by Donald Rapp, Ralf Ellis and Clive Best
A review of the relationship between the solar input to high latitudes and the global ice volume over the past 2.7 million years.
While there is ample evidence that variations in solar input to high altitudes is a “pacemaker” for the alternating glacial and interglacial periods over the past ~ 2.7 my, there are two major difficulties with the standard Milankovitch theory:
(i) The different cadence of the glacial periods prior to the MPT (41 ky) and after the MPT (88 to 110 ky). Mid-Pleistocene Transition.
(ii) The reason why so many precessional maxima in solar input to high latitudes fail to produce terminations in the post-MPT era; yet every fourth or fifth one does produce a rather sudden termination.
Raymo et al. (2006) proposed an explanation for the first difficulty in terms of global ice volume resulting from the sum of an out of phase growth and decline of northern and southern ice sheets. Ellis and Palmer (2016) proposed an explanation for the second difficulty by describing by describing the occurrence of terminations in the post-MPT era in terms of dust deposition affecting ice-albedo on the ice sheets. Raymo et al. used a simple model for the pre-MPT period. That model does not work well in the post-MPT era. Best (2018) then modified the model to include the representation of the dust induced ice-albedo effect.
Our objective is to give a cohesive picture of the driving forces for ice age growth and decay for the obliquity-driven pre-MPT, and the precession/eccentricity-driven post MPT periods. While all three Earth orbit parameters are always acting to affect the solar input at high latitudes, there are underlying reasons why the net effect on the variation of ice volume does not show all three.
In the pre-MPT ice age era, the ice sheets expanded and contracted in concert with the Milankovitch cycle, which is the sum of the ~41 kyr hemispherically synchronous obliquity cycle, and the ~22 kyr hemispherically asynchronous precession cycle. This partial cancelation of out-of-phase precessional ice fluctuations combined with in-phase obliquity ice fluctuations results in a a global (North + South) ice volume that appears to only follow the obliquity cycle. In addition, the geographically small extent of the ice sheets in this era meant that ice-albedo was not a major climatic factor, and so orbital influences were almost completely dominant.
In the post-MPT ice age era, the Earth had cooled to a critical tipping-point roughly 800 ky ago with permanent Antarctic glaciation. At this MPT point, the energy balance of the Earth was reduced such that its natural state favored an ice age, so long as the ice sheets had high albedo. In this domain, ice would continue to grow, perhaps without limit (who knows?), until the albedo of the ice sheets could be lowered via dust deposition. This reduction in albedo resulted in a huge change in the energy balance favoring melting, and the rapid retreat of the ice sheets. At the end of the termination, the ice sheets more or less disappeared from the NH, and the climate system eventually reverted back to its previous mode of slow ice growth.
Due to precession of the Earth’s axis, there is a consistent variation of solar input to high latitudes in alternate hemispheres in regular cycles of roughly 22,000 years. The amplitude of this cyclic variation is modulated by obliquity and eccentricity. The effects of obliquity are usually weaker than precession, but nevertheless subject the amplitude of solar input to high latitudes in both hemispheres to a 41-ky cyclic variation. If, for some reason, the hemispherically asymmetric effects of precession mostly cancel out in their impact on global ice volume during some time period, the global ice volume will vary with the underlying 41-ky obliquity cycle, and the effects of precession will be hidden in the ice volume record.
In the most simplistic interpretation of solar-driven ice ages, one might expect that ice ages would occur every 11,000 years in alternate hemispheres, when the solar input to higher latitudes is a minimum in that hemisphere’s summer, allowing large ice sheets to develop. One might then expect ice ages to occur alternately in each hemisphere, every 11,000 years – in line with the hemispherically alternating precessional cycle. We do not observe this at all in the historical record of global ice volume over the past ~2.7 million years.
Figure 1. The LR04 benthic stack constructed by the graphic correlation of 57 globally distributed benthic δ O records (Lisieki and Raymo, 2005).
What is observed is that from about 2.7 mya to about 1.0 mya, the periodic variability of global ice volume followed a roughly 41 ky cycle, and from about 0.6 mya to the present, the global ice volume followed a much longer cycle of very roughly 90 to 110 ky. Between about 1.0 mya and 0.6 mya, a transition zone occurred in which the cycles gradually lengthened. Yet, the seemingly dominant periodic variation in summer solar input to high latitude (SIHL) due to precession was always in in force during all these eras.
Therefore, further investigation is needed to understand how the Earth system hid the effects of precession via internal dynamics. Note however, that the effects of precession were almost never completely hidden. In both the 41 ky and ~90 to 110 ky cycle regimes, smaller cyclic variations in global ice volume at a 22 ky period were superimposed on the broader cycles that had higher amplitude with longer periods. See Figure 1. Note that the benthic stack records total ice volume, not global temperature. It does record global average temperature, but only by default – and it cannot differentiate between the NH and SH (in both ice volume and temperature), which may alternate significantly.
So, one of the major challenges in understanding ice ages, is to identify why this persistent high-frequency solar signal due to the ~22 ky precessional cycle is rectified to 41 ky obliquity cycles prior to ~1.0 mya and 90 to 110 ky cycles after ~0.6 mya.
The period after about 1.0 mya is covered in the next section.
- The Last Five ice ages
The ebb and flow of northern ice sheets was driven by variations in solar input to high northern altitudes over the past ~400 ky, as shown in Figure 2. Figure 2 shows estimated temperature, but we assume this is also representative of ice volume. Each ~22 ky precession cycle exerts a higher frequency influence on growth of the ice sheets. The NH precessional maxima (up-lobes) tend to increase the northern temperature (reduce the ice volume), while the NH precessional minima tend to reduce northern temperatures (increase the ice volume). But in the period that includes the past five ice ages, there was a seemingly relentless internal drive to increase the ice sheets. Regardless of the precession cycle, the ice sheets expanded (albeit with small higher frequency overtones) until a point was reached where they disintegrated quickly.
Most solar up-lobes due to precession merely temporarily slowed down the rate of growth of the ice sheets, but did not produce a termination. Only about one out of four, or one out of five precession up-lobes produced a sudden, decisive termination. But occurrence of these terminations followed a regular pattern. While it is common to refer to this era as the “100 ky era”, closer inspection suggests that the cycles were spaced by 88 ky or 110 ky. The spacing between the 5th and 4th, and the 4th and 3rd penultimate ice ages was about 88 ky (~four precession cycles), while the spacing between the 3rd and 2nd, and the 2nd and last ice ages was about 110 ky (~five precession cycles). Evidently, the data suggest that the combination of the colder Earth, and the relentless buildup of ice and snow at high latitudes during this period resulted in the ice sheets growing faster during NH precessional minima, but retreating only somewhat during NH precessional maxima. This trend continued through four or five precession cycles, until the next precession up-lobe produced a rapid termination. It should be noted that just prior to initiation of a termination, the northern (and global) temperatures bottomed out, while CO2 dropped below 200 ppm. Ellis and Palmer (2016) provided extensive evidence that deposition of dust on the ice sheets provided a decrease in ice-albedo that acted as a trigger to enable the next precession up-lobe to melt the ice sheets. Only at the depth of an ice age (at the lowest temperature and the lowest CO2) after 4 or 5 precession cycles had evolved, would sufficient dust be deposited on the ice sheets to cause a termination. There is good evidence that Antarctic dust levels peaked prior to each termination, but we only have data for dust preceding the last termination on the northern ice sheets. And while we only have Greenland dust data for the last ice age, that data agrees well with the Antarctic dust record, so it is not unreasonable to assume that Arctic dust flux was closely correlated with Antarctic dust since the MPT.
It seems clear that the effects of higher frequency solar variations due to precession were masked by the albedo-driven tendency toward glaciation in the North, until some trigger (most likely dust) allowed a precession up-lobe to produce a termination.
The pattern during the period from 800 kya to 450 kya was not as regular as that after 450 kya, but the spacing of major glacial-interglacial periods tended to be very roughly 4 precession cycles, or at least it was more than double the 41 ky spacing of the earliest period prior to about 1 mya.
Figure 2. Comparison of mid-summer solar input to 60°N latitude to Antarctic temperature estimate over the past 800 ky.
Imbrie and Imbrie (1980) developed a simplistic model that can be written:
- y = ice volume
- x = SIHL (summer solar input to high latitude)
- T = a time constant (best fit around 17 ky)
- B = a constant to assure that ice volume builds up at a slower rate than the rate that it decays (best fit around 0.6)
This was not stated clearly in the original paper, but the variables x (SIHL) and y (ice volume) can both be positive or negative, and represent deviations from average values, rather than absolute values.
Insertion of the constant B assures that ice buildup will take place more slowly than ice sheet decay. Note that as B varies from say, 1/3 to 2/3, the ratio of effective time constants varies from 2 to 5.
Previous modelers always inserted a term y on the right side to reduce the rate of ice volume growth as the ice sheet volume increased, and increase the rate of growth as the ice volume decreased, but no physical explanations for this were offered. Inclusion of this term provides two benefits in fitting a model to the actual ice volume data:
(i) It shifts the peaks of ice volume slightly to more recent times, which helps to fit actual data.
(ii) It somewhat rectifies the higher frequencies of the SIHL (due to precession) by reducing the rate of expansion of the ice volume when the SIHL is negative, and increases the rate of expansion of the ice volume when the SIHL is positive.
Despite these factors, the application of this model to the North or to the South, nevertheless still results in a relatively “spiky” plot of ice volume vs. time. When the model is applied to the most recent 800 kyrs, the result is as shown in Figure 3 (Rapp, 2014).
Figure 3. Predicted ice volume from Imbries’ theory with T = 22,000, B = 0.6, and starting value 0.2 over the most recent 800 kyrs. (Other staring values shown as thin dashed lines lead to the same end result).
As we shall see in Section 3, this model works better in the pre-MPT period, when the ebb and flow of ice volume responded more directly to SIHL, whereas in the post-MPT period, ice volume continually built up over the years in a colder Earth, until a relatively sudden termination produced an Interglacial. In the post-MPT period, the underlying connection of SIHL to changes in ice volume is less direct and less obvious. Ellis and Palmer (2016) provided good evidence that the trigger that initiated a termination was dust deposited on the ice sheets, thus decreasing their albedo, leading to rapid melting. The Imbries’ model cannot account for this. The Imbries’ model includes the ice volume on the right side of the equation, but this is inadequate to describe events in the post-MPT era where ice continued to build up regardless of the SIHL, and only diminished when dust deposition decreased the ice albedo. In keeping with this picture of regulation of ice ages by albedo changes rather than the SIHL cycle, Best (2018) developed a model to account for this. He began with the simple equation:
dv/dt = – (1 ± b) (S)(1 – a)
where v = ice volume, a = albedo (calculated directly from Epica dust data), S = 65°N insolation, and b is a constant inserted to make the rate of ice growth greater during precessional minima than the rate of ice loss during precessional maxima. The terms S and b require some explanation. It is assumed in this model that there is a long (88 kyr to 110 kyr) period of growth of the northern ice sheets, when the albedo remains high prior to a termination. In this long, extended period of ice sheet growth, S (measured as deviation from the average) oscillates from positive to negative due to precession, but the growth during the 11 ky precessional minima outweighs the loss during the 11 ky precessional maxima due to the high net albedo. Additionally, the plus sign is used during precessional minima, and the minus sign is used during precessional maxima. This results in alternating growth of the ice sheets through many precession cycles as shown in Figure 4 as long as the obliquity remains > 0.9.
Figure 4. Modeled alternating growth of ice sheets through precessional cycles.
At some point in time, perhaps after several precessional cycles when dust deposits have built up over the ice sheets and reduced their albedo to a critical level, the ice sheets can absorb sufficient SIHL to melt during the next precessional maximum since S > 0, and a < 0.3. In this short period (less than 11 ky) the entire termination takes place, as shown in Figure 5.
Best tried two approaches for including decreased albedo due to dust deposition in the equation based on the record of Antarctic dust in the ice cores. This assumes that Antarctic dust levels would be coupled to the dust levels on the ice sheets, but we lack data to confirm this. Best found the best fit if he assumed a 15 ky lag between the dust peak and the onset of termination. His result of integration is shown in Figure 6. While the agreement is not perfect, and could hardly be, the model captures a great deal more reality than the Imbries’ model.
Figure 5. Dust deposition precedes modeled termination.
Figure 6. Comparison of Best’s model result (blue) to LR04 stack measurement of ice volume (black) and smoother LR04 data (red).
- The Pre-MPT 41-ky Ice Age Period
It is clear from Figure 1 that during the extended period from 2.7 mya to the present:
(i) The Earth became generally colder.
(ii) The global buildup of ice and snow increased greatly during cold periods, particularly at and after 600 kya.
(iii) The spacing between cold periods increased non-linearly from 41 ky prior to the MPT to typically 4-5 precession periods after the MPT.
Raymo et al. (2006) provided a very attractive potential explanation for the 41 ky period. First of all, they emphasized that the ocean sediment data measured global ice volume, not merely northern ice volume. Secondly, they emphasized that prior to about 1.0 mya, global ice volume never reached high levels, and the high levels we associate with recent ice ages were not reached until 800 to 600 kya. A transition period (the MPT) existed between these two extremes. They then made a crucial assumption that seems very credible:
Prior to very roughly 800 kya, the buildup of ice sheets in the North was limited, and the northern ice sheets did not exert a dominant control of the global climate, as they appear to have done in the post-MPT era. In particular, the ebb and flow of Antarctic ice was controlled by the local SIHL to Antarctica, and the ebb and flow of northern ice was controlled by the local SIHL to the Arctic.
The East Antarctic Ice Sheet (EAIS) presently is ringed by extensive marine ice shelves. However, in the distant past, according to Raymo et al., “the EAIS behaved glaciologically, at that time, like a modern Greenland ice sheet… A warmer, more dynamic EAIS with a terrestrial-based melting margin, as opposed to a glacio-marine calving margin, is implied. Because such margins are strongly controlled by summer melting, Antarctic ice volume would be sensitive to orbitally driven changes in local summer insolation.”
When did the transition from terrestrial melting to calving of marine shelves take place? Until now it has been assumed that it happened between 3 and 2.6 ma. Raymo et al. proposed that it may not have happened until after 1 ma.
Based on their model, we can hypothesize that in the early period from 2.7 mya to 1.0 mya, during the 41-ky cycle era, and even extending to a rapidly diminishing degree toward 0.6 mya, that:
(i) The ice/snow in the North never built up enough in volume for its high albedo to control the global climate. Buildup and diminution of ice/snow in both the North and South merely responded to local SIHL.
(ii) In the South, ice/snow responded to SIHL much as it did in the North, as a terrestrial-based melting margin.
(iii) The global amount of ice/snow gained or lost during a complete precessional cycle is the sum of gain/loss for the North and the South. The amount of ice/snow gained in the North during the favorable half of the precession cycle is balanced by a reduction in the amount of ice/snow lost in the South. The amount of ice/snow lost in the North during the unfavorable half of the precession cycle is balanced the amount of ice/snow gained in the South. This reduces the higher frequency component due to precession in the ice volume curve, and what we are left with is simply the obliquity cycle, which enhances SIHL at both poles in synchrony.
(iv) The global total ice volume as recorded by the benthic record, is the sum of gains and losses in the North and the South, which therefore appeared to follow the obliquity signal in this era.
(v) During these smaller 41-ky ice age cycles, the total amount of global ice stored in both the North and the South, typically maximized at 50-60 m below present-day sea level and minimized at 0 to 20 m below present-day sea level. This was considerably less than maximum depression of sea level during the last five ice ages, where sea level dropped to well over 100 m below present-day sea level.
The next step is to estimate the ice volume curves in the North and South from 2.7 mya to 1.0 mya. By adding these, Raymo et al. obtained the modeled global ice volume curve.
A fundamental assumption, based on examination of the data in Figure 1, is that the rate of variation of global ice volume is proportional to solar input to high latitudes (SIHL) in the pre-MPT era. That is quite different from the post-MPT ice age era, where the relentless growth of ice sheets through 4 to 5 precession cycles was greatly modified and modulated by ice-albedo feedback influences.
Since the observed pattern of global ice volume shows a pattern with 41 kyr periodicity, and the 22 kyr periodicity only appears as relatively small perturbations superimposed on the main 41 kyr variation, the challenge is to find a mechanism for reducing the expression of the 22 kyr periodicity in the final ice volume curve.
The problem with simplistic models of how ice volume changes with SIHL is that the variation of SIHL with time is dominated by the ~22 ky precession cycles. This, in turn, causes the resultant modeled plot of ice volume vs. time to also show variability with a 22 ky period.
Raymo et al. applied the Imbries’ model to the 41 kyr period from 2.7 mya to 1.0 mya. (There is no need to include albedo in the pre-MPT period, because ice sheet extent was limited, and the effects of albedo were small. In addition, any increase in NH albedo was countered by a reduction in SH albedo, and vice versa.) Their results are shown in Figure 7. It can be seen from the lowermost graphs that the agreement of the model with experiment is surprisingly good. Inclusion of the assumed levels of SH ice volume greatly reduces the higher frequency variation due to precession, resulting in a pattern that follows only obliquity at 41 kyr cycles.
Figure 8 shows a close-up of a portion of Figure 7 from 1.5 mya to 1.4 mya, where the vertical relationship of the various curves can be followed. For example, the vertical dashed line occurs at a precession peak in NH SIHL. Reading downward along this dashed line, it can be seen that at this date, the NH ice volume is on an upward trend, but precedes the peak in NH ice volume by roughly 5,000 years. The SH SIHL is on a downward trend, but precedes the minimum in SH ice volume by roughly 8,000 years. Because the NH and SH ice volume curves are out of phase, the peaks in NH ice volume are balanced by a partial reduction in SH ice volume, and the minima in NH ice volume are balanced by a partial increase in SH ice volume, so the curve for global ice volume shows the effects of precession only as small perturbations to the underlying 41 kyr cyclic pattern.
Figure 9 compares the modeled global ice volume to the obliquity. The global ice volume lags the obliquity by roughly 10 kyrs.
Figure 7. Age versus (A) LR04 stack of 950 benthic d18O records; (B) 65°N summer insolation records for NH (21 June) and SH (21 December); (C) NH (blue) and SH (red) modeled ice volumes; (D) predicted sea level (solid line) and mean ocean d18O (dashed line) and (E) comparison of predicted mean ocean d18O and the LR04 stack detrended by a slope of 0.8° per My from 3 to 2.5 Ma and 0.26° per My from 2.5 to 1 Ma. (Raymo et al., 2006).
Figure 8. Close-up of a portion of Figure 7.
Figure 9. Comparison of modeled ice volume to obliquity.
Previous discussions suggest the following conclusions:
(1) For the whole period from 2.7 mya to the present, precession has exerted a significant higher frequency influence on the SIHL via its ~ 22 ky periodic variability.
(2) Despite the higher frequency input of precession to the SIHL, we do not observe this frequency in the record of ice volume vs. time for the whole period from 2.7 mya to the present, except as smaller secondary perturbations to underlying, more slowly varying major trends in ice volume.
(3) From about 2.7 mya to about 1.0 mya, the fundamental cadence of ice volume variability was paced by a 41 kyr period.
(4) Over the past five ice ages over the past 450 kyrs, the fundamental cadence of ice volume variability was paced by spacings of either four or five precession periods.
(5) The period from 1.0 mya to about 0.6 mya was a transition period from the 41 kyr cadence to the longer cadence, but resembled the longer cadence more closely.
(6) A major problem facing us in understanding the observations made in points 1 to 5 above, is why the higher frequency nature of the contribution of precession to the SIHL never appears in the ice volume record, except as secondary perturbations.
(7) In the earlier period from 2.7 mya to 1.0 mya, the Earth was not as cold, and buildup and decay of ice responded to local SIHL, so Arctic ice-albedo feedbacks could not dominate the Earth’s climate. During this period, the build-up and decay of Antarctic ice volume (at about 30% of northern ice volume) out of phase with northern ice volume, produced a global ice volume (sum of northern and southern) that mostly cancelled out precession variability, leaving the cycle of global ice buildup and decay appearing to only follow the 41 kyr period due to obliquity. Global ice volume never reached a level higher than about 50-60% of that in the recent ice ages.
(8) In the most recent period of the last five ice ages, and to some extent further back as far as 1.0 mya, the Earth was cold enough that the energy balance favored continued growth of the great northern ice sheets, with greater expansion during periods of low precession-induced SIHL than smaller contraction during periods of high precession-induced SIHL. The high albedo of these increasingly larger northern ice sheets could now exert a global influence on the Earth’s climate. In contrast, Antarctic ice sheets had reached their natural continental limit of expansion and so their albedo feedback remained constant, thus allowing the ever expanding NH ice sheet albedo to dominate global climate feedbacks. The ice sheets continued to grow until, after 4 or 5 precession cycles, CO2 was reduced below 200 ppm and the global temperatures reached a minimum. At that point, some factor, most likely dust accumulation on the northern ice sheets due to expansion of deserts, caused the next precessional maximum in the SIHL to relatively quickly melt the ice sheets and bring about a termination.
(9) Precession does not appear as a major factor in the history of ice volume over the past 2.7 my, even though it was always present, always active, and always important. But the final result for ice volume hid the effects of precession for totally different reasons in the early and late regimes.
E. Lisiecki and M. E. Raymo (2005) “A Pliocene-Pleistocene stack of 57 globally distributed benthic d18O records” Paleoceanography 20, PA1003-PA1019.
E. Raymo, L. E. Lisiecki and Kerim H. Nisancioglu (2006) “Plio-Pleistocene Ice Volume, Antarctic Climate, and the Global d18O Record” Science 313, 492-495.
Ralph Ellis, Ralph and Michael Palmer (2016) “Modulation of ice ages via precession and dust-albedo feedbacks” Geoscience Frontiers http://www.sciencedirect.com/science/article/pii/S1674987116300305
Clive Best (2018) “Towards an understanding of ice ages” http://clivebest.com/blog/?p=8679.
Donald Rapp (2014) Assessing Climate Change, 3rd ed., 816 pages, Springer, Heidelberg, ISBN-13: 978-3319004549.
John Imbrie and John Z. Imbrie (1980) “Modeling the Climatic Response to Orbital Variations” Science 207, 943-953.
Moderation note: As with all guest posts, please keep your comments civil and relevant.
Scotese (2018) ‘An Estimate of the Volume of Phanerozoic Ice presents data on the volumes of ice during the past 540 Ma:
I don’t know why you provided this since it is irrelevant to the posting because the posting is limited to 2.7 my, whereas an estimate of ice at 5 my is way outside our period of interest.
Thank you for pointing out that my comment is not strictly on topic for your posting. I accept that. However, from my perspective, the policy relevance of the science needs to be pointed out. Some policy relevant points we can take from the Scotese (2018) data and other sources by Scotese and other are:
1. The volume of land ice over the past 5 Ma has been the largest in the past 540 Ma (since the Cambrian explosion of life, which was during much warmer times than now).
2. Life thrived when the planet was warmer than now and struggled when colder.
3. We are in an extremely cold period compared with most of the past 540 Ma and compared with the periods when life thrived much better than now.
4. Based on the evidence, overall ecosystems will benefit from global warming. Some will struggle others will thrive and take over but, overall, life will probably benefit from global warming. (And yes, I recognise that rate of change is also a factor, but life is adapting well to and benefiting from the current rates of change).
On “thriving” when hot:
“Scientists have discovered why the ‘broken world’ following the worst extinction of all time lasted so long — it was simply too hot to survive”
On rate of recovery:
“Typically, a mass extinction is followed by a ‘dead zone’ during which new species are not seen for tens of thousands of years. ”
VTG, Wrong on all counts. The Permian-Triassic extinction event was caused by a short sudden ice age, not the warming that followed.
Baresel et al, (2017) ‘Timing of global regression and microbial bloom linked with the Permian-Triassic boundary mass extinction: implications for driving mechanisms. ‘
Only abrupt cooling from our present very cold world would cause a mass extinction event, not global warming.
But I’ve explained all this to you many times before and you have never accepted or acknowledged any of it, so I won’t respond to you again on this thread.
Peter Lang above makes four important points:
“2. Life thrived when the planet was warmer than now and struggled when colder.” linked to “4. Based on the evidence, overall ecosystems will benefit from global warming. Some will struggle others will thrive —“. And “Only abrupt cooling from our present very cold world would cause a mass extinction event, not global warming.”
Consider the dinosaurs; particularly their size and the teeth, and variety. The existence of a species require a critical number for continuity and evolution. Their size and teeth meant a ‘top of the food chain’ brute requiring a vast pyramid of support. Earth configuration meant large productive areas from equatorial to high latitudes. That points to a different earth orientation; ie a different obliquity.
There is evidence. 3550bce obliquity change down to ~14.5deg. Sahara cooked abruptly (P DeMenocal research). A change to ~24deg at 2345bce brought the rise of today’s homo sapience. On the other hand a substantially higher obliquity ‘might’ make a habitable earth from equator to much higher latitudes by better distribution of insolation; ie a generally warmer planet without extremes of heat or cold (except at high altitudes to support glaciers), and much more productive.
Whether science likes it or not, a record of obliquity between 5400bce and 2000bce is available, recorded in stone. The dynamics of the earth include abrupt changes along with the Stockwell/Newcomb secular.
“wrong on all counts”.
Well, I’ll politely disagree with that.
Firstly, I did not claim the inception of the Permian Triassic extinction event was warning. Merely that the subsequent lethally warm period unequivocally proves your “life thrived when the planet was warmer” meme wrong.
Secondly, that the geological record, far from supporting your premise that current rapid warming is beneficial, comprehensively falsifies it: We find throughout the geological record that abrupt changes in climate, either hotter or colder, are accompanied by mass extinction events. We also find that recovery from those events takes geologically significant periods of time.
Your simplistic and false ideas are entirely unsupported by any scientific literature. Repeating them ad nauseam does not make them correct.
Still wrong on all counts. The planet is very cold. Near the coldest since animal life began. Cold is dangerous, warming is not. Most mass extinction events, certainly over the past 50 Ma, have been during cold and cooling events, not warming events. Mass extinction when GMST was 20 C and average tropical temps were 17 C higher then now are irrelevant. It’s a strawman argument to raise it and to avoid dealing with the relevant facts.
OK Peter, so let’s lay this out. From your replies here it seems:
1) We agree that there are periods in the geological age which clearly refute a universal conclusion that “life thrived when it was hotter” – for instance, the early Triassic.
2) We agree that abrupt changes in climate, either hotter or colder, are associated with mass extinction events throughout the geological record.
3) We agree that recovery from these events takes geologically significant periods of time.
You now claim that (2) does not apply to the current climate; that uniquely in the geological record, an abrupt increase in temperature now will not have the effects seen with abrupt changes in climate throughout the geological record.
Your position lacks any evidence to support it. I suggest you stop citing the geological record in support and admit it is pure speculation on your part.
Your 1 and 2 are not relevant to the current climate or to GMST up to at least 7 C above current, probably more, so 3 is irrelevant.
I’ve previously provided the evidence to support all I said. You ignored it. I am not going to divert this thread to debate your strawmen.
As usual, you are being grossly misleading and disingenuous.
Cold extermination: One of greatest mass extinctions was due to an ice age and not to Earth’s warming
“The Earth has known several mass extinctions over the course of its history. One of the most important happened at the Permian-Triassic boundary 250 million years ago. Over 95% of marine species disappeared and, up until now, scientists have linked this extinction to a significant rise in Earth temperatures. But researchers have now discovered that this extinction took place during a short ice age which preceded the global climate warming. It’s the first time that the various stages of a mass extinction have been accurately understood and that scientists have been able to assess the major role played by volcanic explosions in these climate processes.”
Read this: https://www.nature.com/articles/srep43630
No mass extinctions due to global warming in the last 50 Ma, and none since GMST was 10 C warmer than now. GMST is not going to get that high again for millions of years.
The last mass extinction that was due to climate change was the Middle Miocene disruption — it was a cooling event.
The “Middle Miocene disruption” refers to a wave of extinctions of terrestrial and aquatic life forms that occurred following the Miocene Climatic Optimum (18 to 16 Ma), around 14.8 to 14.5 million years ago, during the Langhian stage of the mid-Miocene. A major and permanent cooling step occurred between 14.8 and 14.1 Ma, associated with increased production of cold Antarctic deep waters and a major growth of the East Antarctic ice sheet. A Middle Miocene δ18O increase, that is, a relative increase in the heavier isotope of oxygen, has been noted in the Pacific, the Southern Ocean and the South Atlantic.
Yes Peter, we agree that abrupt cooling of the climate is bad. No argument there.
You are proposing that abrupt warming of the climate is not harmful.
Every abrupt warming event in the geological record is associated with massive ecological disruption, and typically extinction events which take geologically significant time to recover from.
You have not, and cannot provide any evidence to the contrary.
If the geological record is your guide, there is no conclusion other than projected warming will cause huge ecological impacts.
Please provide evidence from an authoritative source of mass extinction events in the last 50 Ma that occurred as a result of global warming, and at GMST below 20C (currently GMST is 15 C).
You’re the one making the claims the geological record backs up your extraordinary claims that warming as projected will not be harmful Peter.
Please provide evidence of abrupt warming events not causing mass extinction or other major ecological impacts from the geological record.
Or, stop using the geological record as evidence for your claims.
Yes. As it does. I’ve provided evidence and many links on past threads that support the claim – i.e. that any global warming we get this century will be beneficial not harmful, not dangerous and certainly not catastrophic. On the other hand, if global cooling occurs it will be harmful, possibly dangerous, possibly catastrophic, as is clearly demonstrated by historic evidence. The trend is clear from many lines of evidence – global warming from current temperatures will beneficial for at least this century.
You, on the other hand, keep repeating that rapid warming causes mass extinction events, but you have not provided evidence of mass extinctions that were caused by global warming at the temperatures that might occur this century. Mass extinction events at GMST 10 C above current (e.g. PETM) and more are not relevant and do not refute my statements.
I am not aware of any global warming that has caused mass extinctions when GMST was below 20 C over the past 50 Ma. Are you?
If you have evidence of mass extinction events caused by global warming at GMST below 20 C (i.e. up to 5 C above present GMST) during the past 50 Ma, then please provide it. If you do not, you have not refuted my statements.
You say that in order to refute your claim, there must be an episode of warming essentially identical to the current warming in the geological record.
Yet despite the lack of an episode you deem sufficiently identical, you nevertheless claim the geological record supports your contention!
This is an entirely illogical and contradictory position.
The geological record is clear: abrupt climate change of any sort causes huge disruption to ecosystems and is often associated with mass extinctions. You can show no evidence to the contrary from the geological record, yet you claim that record supports you.
You keep making wild assertion with absolutely no evidence. You also keep twisting what I said and saying I said things I did not say. Please stop being disingenuous.
I get the impression you know very little about the geological record. You have not answered my specific question, because you cannot.
There were no major extinction events caused by global warming in the past 50 Ma. In fact, there are very few major extinction events caused by global warming in the past 540 Ma. And none at GMST below 20 C. Most were due to impactors, volcanism, and cooling events, not warming events.
Some excerpts from https://en.wikipedia.org/wiki/Extinction_event#List_of_extinction_events
Global warming: “The most dramatic example of sustained warming is the Paleocene–Eocene Thermal Maximum, which was associated with one of the smaller mass extinctions. It has also been suggested to have caused the Triassic–Jurassic extinction event, during which 20% of all marine families became extinct. Furthermore, the Permian–Triassic extinction event has been suggested to have been caused by warming.” However, that has recently been found to have been due to an ice age.
More to follow
“The Permian–Triassic extinction event: occurred about 252 Ma (million years) ago, forming the boundary between the Permian and Triassic geologic periods, as well as between the Paleozoic and Mesozoic eras. It is the Earth’s most severe known extinction event, with up to 96% of all marine species and 70% of terrestrial vertebrate species becoming extinct. It is the only known mass extinction of insects. Some 57% of all biological families and 83% of all genera became extinct. It was previously thought to be due to rapid global warming but recently found to be due to an abrupt global cooling and ice age.
“The Cretaceous–Paleogene (K–Pg) extinction event, also known as the Cretaceous–Tertiary (K–T) extinction, was a sudden mass extinction of some three-quarters of the plant and animal species on Earth, approximately 66 million years ago. … it is now generally thought that the K–Pg extinction was caused by the impact of a massive comet or asteroid 10 to 15 km (6 to 9 mi) wide, 66 million years ago, which devastated the global environment, mainly through a lingering impact winter which halted photosynthesis in plants and plankton.”
The PETM: “was a time period with more than 8 °C warmer global average temperature than today. … during the Paleocene–Eocene Thermal Maximum, the planet was essentially ice-free.”
Eocene–Oligocene extinction event:
“The transition between the end of the Eocene (33.9 Ma) and the beginning of the Oligocene is marked by large-scale extinction and floral and faunal turnover (although minor in comparison to the largest mass extinctions). Most of the affected organisms were marine or aquatic in nature. This was a time of major climatic change, especially cooling, not obviously linked with any single major impact or any catastrophic volcanic event. One cause of the extinction event is speculated to be extended volcanic activity. Another speculation is that the extinctions are related to several large meteorite impacts that occurred about this time. One such event caused the Chesapeake Bay impact crater 40 km (25 mi), and another at the Popigai crater 100 km (62 mi) of central Siberia, scattering debris perhaps as far as Europe. New dating of the Popigai meteor suggests it may be a cause of the mass extinction.”
Middle Miocene disruption:
The Middle Miocene extinction refers to a wave of extinctions of terrestrial and aquatic life forms that occurred around the middle of the Miocene, roughly 14 million years ago. This era of extinction is believed to be caused by a relatively steady period of cooling that resulted in the growth of ice sheet volumes globally, and the reestablishment of the ice of the East Antarctic Ice Sheet.
Cold extermination: One of greatest mass extinctions was due to an ice age and not to Earth’s warming
The Earth has known several mass extinctions over the course of its history. One of the most important happened at the Permian-Triassic boundary 250 million years ago. Over 95% of marine species disappeared and, up until now, scientists have linked this extinction to a significant rise in Earth temperatures. But researchers have now discovered that this extinction took place during a short ice age which preceded the global climate warming. It’s the first time that the various stages of a mass extinction have been accurately understood and that scientists have been able to assess the major role played by volcanic explosions in these climate processes.”
once again I agree with you that abrupt climate change to colder conditions is harmful. You do not need to repeat it.
You again state focus on the lack of exact analogues for current warming in the geological record:
If, according to you, there are no analogues in the geological record for current warming then you cannot claim the geological record supports current warming not being harmful.
If there are analogues, please cite them.
Indeed, if you can cite a single example of abrupt climate change of any type not being associated with disruption of ecosystems, please cite it. (you can’t).
VTG, you still have provided no evidence to refute any of what I’ve said (I’ve cited sources for all of this previously). If you cannot refute it with evidence, then you need to acknowledge it is correct. You have not refuted any of the following:
No major extinction events due to global warming in the last 50 Ma
No major extinction events due to global warming in the last 50 Ma
No major extinction events due to global warming when GMST was less than 20°C during the last 540 Ma
No major extinction events due to global warming when GMST was below the optimum for life on Earth
Biosphere productivity increases as temperature moves towards the optimum for life on Earth. The evidence from the Eocene Thermal Maximum (51.9 Ma) suggests that GMST then may be about the optimum temperature for life on Earth. GMST was 9.7˚C warmer, average tropical temperatures were 4˚C warmer, and average polar temperatures about 26˚C warmer than present.
“At the beginning of the Eocene, the high temperatures (GMST 26.5 C, tropics 32 C) and warm oceans created a moist, balmy environment, with forests spreading throughout the Earth from pole to pole. Apart from the driest deserts, Earth must have been entirely covered in forests.
Polar forests were quite extensive. Fossils and even preserved remains of trees such as swamp cypress and dawn redwood from the Eocene have been found on Ellesmere Island in the Arctic. Even at that time, Ellesmere Island was only a few degrees in latitude further south than it is today. Fossils of subtropical and even tropical trees and plants from the Eocene also have been found in Greenland and Alaska. Tropical rainforests grew as far north as northern North America and Europe.
Palm trees were growing as far north as Alaska and northern Europe during the early Eocene, although they became less abundant as the climate cooled. Dawn redwoods were far more extensive as well.”
The end of the Eocene was marked by the Eocene–Oligocene extinction event,”
The Eocene–Oligocene extinction event was a cooling event.
None of your points supports your conclusion (some are wrong too, but leave that aside for a moment).
You claim that the geological record supports a contention that abrupt warming now will be beneficial.
There is not a single abrupt change in climate in the entire record, in any direction, which does not show harm. You have been challenged to find one,
and cannot. The conclusion from the record is that abrupt climate change is harmful.
You then move on to the claim that because there is no exactly comparable event, your claim cannot be refuted.
That’s a total failure of logic; by such mental contortions, *no* claim can ever be refuted.
You cannot cite a single scientific paper which claims current warming is beneficial.
You are entirely alone, without evidence or logic. Ask yourself why.
That’s projection. That’s what you’ve been doing all along. You have provided no evidence to support your claims, have not managed to refute anything I’ve said with evidence, and keep trying to change your arguments to avoid avoid acknowledging you’re wrong,
Before changing your line of argument we need to close the previous points. You have not refuted them, so you need to accept that my points are correct and your assertions were baseless, unsupported and false.
The points that need to be acknowledged before moving on are:
1. Life thrives better when the planet is warmer than present – biosphere productivity is higher.
2. The optimum GMST for life on Earth is higher than present and higher than is projected to be reached this century.
3. GMST during the Eocene Thermal Maximum may be around the optimum for life on Earth. It was about 10 ˚C warmer than present.
4. There have been no major extinction events due to global warming when GMST was below 20 ˚C in the past 50 Ma, and probably in the past 540 Ma.
Don’t try to change your line of arguments until you have refuted with evidence, or state that you accept these points.
Peter, you are the one trying to move things on, from the fact that the geological record does not support your conclusions.
You have attempted to move the goalposts already by claiming that in order to refute your claims, an identical warming episode to current must be demonstrated, whereas to support your claims, no such episode is needed. This is a logical fallacy. Please acknowledge it as such.
Secondly, you have failed entirely to respond to the point that the geological record shows that significant abrupt climate change, from any prior condition, in any direction, results in significant harm to ecosystems, and often mass extinctions. These harms take geologically significant time to recover from.
I will move on to your tedious attempts to repeat your unsubstantiated memes yet again later.
OK, so to go through your talking points one by one.
“Life thrives better” is a meaningless soundbite. Live has existed and evolved throughout geological history. Current ecosystems and species are adapted to current conditions. “Current ecosystems and species thrive better when the planet is warmer than present” would be false.
“Biosphere productivity is higher”. You have no evidence this will the case for an abrupt change. It could be, but it could equally be that an abrupt temperature rise reduces productivity, due to impacts on ecosystems.
You have presented no definition of an “optimum GMST”. Neither have you presented evidence that abruptly reaching it, if it exists, would be beneficial. Finally, the “optimum” you claim was with entirely different ecosystems and species to present. Such an abrupt change to ecosystems is very unlikely to be beneficial, and a sudden change to species is not possible.
And equally, it may not be around the optimum – you are merely asserting a personal opinion. But anyway, so what? Being 1000m lower in altitude may be beneficial to the temperature I experience on a mountainside. It does not make jumping off a 1000m cliff beneficial.
So what? How does this inform whether an abrupt rise now would be beneficial?
*Every* abrupt change to climate in the geological record has a major impact on ecosystems.
You assume that ecosystems and speciation optimise instantly to a change in temperature. This is false; the former will take millenia and the latter more like millions of years. Human induced climate change will, however, take decades or at most centuries to significantly change.
This is endless repetition of your same unsupported assertions. I have responded to them all previously and supported my points with references. You, on the other hand, have not provided links to support your assertions.
You ignored “ biosphere productivity is higher”. Why? Don’t understand what it means? If not, please refer to the references I provided on previous threads. I am not going to waste time posting them again here since it is clear you didn’t read or understand them previously. Furthermore, since you have so much difficulty understanding anything or admitting when you are clearly wrong, I’ve tried to make the point simply so most people would understand it.
It’s the GMST at which live thrives best – i.e. overall biosphere productivity is highest; where the mass of carbon tied up in the biosphere is maximum. That should be obvious.
Your point 3 is really silly. You have not refuted my point, just made silly comparisons with jumping off a mountain. I provided quotes about life during the Eocene Thermal Maximum. On previous threads I gave links to a number of papers about biosphere productivity during the Cretaceous and Eocene Thermal Maximum and also showing how biosphere productivity is negatively correlated with latitude; i.e. it increases as temperature increases.
This was your main point that you kept repeating in many of your comments. I responded each time and showed your assertions are wrong. You now say “so what?” and then try to change the line of argument – again! You have not admitted you are wrong on the previous assertions and arguments you raised.
Please you acknowledge you are wrong on each point in my last comment, especially point 4, or refute them with evidence, before moving on to a new line of argument.
This discussion is over. The reasons are that you have not provided evidence to refute the points I’ve made, clearly you are not prepared to admit when you are wrong, and when wrong, instead of acknowledging it, you try to change your line of argument. So, there is no point continuing.
“by describing by describing” Duplication in last paragraph of the abstract.
Reblogged this on Climate Collections.
Thanks for the article. Still reading but I believe there is a typo in the first line of the abstract..
OK, the article offers a very good explanation for the observed climate for the past ~2.7 my. But where in the equation is the powerful CO2 factor?
What are the contrasts to your work to that of Hansen and Sato (2012)?
The last ice age terminated, with glaciers collapsing, with CO2 under 200ppm but the glaciation started with CO2 at ~300ppm. Is there math explain this if climate sensitivity is sky high as per Hansen?
I like the dirty ice theory. This also must couple with a diminishing precipitation, otherwise the dirt just gets embedded under a new frosty white coating.
BTW, I suggest a Ctrl-F on “altitude,” “in in” and “a a”.
Great work. Thanks.
The dirt does get buried under new snow and ice – for 15 k years, according to the ice core data. So when interglacial melting finally gets going, there is a continual concentration of this 15 k years of dust on the surface. This allows the interglacial to melt, ablate and warm through to completion.
To Ron Graf:
(1) To the extent that CO2 varied up and down with the global temperature change and other factors characterizing the glacial-interglacial transitions, and to the extent that CO2 exerts a forcing (that you refer to as “powerful”), the rise and fall of CO2 would amplify the global effects of these transitions with positive feedback, while the transitions are fundamentally driven by solar input to high latitudes (SIHL). The model in its present state is mostly concerned with the cadence of the glacial-interglacial transitions, which do not directly follow the SIHL. Understanding that is the first step in understanding ice ages.
To estimate the amplitudes, we’d have to include other climatic dynamic effects such as CO2, changes in ocean/land area, global humidity, sea ice, etc. Whether that is practical, remains to be seen.
(2) In regard to Hansen et al. attempting to obtain the climate sensitivity from the transition from the LGM to the Pre-Industrial Era, there are quite a few problems with that calculation.
First, he needs to estimate the global average temperature TG at the LGM. Dwyer et al. (1995) utilized the ratio of magnesium to calcium (Mg/Ca) in fossil ostracodes from Deep Sea Drilling Project Site 607 in the deep North Atlantic to infer that the change in bottom water temperature changed by ~4.5°C in going from the LGM to pre-industrial times.
According to Leroux (2005), the difference in temperature between an ice age and an interglacial was about 10°C in the Antarctic and about 6°C globally.
Taylor et al. (2001) carried out an analysis in which they took into account the reduced CO2 concentration and the extended ice sheets of the LGM in climate models to estimate the amount of cooling at the LGM compared to pre-industrial times. Using six different climate models, they obtained values of 3.5, 3.7, 3.8, 4.4, 5.2 and 5.9, for an average of 4.4°C.
Crucifix (2006) provided a less optimistic view of the precision to which this is known: “The global temperature change is therefore is estimated to be comprised between 3°C and 9°C with 95% confidence”.
Shakun and Carlson (2010) carried out an extensive review of the LGM–Interglacial transition. Their estimate of TG (the temperature at the LGM minus the pre-industrial temperature) was -4.5°C.
Hansen and Sato (2011), based on Hansen et al. (2008), carried out an analysis in which they attempted to utilize the data relating the LGM to pre-industrial times as a basis for estimating the temperature rise in going from pre-industrial times to a doubling of the CO2 concentration (from 280 ppm to 560 ppm). There are several problems with this calculation, including uncertain TG, uncertain greenhouse gas forcings, uncertain surface albedo forcings, lack of consideration of dust (Crucifix (2006) estimated that atmospheric dust would produce a forcing of about 1 W/m2), and lack of consideration was taken of possible changes in humidity or cloudiness.
Crucifix (2006) provided alternate estimates of the forcings. He attempted to use climate models to bridge this gap, but concluded: “… the ratio between LGM and CO2 feedback factors cannot be accurately estimated from current state-of-the-art coupled models”.
Chylek and Lohmann (2008) carried out an independent estimate of climate sensitivity by comparing the LGM to pre-industrial times.
Hargreaves and Annan (2008) wrote a commentary on the paper by C&L.
Kohler et al. (2009) also performed an estimate of climate sensitivity based on glacial-interglacial cycles. They said: “Although water vapor is the most important GHG, the following compilation does not consider any changes in water vapor in the past due to missing constraints on its variability”. In other words, they more or less said: Water vapor may be the biggest factor, but since we have no data on it, we will neglect it!
All of these studies led to significantly different results. I don’t have much faith in any of them. I have documented the details of all this but I doubt anyone cares very much.
Rappolini, thank you for your detailed answer. As to people caring, anyone interested in constraining ECS, must value paleoclimate data as an independent angle for cross-validation. A model must work regardless to the values chosen for the historical or future scenario.
Excellent work and one of the best collection of hypotheses to explain the glacial cycling within our current Quaternary Ice Age. I would add some additional hypotheses to the mix.
1) During glacial maximums, the colder temperatures a mid and high latitudes made air drier over the continents which was a major driver of desertification at mid-latitudes.
2) Very low CO2 levels at the most intense glacial maximums also contributed significantly to desertification.
3) Larger temperature gradients between poles and tropics in the glacial maximums drove much stronger baroclinic storms with much higher wind speeds over larger areas that drove frequent and very intense dust storms in many of the large deserts.
Item (2). This is what this combined theory says. Please see the references – Modulation of Ice Ages via Dust and Albedo.
Milankovitch debunked by Mike Wallace who also debunked the ocean acidification myth. http://www.abeqas.com/missing-milankovitch-100k-cycle/
If the supposed relation only works very occasionally, that’s a negative correlation, or there’s another more causal factor not taken into consideration. I’m inclined to dismiss this paper as wishful thinking clothed in the language of science.
You can dismiss this paper and believe in any old silly stuff you want to. No simple (or complex) model is going to fit all the data perfectly for 2.7 million years. If you set the bar that high, we all ought to just give up. The real question is whether models can capture the main factors that influence the cadence and magnitude of the variations in ice volume, even though they are not exact or perfect. By the way, when you look at it closely, the spacing of the last several ice ages was not 100 ky, but either 4 or 5 precession cycles.
Not pertinent to the overall argument, which looks good, but the permanent Antarctic glaciation began about 25 mya and permanent Arctic about 2.5 mya.
it would be very interesting to show the last Interglacial NH summer forcings at high latitudes and the ice parameters too. I think ice volume decreased until about 4-6ky ago and is increasing over the last milenia…
Thanks for this excellent and important article. Would have expected some mention of Javier’s recent major post here on the same topic.
When a dynamic system undergoes slow transition such that a new attractor emerges, you get flicker. The system flickers between the old and the new attractor. Here the old attractor is nonglacial or interglacial, and the new one is deep glacial. The earth’s climate is slowly heading towards deep, possibly near snowball-earth glaciation and our current flicker is the expected transitional phenomenon.
Here’s an article about flicker:
Flickering as an early warning signal
Dakos V, van Nes EH, Scheffer M. Flickering as an early warning signal. Theoretical Ecology. 2013 Aug 1;6(3):309-17.
Most work on generic early warning signals for critical transitions focuses on indicators of the phenomenon of critical slowing down that precedes a range of catastroph- ic bifurcation points. However, in highly stochastic environ- ments, systems will tend to shift to alternative basins of attraction already far from such bifurcation points. In fact, strong perturbations (noise) may cause the system to “flicker” between the basins of attraction of the system’s alternative states. As a result, under such noisy conditions, critical slowing down is not relevant, and one would expect its related generic leading indicators to fail, signaling an impending transition. Here, we systematically explore how flickering may be detected and interpreted as a signal of an emerging alternative attractor. We show that—although the two mechanisms differ—flickering may often be reflected in rising variance, lag-1 autocorrelation and skewness in ways that resemble the effects of critical slowing down. In partic- ular, we demonstrate how the probability distribution of a flickering system can be used to map potential alternative attractors and their resilience. Thus, while flickering systems differ in many ways from the classical image of critical transitions, changes in their dynamics may carry valuable information about upcoming major changes.
The issue remains controversial despite its simplicity. Everybody cites Milutin Milanković, but very few care what he actually said, with most discussions being about modern (and erroneous) interpretations of his theory.
There is a double periodicity in the Pleistocene, and this confuses people because they try to adjust both periodicities as if it was a single one.
The first periodicity is for interglacials, that were determined in the Early Pleistocene by obliquity, and are determined in the Middle and Late Pleistocene by… obliquity, of course.
I already showed that:
Some people have trouble accepting the evidence that doesn’t fit their hypotheses.
The second periodicity was introduced in the Mid-Pleistocene Transition when the cooling of the planet led to the accumulation of significant extra-polar ice-sheets. If they could not be melted during the following high-obliquity period then one interglacial was skipped.
If the ice-sheets are not melted they grow for two consecutive obliquity periods becoming so large that they turn unstable, prone to periodical collapses (Heinrich events). A host of melting-positive feedbacks become then stronger:
– Insolation that is higher the further south ice-sheets advance
– Rising sea levels as ice-sheets invade exposed continental selves that will be inundated. Water melts ice very fast.
– Increased volcanism and CO2 production from ice-sheet unloading
Some of these feedbacks are very dependent on eccentricity, so a 100-kyr cycle in global ice volume is started at the Mid-Pleistocene Transition linked to eccentricity and reset to zero at every interglacial.
The 41-kyr cycle determines interglacials, that can occur at >23° obliquity and cannot occur at 23°), that paradoxically are higher the higher is the ice volume.
-The temperature of the interglacial, that paradoxically will be warmer the higher the ice volume prior to the interglacial. The Holocene is an exception due to the Younger Dryas sabotage. Otherwise it was scheduled to have been warmer and shorter.
This figure shows the double cycle. Top: Interglacials respond to obliquity, but sometimes they fail to be produced (triangles). Bottom: Global ice responds to eccentricity.
So in order of importance, to decide an interglacial there are three main factors:
1. Obliquity. The most important by far. Interglacials are possible or not depending on obliquity alone.
2. Global ice volume. It can drive interglacials even when northern summer high latitude insolation is low.
3. Northern summer high latitude insolation. When very high (very high eccentricity every 400 kyr) it will bypass the requirement for high ice.
If obliquity is above 23°, interglacials require an ice volume higher than the equivalent to 4.55‰ benthic δ18O or a very high 65°N July summer insolation, above 549 W/m2. If insolation is lower, interglacials require an insolation above 521 W/m2, or very high ice volume, around 4.90‰ δ18O. There are two scenarios when interglacials fail to take place. The most common is case 1, when ice volume is not sufficiently high (8 cases). Case 2 is when there is sufficient ice but insolation is too low, associated to very low eccentricity (3 cases).
When the ice/insolation conditions are not right, there is a failure in the recruitment of ice-sheet melting positive feedbacks and the interglacial is aborted. The Late Pleistocene expanded over three obliquity cycles because on the second there was already sufficient ice, but insolation was not aligned with obliquity and was therefore too low at the crucial time when the feedbacks had to be recruited.
All the rest is just feedbacks, sorry.
“The 41-kyr cycle affects interglacials that can occur at >23° obliquity and cannot occur at 23°), that paradoxically are higher the higher is the ice volume.
-The temperature of the interglacial, that paradoxically will be warmer the higher the ice volume prior to the interglacial. The Holocene is an exception due to the Younger Dryas sabotage. Otherwise it was scheduled to have been warmer and shorter.”
What is your opinion on Raymo’s proposal that the early Pleistocene followed the 41ky obliquity cycle because of a symmetry between ice growth and melt-back of the Northern and Southern hemispheres? This was because the East Antarctic Ice sheet melted back to bare land prior to 1M years ago. So a growth in northern ice sheets was offset by a melt-back of southern ice sheets.
I agree with you that obliquity is the most important factor because it is a global effect that changes the earth’s seasons. Eccentricity just modulates the effect of precession. If the earth’s orbit was circular precession would have no effect at all. You can see how obliquity underpins N/S insolation below.
I talked briefly with Maureen Raymo about it. I don’t think it is correct. I think she was right with her previous hypothesis that variations in the insolation gradient between high and low latitudes control high-latitude climate and ice volume.
Raymo, M.E. and Nisancioglu, K.H., 2003. The 41 kyr world: Milankovitch’s other unsolved mystery. Paleoceanography, 18(1).
Over time and through looking at a huge amount of evidence I have become convinced that gradients are the most powerful factors behind climate changes, both past and present. For example temperature gradients between land and sea are responsible for the monsoons.
The role of moisture transport in building the ice sheets is rarely considered (except by popesclimatetheory that has a fixation on it). Already in 1872 John Tyndall had this to say about moisture transport:
“So natural was the association of ice and cold that even celebrated men assumed that all that is needed to produce a great extension of our glaciers is a diminution of the sun’s temperature. Had they gone through the foregoing reflections and calculations, they would probably have demanded more heat instead of less for the production of a ‘glacial epoch’.”
So when we look at the summer insolation gradient a funny result appears. What are changes in insolation that largely compensate between the northern and southern hemispheres turns into a gradient that is controlled by obliquity. And this insolation gradient is no doubt accompanied by a temperature and moisture transport gradient.
Changes in the summer latitudinal insolation gradient depend on obliquity. a) Mean summer insolation depends mainly on precession. Northern Hemisphere in dark red/dark grey. 60°N (solid curve) and 30°N (dotted curve) are for the month of July (21st June-21st July). Southern Hemisphere in light blue/light grey 60°S (solid curve) and 30°S (dotted curve) are for the month of January (21st December-21st January). Observe how insolation in both hemispheres is in anti-phase, largely cancelling its net effect. b) The summer latitudinal insolation gradient is similar and in phase in both hemispheres, and depends mainly on the obliquity cycle. Dark red/dark grey solid curve is the result of subtracting the 60°N minus the 30°N mean insolation. When obliquity is high both are similar and the gradient is flatter. When obliquity is low the gradient becomes steeper (more negative value). A steeper gradient drives more energy and moisture towards the poles, favoring planetary cooling and ice growth, leading to glacial inception. Light blue/light grey solid curve is the same for the southern hemisphere. Black dotted curve corresponds to obliquity.
Thanks. I love your interglacial decision map!
Why people fight against acknowledging the importance of obliquity and also why they refuse to acknowledge nonlinear dynamics – these are harder to understand than the Pleistocene obliquity driven glacial history itself which, as you say, is quite straightforward.
The obliquity signal on this chart is not evenly spaced at 41kyrs. It has been massaged to make it conform neatly to the Epica series.
The obliquity signal in that chart comes directly from Laskar et al. 2004 solution, downloaded directly from:
the output from the web interface at:
With the following settings:
starting time -1 Myr
ending time 0 Myr
sampling step 1000 Years
solar constant 1368 W/m2
With eccentricity, climatic precession, obliquity, and insolation marked
With mean daily insolation / mean longitude marked
With latitude on the Earth 65 degrees
With true/mean longitude 90 degrees
You are so ignorant about astronomy that you think the obliquity cycle is evenly spaced at 41kyrs.
You are the one dedicated to numerology and engaged in massaging the evidence to the cycles you have adopted.
“The 41-kyr cycle determines interglacials, that can occur at >23° obliquity and cannot occur at 23°”
7e started at <23° obliquity, and 7c started before the peak in obliquity.
“The second periodicity was introduced in the Mid-Pleistocene Transition when the cooling of the planet led to the accumulation of significant extra-polar ice-sheets. If they could not be melted during the following high-obliquity period then one interglacial was skipped.”
So after 15e and 7e there wasn’t an obliquity peak skipped, but after the warmer 11c and 9e interglacials there was one skipped. That’s a problem.
Fair enough, my error, the obliquity period wanders around the mean. Though as for your “numerology and engaged in massaging the evidence to the cycles you have adopted”, that would be you with your adopted Bray and Eddy cycles. I am not guilty of your hearsay, I have uniquely identified cycles whose periods are astronomically defined.
I see the main problem with this chart now. In several cases the smaller interglacial peaks are long before the peaks in obliquity. The curve fitting confuses the timing of the peaks.
It is true that the gradient of insolation depends on obliquity and this changes the extent of climatic zones on earth. The relationship is simple.
1. If you stand at the north pole on midsummers day then the azimuthal angle of the sun is equal to the obliquity.
2. The Arctic circle lies at a latitude of 90-obliquity etc.
So I think this definitely plays a role. I also looked at this some time ago and came to noticed that whenever the gradient is low the ice volume increases.
This plot shows 90-40 lat gradient
Flicker or – indeed slowing down – doesn’t figure here or in Javier’s mechanistically depauperate analysis.
Climate shifts abruptly to new states in response to multiple interactions of powerful and dynamic subsystems. At a pace, to a degree and to a state that are emergent properties of the dynamical complexity of the Earth system.
This Javier dismisses ass feedbacks.
I think you’re dismissing flicker without really thinking about it. What you write:
Climate shifts abruptly to new states in response to multiple interactions of powerful and dynamic subsystems. At a pace, to a degree and to a state that are emergent properties of the dynamical complexity of the Earth system.
is entirely consistent with flicker, as defined in the paper abstract given above.
You misunderstand my intent. I have followed Dakos for more than a decade – the system he discusses has underlying geophysical realities. As to which you will find that people propose and God disposes.
“Generally known as the Dansgaard-Oeschger (DO) temperature fluctuations (Dansgaard and others, 1993), the transition to the high temperature state in the DO occurs in human timescales (years or decades), and hence the urgency in understanding their origin and dynamics. Many researchers have proposed mechanisms for the DO (for relevant reviews see Alley, 2007; Clement and Paterson, 2008, and references thereof) including deep decoupling, nonlinear thresholds in ocean circulation modes, stochastic resonance, freshwater discharges, salt and/or sea ice oscillators, et cetera (Broecker and others, 1990; Winton and Sarachik, 1993; Winton, 1995; Manabe and Stouffer, 1997; Ganopolski and Rahmstorf, 2001; Alley and others, 2001; Seidov and Maslin, 2001; Schulz and others, 2002; Knutti and others, 2004; Ditlevsen and others, 2005; Stastna and Peltier, 2007; Rial and Saha, 2011). However, the causes of the DO and abrupt climate change remain elusive (Broecker and others, 2010; Lozier, 2010).
But flicker is a chaotic abstraction – a ‘simple rule at the heart of climate complexity’. It is truer to say that flicker is consistent with geophysical data.
Now if earth’s climate flickers between glacial and interglacial at transitions to and from major deep glaciations, then we would expect to find geological evidence of this. So I looked for it and this is what I found.
It’s always seemed obvious that Milankovitch driven oscillation between glacial and interglacial represents “flicker” between two chaotic attractors, when the earth is slowly transitioning between deep glacial and non glacial. So the obvious question was – where is the evidence that at the start or finish of previous deep glacial epochs (e.g. Huronian, Marinoan, Sturtian etc.), where is the evidence of Milankovitch driven flicker of interglacials? And sure enough (Popper alert! Hard test of hypothesis) here it is:
somebody didnt read this
That’s a great paper, thanks.
It makes the point that Javier was also making, that regarding interglacial pacing, the most important Milankovitch driver is obliquity.
Lagged by about 6000 years as needed for obliquity related insolation increase to warm the whole ocean water column.
Before the MPR 1 million years ago, interglacials followed obliquity directly, this a cycle every 41k years.
After the MPR, glaciation deepened and interglacials became less frequent. The flicker has started to peter out as the system becomes closer to the glacial attractor and further from the interglacial one. Thus now, interglacials happen only when all three Milankovitch cycles, precession, obliquity and eccentricity, peak approximately together. (When eccentricity and precession modulation peak together midway between two obliquity peaks, then you get a “double header” interglacial as happened 600k and 200k years ago, and will happen again 200k years from now.
That is, if the system is still flickering by then and hasn’t stabilised and entrenched for a few tens of millions of years in the deep glacial attractor.
Its all Greek to me
>>regarding interglacial pacing, the most important
>>Milankovitch driver is obliquity.
On the contrary, this paper-article by Rapp-Best-Ellis demonstrates that precession is still the strongest of the two main orbital forcings. But because it acts hemispherically asymmetrically, its effects can cancel out when averaged globally.
Yes precession is involved since after the MPR an obliquity peak only causes an interglacial if precession is in the right place. Which means only one in every two or three.
I would argue the opposite, Phil.
What you call the flicker is just the response of the climate to cyclical orbital changes, that probably have existed forever, and is detected for many millions of years. The Late Cenozoic glaciation has amplified the climatic response due to the existence of a huge amount of H2O that changes state in response to orbital changes. The poles and high latitudes then become refrigerators that amplify the climatic response.
Since the Mid-Pleistocene Transition, we don’t observe a decrease in the frequency of the oscillations, but a decrease in the amplitude of the climatic response to some of the oscillations, at the expense of the increase in the amplitude of the climatic response to other oscillations.
The glacial maxima of the last 650,000 years represent the coolest the planet has been in 300 Ma, but some of the interglacials that come afterwards are among the warmest the planet has been in 2 Ma. Yes, this interglacial and the previous one are really warm periods for the Pleistocene.
This has had the effect of producing a more extreme climate in the Mid-Late Pleistocene, with huge climate swings over the course of a few millennia. If anything the climate has become more extreme and more variable, no doubt driving numerous species to extinction, and probably helping the evolution of ours.
This increased Pleistocene climatic variability can also be seen in a fractal-style reproduced by Dansgaard-Oeschger events during glacial periods.
Homeostasis is extremely strong in this planet. Otherwise we wouldn’t be here. That means feedbacks are predominantly negative and become stronger the farther from average. The Mid-Pleistocene reaction of producing warmer interglacials than previously means the planet will not cool any further. We have reached the bottom of the Late Cenozoic Ice Age.
As we don’t know what causes ice ages we don’t know how long it will take to get out of this one. Hopefully we don’t have to wait until Antarctica moves out of the South Pole.
“We show that—although the two mechanisms differ—flickering may often be reflected in rising variance, lag-1 autocorrelation and skewness in ways
that resemble the effects of critical slowing down. In particular, we demonstrate how the probability distribution of a flickering system can be used to map potential alternative
attractors and their resilience. Thus, while flickering systems differ in many ways from the classical image of critical
transitions, changes in their dynamics may carry valuable information about upcoming major changes…
As most systems are embedded in highly stochastic environments, they may start to “flicker” between the basins of attraction of their potential alternative states far before bifurcation points at which critical transitions occur.” https://www.researchgate.net/publication/257769537_Flickering_as_an_early_warning_signal
Is there a late Pleistocene flicker? What are the potential alternative fractionally dimensioned state spaces for the hyper responsive, nonlinear Earth system?
I’m not persuaded that either question can be argued convincingly via mechanistically depaurate analysis. But assuredly – a critical transition is coming and we are sure to end somewhere on the hothouse to snowball Earth spectrum.
Oh – and major changes can happen in the span of a human life. God does – apparently – play dice.
read harder and more skeptically.
Javier | March 8, 2019 at 6:49 pm |
Imo ice ages develop when no large (oceanic) magmatic events occur that heat the oceans.
Last big event was the Ontong Jave one, and since that time the deep oceans have been cooling down at 1k every ~5 million years. Consequently the surface temperatures cooled as well.
Milankovitch cycles can only show up as glacials/interglacials when the oceans are cold enough for surface ice to develop.
Extended surface ice actually prevents the deep oceans from cooling, and allows the small geothermal flux the warm the deep oceans at 1K every ~5000 years (entire column). This may explain the rapid exit out of a glacial vs the slow entry into a glacial.
Who is somebody, and what is your point?
The oracle speaks in riddles so nobody knows what he actually means. The future is then safe.
You will do.
Have a look at their proceedure for making a stack. (deterministic)
and practice some skepticism. it will lead you to a more comprehensive stack
The LR04 stack uses a calculation of an optimum fit of series. The prob-stack method allows consideration of uncertainties in fit.
Ultimately if pressed for scientific detail he is a miserable failure.
Well I read Lisiecki and Raymo 2005 carefully.
Yes it is consistent with the interpretation of Javier that obliquity is the dominant forcing Milankovitch component.
And it states that it is a nonlinear oscillation and they model it with a nonlinear formula using the nonlinear term b.
Here is their model:
dy/dt = (1+/-b/Tm)(x-y)
where y is ice volume and x is the 21 June 65N insolation.
And Tm is the lag.
Also – again consistent with Javier – it has a lag Tm between insolation forcing and climate response that increases from 15 kyrs 3Mya to 5 kyrs now. Javier uses a lag of 6.5 kyrs.
They also acknowledge that deep water is the reason for the lag.
So what’s the problem?
Are you concerned with the coherence of the signal prior to the Pleistocene 3Mya? This portion of the data does not concern us much. More recently than that it has good coherence and S/N.
Was the meaning of your cryptic comment that Lisiecki and Raymo 2005 agrees with everything that Javier and I are saying? If so – thanks for noticing. Why is this a reason for skepticism?
“They also acknowledge that deep water is the reason for the lag.
So what’s the problem?”
problem is nobody read the paper skeptically.
You look for what you like.
The energy source never varies?
Sent from my iPad
One possible variable is a drop in patm during the weak geomag fields around the Jamarillo Reversal Events due to the loss of N2 (in particular the lighter N14 with an increase in N15 in the geological record that suggests loss by sputtering). Not only does this imply global cooling, but also a lower seasonal variation between summer and winter T, allowing the less drastic 41ka cycles to predominate.
During stable fields there appears to be a build-up in N2 (as seen in ice bubble data). In the Overview on Researchgate (Giant Bugs etc. version 7) this is looked at in the final sections.
Overall, this cooling process has been going on for about 30 Ma. The fossil biology of giant flying birds of the Miocene (7 m spans) suggests a higher patm, resin/amber data gives a value of about 1.3 bar for the Eocene and 1.2 for the Miocene.
Where can I find this resin/amber data and the description of the calibration method? Thanks.
the amber data is in Tappert 2013. the relation of this to the geocarbsulf model is given in the above overview (version 7) . Version 8 has a new section that looks at cyclone formation in the K over the very high SSTs (they seem not to have existed…)
The ice bubble data showing an apparent loss in O2 over the past 800 ka (which could be read as a rise in N2) is also covered along with all references.
I think I get it now– if not for global warming… there’d be global cooling.
The periods of ice ages and warm periods depend on how much ocean is warm enough to evaporate in cold places and the time it takes depends on the mass involved, Just like vibration problems in our world. Internal cycles resonate differently when the mass and spring rates are different. Spring rates are a function of evaporation rates and snowfall rates and the thawing rate of the ice sheets.
The evaporation rates depend on area of warm water flowing in polar regions. Thawing rates depend on ice extent area, especially where ice is extended into salt water oceans and on land that receives more sunlight. Tropical evaporation and rainfall do not take part in the ice cycles but most of the energy that powers this comes from tropical ocean currents. Tropical evaporation and rainfall do accomplish most of earth’s cooling but that is not the same as polar climate. There is no sudden change that starts cooling, ice volume and weight gets large enough to cause the ice to flow faster than it is thawing and retreating. There is no sudden change that starts warming, ice volume and weight gets too low to cause the ice to flow faster than thawing and retreating. Dust may make a difference, but it does not make a sudden difference.
Ocean and Sequestered Ice Capacitors
Water is abundant, water changes state. Water is stored as liquid with positive energy capacity and water is stored as ice with negative energy capacity. The energy comes from the sun at an almost constant rate. When the energy that is coming from the sun changes there are instant changes where the sun is shining. Milankovich cycles, orbit and axis tilt changes to cause the sun to heat the Northern Hemisphere more part of the time and causes the sun to heat the Southern Hemisphere more at other times. There are internal cycles of the giant capacitors, the oceans and the ice on land. When the warm capacitor is charged, there is the most positive energy stored in the oceans and the least negative energy stored in the ice sequestered on land. When the cold capacitor is charged the most, there is the most negative energy stored in sequestered ice on land and the least positive energy stored in the oceans. Sea ice is a barrier that is placed on top of the warm ocean capacitor to prevent it from losing energy and water by evaporation and to keep it from receiving more energy from sunlight. This switch is turned on and off as needed by thawing and freezing polar oceans. This works in both polar regions and does switch direction of the flow of energy between the capacitors. In the tropics the flow of energy is always turned more toward the oceans. The sunlight adds energy and evaporation removes energy and energy is transported to the polar regions. In polar regions the flow of energy is always removing energy from the warm ocean capacitors or it is turned off by an adjustable sea ice cover. When oceans are cold and frozen the flow of energy in the polar regions between the capacitors is turned off and the oceans stop cooling and the cold capacitor discharges more and the warm capacitor charges more. The process that causes an ice age is evaporation and snowfall when polar oceans are thawed when the warm capacitor has enough charge. This starts during the warming and continues into the cooling. The process that causes an ice age to end is lack of evaporation and snowfall when the polar oceans are frozen. This process starts during the cooling and continues into the warming. The switch occurs when the sea ice covers freeze and thaw. This causes long term cycles that depend on the volume of water and ice that take part in each cycle. These cycles have evolved as continents moved and ocean currents changed and carried more water between tropical and polar regions. The amount of ice that can be sequestered on land depends on the amount of warm ocean that does evaporate from oceans before the switch changes, especially warm currents in colder regions. The cold period length depends on how much ice must thaw to flip the switch again. Shorter term correlations are easier to identify, especially with the abundance of “short term” thermometer records, but those records only cover a small part of the most recent full cycle of cooling out of the Medieval Warm Period and warming into this Modern Warm Period. Thermometers were not available for the last cool period, only for part of the warming.
The cycles grew larger and colder over the last fifty million years with more and more water and ice taking part in the cycles, but every cycle sequestered ice on Antarctic and did not return all to the oceans in warm periods. We have ice cores that are 800 thousand years old to confirm this. Now, enough ice is sequestered in cold places that less converts to water and then enters the oceans, such that warm periods have less warm water energy capacity to create ice ages. Each hemisphere has polar sea ice that thaws and increases evaporation and snowfall to keep enough ice sequestered in that hemisphere. Then the sea ice forms and reduces evaporation and snowfall in that hemisphere until the next cycle reversal. Tropical cooling is constantly cooling depending on ocean temperature. Polar cooling is removing the most energy when the oceans are evaporating and creating ice to be sequestered in the negative energy ice. Polar cooling is removing the least energy when the oceans are frozen and prevented from evaporating.
Over the last ten thousand years, the earth orbit has changed to reduce the solar into the NH and increase the solar into the SH, but in that time ice cores in both hemispheres show the polar cycles adjusted to keep the temperature bounds cycling in the same bounds, but they are not synchronized. Each hemisphere has different heat and cold energy capacity and each hemisphere adjusts the sequestered ice to keep the ocean temperatures around the freeze thaw point that switches the direction of energy flow. These cycles did adjust to the huge difference in energy in that orbit cycles caused, a few watts per meter squared from CO2 will not matter. The SH has more positive capacity in warm oceans and more cold negative capacity in ice on Antarctic. Over the last ten thousand years, Ice was increased in SH as more energy came in and ice was decreased in NH as less energy came in. Yes, our NH glaciers are smaller now than ten thousand years ago because less sequestered ice is needed. The tropics seek equilibrium, the Polar regions cycle. The NH glaciers and ice sheets will grow when the sun comes more to the NH again.
This is a natural cycle and we do not cause it and we can not stop it. Water, in all its states and changing states, regulates temperatures in narrow bounds, in smaller and colder cycles on the top end and warmer cycles on the bottom end, than would happen without water. Without water where the sun was shining would be hotter than now and without water where the sun was not shining would be colder than now. The average might be close to what is now. This does not disprove the suns influence, it supports it, it needs the sun. This does not disprove that more atmosphere might influence the temperature, it supports that, it needs the atmosphere.
This does explain what causes ice ages and what causes then to end, ice and lack of ice.
Ice cools the oceans and earth by thawing and reflecting. The cooling by thawing is totally ignored. Climate people consider it not enough to count and this is a major factor that they do not even consider.
People consider it a disaster when Greenland or Antarctica loses a large ice shelf. That is how these great ice sheets cool the oceans, dumping ice into the oceans. The ice accumulates on the top of the ice and the edges constantly break off and float away, into warmer water, thawing and cooling even faster exposed to warmer salt water. We cool our drinks with ice, so does climate. A large ice sheet cools by reflecting and thawing, maybe half and half, and the thawing half has never been considered. The IR out that cooled the earth happen years before in the warm period when the ice was farmed. Climate people only look at what happens at the same time, they do not have internal natural cycles in proper consideration.
Looking at a, “41-ky cyclic variation,” which is using an abbreviation of an abbreviation (“kyr”)… this is symbolic of attempts to objectify climate by applying the statistics of regression tp cribbed and corrupted databases, which results only in mathematical models that can never be validated.
Long term patterns of solar variability should have times of harmonic agreement with Earth’s orbital variations as they have the same cause. That may help explain why recent glacial cycles are both colder and warmer than 2 million years ago.
Inter-glacial periods of the last 800,000 years, while being nominally at intervals of 85,000 years, have exhibited a longer common and regular compound period of 370,000 years (red lines). This appears to arise from a phase shift of the 85kyr periodicity, and expressed by the additional inter-glacials at 7 and 15. This has produced a symmetrical mirroring of interglacial events centered at 11c, and as far as 5e to 17c. But going by both the symmetry and the 370,000 year cyclicity, the Holocene is some 30,000 years late, meaning that the so called 100Kyr sequence has apparently broken down. This has resulted in a 115,000 year interval since the Eemian.
An important clue is that the precise periods of 85kyr, 115kyr, 285kyr, and 370kyr, occur from various interglacial peaks, to different peaks of the three dominant peaks in 7a-c. The mirror image of 15e follows the same pattern, albeit weaker on its younger peaks.
It is also well apparent that the remaining non-inter-glacial warm features and spikes also conform to these periods and their multiples and differences.
370,000 – (3*85,000) = 115,000.
Graph source with scale, zoom it out large and measure how tight those periods are.
I made a symmetrical measure from the graph scale, marked at 0, 85, 115, 255, 285, and 370 kyr. Placing zero at 5e and then 9e covers all the interglacial peaks between 5e and 17c, and 19c is another 85kyr before 17c.
“Here we show that climate oscillations over the past four million years can be explained by a single mechanism: the synchronization of nonlinear internal climate oscillations and the 413,000-year eccentricity cycle. Using spectral analyses aided by a numerical model, we find that the climate system first synchronized to the 413,000-year eccentricity cycle about 1.2 million years ago and has remained synchronized ever
since. This synchronization results in a nonlinear transfer of power and frequency modulation that increases the amplitude
of the 100,000-year cycle. We conclude that the forced synchronization can
explain the strong 100,000-year glacial
cycles through the alignment of insolation changes and internal climate oscillations.” http://www.dynamicpaleoclimate.org/uploads/2/3/5/4/23543390/ngeo1756.pdf
There is perhaps a third question to be asked about the excessive power of late Pleistocene ~100,000 year cycles. I suspect that changing Earth system resonant frequencies of nonlinear oscillators due to closing or opening of seaways with continental drift or sedimentary shoaling – and the effects of tectonic uplift – on top of chaotic internal responses – figure too little in simple or complex models. You may note that Rial et al have a different answer from above using the same oxygen isotope ‘stack’.
But the real question we are asking is what will trigger the next abrupt transition to a glacial state? Freshening of the North Atlantic by meltwater with reduced Atlantic Meridional Overturning Circulation? As with Gavin’s ‘poster child’ of abrupt climate change?
It seems to have been happening for more than a century.
Is NH summer insolation low enough to allow winter ice sheet persistence? How close are we to an AMOC tipping point? By how much does anthropogenic land and atmosphere changes alter system dynamics? Is a transition to a hothouse state – e.g. https://www.nature.com/articles/s41561-019-0310-1 – more likely?
The true state of Earth system science is having more questions than answers. It’s what makes it so intriguing.
… more than a century…
I agree that the speed of climate change during the DO events was extraordinarily violent – approaching a degree C per decade possibly. Putting in perspective the comparatively trivial climate change happening in the last century or so. I thought DO events happen only during glacial intervals, which are consequently more unstable than interglacial periods?
Or are sudden cooling episodes during interglacials, such as the one ~8000 years ago during the Holocene, are kind of “reverse DO events”?
“In North Atlantic as in other part of the world, many
paleo-climatic records of the last 10 000 years show a pattern of rapid climate oscillations with cyclicity around 1–2 kyrs. Since the pioneering studies of Gerard Bond on IRD from the North Atlantic Ocean (Bond et al., 1995, 1997, 2001) the 1500-year cycle during the Holocene was tentatively attributed to solar activity (Bond et al., 2001), to ocean current intensity variation (Bianchi and McCave, 1999), to tidal forcing (Berger and von Rad, 2002), to atmospheric processes linked to the North Atlantic Oscillation (NAO, Giraudeau et al., 2000) or to modifications of the geomagnetic field (St-Onge et al., 2003).” https://www.clim-past.net/3/569/2007/cp-3-569-2007.pdf
While climate varies on millennial scales – changing perpetually at all scales – I am no great believer in cycles writ in approximate graphology or simple cause and effect.
To quote myself. If as suspected solar activity evolves in response to an incalculable solar system N-body orbital problem – and this is further modulated through internal fluid dynamics of the Sun – cyclic behavior as such is impossible. How far it departs from the cyclical expectations of classical mechanics is unknowable – but depart it does. Solar variability as well triggers nonlinear responses in the planetary system. In climate data the reality is Hurst effects – regimes and abrupt shifts. Wavelet analysis – as above – will give you broad spectral peaks – but this is just math and not proof of anything. Real geophysics is required to understand the climate system and how it may change in future. Nor do cycles say anything about how greenhouse gases may perturb flow and change quasi standing waves in Earth’s spatio-temporal chaotic flow field. It may change them a little or a lot – it depends.
There is a stable 3453 year cycle of centennial solar minima that is well apparent in the GISP2 series as a repeating pattern of warm and cold periods. Easily seen with the three coldest periods. The most recent in the 8th century AD was the warmest part of the MWP for Northern Europe (Esper 2014). The next at 2700-2500 BC was when the Minoans and many other cultures globally flourished and built cities. And during the earliest one at 6.2kyr BC (8.2kyr event) there was an early Harappan expansion, growth of village settlements in Serbia, and villages growing wheat near the Isle Wight England. These periods had warm mid latitude continents, but fast trade winds and cold ocean phases, hence the Greenland cooling.
A “reverse DO event” would be more like the Younger Dryas.
Reblogged this on Tallbloke's Talkshop and commented:
Theorists take another look at the mechanisms that may or may not be important regulators of Earth’s ice ages.
I take it that the following from the abstract (first sentence) is a typo;
“While there is ample evidence that variations in solar input to high altitudes” and ought to be latitudes.
My conjecture about flicker requires looking at the big picture, i.e. back to the Huronian or all our geological history of glaciation. There are deep glacial periods, some possibly approaching “snowball earth” such as the Cryogenian, the Saharan-Andean (end-Ordovician) and possibly now on our horizon. The rest of the time has been in a much warmer non glacial state, including the entire history of the dinosaurs, arguably the high-point of life on earth. (Now the biosphere is struggling with challenges of cold and CO2 starvation and the evolution of for instance C4 photosynthesis and human intelligence can be seen as flails to avoid biosphere extinction – the opposite of what self-hating eco-progressives imagine.)
Anyway – with alternation between these two major states of warm and deep glacial, there are transitional periods in which glacial-interglacial flicker occurs. It it looks compellingly like the transition flicker seen in other systems – even a dodgy light bulb. Yes Milankovitch forcing never stops, all through the dinosaur Mesozoic there was Milankovitch forcing. But there was no glacial-interglacial alternation obviously because the system was far from glacial. Clearly glacial-interglacial flicker is expected only in the transitional periods at both the start and end of deep glacial periods. And sure enough – this has been found at the end of the Cryogenian, just before the Cambrian explosion:
It’s interesting to run with your point about Quarternary glacial-interglacial cycling being instrumental in human evolution, and speculate that this same “flicker” just before the Cambrian explosion might have been a critical stimulus to biological evolution and innovation then also. Maybe generally, geological periods of highly unstable climate, if they don’t cause mass extinctions (and maybe even if they do) cause a big enhancement in evolution of new species (or in the case of the Cambrian, new phyla).
I could agree in that sense with the flicker concept about periods of increased climatic variability. But quite frankly, I think we are just talking about water changes of state. When the planet is at a temperature close enough to the water→ice transition, small changes in temperature cause large volumes of water to change state, and that has a huge effect on climate, amplifying the amplitude of climatic changes and producing that flickering. Obviously as you get farther from that temperature towards both sides, the volume of water that changes state decreases reducing the amplitude (flickering).
This is a water planet. If we don’t give water a central role in its climate we won’t understand it.
This turned up today…
Why ice ages became longer and more intense
MARCH 9, 2019
Researchers say they have confirmed the crucial role of the Antarctic Ocean during periods of climate change. Over the past million years, less frequent mixing of deep and surface waters may have influenced the transition to longer, more intense ice ages.
During the mid-Pleistocene transition period, which began one million years ago, ice ages extended and became more powerful; the frequency of their cycles increased from 40,000 years to 100,000 years.
One of the keys to this phenomenon lies in the deep waters of the Southern Ocean surrounding Antarctica, according to research published in the journal Scienceexternal link, carried out by a team led by professor Samuel Jaccard from the University of Bern.
‘The scientists looked in details at the difference in salinity and temperature between the surface and deep waters, which determine the intensity of the mixing. During the transition to longer ice ages the surface waters became simultaneously colder and less salty. Consequently, the mixing of water layers decreased considerably during ice ages.’
Lots of theories about the MPT, but the simplest one (Occam’s razor) is that as the planet was cooling for several million years, it reached a point when the amount of extra-polar ice built during low obliquity periods was too high to melt during high obliquity periods, leading to ice accumulation from an obliquity oscillation to the next, and requiring the concerted action of obliquity, precessional insolation and eccentricity to produce an interglacial and melt all the extra-polar ice. The result from such a simple mechanism is longer and more intense glacial periods.
There isn’t much change in amplitude between the 41kyr signal at 1.5 ma, and the longer glacial cycles from 1.2 ma.
“The mid-Pleistocene transition (MPT)
occurred in the absence of any discernible changes in the orbital parameters that control the seasonality and meridional distribution of incoming solar radiation (1), driving decades of research into the possible contributors to the change (2–7).”
Thanks for this. It shows something other than orbital variability contributing to glacial/interglacial transitions.
I am quite sure that we have not got to the bottom of it yet – pun intended.
Considering as a possibility that the warmer periods had a higher patm (from a slow build up of N from sea floor spreading) and that N2 is lost (preferentially as N15 due to sputtering during low geomag fields, (reversals), I note that there were multiple reversals during the Pliocene and at the start of the Pleistocene which may have led to a drop in patm. Certainly, some very large birds (7 m spans) went extinct at this time…
With a slightly higher patm, seasonal differences in T are less (witness the T at the Dead Sea compared to Llasa. The 41 ky cycle dominates as winter freezing is less severe. After the last major reversal event, patm may have dropped (so the N15 tells us) and winters became more severe leading to the overall major and more violent 100+ ka cycle.
The earth does not respond to obliquity. The earth does not respond to eccentricity. The earth does not respond to precession. The earth responds to solar input to high latitudes (SIHL), which in turn, depends on all three. In particular, precession contributes a significant variability to the SIHL, and if the effect of precession is not readily apparent in the ice volume record, it is necessary to find reasons. In the pre-MPT era, apparently, the reason is that out-of-phase variability of ice volume in the North and South, together with a delay in building ice, generates a pattern of ice volume vs. time where the effects of precession appear only as small “blips” superimposed on 41-ky cycles attributable to obliquity. In the post-MPT era, the energy balance favors growth of ice sheets and this continues unabated for 4 to 5 precession periods. During that period of growth, precession affects the rate of growth, with precession maxima slowing down the growth, and precession minima increasing the rate of growth. But nevertheless, the ice sheets grow and grow, until some cataclysmic event causes a termination to occur in one period of a precession maximum. We believe the missing factor is dust deposition. That is what this posting is about. Almost all of the comments on this website are irrelevant to the posting. As I read the many postings on Judith’s website, I find most replies are irrelevant, and merely re-re-re-re-state the views of various individuals. For example, when Judith posted on uncertainty, very few replied on uncertainty, while many resorted to the endless discussion of alarmism vs. denialism.
Speaking as a casual observer, who is probably saying completely obvious things, the question I would want resolved better is just how that dust deposition switch works. If I understand the situation correctly at the end of one of these ice pulses you have these massive ice sheets that are basically stable or only very slowly melting, and then within a relative short amount of time these massive ice sheets melt.
Now this can only happen when solar input at high latitudes is high enough, but it doesn’t happen every time the solar input is high. Instead you postulate that this only happens when in addition there is enough accumulated dust in or on top the ice sheet.
I get the hypothesis but I’m puzzled about the dramatic difference in melting rates driven by the dust. I assume there will always be some dust present, but is it that this switch is driven by a dramatic difference in the dust load? Or is it that the melting rate changes dramatically for a small change in dust load?
If it were the latter case, then that maybe could be demonstrated experimentally. Or if it’s the former case then it needs to be explained why there’s such a dramatic variation in dust load from precession period to precession period.
“(8) In the most recent period of the last five ice ages, and to some extent further back as far as 1.0 mya, the Earth was cold enough that the energy balance favored continued growth of the great northern ice sheets…”
Yet the most recent interglacials were warmer than those back to 2.5 million years ago.
An interesting observation. But what do you make of that? I don’t have any great ideas here. I suppose that there are positive feedbacks during a termination (CO2, water vapor, land transitioning to water, etc.) and maybe that produces an overshoot? But that is just hand waving?
No, your explanation is probably correct. Melting positive feedbacks become stronger the larger the ice-sheets. It is like pulling a spring harder. When released it produces a stronger effect in the opposite direction. The result is that Mid-Late Pleistocene glacial terminations are faster and proceed further, producing a saw-tooth effect on δ¹⁸O measurements, and resulting in warmer interglacials.
You have an analogy of a spring but no explanation for why it would behave like that. Are the Mid-Late Pleistocene glacial terminations actually faster?
Why ask me something that you can answer yourself? Measure the rate of δ¹⁸O change over time in LR04. Then you have two choices to explain it:
a) The strength of feedbacks depends on the distance from the average, as it always does.
b) Something mysterious that needs explaining (insert your pet theory).
They are not faster so there is nothing to explain.
I have always thought that the saw-tooth shape is evidence for a solar forcing component. It’s just how multi body synodic cycles progress. The three inner gas giant’s primary synodic period is 317.666 sidereal years, and that slowly drifts out of sync over 12 steps until a shorter step of 138 years can bring them rapidly back into sync again. With four gas giants it’s obviously much longer with their primary synodic period at 4627 years. The evidence for why that matters is found at the noise level. The phase relationships with Earth and Venus must be considered too as they are intrinsic to the ordering of sunspot cycles and centennial minima. They are in phase with the three inner gas giants at 3*317.666 years, and this can be seen to work at the scale of seasonal weather, with the winter of 1010 AD when the Nile froze, and the winter of 1963.
Furthermore, despite the fact that these recent interglacials were warmer, and oceans were higher, CO2 was still ~ 280 ppm, which suggests that something other than CO2 was playing a role here?
I cannot think of an internal reason why extra cooling would lead to extra warming. It doesn’t hold true anyway, interglacials 13a and 15e are close to the same temperature, but the glacial maximum before 15e was far deeper, and 15e didn’t get as warm as 11c either.
I suspect that shifts in orbital resonance patterns ordering a solar variability component has increased the variability and terminated the ~41kyr dominance. The periods that I noted in my comment at March 8, 2019 3:18 pm do not belong to Earth’s orbital variations, though where they do roughly coincide with the effects of Earth’s orbital variations they could combine constructively and act to increase the variability.
Having given this further study, I can see that the interglacials are constrained by obliquity cycles in the long term, and the regular ~370kyr compound period would be when obliquity cycles and a solar cycle component are in tighter phase. The solar-obliquity phase relationships could then account for why warming occurred only at certain peaks in obliquity, for the variance of interglacial peaks from the obliquity peaks like between 15a and 7e, and provide an explanation for the neat symmetry around MIS 11.
I have found the 115kyr period. It would be a product of a 1726.62yr cycle of centennial solar minima, and the 4627.5yr grand synodic period of all four gas giants, and with both phase coherent with Earth-Venus syzygies. The latter would modulate the former. A 67:25 ratio with a 4 year error and close to 115684 years.
In some epochs every other 1726.62yr node of the solar minima cycle is displaced due to the non-circularity of the orbits, resulting in a 3453.24yr cycle instead. Which is apparent in the GISP2 series as a repeating pattern of warm and cold events at that pitch.
The 370kyr compound interval that I noted by solar minima cycles would be 214*1726.62 = 369947 years, which when divided by 9 gives 41055.25 yrs rather than the quoted 41040 yrs for the mean obliquity period. Deducting the 115684 and dividing the remainder by 3 gives 84604.5 years. That would predict that the time between interglacials 7e and 15a is 4*84604.5 = 338.4kyr rather than the 328.3kyr with 8 mean obliquity cycles.
typo, 369497 not 369947 years. (5.20am)
There is a disparity of around 46.6 days between 2160 Earth-Venus synodic periods and 250 Jupiter-Uranus synodic periods at 3453.1116 sidereal years in the modern epoch. This would result in losing one Earth-Venus synodic period about every 43 thousand years to maintain parity if the orbital periods were stable.
The 4627.33 year cycle of all four gas giants does slip out of sync, and would self correct with a smaller step at points in a similar manner. I do not have that calculated yet, but I suspect that the finer harmony at 115684 years (-24yrs) will still hold. It is interesting to note that all four gas giants in finer inferior conjunction occurred during the Holocene in 3322 BC and 1306 AD. It would have to take several tens of thousands of years for a series of those to reoccur.
I wonder if clouds and precipitaion should not be a part of the analyzing.
I am not sure they can be averaged out even over 20 to 100 thousand years.
Perhaps this will shed some light on why the authors found that precession had less effect than they expected.
I would be interested to reread the article with precession removed from consideration.
Like pizza without the cheese, meat and sauce?
Precession of the earth does not exist, or at least it is minor. The solar system precesses. Read through the link.
Synchronisation of driven nonlinear oscillators:
Milankovitch forcing is best understood as a periodically forced nonlinear oscillator. Before the MPR ~ 1Mya, the forcing was strong and thus the system was “mode-locked” such that the interglacial pacing matched closely that of obliquity – with a suitable lag for ocean heat “inertia”.
But after about 1 MyA, at the MPR, the forcing weakened and mode-locking ceased. Now the pacing of interglacials shows the more complex pattern characteristic of weakly, rather than strongly forced (and mode-locked) nonlinear oscillation.
What are the Melnikov functions of precession, obliquity and eccentricity?
This would be a good starting point.
There is another way to interpret the LR04 “ice volume” data – namely as a fall in global sea level.
The threshold between the 41k world and the ‘100’k world now appears as a minimum in sea-level instead of a maximum in ice cover. A greater expanse of exposed land (initially mud flats) combined with increasing aridity then prime the next (obliquity+precession) NH summer maximum to melt back the ice sheets, aided by thousands of years of dust reducing albedo.
High obliquity shifts the 18.6 year tidal Maxima to higher latitudes, increasing diurnal tidal flows around ice sheets.
The eccentricity cycle looks a lot like the harmonic mean of two and three obliquity periods, i.e. the square root of 82*123 kyr = 100 kyr.
Correction – geometric mean.
On that basis, inclination matches the geo. mean of one and three obliquity periods at 71 kyr.
The problem with simple answers (the Occam’s razor fallacy for complex systems) is that it obviates the need to ask simple questions.
This from a study suggesting an association between large terrestrial lifeforms and the development of an ozone layer. Note the steadily increasing solar intensity.
If solar intensity is increasing – what causes long term cooling? Why extreme change and shifts in mode? Why do post MPT interglacials terminate?
Nor – on another point – can we generally correlate reductions in species richness with global temps. The Ordovician-Silurian extinction, about 440 million years ago, is the one most clearly linked to cool conditions.
The world today is not conspicuously deficient in flora or fauna. The wonder of our blue-green planet is the diversity of its habitats and life will find a way. But climate is an emergent property as tremendous energy cascades though powerful subsystems – collectively hot or cold – and the future is unknowable. Our best policy is to build prosperous and resilient communities in vibrant landscapes. Ironically – this is the path to reduce and reverse human emissions of greenhouse gases and aerosols to the atmosphere in our dynamically complex Earth system.
Pingback: Weekly Climate and Energy News Roundup #351 | Watts Up With That?
“In the most simplistic interpretation of solar-driven ice ages, one might expect that ice ages would occur every 11,000 years in alternate hemispheres, when the solar input to higher latitudes is a minimum in that hemisphere’s summer, allowing large ice sheets to develop. One might then expect ice ages to occur alternately in each hemisphere, every 11,000 years – in line with the hemispherically alternating precessional cycle. We do not observe this at all“
Found this the most entertaining comment.
For some reason we all expect both poles to melt in a warming world. But what is a warming world when one pole can go colder and the other hotter at the same time.
Why if the theory is correct does the result not follow?
Continental drift and weight change with thick continental glaciers affecting the precession and obliquity and albedo position are not even touched on though they may be just as relevant as dust to ending glaciers.
EM is right, change can occur quickly and unexpectedly.
There are too many unknowns to do any form of meaningful pattern prediction.
That is so backward. ice sheets form in warm times to counter the warming. ice sheets deplete in cold times to counter the cold. The poles do not become warmer and colder in these cycles each pole promotes evaporation and snowfall in warm times to counter it and promote less evaporation and snowfall in cold times to counter it, Temperatures in both hemispheres stay well bounded in spite of the large changes to energy in.
EM is right, change can occur quickly and unexpectedly.
There are too many unknowns to do any form of meaningful pattern prediction.
Change should be expected, what happens now and next repeats the change that has occurred over the last ten thousand years. If you are surprised, you have not studied ice core data and history. What has happened will happen again. What has not happened in the past is really stupid to expect in the near future.
popesclimatetheory | March 11, 2019 “That is so backward. ice sheets form in warm times to counter the warming. ice sheets deplete in cold times to counter the cold.”
While not having read a lot of your posts I get the gist of your belief that feedback loops exist that automatically tend to slow down and reverse actions or changes in the climate.
I do take exception to the way you have worded your idea above. It is not even counter intuitive. In effect you have forced yourself to say that heat causes ice sheets to form and cold causes melting.
Science and scientific explanations just do not work like that.
Secondly the use of the expression “to counter” is also wrong. It implies an element of anthropomorphism to your argument. The idea that there is a purpose to the action, as if nature has decided to act in a certain way.
Thawed north polar oceans provide evaporation and snowfall on Greenland.
Thawed south polar ocean provide evaporation and snowfall on Antarctica.
in cold times the polar oceans are frozen and there is no evaporation and snowfall, ice flows and dumps in the oceans and ice depletes in cold periods. Lake effect snow occurs when lakes are warmer and thawed. Oceans effect snow occurs when oceans are warmer and thawed. Frozen water does not allow evaporation and does not cause snowfall.
It is really amazing how many people do not understand simple science.
While not having read a lot of your posts I get the gist of your belief that feedback loops exist that automatically tend to slow down and reverse actions or changes in the climate.
Yep, that is why every warm period in data from all of the last fifty million years ended with colder. That is why every cold period in data from all of the last fifty million years ended with warmer. Tropical climate is like this in that when it is warmer, the temperature to the fourth power does cool the tropics more and less as temperature changes but there are not over shooting cycles involved. In Polar climate, there are ice cycles that over shoot. It snows until there is too much ice volume, but it keeps snowing until oceans freeze. Then it snows too little , but it does not start snowing until the oceans thaw. This ice feedback causes cycles in polar climates. In major ice ages most of the extra ice was placed to far from the north pole and it all did thaw and reenter the oceans. Each cycle kept some ice sequestered on Antarctica and now there is not enough available for a major cycle. Therefore, we now have a new normal.
“What has happened will happen again. What has not happened in the past is really stupid to expect in the near future”.
Depends on whether you are talking about the sun coming up in the morning or having tea ready on the table when you get home I guess.
Taleb gives an example of the Turkey on Thanksgiving day.
Does not expect it’s head to be cut off after all those days of free food and water.
No ice core data for the dinosaurs? So we can only deal with 10,000 year chunks? The argument seems to have a problem in there is a pattern in this segment of time but not in others.
Ice core data covers 800 thousand years and in that 800 thousand years this is the only ten thousand year chunk that looks like the one we are in, This is the new normal. There has never been this much ice sequestered in the cold places where the ice did not thaw and reenter the oceans. less ice and water is being exchanged between warm periods and ice ages. New smaller warm periods and smaller little ice ages is the new normal in both hemispheres. These are not synchronized with each other. different amounts of ice and water are taking part in cycles in each hemisphere and it takes different amounts of time. One part of a Milankovich cycle has removed a lot of energy into the far north above 60 degrees and added it to the far south below 60 degrees and the ice sequestered in each hemisphere was adjusted to counter the difference. it over came much more watts per meter squared change than CO2 is reported to cause.
The argument seems to have a problem in there is a pattern in this segment of time but not in others.
No, the argument seems to have a solution in that there is a pattern in this ten thousand year segment of time that is not in any others over 800 thousand years. We do live in the best of times.
The ice is finally right. Glaciers in the north have decreased as the north needed less ice for cooling. Ice on Antarctica has increased as more solar in came the south. The answerers were given to us by analysis of ice cores. Much time and money has been spent studying only the influence of CO2 and trying to stop something that has helped life by growing our food better using water more efficiently.
Thanks for the interesting comment.
Thanks. You are one of the few calm people here. I am not. Appreciate the measured response despite my difference of opinion in some areas.
Pingback: Smutsig luft, tipping points och klimatkänslighet - Stockholmsinitiativet - Klimatupplysningen
I don’t understand the concern about “recent rapid global warming”. The warming from 1910-1941 was as steep as the current. As was the warming from 1850-1880. And the rise after the Younger Dryas 12000 BP was much much steeper.
We’ve been cooling since the Holocene Optimum 6000 years ago, with interval warmings at the Minoan, Roman, and Medieval times, and the current.
I also don’t understand the reference to Ice Ages in the last 800,000 years, since we have been in an Ice Age for the last 2.6 million years, with perhaps eight glaciations in the last million, interspersed with interglacials including the current. These can be further subdivided into stadials and interstadials for isotopic analysis.
I don’t see any evidence that Peter Lang is wrong. Help me out.