
            Second I have had many conversations with those in charge of these adjustments.
            The Berkeley Earthe people have actually admitted that if you want to know the Actual Historic Temperature you should look at the RAW data.
            If you want to know what the Temperature should have been based on modelling the data use their “Final Product”.
            Both Land & Water Temps have been adjusted by the same people.

            What advance of springtime exactly?
            Nothing unusual going on with sea level rise.
            Nothing unusual with Ice.
            Nothing unusual with Coral, other than bleached by “Cold” water in the last ten years or so.

            How can you rest your case when the Temperatures are not real?

            From you other comments I have to ask have ever bothered to read the Climategate emails?

            Have you looked at the Global weather during these so called Hotttest Ever Months and years?
            More Cold Records than ever before, Record Snow Depths, Snow for the first time ever in some places.
            Massive hailstones in Vietnam, the list goes on and on.
            You won’t find them in the western MSM though but they have all been happening, do they sound like a hottest ever world to you?

          • A C Osborn says:

            The other thing I find extremely odd is that you would accept the word of Scientists with a vested interest over that of all the Scientists, Engineers, workers and people since the Victorian age where the Temperature Record is concerned.
            Are you saying that those people were just plain wrong, incompetent, lying, having mass delusions?
            They measured temperatures which are completely backed up by anecdotal evidence and yet somehow they were wrong and “Climate Scientists” 100, 50, 20, 10 years later were right.
            Because that is what those adjustments mean, they have even adjusted up to last year, let alone 100 years ago.

            Did the Australians not die of heat stroke, along with millions of animals and birds in the late 1800s?
            Did the Americans not experience the 1930s highest ever temperatures and the Dust Bowl?
            Did both countries not have more Wild Fires then than now.
            Have Americans not experienced worse and more Hurricanes and Tornadoes in the past than now?

        • Owen says:

          Polar Scientist :
          There may be need to adjust temp records but all the adjustments Ive seen exaggerated warming. On the balance of probability there should be adjustments that reduce the warming. But we dont see any.

          The AMO doesnt show any net warming, just oscillating cold / warm periods which fit well with Valentia record.

          The image at bottom of this thread shows 23 out of 28 scientists in 1977 published papers that predicted the onset of long term global cooling i.e 82%.

          Large organisations can get things very wrong, the recent global financial collapse of 2008 proves that.

          The other image below shows that since 1945, there was a 1.5C drop in temperature to 1977 so bringing us back to post LIA temps, i.e. no net warming since then.

          • polarscientist says:

            There were a great many more than 28 scientists publishing on global climate in 1977. Indeed there was a significant climate science meeting in 1971 near Wijk, Sweden, organized by MIT with the heading “Study of Man’s Impact on Climate” (SMIC), and designed to look into “The implications of inadvertent climate modifications”. Results were published as SMIC, 1971, Report of the Study of Man’s Impact on Climate (SMIC). MIT Press, Cambridge, Massachussetts. 308pp. It involved 30 scientists from 14 countries, and concluded that increasing emissions of CO2 and related greenhouse gases would warm the climate significantly. As early as 1965, one of President Lyndon B Johnson’s Science Advisory Committees warned that burning fossil fuels would lead to a rise in CO2 of 25% by the year 2000, which might be expected “to produce measurable and perhaps marked changes in climate…[that could be]… deleterious from the point of view of human beings”. That bit was written by Roger Revelle, aided by Wally Broecker of Lamont, and Charles Keeling among others. The actual rise has been close to 40%. That opinion was reiterated in the Club of Rome’s report ‘Limits to Growth’, in 1972, which forecast a rise to 380 ppm by the year 2000. These reports by groups of experts in climate are far more significant than a bankers arbitrary survey of 28 scholars, who may or may not have known what they were talking about (indeed, they may have merely been expressing opinions – without knowing who they were or what their papers said, we have no way of knowing).

          • Owen says:

            Polar S,
            either the world cooled in the 60s and 70s or it didnt, it cant be both (unless of course regional variation). I have a lot of facts in my possession which says it cooled, and quite significantly too.

            So CO2 cant be the main driver of climate

  15. Andy Dawson says:


    lacking any other way to contact you, I’ve written a piece on feasibility of a (near) all-nuclear/pumped storage UK grid which I think might be suitable for “Energy Matters”.

    It’s available here:

    Is it of any interest?

  16. Euan Mearns says:

    A couple of images from Owen dated 1977.

    • Euan Mearns says:

      Referring to my Figure 9, I’d note that 1977 fell just past the solar minimum between sunspot cycles 20 and 21. We then headed into 3 very active cycles that included the grand solar maximum (Lockwood and Frolich which I don’t have to hand).

      • Owen says:

        In theory, Polar Scientist seems to be right but the point is that in practice, in reality it is not how the planet’s climate works.

        Its a bit like Einstein, in theory the universe should not have been expanding, that would made things very neat, but in reality the universe was expanding. Nature doesnt really care what we think in theory should happen.

      • Javier says:


        One thing. The 20th century grand solar maximum no longer exists. I know it is in a lot of papers, but apparently there was a change in the way sunspots were counted after Wolf retired and the numbers increased. Since 2011 there were annual workshops were all the experts reviewed all the data. They could have lowered the modern numbers to match Wolf’s numbers, but since they are used by a lot of weather software and programs they decided to raise the old numbers in July 2015. The main effect was that the 20th century grand solar maximum disappeared. That was bad news for a lot of old solar theories that relied on that maximum, but the evidence seems pretty strong and most if not all expert agree on the change.

        Make sure that you use the new graphs and data. You can get them at SILSO (Sunspot Index and Long-term Solar Observations).

  17. Javier says:


    I don’t think I am missing anything relevant about sea level rise. Old sea level data is very subject to interpretation. What is to me the best data available indicates that modern sea level rise started around 1700, together with global warming and has accelerated every century since. See the following figure from Grinsted et al., 2009:
    Notice that it shows sea levels below those during MWP.

    Sea level rise has been essentially linear for the entire 20th century. The claim that it has accelerated during the latest decades is not supported by satellite altimetry data, that shows a constant increase from the entire period 1993-2016:
    See graph at the bottom of the Laboratory for Satellite Altimetry/Sea Level Rise webpage:

    When one examines the claim that sea level rise is accelerating, it turns out the entire acceleration is coming from the Global Isostatic Adjustment (GIA) that appears contaminated by large sea bottom adjustments.

    A recent acceleration of sea level rise would not make sense, because the Argo system is reporting that the oceans are not warming much since 2003 and the steric contribution is believed to be increasing equally little:

    Regarding the assumed contributions to sea level rise, that is not science but fantasy. Zwally’s paper on Antarctica’s ice increase, if correct indicates that the +0.27 millimeters that Antarctica was suppose to contribute would actually be a -0.23 millimeters. An accounting error of 0.5 mm over a total amount close to 3 mm.

    Grinsted, A., et al. (2010). Reconstructing sea level from paleo and projected temperatures 200 to 2100 AD. Climate Dynamics, 34(4), 461-472.

    Zwally, H.J., et al. “Mass gains of the Antarctic ice sheet exceed losses.” Journal of Glaciology 61.230 (2015): 1019-1036.

    • polarscientist says:

      There is a question mark over Zwally’s results. See Jonathan Bamber at

      • Javier says:

        I don’t think Bamber’s criticism is valid.

        Bamber criticizes Zwally for being alone in defending Antarctic growth, not a scientifically valid point; he is either right or wrong and scientific democracy has nothing to do with that. He then criticizes the method because although altimetry is very precise, estimation of snow densities is not and defends that changes in firning could invalidate the conclusions although he does not have any data that actually contradicts Zwally’s model. Then he criticizes the measurement of altimetry from satellites and he is clearly wrong on that because with those techniques me measure the distance to the moon with a precision of a few millimeters. Then he goes to defend the satellite measurement of gravimetric anomalies as his preferred choice, a technique clearly prone to a lot more error than altimetry in the interpretation of the data and that has been used to measure mostly changes in liquid bodies, that experiment local changes much larger that what can be expected from snow deposition from year to year. That is seeing the mote in the eye of one technique and not seen the beam in the other technique.

        Have you read Thomas et al., 2015 (article and press release from AGU)? They invalidate Bamber’s criticism almost point by point. Since they measure in ice-cores they don’t need a model of precipitations, since they measure amount of water in the ice core, they have no problem with firn densities. Since they measure mass directly, they have no problem with satellite errors. It is in any way superior to both Zwally’s and Bamber’s and it supports Zwally. They find that annual snow accumulation on West Antarctica’s coastal ice sheet increased dramatically during the 20th century. Annual snow accumulation increased 30 percent between 1900 and 2010. They found that in the last 30 years of the study, the ice sheet gained nearly 5 meters (16 feet) more water than it did during the first 30 years of the studied time period. Since the record is 300 years long, we can see that the amount of snow that has been accumulating in this region since the 1990s is the highest we have seen in the last 300 years. This is the same West Antarctica that is supposed to collapse any time now.

        Thomas, E.R., et al. “Twentieth century increase in snowfall in coastal West Antarctica.” Geophysical Research Letters 42.21 (2015): 9387-9393.

        • polarscientist says:

          Yes, I have seen the Thomas et al paper. It points to a local accumulation of snow in response to a change in the intensity of the Amundsen Sea Low Pressure Cell. This is a continuation into Ellsworth Land of a similar increase seen in the Antarctic Peninsula. But the paper also points out that there has been no similar increase nearby, on the Thwaites Glacier, and a negative trend in accumulation elsewhere in West Antarctica. Hence it is a ‘local’ trend. Nearby, at the coast there is ample evidence that the ice shelves are being thinned from beneath by warm water erosion, and glacial discharge to the Amundsen Sea is speeding up as a result. The paper further points to the paradox “that an increase in snow accumulation over Antarctica will result in additional dynamic ice loss”. It is the changes at the coast, leading to the speeding up of glaciers like Pine Island and Thwaites, that will drain ice from the interior, much as is happening along the Greenland coast. These are observations, not models. Zwally is dealing with models based on observations. The key question, as Bamber points out, is whether or not Zwalley’s model is right. Anyway, Zwalley’s maps of accumulation confirm that there has been accumulation where Thomas finds is, as a southward extension of the accumulation in the Peninsula, and also agrees that there is significant ice loss over the hinterland being drained by these major glaciers. The key question would seem to be the extent to which Zwally is right (or not) across the whole of the continent. Neither you nor I have the answer.

          • Javier says:

            Of course we don’t have the answer, but the logical conclusion to that is that we don’t know, and that is not the logical conclusion that is being transmitted. We are being told that science is settled when it clearly isn’t.

            Antarctica is not warming. In fact most of it has been cooling for the last decades. This is consistent with gaining mass, not losing it. But at the same time that we are being told that Antarctica is melting, without having clear evidence for it (as you recognize when you say that we don’t have the answer), the CO2 greenhouse hypothesis is turned inside out, like a sock, to explain that it is acting backwards to explain that increased CO2 is actually cooling Antarctica instead of warming it (Schmithüsen et al., 2015). So much for polar amplification. This is fantastic, isn’t it? This hypothesis explains any result, warming or cooling, less snow or more snow, less precipitation or more precipitation. What I don’t know is why people still thinks it is a scientific hypothesis. An absolute requirement of any scientific hypothesis is that it has to be falsifiable, and the CO2 hypothesis is clearly not.

            Schmithüsen, H., et al. “How increasing CO2 leads to an increased negative greenhouse effect in Antarctica.” Geophysical Research Letters 42.23 (2015).

          • A C Osborn says:

            Yes Javier, never in the history of science has a hypothesis Flip Flopped so many times in such a short period.
            Every time they make a prediction that is proven wrong a flip or flop occurs where they find an Excuse for why they were correct but Real Life did not react in the predicted way.
            Where they really can’t find an alternative they either ignore the problem or use the press to lie to the public.
            The lies told by Obama have been some of the worst.

            How anyone can even consider the science anywhere near “settled” is a joke and it is worse when it comes from Scientists.

            There have been hundreds of Peer Reviewed contradictory papers issued over the last few years.
            But you will not see anything in the MSM about it.

            Add to that at the same time the majority of “headlining” supporting papers have been totally shredded by Sceptics from both sciencific and statistical backgrounds, that is also never in the MSM, only the original headlines.

            If the Science is so settled and so solid why are many many supporters calling for sceptics freedom of speech to be revoked with threats of prison etc.

            Do “Scientists” in general support this inquisition?

        • polarscientist says:

          Javier – another comment on Antarctica (there was no reply space for your April 17 comment below). The evidence is that West Antarctica is warming slightly while East Antarctica is either stable or cooling slightly. The reasons are that the two are under different weather systems. West Antarctic weather and climate are driven by the Amundsen Sea Low Pressure Cell that brings warm air from the north down the Peninsula and over West Antarctica. However, in general, Antarctica as a whole is also under the ozone hole, which has the effect of speeding up the southern hemisphere westerlies around the continent, which helps to keep warm air away, thus cooling East Antarctica. These are well understood meteorological phenomena.

          Aside from that, under-ice investigations show that warm water is welling up onto the continental shelf of West Antarctica beneath the ice shelves there, and melting them from beneath, which removes their buttressing effect and allows the glaciers arriving from the interior to move faster there, draining more inland ice into the ocean as icebergs.

          You give the impression that we don’t know anything about Antarctic climate, but in fact we know a lot – even if not everything.

          • Javier says:

            I am sure that we know a lot, but only very recently. Alarmism is baseless if we don’t know the time derivative of what we are studying.

            It is assumed that warm water up welling is something new, but for what we know could have been happening for thousands of years. Places of up welling of oceanic currents are not something that changes from a decade to the next, and the temperature of that up welling water was probably determined centuries ago.

            To that we have to add that since 2004 the Argo system shows no warming of the Southern Ocean 55-65°S:

            Crosta, 2009, has shown that a warm/less-sea-ice loving diatom species did great during the Holocene Climate Optimum and it is now at its lower limit, while a cold/more-sea-ice loving species has been doing better since the HCO ended and is now at its best.

            All this evidence supports the notion that the cooling of Antarctica has proceeded for thousands of years and did not join the recovery from the LIA, except for West Antarctica that is under the influence of a different climate linked to Patagonia.

            So there is probably nothing wrong with Antarctica and however we get in the news every Tuesday and Thursday how dangerous is the climate situation of Antarctica. This does nothing to boost our confidence on cryologists. Scientists usually complain of lack of communication with the society. Well lying to people is not the best way to improve communication.

            Crosta, X. (2009). Holocene size variations in two diatom species off East Antarctica: Productivity vs environmental conditions. Deep Sea Research Part I: Oceanographic Research Papers, 56(11), 1983-1993.

          • A C Osborn says:

            And very little mention of the undersea Volcanoes off the west coast of Antarctica.

    • polarscientist says:

      Response to Javier on sea level. When referring to Grinsted et al 2010, what you did not explain to the reader was that they were discussing a MODEL of sea level change with time, in which they supposed that there was a specific relationship between past temperature and sea level that would enable them to DEDUCE a past sea level curve from palaeotemperature data. So it doesn’t actually provide us with FACTS about sea level in the medieval warm period.

      • Javier says:


        My point was that global sea level rise started in 1700, and although the model confirms it, that conclusion is grounded on evidence. From the paper:

        “The model is calibrated against the ‘virtual station’ GSL reconstruction (Jevrejeva et al. 2006) from 1850 to 2001, as it has published standard errors (see Fig. 1) and preserves volcanic signatures (Grinsted et al., 2007). This GSL reconstruction stacks 1023 tide gauge records using a stacking algorithm designed to minimize spatial bias while being independent of satellite altimetry.

        We also extend the GSL reconstruction prior to 1850 using the record of annual mean sea level from Amsterdam since 1700 (van Veen, 1945) correcting it for the post glacial land submergence rate of 0.16 mm yr-1 (Peltier, 2004). The issue of how well the Amsterdam record represents global sea level was treated in some detail in Jevrejeva et al. (2008b), and the representativity error dominates the uncertainties in local station vertical movement”.

        I just made a passing comment on the sea level during the MWP, because that is what the model and the research from one of the main groups working on paleo sea levels (Jevrejeva’s) support. The effect of temperatures on sea levels through the steric effect and ice forming/melting is not the least controversial. For the impact we will have to rely on evidence and models, as always.

        So this is Figure 1 from the paper with my blue line. Do you really think the evidence supports that global sea level rise is accelerating? How much? Since when?

        There is a lot of people commenting about the dangers of sea level rise in Roger Andrews Blowout Week 120 post in this blog. Given that we are near the end of the interglacial, I would be very worried if global sea level stops rising and starts dropping. This modest nearly constant rise doesn’t worry me the least.

        • polarscientist says:


          On your diagram you could, if you were so minded, put one line from bottom left to 1930, which would have a low slope, and another from 1930 to present, which would have a steeper slope. In fact that is more or less what the Jevrejeva paper does (Jevrejeva et al, 2008 Geophysical Research Letters 35.

          • Javier says:

            Yes Colin,

            I could “put one line from bottom left to 1930, which would have a low slope, and another from 1930 to present, which would have a steeper slope.” And I would do cherry pick that date to split the sea level rise if I were politically motivated, I wanted to increase my chances of getting my paper in a better journal, or I wanted to get a grant to support my lab for the next years to study sea level acceleration.

            Otherwise I would just go with the data that clearly shows that changes in the sea level rise rate follow an irregular sinusoid that agrees extremely well with the rest of the climatic data that shows the same sinusoid with different lags.

            In the next figure you can see δ18O in the Gomez glacier in West Antarctica that follows the Southern Annular Mode (top blue and red), the AMO (top green), and HadCRUT4 temperatures (top yellow) compared to global sea levels (bottom).


            As you can see with your own eyes, global sea level rate of rise is following temperatures with a 15 years delay.

            As global average temperature is not increasing its rate of change, sea level is not going to increase its rate of change. In fact the 15 years delay means that global sea level rise is entering the pause in warming about now, so I can predict that global sea level rise is going to slightly decelerate over the next couple of decades.

            That is the sorry state of climatology that a casual observer can look at the evidence and come up with better predictions for free when the IPCC and global warming climatology is costing us hugely to produce what essentially amounts to bullshit.

          • polarscientist says:

            The Jevrejeva paper to which I referred did not cherry pick the data to show that the rate of rise of sea level had increased over time since about 1850. Instead, they used standard statistical methods to put a curve through the sea level data, which seems entirely reasonable to me. All I was doing in response to your blue line was pointing out that even by eye one could see that the older data followed a slightly different path from the younger data. I was not trying to cherry pick anything. If you want to fight about it you should really take it up with the authors of the Jevrejeva et al paper of 2008, not with me.

            As to the Atlantic Multidecadal Oscillation (AMO), it is not surprising to see that it is evident in the Antarctic Peninsula data (though superimposed on the rising trend of global warming). After all it is warm air from the north (the Atlantic) that is brought down the Peninsula by the northerly winds of the Amundsen Sea Low. It would be instructive to carry out a definitive statistical examination of the relationship between the temperature and sea level. Just drawing blue lines on a graph doesn’t really cut it.

          • Javier says:

            “It would be instructive to carry out a definitive statistical examination of the relationship between the temperature and sea level. Just drawing blue lines on a graph doesn’t really cut it.”

            What? Do you think this is really not already perfectly known?
            This is an overlay of two official graphs. Bottom graph is IPCC AR5 WGI Fig. 3-14. “18 Year mean global sea level trends”. Top graph is “Rate of change of global average temperature” from the European Environment Agency, here:

            So this is the overlay:


            The sea level and temperature lines are so close that it is hard to follow them.

            Since we know that global warming is not accelerating, the conclusion is clear. When a climatologist says that sea level rise is accelerating either he is lying or he doesn’t know what he is talking about. And James Hansen is the biggest liar when he talks about sea level rises of several meters.

          • polarscientist says:

            Interesting graphic, thanks. But it is quite different from the one you used for Antarctica where you said “global sea level rate of rise is following temperatures with a 15-years delay.” There’s no 15 year delay in the data in your overlay of sea level and temperature. Even you note that “The sea level and temperature lines are so close that it is hard to follow them.” You can’t have it both ways. Either there is or there is not a 15-year delay.

            There’s another problem too. The data in your graph are rates of change, not absolute amounts of change. It is the absolute amounts, showing increases in both temperature and sea level, that tell us that the globe is warming.

            Finally, you accuse climatologists of lying (or being ignorant of reality) when they say the rate of rise of sea level is accelerating, which I assume you would apply also to climatologists who say the rate of temperature increase is itself increasing (since temperature and sea-level are so closely linked). But, a careful examination of your overlay diagram shows that the minima of the rates of rise of both temperature and sea level were lower in 1910 than they were in 1960 or 1975, while the maxima were higher in 2010 than in 1930 or 1905. My question is – what would smoothing do to these curves? It would flatten out the peaks and troughs to show whether or not there was an underlying trend. I have a sneaking suspicion that that trend might well be upwards.

          • Javier says:

            The delay in the first figure was from the sinusoidal adjustment of two curves that were not normalized, comparing only two points. The rate of change graph shows three temperature and three sea level curves and provides a better comparison over the entire curve. Taking only HadCRUT and Jevrejeva’s curves, the delay appears to be smaller but still clear. So on closer examination I would say the evidence supports a delay but of only a few years.

            “It is the absolute amounts, showing increases in both temperature and sea level, that tell us that the globe is warming.”

            Yes. As the rate of change curves spend most of the time above zero, it is clear that both temperatures and sea levels have been rising… so far. Despite massive amounts of CO2 released to the atmosphere the evidence shows that both sea levels and temperatures have not significantly accelerated their rise with respect to a time when little CO2 was being released. A small acceleration is still possible (see for example ), but given frequent significant changes to the temperature record, it is within uncertainty.

            The CO2 hypothesis cannot explain why the warming rate has not accelerated significantly. It conveniently defends that natural forcings that were acting prior to human CO2 increase to warm the world (mainly increased solar activity, lack of volcanic eruptions, and lack of aerosols) were seamless replaced by GHG forcing (plus volcanic eruptions and aerosols). It uses the right aerosol fudge factor to match evidence and hypothesis, but since the warming rate during the 80’s and 90’s was higher, it has to get rid of adjusting cooling factors, and runs out of excuses to explain why it is still not accelerating.

            This compendium of ad hoc explanations tries to hide the inevitable but unacceptable conclusion that CO2 climate sensitivity must be quite low, and that we know so little about climate change, that CO2 forcing has been overestimated, while solar forcing has probably been underestimated. The confession that control over the world energy resources and policies has been placed on such shaky ground is also unacceptable and so the illusion of control has to be maintained. What previously were mistaken claims based on wrong assumptions from imperfect knowledge are turning into a chain of lies once they become unsupported by evidence.

            Nobody knows the contribution of natural factors to global warming, and as our emissions are no longer increasing much, even before we reach the end in fossil fuels growth, nobody knows if in 20 or 50 years the world is going to be still warming or is going to be cooling. Meanwhile once more society takes the wrong path when trying to support its favored policies on scientific grounds, like it did with eugenics in the early twentieth century. Science doesn’t come out with a good reputation from its marriages to politics. Some lessons are not learned. Climatologists are at the center of the dispute this time, and if they don’t start publicly speaking against the alarmists within their ranks and emphasizing the uncertainty, if they wait for nature to disprove them, they will become scapegoats for the politicians.

          • polarscientist says:

            I am sure you would agree that what you are citing is your opinion, not hard evidence.

          • clivebest says:


            May I ask you two questions?

            1. Is global warming serious enough of a problem to abandon fossil fuels before they become (economically) exhausted?
            2. If so then how would you propose that we generate sufficient energy to support 9 billion people with currently available technology ?

          • Good questions, Clive. My perception is that we should anticipate winding down our use of oil, gas and coal, and spinning up their renewable replacements LONG BEFORE we run out of coal, oil, or gas. It is self evident that we would be better with gas than with coal or oil, because burning gas produces less CO2. However, for some activities (flying airplanes) there may be no alternative to gasoline (unless we revert to airships!). Why do I think this? (i) Continued addition of CO2 to the atmosphere will further acidify the ocean, with deleterious effects on things like pteropods at the base of the food chain, which will not do us any good; (ii) continued warming will raise sea level, which will continue to rise long after we stop putting CO2 into the atmosphere. Given that man is a largely coastal species, this is also not a good thing. Evidently the crux of the matter concerns rates of change. But right now they are rapid compared with the geological past. How would I propose generating suitable supplies of energy to replace what we get from hydrocarbon sources? Well, we are already globally expanding wind and solar power; water power still has some legs in it; so too does nuclear (provided you don’t site your power plant on a tsunami coast). A key problem for the immediate future is increasing electric storage capacity, and dozens of engineers are working to get around that challenge. For the young engineers of today and tomorrow, the answers to your questions are what will make them the Fellows of the Royal Society of tomorrow, or make them very rich. We don’t yet know exactly what form the future will take, but believe me, it is occupying the minds of the brightest people on the planet, most of them outside the hydrocarbon industry. And that is a shame. If only the hydrocarbon companies would see themselves as energy companies, we might have cracked this a long time ago.

          • I am responding to Javier’s point of April 21, 2016 at 10:59 am. This is about the extent to which we do or do not know what is changing today’s climate. Javier thinks we have no idea what’s going on. But if you put a powerlaw curve through the temperature data since 1880 you will derive the long term trend of modern temperature. If you then plot that against the curve of CO2 for the same period, you will get a straight line. The two are related.

            Then again, as you yourself point out, when you subtract the long term trend from the temperature, you get the deviations of temperature, otherwise known as the ‘residuals’, which show a 60-year cycle of natural variability (probably driven by the Atlantic Multidecadal Oscillation and its relatives like the Pacific Decadal Oscillation). But these merely modify the underlying upward trend. They cause the ‘flat spots’ on the modern temperature curve. There’s nothing mysterious or magical about this. What we are seeing is the interaction of natural variability with an imposed emissions driven trend.

            We can even project the natural variability trend into the future, for, say 50 years (longer might be pushing the data rather hard). If we do so, and add the known annual increments of CO2 what we get is a rise of temperature by 2 deg C over 1900 values by about 2060. This is without using one of those supposedly suspect climate models. Do we want global temperatures to be 2 deg C above 1900 values by 2060, knowing that this implies a further rise in sea level plus further ocean acidification? That’s the key question.

          • Javier says:


            I am sure you would agree that what you are citing is an opinion, not hard evidence.

            From 1900 to 2013 we believe the planet must have warmed about 0.6-0.8°K. As I have shown, there is no significant acceleration in warming. Therefore by 2060 ceteris paribus we could expect temperatures to be about 1.2°K above 1900.

            BUT it is not going to be ceteris paribus. Emissions have been showing signs of reduced increase for the last five years, and both oil and coal show a reduction in production since the summer of 2015. If you are using IPCC RCP 8.5 you are hopelessly wrong. We might actually follow RCP 2.6 for all we know.

            And no, we cannot project natural variability into the future 50 years until we can show the ability to make successful climate predictions 10 or 20 years in the future, and we have demonstrated that we cannot, because we don’t understand natural variability. I mean, in the past we have successfully projected both our emissions and CO2 atmospheric rise. We have failed spectacularly in anticipating both methane increase and temperature increase. Do I have to remind you of IPCC AR4 0.2°K/decade? It is inescapable given that disparity that we conclude that we do not understand the relationship between CO2 increase and temperature increase.

            Do not pretend that you know more than you do. Making 50 year projections at this point is ludicrous. And 2060 temperatures won’t be 2°K above 1900 even if we want them to. That requires a sustained rate of warming that has not been observed during the instrumental era and thus is unrealistic.

          • Response to Javier’s message of April 22 at 3.44 pm.


            I am not sure where you got your figures about the lack of significant rise in CO2 in recent years, but the following comes from the Scripps site that manages the Mauna Loa CO2 data:

            “The annual growth of atmospheric carbon dioxide (CO2) during 2015, above three parts per million (ppm) per year, was the largest ever recorded at the Mauna Loa Observatory in Hawaii….. Independent observations by NOAA’s Earth System Research Laboratory and by Scripps Institution of Oceanography at UC San Diego show that not only did 2015 have the largest increase, but also that the annual increase was larger than two ppm for each of the last four years, another first.”

            Furthermore, the standard NOAA temperature plot from 1880 to 2015 clearly does show that the rate of temperature rise has increased with time. It is not a straight line.

            Clearly you and I cannot be using the same data sources if what you say is correct.

            And, No, I am not using the IPCC for my data or statements. Given the growth in world population to 10 billion, and its need for growth in energy, and the fact that a great deal of that energy will inevitably come from burning hydrocarbons, it seems highly unlikely to me that the CO2 emissions curve will flatten significantly in the next 40-50 years, no matter how much I wish that were not so. And despite what you say, anyone can plot the curve of rising temperature and the curve of rising CO2 and see that they are related once the annual temperature fluctuations are removed. The underlying trends do match. Hence, if energy is to increase, CO2 will increase (even if we do use more renewables), and thus temperature will increase. Denial is of course an option.

          • Javier says:


            We are clearly not using the same data as I clearly talked about emissions and you answered with Mauna Loa atmospheric data that does not measure emissions.

            The data for the lack of significant rise in CO2 emissions in recent years is everywhere. Why don’t you try the IEA for example?

            You can get a figure with yearly rates here:

            Since 2012 the average rate of growth is ~0.6%, the lowest by much for a 4-year period in decades.

            I always find it funny that when talking about good news like the pause, scientists will always say that climate trends are only significant when they are sustained for decades and the rest is natural variability, yet when a year comes with above average bad news it becomes significant enough to make the news and sustain alarmism. I am sorry to spoil your alarmism, but the 2015 increase in atmospheric CO2 is properly assigned to natural variability, in this case the strong El Niño. There is a chance that it could be corrected by a coming La Niña over the 2017-18 period.

            Two things that are increasing will always correlate and can always be drawn at a scale as to match, Yet temperatures show long periods 1950-1975, 2001-2013, of decrease and little grow respectively, so the correlation in this case is poor. A comparison between rates of change in both temperature and atmospheric CO2 (averaged to eliminate seasonality) clearly shows the poor correlation (unlike temperature and sea level). The match is made based on CO2 warming hypothesis and fudge factors like aerosols.

            “a great deal of that energy will inevitably come from burning hydrocarbons”
            This is an assumption. Since 2010 nearly all growth in oil production has come from Light Tight Oil and bitumen, but their production has never been profitable, as producers of un-conventional oil operated as Ponzi schemes paying dividends from increasing debt. With the current price crisis they have become zombie companies that can walk for as long as good money is thrown to rescue bad money. At the same time capital investment in future oil has plunged compromising future oil production for a decade. We are set for a multi-year decline in oil production, and Peak Oil could very well have arrived already. Coal production is also going down. I seriously doubt that Natural gas increase will compensate for oil and coal reductions. Perhaps we will have to do with less energy. Already many countries have reduced their energy consumption below levels eight years ago.

            “Hence, if energy is to increase, CO2 will increase (even if we do use more renewables), and thus temperature will increase. Denial is of course an option.”

            WRONG. There cannot be denial of the future as the future is unknown and we cannot deny what has not taken place. Every time we think we know the future, we are proven wrong, yet we do not learn.

          • Yes, I am very familiar with the International Energy Agency. But they tell us that CO2 emissions are still running at 32.5 GT/yr, which is a huge amount. It equates to a rise in CO2 in the atmosphere of 2 or more ppm per year, which constitutes a significant increase in atmospheric concentration and the inevitable rise in temperature. And the rest of the CO2 is still going into the ocean, adding to ocean acidification. Besides, other activities (land use change and cement production) are also adding to CO2.

            Yes, there is a slowing of the growth of emissions. But how temporary is it? Oil companies are investing less because of the oil price drop. Coal production is down because China is getting richer and rich Chinese want to lessen the air pollution that comes with coal. Gas is on the rise everywhere as a result. And it produces less CO2. Good. Renewables are also on the rise, not least in China. But the rise is slow.

            I am sure you are right that SOME of the current rise in CO2 is due to El Nino warming. But don’t get your hopes up too far. That is only a small part of the story.

            There is no doubt in my mind that the rise in temperature since the 1880s is correlated with the rise in CO2, and that superimposed on this pattern we see fluctuations caused by natural oscillations like the Pacific Decadal Oscillation, El Nino and the AMO. But that’s all they are – just as volcanoes are – temporary deflections on an underlying ever upward trend.

            Given that our investment in the hydrocarbon infrastructure is huge, and that renewable development is slow, and that we have no alternative but hydrocarbons to power aircraft, I am sure I am right that we will still be using a lot of hydrocarbons in 2050. The fact that investment on oil has slowed is just a blip driven by the low oil price. We are not yet at peak oil though we may be on an oil production plateau. Mark my words, this is temporary. As population continues to grow towards 10 billion, the demand for energy will make the use of hydrocarbons soar until such time as the price of renewables drops and batteries develop to provide adequate storage. That will be some time coming, no matter what the good intentions were of governments meeting in Paris last year.

            We need to learn to waste less energy, to make cars more fuel efficient, to make coal-fired power stations MUCH more efficient. All this costs money, and nobody says they have any.

            No we do not KNOW the future, but within the short to medium term we can make some well-informed estimates. You can bet your boots things are not going to change overnight.

            And as far as I can see none of that is alarmist – just plain old fact as Mark Twain would say.

          • Euan Mearns says:

            Over the decades we have become MUCH more energy efficient. And the more efficient we become the more energy we use. Have you heard of Jevons’ Paradox?

            Improving efficiency makes energy services cheaper for us hence we use more. Deep Greens actually argue that improving efficiency is our worst enemy, which of course it is not.

          • polarscientist says:

            Yes, Euan, I am familiar with Jevons’ Paradox. Nevertheless we should strive to waste less. If you examine the IEA figures you will see that in recent decades the OECD countries use of energy and output of CO2 emissions has stayed more or less flat, suggesting that they are managing to be efficient and without much increase. The big increases have come from the less efficient environments of India and China, where there has been rampant growth and a lot of inefficiency. That pattern is likely to continue if not in India and China then in other parts of the developing world that are trying to play catch up and where the main population growth for the next 40 years is likely to be focused.

          • Euan Mearns says:

            The pattern of energy use is complex with the following factors in play with approximate rank:

            1) Rise in energy prices due to scarcity
            2) Rise in energy prices due to policy
            3) Offshoring heavy industry
            4) Recession due to all of the foregoing
            5) Use of imagined low C energy
            6) Improved energy efficiency


  18. polarscientist says:

    Javier, We agree that the climate of West Antarctica differs from that of East Antarctica. West Antarctica is warming. The upwelling in the Amundsen Sea is a product of the change in the winds that produced the Amundsen Sea Low, and which led to warming of the Peninsula over the past 50 years. This is the period in which the Pine Island and Thwaites Glaciers can be seen to have moved faster due to erosion of their buttressing ice shelves from underneath by upwelling warm water.

  19. Alistair Buckoke says:

    Bond cycles and the dendrochronological record

    With regard to the discussion concerning Bond cycles and their Dansgaard/Oescher counterpart, it has to be asked if enough attention has been paid to the dendrochronological record and to the work of Mike Baillie (1999, 2006, with Patrick McCafferty 2005) in particular. The principle reason for this is clear; his identification of significant tree ring events at 3195 BC, 2350 BC, 1628 BC, 1159 BC, 207 BC, 44 BC, AD 540, and the AD 1294-1361 period tallies well with the main Bond cycle peaks since the MHT, with the partial exception of the 1628 BC event (nominally the date for the Santorini eruption).

    These ‘events’ actually mark the onset or significant junctures in whole periods of narrow or virtually non-existent tree-rings, which as Baillie has shown correspond to multiple calamitous events ranging from temperature downturns and increased precipitation causing famine or floods, drought, disease epidemics, pest plagues, earthquakes, volcanic eruptions, tsunamis, seiches, meteor impacts, sightings or references toward unusual atmospheric or chthonic phenomena and sightings of comets. These events usually have worldwide counterparts, strongly evident in the international dendrochronological record. Baillie remarks on the almost ridiculous closeness of this record between AD 1340 and AD 1400. The long-established conventional view that vulcanicity or seismic events are the main culprit has been challenged by Baillie, whose two main theses are that dust and impacts or airbursts deriving from comets principally (or asteroids?) is the main culprit, and that the pathogen or toxic agent causing the Black Death in AD 1348 was of extra-terrestrial origin, a revival of the thesis of Fred Hoyle and Chandra Wickramasinghe.

    Unfortunately the extra-terrestial theme combined with widespread use of literary and historical source material, particularly in areas concerned with mythology or astrology, has caused his work to not be taken as seriously as it might have been. This ‘skepticism’ has surfaced in another example, that of the ongoing dispute over the dating of the Santorini eruption (and therefore on the accuracy of long-established archaeology-based Mediterranean chronologies). Work on the high-precision radiocarbon dating of an olive branch buried in volcanic tephra on Santorini, yielding a date range of 1627-1600 BC (Baillie 2010, Manning et al. 2006, Friedrich et al. 2006), has helped to support the dendrochronological case. As Baillie (nb 2006) reminds us, dendrochronology is quite simply by far and away the most accurate and reliable dating technology which we currently have, with the extra advantage of carrying environmental information. His comet thesis derives very largely from long-term dendrochronological research, in which discipline he has an international reputation. In seeking answers to the questions posed by the evidence, he has quite properly had scholarly recourse to any material which might throw light on the problem, including literature.

    An initial view of the Bond cycle graph provided by Euan suggests a regular, long periodicity up to and including Bond 5 (Mearns 6), somewhere around 950 years. A shorter periodicity shows through these peaks and troughs to some extent, but this becomes much more evident after 7000 BP (5000 BC). The departure of Bond 4 (Mearns 5, the MHT) from the regular long periodicity is striking, as is the emergence of the shorter periodicity, which appears to be close to a 470 year interval. After Bond 4 the long periodicity peaks are diminished and much more broken up, and do not yield any appreciably consistent periodicity, figures ranging from approximately 780 years to 1450 years. As Bailey, Clube and Napier (1990) and Clube and Napier (1990) could intimate, this picture can be accounted for by the regular visitation of a giant comet up to c.5000 BC, a significant break-up during the MHT (with resulting mayhem), and a resulting or propagated shorter period comet visitations, combined with the remnants of the long period phenomenon in the form of a debris trail. Baillie suggests that the passing of the earth’s orbit through the debris trail of a recent near-pass of a large comet (or a dead comet) may be where the main source of problems lie. The seasonal Taurid meteor showers are best explained as transit through trails of past phenomena. Every so often of course, the lumps of debris are of a size which is problematical for the earth.

    Reflecting on the work on Irish mythology (2005), it occurs to one that the period between 5500 BC and 3000 BC (ie the MHT) is the great period of long houses or long monuments in Europe. Is this association purely accidental, or solely related to social developments, connected with the adoption of domesticates?

    While this thesis might be an over-reading of the smoothed graph for some, Euan has discussed the good correspondence of the Bond cycle trace with the ice core record for atmospheric 10 Be and, correspondingly, 14C. As may easily be imagined, Baillie (2006) has a fair amount to say about atmospheric 14C. Looking at the LIA context, in his view the marked depletion of 14C between 1320 AD and 1375 AD results from dilution of 14C through the wholesale addition of old/reduced carbon, largely through significantly increasing atmospheric CO2 after c.1250 AD and until c.1350 AD. He suggests that this derives from a ‘massive ocean turnover event of some kind’, which would, in his thesis, most likely result from oceanic impacts. Disturbance of tectonic plates by a large impact might be another route which would lead to this, particularly through increased vulcanicity. The work of Ted Bryant (2001) on tsunami evidence in Australian coastal areas hints that there may have been many more serious oceanic impacts in the Holocene than we currently realize. Some of the dates Bryant has identified (7500 BC (Mearns 8), 5000 BC, 3300 BC (Mearns 5), 2000-500 BC (Mearns 3), 500 AD (Mearns 2), 1500 AD (Mearns 1)) correspond well to Bond cycle chronology. Peaks at 500 AD, 800 AD, the 1330s AD, and 1500 AD are of particular significance. Some of these events appear to make the 2004 Indian Ocean tsunami look like a ripple in a millpond.

    As part of use of any research material which can bear on the dendrochronological record, Baillie (2006) has looked at the atmospheric ammonium and nitrate signals as revealed in ice core data, particularly when and where there is no smoking volcano to identify blame. The ammonium signature for 1908, the year of the Tunguska airburst, is noted. The ammonium and nitrate signals in the period between 1292 AD and 1350 AD are remarkable, consistent with the best part of a century of one natural disaster after another. Baillie identifies the periods 1293-1296 (no volcanoes), 1303, 1309-1311, 1317-1318 and 1347 as associated with either ammonium or nitrate or both, but with vulcanicity signals for 1333 (Etna) and in the period 1343-1349. The 1290s were notable for meteors (including impacts) and unusual atmospheric phenomena. Comet Halley was sighted in 1301, and Comet Negra in 1347. The same overall horizon is punctuated with inundation (tsunami?) and earthquake records, the latter notably in 1303 (x2), 1318, 1337 (China), 1338 (China), and between 1338 and 1345. The 1338 earthquake was noted for continuing over 10 days (compilations from Hecker, Ziegler and Baillie).

    The positioning of the 1628 BC event in relation to the 1159 BC event is referred to at length, noting the 470 year interval between (ref. the anomalous periodicity of 1628 as above). The Santorini eruption can be linked to the Biblical Exodus, and the later event is linked to not only King David but also to the demise of most Bronze Age Mediterranean civilizations. Dynastic ructions in China appear to have been happening at the same times (the Mandate of Heaven withdrawn for the Xia and Shang Dynasties, respectively). Baillie (1999) observes that both 1628 and 1159 were situated in pre-existing temperature downturns, making Santorini less likely to be a principal cause. He notes that Anatolian tree ring data and acid occurrence in ice core data have that same periodicity found in Chinese, Egyptian and Irish history and mythologies, ie 470-490 years. If a 470 year interval is plotted on Euan’s graph, from 3195 BC onwards, it is noticeable how the periodicity occurs in troughs and slopes just preceding Bond downturn peaks. This same pattern also occurs in 3 of the MHT sub-peaks.

    The evidence justifiably points towards a recurrence of debris bombardment. The loose periodicity is consistent with what is known about comets and their fragmentation, the latter resulting in shortening of orbital periods.

    This research tends to show that that which causes cooling episodes might be much more important to worry about than that which, supposedly, causes warming.


    Bailey, MF., Clube, SVM., Napier, WM.. 1990. The Origin of Comets. Pergamon Press. London.

    Baillie, MGL. 1999. Exodus to Arthur: Catastrophic Encounters with Comets. Batsford, London.

    Baillie, MGL. 2006. New Light on the Black Death; The Cosmic Connection. Tempus Books.

    Baillie, MGL. 2010. Volcanoes, ice-cores and tree-rings: one story or two? Antiquity 84, 202-215.

    Baillie, MGL. and McCafferty, P. 2005. The Celtic Gods: Comets in Irish Mythology. Tempus Books.

    Bryant, E. 2001. Tsunami: The Underrated Hazard. Cambridge University Press.

    Clube, SVM., and Napier, B.. 1990. The Cosmic Winter. Blackwell, Oxford.

    Friedrich, WL., Kromer, B., Friedrich, M., Heinemeier, J., Pfeiffer, T., Talamo, S.. 2006. Santorini Eruption Radiocarbon Dated to 1627-1600 BC. Science 312.

    Manning, SW., Ramsey, CB., Kutschera, W., Higham, T., Kromer, B., Steier, P., and Wild, EM.. 2006. Chronology for the Aegean Late Bronze Age 1700-1400 BC. Science 312.

    • Euan Mearns says:

      Alastair thanks for this magnum opus. I don’t have time to go into details right now. But can you perhaps say what the MHT is. And it will be important to separate observations of historic events from speculation about the process. I’m not averse to the idea of comet dust but it is difficult to reconcile this with the 14C and 10Be evidence. But Iceland volcanoes certainly played a role in the LIA in Europe.

      It is possible that plate tectonics, earthquakes and volcanic eruptions are episodically more active since they represent the build and the release of strain. And there is a problem with these quasi periodic cycles that we like and that is that they are quasi.I remain of the opinion that there is a big piece of the jigsaw missing and dust may be part of it. The invariable plane of the solar system I believe has a lot of dust.

      I will one day get around to plotting all your dates on the Bond chart.

      • Alistair Buckoke says:

        Mid Holocene Transition.

        I think I should emphasize that I have been providing a precis of Mike Baillie’s work here, with an accent on areas which seem relevant to the discussion. I’m not an expert on climate science and would obviously defer to Baillie’s expertise. I suggest that you look at the chapter on the ‘Radiocarbon Story’ (2006, 41-50).

        That having been said it is important to remember that the dendro research community are operating out of data sets which are not only accurate and reliable but also pretty direct and concrete, and which easily translate into actual history. Indeed, the 3 books are not marked by a clear cut differentiation between the dendro record and the historical record. If you have an issue with that you need to take it up with Baillie.Though climate science questions obviously apply, the primary focus is not on climate science but on dating. The dendro record is at the same time a good foil to climate science questions and this is one of the main reasons why I have put this comment in. The trees have seen everything and store up that experience in their substance, and have a way of handing on that experience to succeeding generations!

        The background for the 14C question is the establishment of the radiocarbon calibration curve, though with obvious spin-offs. Using data from Stuiver, Pearson, McCormac and Hogg with respect to the 14th Century AD, Baillie was struck by the sharp change from the enrichment background (deriving from a less active sun, in contrast to the deflecting effect of increased solar wind) to a depletion mode at c.1325 AD, and then the sharp change back to enrichment at c.1375 AD. In his view the change is too sharp to be accounted for by the sun becoming more active, and needs another explanation. One might add that our 20th Century experience of a more active sun has not yielded the sort of problems they had in the 14th C. Casting about for ways of explaining the change, he came across the Siegenthaler et al. (1988) data on CO2 in ice cores, which show a dip in atmospheric CO2 beginning c.1100 AD to a trough c.1180 AD (cooling?), followed by a steep rise to a significant peak at c.1350 AD, and then a fall back to moderate levels by 1500 AD. He suggests that the sharp rise results from a dramatic increase in ocean turnover, and the excess old/depleted CO2 in the atmosphere diluted the production of 14C. One notes also that the upper part of this rise was happening in the 1290’s, an aspect not illustrated by the 14C data, which seems to show a time lag. Nevertheless, the CO2 trend began nearly a century before 1290, which is interesting ( assuming that there are no missing sections of ice core data here to affect the date count, a subject which Baillie has commented on extensively). The 14C story is much more compact, suggesting something pretty dramatic happening in the 1320s AD, on top of earlier events or downward influences. This is supported by the tree rings.

        I daresay some of the science will have moved on since 2005, and I would welcome any observations about the models Baillie was using then.

        I would fully agree that earthquakes and volcanoes are sometimes just that, but Baillie has matched obvious and drastic episodes in the tree rings with periods where just about everything went wrong, and where the onset of problems is often not clearly marked by an eruption, or by an earthquake. An interesting factor about the 540 AD event was that there are no volcanic eruptions associated with it. In the 2006 book there is, on balance, more emphasis on impacts or airbursts rather than dust, and I happen to know that Ted Bryant’s work is taken pretty seriously by the relevant research communities. An oceanic impact will have the immediate tsunami and secondary tectonic effects but may not produce that much dust, so much as huge quantities of water droplets and water vapour being pumped into the upper atmosphere.

        Please do plot the Baillie dates on your chart. The 470 year periodicity is also pretty interesting, though I would suggest starting from the 1628/1159 BC pair rather than from 3195 BC.

Comments are closed.