Sunday, June 23, 2013

Are the Dansgaard-Oeschger (D-O) Warm Events driven by Lunar Tides?

What are Dansgaard-Oeschger (D-O) Warm Events ?

D-O warm events are abrupt increases in temperature to near-inter-glacial conditions that occurred during the last Ice-Age. These temperature increases occurred in a matter of decades and they were quickly followed by a period of gradual cooling.


What are some possible explanations for the D-O Events?

#    Two types of explanations have been advanced:

           a) periodic external forcing
           b) internal oscillations within the climate system.

#     If the 1,470 year cycles originate within the Earth system, we would also expect the period to change as the background moves from full glacial to inter-glacial conditions.

#     In contrast, orbital cycles are highly regular and so they would not be expected to change between glacial and inter-glacial conditions. 

Some Important Conclusions about D-O Events

§ The D-O events are discrete events paced by a regular cycle of 1470 years.

§ The five most recent events, arguably the best dated, have a standard deviation of only 32 years (2 %) about a 1470 year spacing.

§ This level of precision points to the orbital cycle explanation.

§ The 1,832 Lunar tidal cycle proposed by Keeling and Whorf (1998) cannot be reconciled with the 1,470 year spacing found in the Greenland ice-core data.

§ The origin of the regular pacing of this phenomenon remains a mystery. 


Are D-O Events Still Present in the Holocene?

Yes! They are called Bond Events and the next 
D-O/Bond Event should begin about 2150 A.D.! 

Reference: Bond et. al. SCIENCE , VOL. 278,  14 NOVEMBER 1997 A Pervasive Millennial-Scale Cycle in North Atlantic Holocene and Glacial Climates

So if the 1832 year cycle in the ABSOLUTE lunar tidal strength does not appear to provide the right external synchronization time needed for the 1470 year D-O Events, can the Lunar tides still play a role ?

The answer is yes, if we are prepared to make a paradigm shift!   

What happens if instead of looking for cycles in the absolute strength of lunar tides, we look for cycles in the strength of the lunar tides that are synchronized with the seasons?

        This means that if we start out with a New Moon (i.e. Syzygy - when the Earth, Moon and Sun are aligned) at closest Perigee (i.e. when the Moon is closest to the Earth) at the time of Perihelion (i.e. when the Earth is closest to the Sun) on or about January  1st, how long does it take before the New Moon returns to the same precise alignment with the seasons?  

In order to answer this question we need to consider a few definitions: 

§ One Full Moon Cycle (FMC) is the time required for the point of Perigee in the Lunar orbit to re-align with the Sun.

§ As the Earth revolves around the Sun, the Line-of-Apsides very slowly turns in a clock-wise direction. This motion is caused by the precession of the Line-of-Apsides of the Lunar orbit around the Earth, once every 8.8502 Sidereal years, as measured with respect to the stars.  It is known as the Cycle of Lunar Perigee.

§ The Perigee-Syzygy-Perihelion Cycle is the time required for a Full (or New Moon) at Perigee to re-occur at or very near to the time of Perihelion.

       The Perigee-Syzygy-Perihelion Cycle is one lunar tidal cycle that is known to precisely realign with the seasons.

§ This cycle repeats itself at the following times:

0.00 FMC=0.00 Tropical yrs=New Moon at Perigee & Perihelion
27.5 FMC=31.00 Tropical yrs=Full Moon at Perigee & Perihelion
55.0 FMC=62.01 Tropical yrs=New Moon at Perigee & near Perihelion
82.5 FMC=93.01 Tropical yrs=Full Moon at Perigee & near Perihelion
157.0 FMC=177.00 Tropical yrs=New Moon at Perigee & at Perihelion

        The realignment of Perigee with the Sun on January 1st resets itself with respect to the stars once every:

157.00 FMC’s = 177.00 Tropical years.

This happens because:

157 x FMC   = 176.999 Sidereal years
20 x 8.8502 = 177.004 Sidereal years

         What happens when we extend the 177.00 year Perigee-Perihelion Cycle over longer time periods?

§ The following plot shows the Earth’s position in its orbit when it is ≤  seven days from the 1st of January near Perihelion.
§ All FMC's (where Perigee either points directly at the Sun or directly away) are shown  up to 354 (= 2 x 177.00 ) years.
§ The FMC's that are separated from their predecessor by 9.0 years are shown in the  same colour.
§ The FMC's in the sequence 31, 208, 385…. years are extended until the Perigee- Perihelion cycle is almost precisely reset after 916.00 years.
§ Of course, this is only half of the Full reset cycle since the perigee points directly at the Sun at 0.00 years and it points directly away from the Sun after 916.00 years.
§ Hence, the full reset time for the Perigee-Perihelion cycle is 1832.00 years.  This the famous Keeling & Whorf 1800 year tidal cycle.

The next graph re-plots the data in the previous graph to show how the proximity of a given FMC event  is to Perihelion changes over time.

Note: The strongest lunar tides occur when the FMCs occurs at or very near to Perihelion, once every 177 years. These times are marked in the following diagram with vertical arrows.

How do the phases of the Moon re-synchronize 
with the 177.0 year Perigee-Perihelion Cycle?

§ When the Perigee of the Lunar Orbit is pointing at the Sun at (or very near to) Perihelion it does not necessarily mean that the phase of the Moon is either New or Full (Syzygy).

§ The next slide shows the number days that the phase of the Moon is from being New or Full, for each of the FMC's that are at (or near to) Perihelion. The graph starts out with a New Moon at Perigee on January 1st (near to Perihelion on January 3rd) in the year 0.00.

§ New or Full Moons that re-occur for FMC's at (or near to) Perihelion that are whole multiples of 739 years (i.e. 0.0, 739.0, 1478.0 and 2217.0 years) after the starting date, always occur at lunar Perigee.

§ In contrast, New and Full Moons that re-occur for FMCs at (or near to) Perihelion half way between whole multiple of 739 years (i.e. 370, 1109 and 1848 years) always occur at lunar Apogee.

§ Hence, we end up with the following 739.0 year repetition sequence for the times where FMC's are at Perihelion:

                            0.00 Years  è New or Full Moon at Perigee
                        184.75 Years  è First or Last Quarter Moon
                        369.50 Years  è New or Full Moon at Apogee
                        554.25 Years  è First or Last Quarter Moon
                        739.00 Years  è New or Full Moon at Perigee

§ Careful study of the New and Full Moons near 739.0 years shows that the strongest alignment between the phases of the Moon and the 177.0 year Perigee-Perihelion cycle occurs at the FULL MOON at 739.001 years. This contrasts with the NEW MOON at 0.000 years.

§ What this is telling us is that it actually takes 1478.00 years (= 2 x 739.00 years) to complete the cycle with a New Moon at Perigee when a FMC is close to Perihelion once again.

§ The FMC cycle is closest to perihelion at ((1447+1478)/2) years = 1462.5 years, while the lunar phases are most closely aligned with the Perigee-Perihelion cycle at 1478 years – producing a best synchronization at roughly (1478+1462.5)/2 = 1470.3 years.

§This is in extremely good agreement with the measured spacing of the D-O climate warming events of 1470 years!

         Hence, if we look for cycles in the strength of the lunar tides that are synchronized with the seasons, rather than cycles in the absolute strength of lunar tides, we find that there is a natural 1470 year tidal cycle.

This supports the contention that Dansgaard-Oeschger (D-O) Warm Events are being driven by a 1470 year periodicity in the long-term Lunar Tides!


  1. You may find the comments of E.M. Smith (Chiefio) interesting

  2. Thanks unknown - E.M. Smith has clearly not made the paradigm shift that I promote in the above post.

  3. So reading between the lines we can expect cooling to 2150. Is that reasonable to conclude ?

    It does tie in nicely with one of those elephant trunk wiggling graphs done strictly for amusement that show cooling to 2140.

  4. No, there will be a mild warming event that will start around 2150 and which could take anywhere from a couple of decades to 500 years to warm to a maximum. [N.B. down is warming in the Holocene data, not up].

    Here are some examples of what happened during the last three Bond/D-O events.

    The Minoan warm period (started ~ 2300 B.C and peaked in warmth ~ 1700 B.C.)

    The Greek warm period (started ~ 800 B.C. and peaked in warmth ~ 500 B.C.)

    The Medieval warm period (started ~ 700 A.D. and peaked in warmth ~ 1150 A.D.)

    The D.O events were in ~ 2250 B.C., ~ 780 B.C. and ~ 690 A.D. i.e. separated by ~ 1470 years.

    If, we follow this pattern in future, the Bond/D-O event will start ~ 2150 A.D. and the warming will reach a peak ~ 2600 A.D.

    [It is interesting to note that sometimes there is smaller warming event half way between D-O/Bond events. One example of this is the Roman Warm Period.]

  5. Also of interest - the Greek Dark Age lasted from ~1200-1100 B.C. to 750 B.C. ending with a D-O/Bond event in ~ 780 B.C.

    Interestingly, the European Dark Ages were 1500 years after the Greek Dark Ages, spanning the period from 300-400 A.D to 750 A.D.

    Both of these events where thought to be periods that were colder than normal.

  6. @Ninderthana:

    Um, not seeing the need for a 'paradigm shift'. While I like your analysis, I'm mostly just quoting other folks works and trying to line them up with the known data.

    Some of the better stuff points out that the 1470 is just an average and that actual D.O. / Bond Events have spacing offset to either side of that point. Were I to speculate, I'd speculate that often your analysis is what happens, but also that sometimes a cycle is offset toward the 1200 or 1800 year ends of things as some alignments have "slipped" too far from sync at that point. Then it all gets back toward the closer 1470 once that process gets resynced. In essence, that the times when "almost the same" have drifted out of sync, you get a 'skip beat' and the event is either closer or further out in time. Then the drift slowly puts them back in sync at 1470 (ish).

    FWIW, I especially like your 739 year finding. I've found a similar "half Bond event" cycle in historical records. Not a full on Bond event, but "something happens" that's not very good...

    You might want to look at the exact dates of some of the older D.O. and Bond Events and note how they vary from the 1470 average.

  7. Chiefio,

    Thanks for you comments. I think we will have to agree to disagree on this one. I believe that you are discounting a possible explanation before it has properly been tested.

    The real world lunar cycles are not precisely 1470 years. The quoted figures in this post are not mean to be taken as representative of what exactly happens in the real world - only what happens as a long term average.

    The complex nature of the lunar orbit means that a high quality ephemeris needs to be used to see how specific DO events align with lunar tidal events.

    At this stage, I do not have the time to followup this possible avenue of research but others are welcome to give it a try if they feel it is worth it.

    1. Um, l am not 'disagreeing'. Just suggested a refinement or two (and didn't see the need to call your observation a paradigm shift... important step forward is likely a better fit).

      I have pointed out that the earth rotation matters (which ocean is under the tidal bulge repeats about 3 x 18.x year lunar cycle), and adding seasons is likely as or more important.

      I see all of it as mattering, with how much not yet known. Seasons could easily be very important with Artic ice breakup and flushing as the means by which the tides shift things.

      At any rate, I think your observation needs integration to the general understaning, though I still am working on fitting it in (if slowly :-)

      I'm also pretty sure the spacing of actual events is not right on 1470 yrs and the difference could lead to further understanding. The variable spacing is visible in the red tick marks on the DO graph above.

      So is that variance an error (so what can be learned is that the dating is dodgy) or material (so what can be learned is another minor influence)?

      In either case, I don't see that as a dissagreement with this work. More of a next step, perhaps.

  8. Very interesting. The moon has held a very special interest to me for some years now.