A Deterministic Forecast of Future Climate Changes

By Renee Hannon


In my previous WUWT post, I proposed a conceptual process model describing the influences of astronomical controls on interglacial warm periods and glacial cycles (Figure 1). The warm onsets with rapidly increasing temperatures are triggered by an increase in the eccentricity of Earth’s orbit, an increase in obliquity and an increase in summer insolation. The nearly concurrent increase in these astronomical forces is a necessary component for a significant interglacial warm onset. Obliquity is the dominant control on duration and cooling of these major warm periods (Javier). Summer insolation which retains the higher frequency signal of precession may modify warm duration patterns or influence stadial and interstadial occurrences.

Figure 1: Conceptual process model of Milankovitch cycles influence on interglacial warm period temperature variations. Horizontal axis is relative time for kyr. Warm onset occurs during increasing Eccentricity, Obliquity, and NH summer insolation. Obliquity is the dominant control over cooling and thus, warm duration.

In this post, I attempt to address the conundrum “It is not a question if cooling will occur but simply a question of when”. By using a deterministic approach based on the astronomical process model and paleoclimate analogs, approximations are developed for future significant climate events. These events include the end of our current Holocene warm period, length of the current glacial cycle and onset of the next warm interglacial period. This deterministic forecast is not intended as a prediction of the future. Rather, it is an attempt to investigate a set of discrete assumptions using astronomical forces as the key trigger for major climate temperature events.

A deterministic model helps tune primary controls and evaluate a range of outcomes for a climate forecast. This skeleton climate framework displays large scale global temperature patterns that may enable others to infill additional primary and secondary higher frequency temperature events.


Antarctica Dome C ice core deuterium isotope ratios converted to temperatures are used as the central dataset for this study. These data were chosen because less temperature sensitivity to northern hemisphere events are evident. This allows for identification of global temperature events versus higher frequency temperature variations such as the North Atlantic Ocean oscillations as seen in the Greenland ice core data.

Calculated temperatures from the Antarctica Dome C data are multiplied by 0.5 to approximate global temperatures rather than polar temperature regions. Uncorrected, the magnitude of the delta degree C would be double than what is stated throughout in this post.

Past and future orbital parameters of the Milankovitch Cycles for eccentricity changes in Earth’s orbit, obliquity or tilt of Earth’s axis, and precession which affects seasonal distribution of radiation are from J. Laskar (2004). Insolation or solar radiation calculations from these orbital parameters for various latitudes and seasons are also from Laskar (2004). Northern Hemisphere summer insolation at 65 degrees which is strongly influenced by precession is used in this post.

Past Glacial Cycles Overview

Glacial cycles over the past 450 kyrs are demonstrated to vary in duration from 89 to 119 kyrs and consist of a warm interglacial period, a mild glacial period and a full glacial period as shown in Figure 2. This variation in past glacial cycle durations correlates to eccentricity and northern hemisphere summer insolation cycles (see Hannon Figure 8).

Figure 2: A traverse of the past five interglacial-glacial cycles is datumed (the datum for the curves is the red line at the bottom of the traverse) on the Warm Onsets. EPICA Dome C isotope temperature estimates are plotted as curves in 1 degree C increments on the horizontal scale (cold to left and warm to right). The vertical scale is time in 20 kyr increments. Actual age is plotted on each cycle. Interglacial warm periods are highlighted in red and the coldest portion of the glacial period in blue. Dark red is used above Δ-1°C, this cutoff is also used to calculate the warm period duration.

The key glacial cycle event is the termination of the full glacial period and the onset of the warm interglacial period. This event bounds each glacial cycle as shown in Figure 2. This has happened five times over the past 450 kyrs. These warm periods exhibit an onset of rapid warming with significant global temperature increases of 5 to 7+ degrees C in just 5 to 7 kyrs. The cooling of warm periods is more gradual with temperatures dropping by 3 to 4 degrees C over 8 to 14 kyrs. The Holocene present day is currently within a warm interglacial period as it has been for the past 12 kyrs.

Several smaller warm events (interstadials) and abridged cooling events (stadials) which last only thousands of years occur within each warm period and mild glacial period. Many of the Holocene intermediate events include the Boiling/Allerod warming interstadial followed by the Younger Dryas Cooling Stadial, and the intervening 8.2 kyr Stadial cooling during Holocene optimum. The phasing of obliquity and higher frequency summer insolation influences the warm duration patterns and interstadials/stadials during the mild glacial periods.

Forecast Assumptions

A deterministic model is constructed for the current glacial cycle using astronomical data as a control and paleoclimate analogs over the past 450 kyrs (Figure 3). Astronomical cycles are reasonably well synchronized during the next 100 kyrs; meaning several insolation and obliquity highs and lows are close in timing. The deterministic model developed for the current Holocene interglacial duration and future cooling as well as the next interglacial warm onset uses these primary assumptions:

  • Astronomical curves are shifted by 6.5 kyrs as determined by Javier. This shift advances the curves in time and improves correlation to Earth’s past temperature changes. Astronomical forces initiate dramatic temperature changes that are buffered by Earth’s expansive oceans causing a delayed reaction to temperature shifts. Additionally, deep oceans are becoming colder and more energy is necessary to exit a glacial state as recently discussed by Röst. The delay of global temperature changes due to solar impacts warrants additional investigation. Shifting the astronomical curves is probably the largest uncertainty in the development of this deterministic forecast.
  • Temperature variations due to oceanic, atmospheric and geologic events have not been specifically incorporated. These include ocean oscillations, greenhouse gases, meteor impacts and volcanoes. These factors should be captured within bounds of the deterministic cases presented here. Their omission is a secondary uncertainty within this deterministic forecast.
  • The previous four interglacial warm periods provide temperature analog data, boundary conditions for warm period durations, as well as calculated slopes for warming and cooling trends.


Figure 3: Projected Holocene warm duration, cooling descent and subsequent interglacial onset curves plotted with Milankovitch cycles. Dome C temperature curves are in red with scale on left hand side. Astronomical cycles in blue with various scales on right hand side. The bottom horizontal axis is time in kyrs and the top horizontal axis in blue contains the M cycle values.  Calculated cooling and onset curves based on past Warm periods slopes in dashed lines with best estimate in red solid line.

Present-Day Warm Duration Forecast

Three discrete cases are constructed for the duration and eventual cooling of the Holocene interglacial warm period. Deterministic estimates are built for a minimum (P10), maximum (P90) and best estimate cases. The P10 case expects that less than 10% of the time a case could occur where cooling will be sooner. The P90 case expects that less than 10% of the time cooling will begin later. The best estimate case usually falls near the mean of the distribution.

Cooling slopes calculated for the past warm periods are shown in Table 1. Warm IV cooling slope is used for the minimum case and Warm V cooling slope for the maximum case as they are endmembers for the four past cooling slopes. Warm IV’s duration of 14 kyrs is used for the Holocene minimum warm duration case suggesting cooling will begin within a few hundred years. The minimum case cooling descent is bounded by summer insolation which is rapidly decreasing (Figure 3). Obliquity is also decreasing but has not decreased below 23.5 degrees tilt.

Table 1a: Interglacial durations (greater than minus 1 degree C) and cooling slope calculations (Hannon).

Interglacial Period

Warm Duration (kyr)

Cooling Slope

Warm I (Holocene)


see Table 1b

Warm II (MIS 5)



Warm III (MIS 7e)



Warm IV (MIS 9)



Warm V (MIS 11)



Table 1b: Deterministic cases for projected Holocene Interglacial duration and cooling slopes. Duration is defined as greater than minus 1 degree C.

Holocene Warm I

Warm Duration (kyr)

Cooling Slope

Minimum Case



Best Estimate



Maximum Case



The best estimate case for the Holocene warm period uses a duration that fits obliquity when tilt begins decreasing below 23.5 degrees (Figure 3). The Holocene warm duration, in the best estimate case, is interpreted to be 19-20 kyrs long and will begin a cooling descent about 4 kyrs from present day. The cooling descent is expected to last approximately 10 kyrs as observed in past warm endings which would result in reduced global temperatures of 3-4 degrees C.

The maximum case for the Holocene warm duration is 27 kyrs with a cooling descent about 8+ kyrs from present day. This case roughly corresponds to decreasing eccentricity. Obliquity and insolation have already decreased. The maximum case is expected to be shorter in duration than Warm V. Warm V has a pronounced insolation increase during the middle of its warm duration which extended the warm period. A strong insolation increase is not present during the later portion of the Holocene warm period and therefore, Warm V’s extended warm duration was not used as a reasonable maximum case. However, it would not be surprising to see an increase in average global temperature of 1-2 degrees C during the remaining Holocene warm period. This is because our current interglacial warm period has been cooler than the warm optimum of past warm periods by 1-2 degrees C (Hannon).

Glacial Cycle I and Future Warm Onset Forecast

Significant warm onsets occur when eccentricity, obliquity, and summer insolation are concurrently increasing. Therefore, the potential onset of the next warm interglacial period was positioned at approximately 62 kyrs (+/- 3 kyrs) from present day (Figure 3). This assumes one obliquity cycle is skipped because eccentricity is decreasing and near circular during this time. Note the future warm onset has both obliquity and insolation increasing simultaneously.

The slope of the future Warm onset at 62 kyrs was constructed in a similar fashion as the cooling slopes and utilizes observations from warming slopes during the past 450 kyrs (Table 2). Only two discreet cases are developed for this warm onset slope; Warm II slope is the minimum case and Warm V slope is the maximum case. Since obliquity and insolation are increasing at the same time, this may result in an onset more like Warm II rather than Warm V. The steep slope of Warm III was not used because it was initiated during a predominately elliptical eccentric cycle where obliquity and insolation are amplified.

Table 2: Warm Onset Slopes for the past 450 kyrs and cases used in deterministic onset slope calculations.

Warm Period

Warm Onset Slope

Future Warm Onset Slope

Warm II (MIS 5)


0.89 min Case

Warm III (MIS 7e)


Warm IV (MIS 9)


Warm V (MIS 11)


0.68 max Case

If the future interglacial warm onset occurs in 62 kyrs and the onset of the current Holocene Warm period was 17 kyrs ago, then the duration of Glacial Cycle I will be approximately 79 to 80 kyrs. This will make the current glacial cycle the shortest glacial cycle in the past 450 kyrs (Figure 2). This is plausible because current eccentricity cycle is predominantly circular and shorter in duration than the past three eccentricity cycles.

As demonstrated in my previous post, eccentricity and summer insolation cycle length correlate reasonably well with the duration of glacial cycles. Glacial Cycle V occurs when eccentricity cycles are predominately circular. Glacial Cycle V is a short glacial cycle and lasted only 89 kyrs shown in Figure 2. Because Earth is currently in a predominantly circular orbit, Glacial Cycle I is more likely to be analogous to Glacial Cycle V with a potentially shorter glacial cycle. Predominantly circular eccentricity orbits occur approximately every 400 kyrs which includes the present-day Glacial Cycle I, Glacial Cycle V (MIS 11) and MIS 19 glacial cycle as discussed by Javier.

A Conceptual Forecast

Now that the key framework onset and cooling components for the future Glacial Cycle and Holocene warm period duration have been deterministically constructed, the conceptual forecast showing various glacial periods and phases are interpreted as shown in Figure 4. For simplicity, only the best estimate for the Holocene warm duration and cooling are plotted.

Figure 4: Forecast for the Holocene warm period and Glacial Cycle I overlain on obliquity and summer insolation. Glacial periods and components are shown as colored bars in the middle of the graph and are approximations with a +/- error bar of 3 kyrs.

The mild glacial period and glacial maximum are estimated for the current Glacial Cycle I by using key astronomical forces as a guide. After the current warm duration and cooling, a mild glacial period is expected to follow. Several interstadials and stadials are interpreted to occur during the mild glacial period caused by increases and decreases in summer insolation. Around 30 kyrs from now both obliquity tilt along with summer insolation increase perhaps creating a substantial interstadial. Interstadial and stadial events can be correlated within past mild glacial periods suggesting a recurring and predictable forcing such as interplay between summer insolation and obliquity as shown in Figure 2 (yellow lines).

The next glacial maximum is placed at approximately 55 kyrs from present when both obliquity and insolation are decreasing in phase and reach their lows. Glacial maximum’s can last up to 10 kyrs in duration. As discussed, the future warm onset occurs at approximately 62 kys ending Glacial Cycle I’s glacial maximum. The warm duration and cooling for the future interglacial have not been calculated in any detail and are simply blocked in using obliquity as a guide.


A deterministic framework for forecasting the current glacial cycle is constructed using astronomical forces which are correlated to past paleoclimate isotope/temperature data. This skeletal framework, although uncomplicated in design, provides a basic forecast of future major climate changes within our current glacial cycle.

Based on the past four warm periods, the present-day Holocene warming is expected to range in duration from 14 to 27 kyrs with a best estimate of 19 kyrs. The best estimate case is also controlled by the obliquity cycle and results in a Holocene cooling descent beginning in about 4 kyrs.

As seen in past warming onsets, an increase in eccentricity, obliquity, and summer insolation are required. The next concurrent increase of these astronomical cycles is interpreted to initiate a future interglacial in approximately 62 kyrs from present day. This will terminate the current glacial cycle making it one of the shortest glacial cycles in the past 450 kyrs lasting approximately 80 kyrs. A short glacial cycle is a reasonable forecast given the current eccentricity cycle is a predominately circular orbit like Warm V (MIS 11) and MIS 19 which also have short glacial cycles.

This model is one scenario and intended more to stimulate discussion on important climate premises rather than be a definitive forecast. Key uncertainties include the 6.5 kyr shift in astronomical data and exclusion of other climate drivers. The model does illustrate how astronomical forces can be used to help explain both observed and forecasted temperature variations that significantly impact our planet.

Acknowledgements: Special thanks to Andy May and Donald Ince for reviewing and editing this article.


90 thoughts on “A Deterministic Forecast of Future Climate Changes

  1. Think you mean “dominant” instead of “dominate” modifying “control” in the first paragraph.

  2. An estimate of several thousand years until the next Ice Age does not seem that bad, unless the green blob stops all technological progress, as they seem wont to.

    • Fusion-powered, giant blow-driers aimed at snow in Canada should do the trick.

      Or snow removal on a massive scale.

      • Gloateus–you just gave me a flashback, and it wasn’t very pretty. Growing up in Philadelphia during the 50’s-early 60’s the plowed snow would remain piled up well into spring–April and if memory serves me right even early May. And of course, it was black. Just wondering–if the atmospheric pollution hadn’t covered the snow with soot, and the effect of the UHI even back then hadn’t been a factor–could those piles of snow have possibly lasted clear through the summer into the next snow season???

  3. To me the worst and most likely Climate catastrophe is not warming but the impending decent into Little Ice Age conditions which with a quietening sun seems wholly likely this century, next century or this millennium.

    Maybe as soon as next decade. Some of the signs seem to be already there.

    That could be a true climate catastrophe.

    According to ice core records, the last millennium 1000AD – 2000AD has been the coldest millennium of our current Holocene interglacial.

    But it seems that, driven by the need to continually support the Catastrophic Anthropogenic Global Warming thesis / religion Climate scientists and Climate alarmists examine the temperature record at too fine a scale, weather event by weather event, month by month, year by year.

    Our current, warm, congenial Holocene interglacial has been the enabler of mankind’s civilisation for the last 10,000 years, spanning from mankind’s earliest farming to recent technology.

    Viewing the Holocene interglacial is much more fruitful, on a century by century and on a millennial perspective.
    Each of the notable high points in the Holocene temperature record, (Holocene Climate Optimum – Minoan – Roman – Medieval – Modern), have been progressively colder than the previous high point.
    The ice core record from Greenland for its first 7-8000 years, the early Holocene, shows, virtually flat temperatures, an average drop of only ~0.007 °C per millennium, including its high point known as the “climate optimum”.

    But the more recent Holocene, since a “tipping point” at around 1000BC, 3000 years ago, has seen temperature fall at about 20 times that earlier rate at about 0.14 °C per millennium.
    The Holocene interglacial is already 10 – 11,000 years old and judging from the length of previous interglacial periods, the Holocene epoch should be drawing to its close: in this century, the next century or this millennium.

    Nonetheless, the slight beneficial warming at the end of the 20th century to a Modern high point has been transmuted by Climate alarmists into the “Great Man-made Global Warming Scare”.

    The recent warming since the end of the Little Ice Age has been wholly beneficial when compared to the devastating impacts arising from the relatively minor cooling of the Little Ice Age, which include:
    • decolonisation of Greenland
    • Black death
    • French revolution promoted by crop failures and famine
    • the failures of the Inca and Angkor Wat civilisations
    • etc., etc.

    As global temperatures, after a short spurt at the end of the last century, have already been showing stagnation or cooling over the last nineteen years or more, the world should now fear the real and detrimental effects of cooling, rather than being hysterical about limited, beneficial or probably now non-existent further warming.

    Warmer times are times of success and prosperity for man-kind and for the biosphere. For example during the Roman warm period the climate was warmer and wetter so that the Northern Sahara was the breadbasket of the Roman empire.

    But the coming end of the present Holocene interglacial will eventually again result in a mile high ice sheet over much of the Northern hemisphere. As the Holocene epoch is already about 11,000 years old, the reversion to a true ice age is becoming overdue.

    That reversion to true Ice Age conditions will be the real climate catastrophe.

    With the present reducing Solar activity, significantly reduced temperatures, at least to the level of another Little Ice Age are predicted quite soon, later in this century.

    Whether the present impending cooling will really lead on to a new glacial ice age or not is still in question.

    This point is more fully illustrated here:

    • IMO this century is too soon for another LIA. Prior warm periods have lasted a few to several centuries. The current one only started around AD 1850.

      • Still, its good to be alive in the Modern SLIGHTLY WARM Period.

        I do wish people would appreciate that fact, and stop trying to take us back to the dismal and desperate LIA !!

      • The geologic record shows many warming and cooling periods, mostly lasting about 25-30 years (i.e. ~60 yr cycles)–not hundreds of years. 1850 to 1880 was a warm period, but 1880 to 1915 was a profound cool period during which many cold temp records were set in the US. 1915 to 1950 was a warm period slightly warmer than the most recent one (1980-2000), and 1950 to 1978 was a strong cool period. As we head into another cool period, it would appear that we are not yet out of the Little Ice Age.

      • Perhaps we should have multiple scenarios we should be planning for instead of focusing on just an Anthropogenic warm scenario. History lessons from the LIA would be good preparation for an eventual cooling scenario.

      • Cracker, i started taking an interest in all this “climate crap” right around the time that AIT came out. i used to go to wikipedia to get global temps (and read about attribution, etc). Back then the data showed .4C warming, .2C cooling and .4C warming over the breadth of the 20th century. And i’d go there every january just to get the latest data point. (this, of course, was the beginning of all the talk of ‘the pause’ being the latest buzz) After several years, i began to lose interest in wiki, so i stopped going there. Eventually, these latest interpretation data sets began cropping up where e’er i went. A feeling of orwellian creepiness came over me that has stuck with me to this day. Believing in these latest data sets is a leap of faith that’s a little too great for yer humble (and generally faith filled) fonz. What’s worse is that my scant efforts to find out why have usually ended up going nowhere. It’s as though the past has been erased, so there’s really nowhere to go to find out about it. A recent post by stokes has got my interest in it up again. i figure there has got to be a decent critique of the evolution of these data sets somewhere. It would be interesting to see how data sets evolved even before a decade ago (as it appears that i came late to the “adjustocene” party). And then, of course, there’s the “callender consp*racy” where it’s thought that the godfather of agw sought to manipulate data collection to get a faux .75C/century warming trend to prop up his (then) failing theory. (so, there’s a lot to keep this layman busy for a while)…

      • Don Easterbrook wrote:
        “The geologic record shows many warming and cooling periods, mostly lasting about 25-30 years (i.e. ~60 yr cycles)–not hundreds of years. 1850 to 1880 was a warm period, but 1880 to 1915 was a profound cool period during which many cold temp records were set in the US. 1915 to 1950 was a warm period slightly warmer than the most recent one (1980-2000), and 1950 to 1978 was a strong cool period. As we head into another cool period, it would appear that we are not yet out of the Little Ice Age.”

        I tend to agree with you Don, particularly re the “strong cool period” from ~1950 to ~1978, I wonder if Earth is about to revisit that moderate global cooling again within the next decade, as SC24 and SC25 are both expected to be weak.

        We predicted this moderate global cooling to start by ~2020-2030, in an article published in 2002. To be clear, this is a much less severe cooling than is being discussed in the posted article, but it could cause significant disruption and hardship in a world where energy systems and agricultural policies have been significantly compromised by false global warming alarmism.

        To be clear, I do not believe that increasing atmospheric CO2 is causing significant global warming, because ECS is very low. I am not at all worried about imminent global warming, but I am concerned about imminent moderate global cooling. I hope to be wrong about this cooling prediction.

        Best, Allan

  4. Observing figure three I note two interesting things I will call Lag and Limit:

    1. Lag
    Eccentricity and obliquity determine insolation. It is changes in insolation that causes temp. to change. Since the insolation rise at 65 degrees north lags the temperature rise, some other factor must be involved. Perhaps the insolation plot for a different latitude has a better match? Or is there simply a time error in the plot?

    2. Limit
    The temperature rise levels out about 12K years ago even though eccentricity and obliquity factors continue to cause an increase in insolation at 65 degrees north for several thousand more years. I would have expected temp to rise in corolation with increasing insolation until continental ice sheets began to melt, to run more level until melt-back ends, then rise again. Instead, temps rose as ice sheets melted, then stabilized. Any suggestions out there by other readers as to what limiting factor prevented further temp. rise? I think warmer oceans causing an increase in cloud cover may be the dominant factor. Unless the timeline is all wrong.


    • What you are calling limiting factors, probably means negative feedbacks in engineering terms. The primary one is the radiative Planck f/b .

      The observation that obliquity is a trigger ( which I remain less than convinced about ) also requires positive f/b to make it powerful enough to break out of a glaciation. It is possible to propose +ve f/b , like CO2 out-gassing from the oceans , increased atm water vapour that could lead to a ( geologically ) rapid swing from glaciated state to interglacial.

      Positive f/b either cause a system to go unstable or are dominated by more powerful negative feedbacks. The evidence of long term stability is clear evidence of the latter.

      Some +v/e f/b causes the rapid flip but is ultimately dominated by Planck radiative f/b

    • Stevan. Addressing 1. Lag. My understanding is that precession dominates the 65 degrees NH summer insolation calculation over Eccentricity and obliquity. Therefore it has a higher frequency cycle than obliquity and can be slightly offset as seen in the Holocene interglacial onset. Perhaps the slight delay of increasing insolation is what brought us out of the Younger Dryas stadial event? Just a thought.

      • this idea of using a specific latitude at a specific time of year because it is convenient to build a hypothesis worries me more than a little as a logical process.

        This seems to be current practice in supporting Malankovich forcing but is never justified.

        So , fine, if insulation is greater in that specific time and place why does that matter and why you get away with ignoring the rest of the year and the rest of the planet, which is actually doing the opposite but apparently that does not matter. The reason it does not mater is so evident that it does not need justifying !?

      • Perhaps the slight delay of increasing insolation is what brought us out of the Younger Dryas stadial event? Just a thought.

        No, I don’t think so. There are many processes at work and there is much too much time wasted by people who latch into a specific hypothesis at the exclusion of all else. Everyone wants : it’s CO2; it’s the sun ; it’s orbital cycles ….

        The transition between the two states : glacial and interglacial is the unstable regime between two quasi-stable states. During this transition it can snap back and forth because it is a regime where +ve feedbacks are significant.

        A trigger event starts deglaciation, CO2 outgassing provides a +ve f/b and more warming. LIfe wakes up and eats lots of CO2 causing cooling which reabsorbs CO2 and causes rapid cooling.

        There are a number of other processes. Don’t expect a single variable or even two variable explanation to work.

      • I agree with Greg August 19, 2017 at 12:42 pm in doubting that insolation level at a specific latitude and time of year is the controlling factor. Surely total insulation on a hemisphere corolated with time of year matters more.


      • Greg,

        “There are many processes at work and there is much too much time wasted by people who latch into a specific hypothesis at the exclusion of all else.”

        Nobody thinks there is only one factor at play. However:

        1. Interglacials do not take place outside the rising->decreasing half obliquity cycle. Ever.

        2. Falling obliquity always puts an end to interglacials. No exceptions.

        3. During most of the Pleistocene interglacials take place at 41 kyr obliquity intervals.

        These three simple facts demonstrate two things:

        A. Obliquity is the main determinant for interglacials. All the rest are secondary or feedbacks. We only have to explain why some obliquity increases do not produce an interglacial and the answer is trivial: obliquity alone is not enough.

        B. One can never underestimate the capacity of scientists to come up with the wrong answer to the most evident question. By applying an inappropriate mathematical procedure they came up with the wrong number: 100 kyr, messing up the entire field for decades and teaching each other to persevere in the error. They will also resist to the death correcting their mistake even when confronted with clear evidence.

        “LIfe wakes up and eats lots of CO2 causing cooling which reabsorbs CO2 and causes rapid cooling.”

        This is wrong. Glacial cooling is slow and progressive over tens of thousands of years and with a huge disconnect with CO2, as the cooling can take place for 14,000 years without any decrease in CO2, as it happened after the Eemian.

      • this idea of using a specific latitude at a specific time of year because it is convenient to build a hypothesis worries me more than a little as a logical process.

        This seems to be current practice in supporting Malankovich forcing but is never justified.

        Again lack of knowledge. 65°N summer insolation was chosen because ice sheets occur in the NH, not the SH, and they reach 65°N at maximum extent. It is considered that the melting of ice sheets starts during the summer at its southernmost border and then it is pushed North by feedbacks like decreasing albedo that enhance the effect. Thus maximal insolation at 65°N during the summer is considered determinant to start the termination.

        It ends up not working like that but the reasoning is not faulty nor unjustified.

      • Javier, how much of the 4-5C rise in temperature from the last glacial to the current interglacial would you attribute to CO2?

      • Afonzarelli,

        I think it would be easier to determine the contribution of CO2 to current global warming, as CO2 change is much higher for the temperature change, and there are less confounding factors. And nobody has been able to do it convincingly.

        I do not have any idea of how much of the warming was due to CO2 feedback. An upper limit can probably be placed at about 1/3 of the glacial-interglacial warming (assuming all modern warming since 1950 is due to CO2), but the lower limit is completely unknown.

      • Does even 1/3* sound feasable? Given massive ice albedo, water vapor feedbacks (and whatever else) should we expect CO2 to compete with those?

        *(i get the 1/3; half way to a doubling equals 1.5C)

    • Greg August 19, 2017 at 12:23 pm

      Yes, I would call “warmer oceans causing an increase in cloud cover” a definite negative feedback.

      However, radiative Plank f/b would have given the Earth a higher equilibrium point once insolation ceased increasing. On graph 3 temperature ceased rising before insolation did.


      • Steve , why do you put that in quotes when it is NOT what I wrote. I said WV would act as a +ve f/b not a negative one. If you want to propose cloud formation that is yet another aspect Then yo need to look at what cloud type , what altitude and whether it has a warming or cooling effect.

        Our best knowledge seems very inadequate on all these questions which is why they have to much freedom to tweak model output by fudging all these poorly constrained parameters.

      • Greg August 19, 2017 at 12:55 pm

        Sorry if i mislead. “warmer oceans causing an increase in cloud cover” was what I wrote.

        Do you have an opinion on my response concerning radiative Plank f/b?


      • Greg, I agree that WV, when transparent, is indeed a GHG. Its strongest effect is slowing cooling at nightime. I also think condensation of WV into droplets greatly reduces daytime heating, overpowering any GHG effect.

        I think CO2 plays a very minor role in determining temperature.


    • Stevan,

      “1. Lag
      Eccentricity and obliquity determine insolation. It is changes in insolation that causes temp. to change. Since the insolation rise at 65 degrees north lags the temperature rise, some other factor must be involved. Perhaps the insolation plot for a different latitude has a better match? Or is there simply a time error in the plot?”

      Insolation is determined by obliquity and precession. Eccentricity has a very small contribution, but its main role is allowing precession to occur. Insolation at 65°N is a red herring, Milankovitch used half-summer caloric content, because when insolation at 65°N is very high, summers are significantly shorter due to Kepler’s second law. Peter Huybers and Paul Tzedakis have been using entire summer energy calculations which are a much better temperature indicator and interglacial predictor.

      “2. Limit
      The temperature rise levels out about 12K years ago even though eccentricity and obliquity factors continue to cause an increase in insolation at 65 degrees north for several thousand more years.”

      We don’t have an explanation for that. The temperature profile of every interglacial is different, and not all have higher temperatures at the start. Other factors must be important for the final temperature determination. In the case of the Holocene I believe that its temperature potential was bigger, but was sabotaged by the Younger Dryas cold relapse at a most inopportune time.

      • This is what worries me about this whole process. Linking various variables by a ” sometime it works, sometimes it doesn’t ” basis is usually a pretty good indication you are chasing fairies. This is so often done with 11y solar cycles. Sometimes it aligns : go for it! Then when it doesn’t and it’s anti-phase , we look the other way.

        What this really indicates is that there is another driver of similar but somewhat different period.

        Much of what is claimed to be 11y solar is actually 9y period. It aligns every 50 or 60 years with solar then drifts out of phase again.

      • If you need simple easy answers then science might not be for you, as sometimes the answers get really complicated and take a long time to come.

        Part of your problems with the 11-year solar cycle might come from not knowing that is the average of several periodicities.

      • “Part of your problems with the 11-year solar cycle might come from not knowing that is the average of several periodicities.”

        I estimated dominant quasi-periodicities at 10, 10.8, 11.8, and 131 years. These come about because of rectification of Hale Cycle terms at about P1 = 20 and P2 = 23.6 years. Rectification splits these into P2*P1/(P2+P1) = 10.8 years, P2*P1/(P2-P1) = 131 years, P1/2 = 10 years, and P2/2 = 11.8 years.

      • Obviously, the 131 year estimate is most sensitive to precise knowledge of the underlying cycles, and has the widest error bars.

      • Thanks Javier, I’m familiar with spectral analysis and the variable nature of the solar cycle.

        Can you now show me a similar periodogram of some climate parameter which has a similar set of peaks. That would be specific enough to provide a very conclusive argument for a solar driver. I’m not aware of anything and I’ve done lot of spectral analysis on dozens of climate datasets.

      • That’s your particular requirement for your particular conviction. Most researchers find a lag in the climate response to solar forcing. Are you familiar with what a slightly variable lag would do to that periodogram’s peaks?

      • Javier
        I agree, a great post.

        I would have thought that eccentricity may have a significant influence on atmospheric circulation patterns. Both volumetric and direction. The effects on ice reduction could be significant.

        As the ice cover over continents reduces a natural amplification occurs accelerating the process.

    • Stevan

      It is possible that having Antarctica covered in ice is a factor. It may be ‘too big to melt’ and as long at the continent sits on the South Pole, it may affect global temps via a variety of means.

    • “Any suggestions out there by other readers as to what limiting factor prevented further temp. rise? I think warmer oceans causing an increase in cloud cover may be the dominant factor.”

      The physical property of water called, the Latent Heat of Fusion. It takes 144 Btu/lbs to either melt ice or create it for the state change. There’s your time lag. It takes 144 x the energy transfer to form ice than to merely cool liquid water.

      It takes only approx. 0.5 (0.53) Btu/lbs to lower the temperature of ice once formed, It takes 1 Btu/lbs to raise or lower the temperature of water in its liquid state.

      The Latent Heat of Evaporation is your limiting factor on the warm side as it takes 970 Btu/lbs to change water’s state from liquid to gas/vapor. Where does the heat come from??? The remaining body of water… If you evaporate one lbs of water, that means you pull 970 Btu/lbs from the remaining water or one degree Fahrenheit. We use this physical property of water in cooling towers for HVAC.

    • Tom, as your linked article refers to glaciation inception, I assume your reply was directed to Renee Hannon?

    • Tom,
      Thanks for the reference. Ganopolski’s climate model is one of the first models I’ve seen that history matches entire glacial cycles.

  5. Oh my, Renee. Using the past to try and predict the future? That’s not permitted in climate science, you know. You’re supposed to identify ONE factor that you suppose controls climate, then you write complex programs to simulate its supposed effects on climate, that require the biggest and fastest computers ever built (except the ones at NSA and GCHQ of course). Then you tinker with adding volcanoes and aerosols in at judiciously selected intervals, so that your program can simulate the past, then you re-run your programs (with no future aerosols or volcanoes) and – surprise!! – conclude that disaster awaits us, and it’s all our fault.

    Original thinking is not allowed. (/sarc)

    Seriously, great post. Very thought-provoking. One has to ask, why isn’t the climate establishment doing stuff like this? (and of course, we do know the answer to that one).

    • Then you tinker with adding volcanoes and aerosols in at judiciously selected intervals, so that your program can simulate the past, then you re-run your programs (with no future aerosols or volcanoes) and – surprise!! –

      You are on the right lines but you should have said you judiciously tinker with the magnitude of the volcanic forcing , using no other criterion than to engineer a similarity to the recently cherry picked part of the climate record which is a reasonably mono-tonic rise and has the potential to fit the AGW narrative.

      From Hansen et al 2002

      3.3. Model Sensitivity

      The bottom line is that, although there has been some
      narrowing of the range of climate sensitivities that emerge
      from realistic models [Del Genio and Wolf, 2000], models
      still can be made to yield a wide range of sensitivities by
      altering model parameterizations.

      for links to source papers:

  6. Great article, Renée.

    I agree with everything you say. I had also calculated 70,000 years to the next interglacial. A very long winter is coming to the planet and it will put a permanent wall of ice up in the North.

    The proposed solutions aren’t likely to work. The default situation during an ice age like the one the planet has been for over 3 million years is glacial conditions. Interglacials are short periods that require a conjunction of special circumstances. The planet is striving to get back to glacial conditions and it is a huge force.

    Global warming is ~ 350 years old and ending too. It should end by about 2100. In 160 years the 1500-year cycle is due (~ 2180 AD) and it should produce moderate cooling (~ 0.1-0.2°C) then temperatures should start decreasing very slowly as they move towards what current obliquity determines which is lower than what we presently have. By 2400 AD temperatures should decrease faster towards the 2600 low in the 1000-year Eddy cycle. Another LIA should not be expected until ~ 3800 AD when the next low in the Bray cycle will approach. That LIA should be colder than the previous one and the first possible glacial inception. If the next glaciation doesn’t start by 4000 AD it is very likely that it will start at the next low ~ 6500 AD. This time frame including solar cycles agrees very well with Renée Hannon deterministic forecast.

    • Thanks Javier,
      Was your 70,000 years calculated to the next interglacial peak or inception of the interglacial onset? That would explain the 8,000 year discrepancy. Thanks for providing infill details of the Holocene shorter temperature cycles of 1500-years and 1000-year Eddy cycles.

      • Hi Renée,

        The operational definition of interglacials is not based on temperatures but on the absence of significant permanent ice in the NH outside Greenland, and are therefore defined in terms of sea levels and/or marine δ18O levels.

        According to this definition [from: Berger, A., et al. “Interglacials of the last 800,000 years.” Reviews of Geophysics 54.1 (2016): 162-219.], an interglacial starts when sea levels reach -20 m. (zero at present levels), after having been ≤ -50 m. (glacial conditions). This is equivalent to LR04 benthic stack δ18O levels reaching a value between 0.73-0.50 ‰.

        According to this definition the Holocene started ~ 11,700 BP (~ 9750 BC), and the next interglacial should start ~ -65 to -70,000 BP (~ 63,000-68,000 AD). That’s why I say that the next interglacial should occur in ~ 70,000 years.

        Another way to look at it is to set interglacials according to the obliquity cycle. As the next obliquity increase is likely to be skipped, the temporal span should be 82,000 years (2 periods) counting from 11,750 years ago. That gives 70,250 years to the next interglacial start.

        Note that as glacial termination is a process that takes several millennia, glacial termination dates are a few millennia before interglacial conditions are reached (-20 m. sea levels).

  7. I believe there is one other factor necessary for the occurrence of glacial epochs. Going to be hard to model, though. Currently the South Pole is covered by a continent, the North Pole is not completely land locked, but enough to block oceanic gyres from transferring heat from the equator to the poles. When plate tectonics moves land masses around and the poles are warmed by equatorial heat, the polar regions don’t get cold enough for ice to form and last year around. But when the heat transfer is blocked, the Milankovitch cycles can bring the polar regions cool enough to form ice caps, and the Earth enters a glacial epoch.

    • Well if you start moving the continents, you may as well change solar activity. If you take a large enough view anything is possible. I think that is neither fruitful , falsifiable nor interesting.

  8. >>Significant warm onsets occur when eccentricity, obliquity,
    >>and summer insolation are concurrently increasing.

    This and other sentences indicate a misunderstanding about the role of eccentricity.

    Eccentricity in itself has no role in increasing polar insolation (and warmth). Eccentricity can only envigorate the precessional cycle, and so an eccentricity graph without a precessional graph is quite meaningless.

    Perhaps because of this, you fail to mention that most glaciations occur because of precession, not obliquity. We know this because most interglacials are very short, and plunge immediately back into an ice age because of the next NH precessional Great Winter. The only time we get extended interglacials, like now and MIS 11, is when eccentricity is low and therefore precession is low. In the absence of a strong NH presessional Great Winter, obliquity is more dominant and the interglacial can be extended for the full length of the obliquity cycle.

    Likewise we know that precession is normally dominant, because interglacials only occur during a NH precessional Great Summer (and not during a SH Great Summer). This is because the large land masses and therefore the large ice sheets are all in the NH. So we know that precession is the dominant cycle, because obliquity effects both poles simultaneously, while precession is selective. Since interglacials only occur during precessional NH Great Summers (not SH Grt Summers) demonstrates that precession is the dominant factor. (But slightly less dominant during low eccentricity).

    This NH bias also demonstrates that CO2 is not the primary feedback mechanism. Again CO2 is a global feedback mechanism, and should operate during either NH or SH Great Summers, and cause interglacials on either. But that is not the case, they only occur on NH Great Summers. (The insolation graph in this article is for the NH, so it is being implicitly acknowledged that only the NH is important in interglacial initiation.). However, the well known feedback mechanism that IS selective is albedo, which has a strong NH bias because of all the NH ice sheets (extending to low latitudes). So the primary feedback for glaciation and interglacials is albedo, not CO2.

    In summary, you cannot predict the timing if the next ice age unless you understand the close interplay between obliquity and precession, with the latter being guided by exentricity.

    The Next Ice Age:

    Since eccentricity is low at present, and there is no strong NH Great Winter to initiate NH glaciation, it is touch and go as to whether we will enter glaciation at all. We may have an extended interglacial, for the next 50 – 100 kyr. However, the interplay of obliquity insolation and precessional insolation (not the same as your combined insolation graph) suggests we will reach the required conditions for an ice age in 500 to 1000 years.

    Once we enter ice age conditions, and ice sheets extend, nature finds it impossible to reverse them. Ice ages are initiated by orbital considerations, but enhanced by albedo. And once the northern lands are covered by even 30 cm of snow-ice, the effect of high albedo reflection and cooling ensures that the ice age will continue and enhance.

    This is the reason for the missing precession and obliquity cycles in the ice age record. Whether you champion obliquity or precession as the orbital influence, the fact remains that many high insolation NH Great Summers fail to give an interglacial. This is because the high albedo ice sheets are able to reflect and nulify all that extra high latitude insolation. And so the ice age will continue to exteand and strengthen, until we either get a snowball Earth, or the ice sheets can be darkened with dust.

    But this new feedback mechanism for ice age initiation and termination does mean that Man can easily control ice ages from now on. Since the primary feedback mechanism is albedo, we only need a fleet of carbon-soot spray aircraft to darken the extending ice sheets, and prevent glaciation continuing. And if you believe in CO2 as a warming feedback, then higher CO2 conditions would assist. The world has cooled over the last 25 million years, and is currently predisposed to ice age conditions – hence the long ice ages and the short interglacials. Any additional warming, if CO2 does indeed give any warming, would be greatly appreciated.


    • “… when eccentricity is low and therefore precession is low.”

      What do you mean by “low” precession? precession is period of rotation of the earth rotational axis. while the tilt ( obliquity ) changes, I don’t think this alters the period of 26ka much.

      • Precession is precession, that is true.

        But when the Earth’s orbit is circular (low eccentricity), there is no change in high latitude insolation at all. High latitude insolation can only change when the Earth is closer to the Sun in winter or summer, than it is in the opposite season (ie: when eccentricity is high).

        And high latitude insolation actually changes over about 22 kyr, not 26 kyr. You need to take apsidal precession into account, when looking at the Seasonal Great Year (the precession in the Milankovitch cycle), which shortens the cycle.


    • Ralph,
      “The only time we get extended interglacials, like now and MIS 11”.

      I don’t consider the current interglacial extended. Table 1a shows past interglacial durations and the current Holocene is less than both Warm II Eemian and Warm IV. The best estimate duration of the current interglacial is 19 kyrs only slightly longer than the Eemian Warm II duration.

      • The peaks of ‘warm 2, 3, and 4’ are very short indeed, and then there were subsequent plateaus, caused by the precessional cycle. But the low eccentricity ‘warm 5’ interglacial was quite different, and stretched to the full width of the obliquity cycle.

        The Holocene has not reached its decline as yet, so it is on course to be another extended interglacial. But it cannot be as long as ‘warm 5’, as precession is slightly more active and is dragging high latitude insolation down earlier. But in general orbital arrangement, initiation and extension, the Holocene is very much a copy of ‘warm 5’ – the previous low eccentricity interglacial.


      • R.
        The width of Warm V exceeds the obliquity cycle, was initiated slightly prior to obliquity and was extended by a secondary increase in insolation. A secondary insolation increase is not present in the Holocene interglacial and therefore it may be shorter in duration than Warm V. I agree that we are currently in a similar eccentricity cycle as Warm V, however obliquity and insolation appear to be more in phase.

      • >>The width of Warm V exceeds the obliquity cycle.

        The obliquity cycle is 41 kyr, which is pretty much the length of ‘warm 5’.

        What do you mean by ‘obliquity and insolation’? Obliquity IS an insolation increase on its own. The Milankovitch insolation plot displayed here is a combinarion of precession and obliquity.


  9. This is a different time perspective, but how about the following? My thoughts and research have been in the shorter term, not the long term as in this study. I have been trying to relate sunspot cycles to climate, i.e. temperature ups and downs. I have been trying to visualize cycles within cycles which moderate or accentuate cooling/warming due to anti-phasing and phasing, respectively. It seems to me that there are different effects working on different time scales. Longer term cycles, as referenced in this article, logically seem to be governed by the mentioned eccentricity, obliquity, and NH summer insolation. But what of the shorter cycles, i.e. the 60-70 year cycle, the 200 year cycle and the 800-1000 year cycle? Leaving possible mechanisms aside for now, I suspect and hypothesize that groups of 3 individual sunspot cycles (say 10 or 11 years each, on average) comprise half the 60-70 year cycle (30-35 years), and that 3 groups of 3 sunspot cycles (about 100 years) comprise half the 200 year cycle, which perhaps has a similar magnitude of effect as the 60-70 year cycle (but both less than the 800-1000 year one), and that the longest 800-1000 year cycle (half-cycle 400-500 years) is the most dominant cycle. (I’m sorry – I wish I knew how to draw graphs here.) I termed the 60-70 year cycle the Ludlum cycle about 15 years ago, after David Ludlum, who contributed pioneering works on early American weather.

    To me, it seems that these cycles are all (nearly) peaking as well as (nearly) phasing now, giving rise to our current natural late 20th century and early 21st century Modern Warm Period. The warm portion of a warm/cool 65-70 year couplet has just ended, in about 2010 (1975-2010), and we are past its peak, but still warm. The same is true for the warm portion of the 200 year cycle, which is coincident with the centuries; the warm phase was 1900 to 2000, and, while we are past its peak, we are still near the warm maximum. And, the dominant longer 800-1000 year cycle is still warming out of the Little Ice Age minimum reached in 1600 or 1650 or so, with its peak not expected until 2050 or 2100 or so. So, it would follow that the two more minor cycles would serve to offset the longest one through the rest of this century, and we might not see real cooling until we go through the 2100s and 2200s. However, we have 20 to 25 years of the current “cool” 3-sunspot group cycle to go, so that could provide slight cooling from our current plateau. Increasing human-generated CO2 is probably a (very) minor player, if it contributes any increased effect to temperatures at all. Obviously, the above cycles would nest well within the conceptual framework posited in the above article.

    A possible mechanism? Could the odd-even polarity in alternate sunspot cycles manifest in an upper atmospheric effect somehow (ozone?) which cascades eventually to the surface, such that dominant-odd followed by dominant-even polarity “3-groups” generate a see-saw temperature effect, especially in PDO and/or AMO-related effects? Hey, be kind, I’m just thinking out loud here. Perhaps a solar physicist could team with an atmospheric chemist and an atmospheric dynamicist to work on such an idea.

  10. In support of 65N , we could argue that the tropics are very insensitive to changes in radiative forcing ( there is not long term trend in Nino 3.4 ) , this is Willis’ tropical thermostat. Extra-tropical zones are more sensitive to radiative forcing and are stabilised by exchanges of water with the tropics via the major ocean gyres.

    Temp swings over land are about twice as large as ocean for the same change in dRad. ( due to lesser specific heat capacity ). If this is melt snow or ice the NH will be more inclined to swing than the SH which is less volatile due to larger proportion of water.

    • The tropics are less sensitve to Milankovitch insolation cycles for two reasons:

      a. Milankovitch insolation does not change much in the tropics. Obliquity steals some insolation from the tropics, and donates it to the poles. Precession has little or no effect on the tropics.

      b. The true feedback controlling ice age temperatures is ice-sheet albedo, not CO2, and there is no ice in the tropics.

      This is why the poles varied 12 to 14 degrees ‘c’ over each ice age cycle, while the tropics only varied by 3 or 4 degrees.


  11. Interesting and good effort. I like conceptual models but prefer a simpler one.

    The implications from CERN CLOUD experiments are that CO2 does not play a significant role in global warming, climate models used by the IPCC to estimate future temperatures are too high, and the models should be redone. A likely conclusion is that Newtonian physics cannot model earth processes adequately to produce useful results.

    A thought experiment tells me attempting to predict the earth’s temperature either by modeling a myriad of complex interactions in the solar system or by using astronomical forces to develop a deterministic framework is unnecessarily complex and possibly futile. If one cannot model the output of a complex system, what is the likelihood that the complex system itself can be modeled? The deterministic model involves a lot of assumptions and relationships between several variables that seem to be ad hoc, i.e., not adequately justified.

    I like the idea of dealing with one variable, temperature, represented by known time-varying observations, which sums up the results of complex but indefinable processes in the solar system. My simple analysis of the HadCRUT4 time-temperature data indicates a high likelihood of the beginning of an absolute decline in the global mean surface temperature trend line within the next decade. The first derivative of the temperature trend line is positive but has decreased in value every month for the past 20 years. The rate is likely to become negative in the mid-2020s and increase in negative value well into the 2030s, i.e., the mean global surface temperature will decline. I leave it to others to check out if greater focus on trend analysis has merit for long-term forecasting.

    • A likely conclusion is that Newtonian physics cannot model earth processes adequately to produce useful results.

      That may be possible over short durations. The problem is that with a basket full of poorly constrained “parameters” you can easily end up with a fit which works reasonably well over the calibration period, but is physically wrong and deviates rapidly outside that period.

      This is what is called an unconditioned set of equations. It cannot be solved uniquely so you are left guessing ( or cherry-picking ) which one of thousands of combos of parameter values which fit the data equally closely you decide to go with.

      You then present this to the world as an urgent problem and try to get your policy solutions cast into legally binding international treaties before your chosen set of parameters is shown to be wrong.

  12. Maybe you missed my point or I missed your point. 1. I don’t see evidence of an urgent problem. 2. My thought experiment leads me to conclude that your argument would apply to any of the universe of poorly constrained parameters requiring estimates in GCMs. If one concludes GCMs are viable, then a model reduced to one variable which is the summation of the effects of all the variables required in a GCM should be a useful simplification and also viable. 3. If you are suggesting that Newtonian physics can produce useful results, where is the evidence? 4. If you never see a solution in our lifetimes, I might agree with that also.

    “This is what is called an unconditioned set of equations. It cannot be solved uniquely so you are left guessing ( or cherry-picking ) which one of thousands of combos of parameter values which fit the data equally closely you decide to go with.”

    This statement is too general. Has the question of how far forward an earth temperature time-series can be projected with some degree of confidence been addressed? I would think that a Fourier Analysis of trend lines might produce usable results.

    My problem with all predictions is that long-term ones cannot be tested within the lifetimes of the prognosticators.

  13. Sorry, but MIS 11 is probably the best analogue for the Holocene. That inter-glacial lasted about 40,000 years and sea level reached about 20 meters higher than at present.

    Even though there were no SUVs. I wonder if “Anthropocene” in the title is intended as a joke.

    About past interglacials as analogues to the Holocene and Anthropocene
    Andre Berger and Qiuzhen Yin
    George Lemaitre Centre for Earth and Climate Research, Earth and Life Institute, Université Catholique de Louvain, Louvain-la-Neuve, Belgium

    Geophysical Research Abstracts, Vol. 16, EGU2014-3326, 2014, EGU General Assembly 2014

    • MIS 11 is probably the best analogue for the Holocene.

      No, it is not. MIS 11 is a very, very unusual interglacial and there is no other like it in over a million years.

      Compared to the average interglacial MIS 11 started too early in the obliquity cycle and so its start was not driven by obliquity, but by eccentricity and precession with other factors. This produced a relatively cool interglacial that was even cooling further when finally obliquity increased enough and gave it a second push upward. When obliquity had peaked another increase in insolation from precession gave it a third push upward. By then the interglacial had been so long that the thermal inertia was bigger than usual and carried the interglacial further to an almost double the average duration.

      You can see that in this figure in red:

      In stark contrast the Holocene is an average interglacial. It has started on time due to increasing obliquity helped by increasing insolation, like most interglacials. It will end due to decreasing obliquity regardless of precessional changes in insolation, like most interglacials have done before.

      MIS 11 is about the worst analogue to the Holocene. The orbital parameters are so different that it is impossible that the Holocene will follow its path. The best orbital analogue to the Holocene is MIS 19, and this analogue says the interglacial should end in a few thousand years.

      Berger is mistaken.

      • Javier
        An impressive and cogent refutation of the quite established idea of MIS11-Holocene equivalency.
        In addition to the unnaturally stretched out MIS11, some interglacials at maximal eccentricity where the eccentricity peak was half way between obliquity peaks, were double-headed, such as 200k and 600k years ago.

  14. There must be hundreds of factors which influence global climate(most of which today remain unknown). Short term, until the sharp 1920-1940’s warming and the subsequent
    1950”s -1970’s global cooling are somehow satisfactorily explained, I just don’t think man kinds present understanding is at all clear.

  15. The “forecasts” are made by a model possessing no underlying statistical population. “Forecasts” of this type can be proven to generate no information about the outcomes of events thus being unsuitable for the purpose of making public policy. It would be better to call these “forecasts” projections as a projection has no underlying statistical population but a “forecast” has one.”

    • Terry,
      This forecast is deterministic and was mostly based on statistical analyses of past analog data. It also uses assumptions such as shifting temperatures by 6.5 kyrs. The term forecast was chosen here as it is used, for example, when USGS seismologists forecast a major earthquake will strike Southern California in the next 5 years or volcanologists may forecast another Mt. St. Helens eruption. My use of the word forecast was to imply there is uncertainty associated with assumptions and a range of outcomes exist. I agree these outcome of events should not be used for the purpose of making public policy.

  16. Javier:

    “According to this definition [from: Berger, A., et al. “Interglacials of the last 800,000 years.” Reviews of Geophysics 54.1 (2016): 162-219.], an interglacial starts when sea levels reach -20 m. (zero at present levels), after having been ≤ -50 m. (glacial conditions). This is equivalent to LR04 benthic stack δ18O levels reaching a value between 0.73-0.50 ‰.”

    Ok, let’s get to the nub of this, at what degree of obliquity did the -20 m sea level occur? From that we can extrapolate how long from our current obliquity.

    • Dscott, do you mean 3.73-3.50‰?

      In my post, I used Dome C global temperature cutoff of -1 deg. C for interglacial duration calculations. This is equivalent to a δ18O value around 3.70‰ which fits nicely with the definition of when an interglacial starts (sea levels reach -20 m. and ice is absent in the NH with exception of Greenland.)

      This -1 deg. C temperature cutoff occurs at approximately 23.25 degrees obliquity according to my plot. However, the cooling periods of the interglacial are not as steep and are shorter in duration than warm onset slopes. This asymmetry makes using a single degree of obliquity for -20 m sea level more complicated.

      I’m very interested in Javier’s response.

      On Tue, Aug 22, 2017 at 12:38 PM, Watts Up With That? wrote:

      > dscott commented: “Javier: “According to this definition [from: Berger, > A., et al. “Interglacials of the last 800,000 years.” Reviews of Geophysics > 54.1 (2016): 162-219.], an interglacial starts when sea levels reach -20 m. > (zero at present levels), after having been ≤ -” >

      • Sorry about symbols being converted 3.73-3.50‰. See if images below worked.


      • Our current obliquity is 23.45 according to this calculator

        An obliquity of 23.25 degrees (decimal notation not the same as degrees, minutes) according this calculator will be achieved in the year 3470 AD. Meaning that 1,450 years in the future, as time progresses, sea level will steadily drop some 20 meters (roughly 60 ft). 1450/20 = 72.5 years/meter IF all things are proportional (which they are probably not).

        We know that during the depth of the LIA, sea level was a 1/2 meter lower than it was today (which was probably not due to obliquity but insolation effects). IF sea level is somewhat proportional to GAT, then we have a reasonable analogy of a climate time table. It therefore follows that as obliquity decreases, periodically moderated by the solar cycle, deterministically that we can expect a LIA type GAT in 30 years.

        We now have a forecast that we can falsify.

      • There is a 6500 year lag factor to allow earth’s temperature to react to obliquity. That would place your 1450 years to 7950 years in the future for 23.25 degrees and 20 meter sea level drop. 7950/20 = 397.5 years per meter which assumes they start lowering now and are portional.

      • btw – according to the obliquity calculator, nutation will either moderate or increase the rate of decline (a harmonic) depending on its own shorter cycle.

      • To put a date on it using the obliquity calculator, by the year 2040 AD (23.434 degrees), we should see sea level 1/2 meter lower than today. Intriguing isn’t it? So are we on the knee of the curve in effects on GAT and sea level or are we now on the negative slope of the curve??? Our hint here maybe is that for the last 2 years, sea level has dropped, we maybe past the knee.

        This is predicated on the solar cycle having zero effect, which we know it does, but Russian scientists are predicting the next 2 solar cycles being depressed, below normal.

  17. Renee said: “There is a 6500 year lag factor to allow earth’s temperature to react to obliquity. That would place your 1450 years to 7950 years in the future for 23.25 degrees and 20 meter sea level drop. 7950/20 = 397.5 years per meter which assumes they start lowering now and are portional.”

    Except that the 6500 yr delay is already incorporated… IF 23.25 degrees obliquity is the point at which a – 20 m sea level drop has occurred in the past. This becomes a straight forward graphing solution of y=mx + b, IF all things are equal.

    Based on your research, what were the obliquity values of the onset of the last 5 glaciations where – 20 m was the standard definition of an ice age?

  18. dscott said: “Except that the 6500 yr delay is already incorporated…”.

    The x-axis time values on the bottom of figures 3 and 4 are for the Dome C isotope/temperature data. I did not show the actual shifted x-axis values for obliquity. Obliquity at 23.25 to 23.30 degrees (I prefer to use a range at this time) actually occurs in 1.5 to 2 kyrs in the future. On figures 3 and 4, I shifted obliquity by 6.5 kyrs so it plots at 8 kyrs in the future on the graph. Perhaps, I should have included actual obliquity time values on the plot. If this is still unclear I can update the graphs in my post.

    I will go back and determine obliquity values of the onset of the last 5 glaciations (shifted and unshifted) using -20 m correlated to temperature. Thanks for the suggestion.

    • Thank you, anything that lends clarity reduces the misinformation and sensationalizing to help to tamp down the catastrophism that some are very prone to do. At the same time we need sensible decision making based on facts. What we don’t need is yet another Global Cooling scare (1970’s) where huxters scam the taxpayers as they are doing now with Global Warming.

Comments are closed.