Climate and Solar Regularities and Global Cooling
I was thinking recently about the downside of the 60 year cycle that we seem to be into and wondering if there might be any other cycles in phase with the 60 year cycle that would make the cold period worse than it would otherwise be. That led me to some research on climate cycles that would seem to be driven by solar effects related to planetary orbits. It was not difficult to find several such climate cycles, including some that seem to have little or no connection with planetary motion. Perhaps more interesting was the number of papers that identify climate periodicities, without any attempt at explanation, that do have clear connection to planetary/solar cycles.
The “no connection” cycles led me to the question of whether some apparent cycles were simply sidebands or beat frequencies of the well-established cycles. I had just begun to struggle with that issue when I found Wilson's paper at www.alp.org.au/Congress 2006/625.PDF and discovered that he had already done most of the work for me. His other works then led to several other mysterious connections.
Following is a summary of the more or less well-known regularities that I could find.
Recognized regularities or cycles
• 9 year – Lunar cycle – evident in weather if not climate.
• Schwabe 11 year sunspot cycle – a bimodal cycle of approximately 10 years and 12 years synchronized to the Venus/Earth/Jupiter cycles. Can be from 9 to near 14 years.
• 18.6 year lunar nodal cycle – clearly apparent in Arctic proxies
• Jupiter/Saturn synodic period - 19.86 years average. Possibly confounded with the lunar nodal cycle in some proxies.
• Hale 22.3 year average solar magnetic polarity cycle. Every second Schwabe cycle, but no other specific link to planetary orbits. Could be influenced by the lower beat frequency of the Jose cycle and the Jupiter/ Saturn synodic cycle as a strange attractor. No significant climate signature.
• Saturn orbital 29.5 years – some weak evidence of thirty-year cycle in climate records.
• Unnamed 60 to 62 year cycle – three Jupiter/ Saturn synodic cycles – time for the sun to repeat one complete tour of its track around the SSB. Very clear climate cycle in the last several hundred years.
• Gleissberg cycle – variously expressed as 72 years to 120 years. Also seen as 72 to 83 year sunspot cycles. Climate sources used by Wilson give an average of 88.7 years. The lower sideband (beat frequency) of the 60 year cycle and the Jose cycle is 90 years. Use 90 years. Wilson describes why the Gleissberg cycle changes so widely. http://www.scribd.com/doc/16785737/Does-Spin-Orbit-Coupling-Between-the-Sun-and-Jovian-Planets-Govern-the-Solar-Cycle-
• José cycle – 178.7 years – the shortest time for each near alignment of the Jovian planets. Also the beat frequency of the 22.34 average length of the Hale cycle and the 19.86 year average length of the Jupiter/Saturn synodic cycle. The variable length of these two cycles can give a fairly wide range for this beat frequency, from about 172 to 189 years.
• DeVreiss cycle – 200 to 210 year cycle. Wilson's sources average 201 years, but 205 and 210 years are more commonly mentioned. No obvious driver. Wilson notes a beat frequency between the first harmonic of Hallstatt cycle and the Jose cycle, but that seems like a long reach.
• Deep Grand Minimum (DGM) period – 363.3 years - the length of a Deep Grand Minimum cycle that meets the “golden proportion”. See “Solar Grand Minima” below. (My contribution – Duffin cycle??)
· .950 – 1100 year cycle - A cycle of near 1000 years shows up strongly in some reconstructions. See http://scienceandpublicpolicy.org/images/stories/papers/originals/climate_change_cause.pdf page 33. Since originally writing this I have come across the Climate theory presented here http://climaterealists.com/index.php?id=6482 . This cycle could be a terrestrial thermohaline cycle.
• 1500 year - Dansgaard/Oeschger (D-O) events recur in last Ice Age cores at 1470 years, + -12%. Bond events appear in Holocene records as about 1500, + -500 years. The last four Bond event are spaced about 1400, 1400 and 1700 years apart. (4, 4,and 5 DGM periods? )
• 1800 year tidal cycle based on shifting lunar declination ??
• 2300 year Hallstatt cycle - seen in C14 tree ring data. Wilson notes that it takes 121 Jupiter/Saturn synodic periods or 2403 years for the alignment to return to the same position with respect to the stars, which doesn't seem very meaningful in climate terms. Charvatova notes that every 2400 years the Jose cycle drops to 160 years for a period of about 370 years, which does seem meaningful. The next occurrence will be between 2240 and 2610. Strangely the 370 years is very close to one 363 year DGM period.
Solar Grand Minima
Name Period Cycles from prior Type
Oort 1010—1050 Shallow
Wolf 1280--1340 24 Deep
Sporer 1420—1530 12.6 Shallow
Maunder 1642—1705 20 Deep
Dalton 1790—1820 13.3 Shallow
SSC24/25 2009--?? 19.7 Deep
What about the date of the Oort minimum? It seems to have been determined by radiocarbon dating. How accurate is that? Three laboratories working independently provided a consensus date for the Shroud of Turin as 1260 to 1390. That is 1325 +- 65 years. The Oort minimum occurred about 300 years earlier, so might have an uncertainty range of +- at least 80 years. If the start of the Oort was 50 years later (1060) we could have a pattern of 20,13,20,13,20 cycles between grand minima. (In the curve shown here http://www.landscheidt.info/ the bottom of the Oort is just about exactly at 1060).
Given the 20,13 repetition, it appears that each grand minimum at the end of a 20 cycle period is deeper and longer than after a 13 cycle period, and has a greater impact on climate. Call them deep and shallow grand minima (DGM and SGM). Wolf and Maunder were DGM and have clear signatures in the proxy record. Oort, Sporer and Dalton were SGM. The proxy records display the Sporer but not the Oort. We are probably now entering a DGM. Curiously the DGM cycle is near 360 years or near 2 Jose cycles, and given my assumptions below, they seem to align pretty closely during the last 2000 years.
Could there be an orbital/SSB reason for such regularity? 221 years (20 Schwabe cycles) is very close to the DeVreis regularity. Also 13 cycles at average length of 10.68 years and 20 cycles at average length of 11.23 years is the golden proportion (38.2:62.8), with a DGM cycle length of 363.3 years. (The 20 11.23 year cycles take 224 sidereal years, which is strangely close to Wilson’s 224 year period noted here http://astroclimateconnection.blogspot.com/2010_05_01_archive.html). The average solar cycle length for the 363.3 year period is 11.01 years. Wikipedia gives an average length of 11.04 years (ignoring Usoskin et al’s extra cycle at the beginning of the Dalton), which is close enough given the degree of uncertainty in estimating prior sunspot cycles. Four such DGM cycles = 1453.2 years, very nearly one D-O event period.
There is another strange factor appearing now. Usoskin et al invoke the Gnevyshev-Ohl (G-O) rule that says odd numbered sunspot cycles are always stronger than the preceding even numbered cycle. Unfortunately for G-O this rule has only been valid for the period since the Dalton minimum. The last 2 cycles at the beginning of the Maunder and Dalton grand minima had the odd number intensity lower than the preceding even number. We now have cycle 23 lower than cycle 22, entering another grand minimum. It seems that the G-O rule needs to be refined, and Usoskin et al’s extra cycle at the beginning of the Dalton may be invalid. Maybe SGM are always preceded by extra long cycles.
The biggest problem with the cycles and events in the proxy climate records is inconsistency. They are confounded by contributions from plate tectonics, volcanoes, meteorites, ocean currents and variable delay times in earth systems. When such factors were isolated, there were still major irregularities that are just now becoming understood. Trying to get a layman’s understanding of the mechanisms required more digging. Nearly a decade ago I had found Niroma, DesMoulins and Landscheidt, and while they provided insight on correlations, they didn’t help much with mechanisms. Now one finds several gifted amateurs (Tallbloke, Stephen Wilde, Geoff Sharp, Ian Wilson, Nicola Scafetta) with various particular skills collaborating informally on the internet to produce quite profound analyses of probable mechanisms, including some realistic quantification. The most helpful (for me) of these was I. R. G. (Ian) Wilson. The seminal work seems to have been done by P. D. Jose, published in a 1965 paper. As with other pioneering iconoclasts (think evolution and plate tectonics) his work was largely ignored until now. Based on several days of digging I have tried to write a simplified description that can be understood to a reasonable degree by the “great unwashed”, like myself.
Non- Technical Description of Solar Mechanisms
Viewing a diagram of the sun’s motion relative to the solar system barycenter (SSB) (http://www.orbitsimulator.com/gravity/articles/ssbarycenter.html) it is obvious that sometimes the Barycenter is near the center of the sun, and sometimes well outside its circumference. Sometimes the sun is moving toward, sometimes away from the barycenter. Conversely, viewing the barycenter moving relative to the sun it is intuitive that the barycenter accelerates and decelerates the sun, and imparts angular momentum varying from substantial to near zero. Simultaneously the planets exert gravity on the sun that varies with their orbital distances (elliptical orbits), and alignments. All of these forces perturb the rotation of the solar dynamo in patterns that repeat, but not exactly. For a pretty good animation go to http://www.youtube.com/watch?v=1iSR3Yw6FXo . As a thought experiment, imagine a snowglobe rotating around an axis that in turn is being swung around on the end of a string. Now imagine that the strings anchor is being moved back and forth, while the string is getting shorter and longer, and both the string’s and the globe’s rotation rates are being speeded and slowed. One can imagine the complexities imparted to the rotational movement of the liquid inside the globe. Think of it as swirl, stir, shake and slosh.
Missing from this thought experiment is any sense of scale. Imagine that the snow globe is ginormous and is a vast distance away, so that all of the forces are attenuated almost to the point of triviality in relative terms, but still able to have some effect. That is something like what is happening to at least the outer layers of the solar plasma. Agitation of the plasma, and variations of the strength and speed of the solar dynamo cause waxing and waning of sunspots, increases and decreases of solar uv output, strengthening and weakening of the solar magnetic field, and periods of stronger or weaker CMEs, all of which drive our climate change. However the sun is not uniform in density and the state of agitation will never be the same at the time of repeated planetary alignments or position relative to the SSB, so the solar reaction does not repeat exactly in timing or intensity.
If one follows the work of the a/m group of amateurs, one can see an holistic understanding of the complexities of the solar system, and their impacts on our climate developing, with growing elucidation and quantification of the mechanisms involved. This work is already so advanced that it is very difficult for a rational person to follow it with an open mind and not agree that the sun is the primary driver of climate change and that the 30 year warming from 1975 to 2005 was simply the upside of a natural climate cycle, with a cooling to follow.
What might we expect for this cooling?
Start with the probability that the 1976 to 2006 surface instrument warming peak is overstated, and was probably less than the prior 1938 to 1944 peak. The 60 to 62 year cycle peaked in about 1820, 1880, 1940 and 2000. At 62 years it would be 1820, 1882, 1944 and 2006 which might be a little better. It is now on a downslope to about 2030 – 2040. The José cycle seems to be near to in phase with the 60 year cycle. If the José and 60 year cycles bottomed about 1650 and 60 year cycle about 1600 and 1710, it would partly account for both the depth and length of the bottom trough of the L I A. If so, it (the Jose cycle) bottomed again 1850 – 1860 (cool period) and can bottom again 2030 – 2040. It could have peaked near the 1938 – 1944 peak warming contributing to that peak being higher than the recent peak. The cycle from 1910 to 1974 warmed 1910 to 1938 (28 years), flatlined to 1944 (34 years overall), then plunged 0.3 degrees C to 1953 (9 years) then flatlined to about 1976 (23 years). The current cycle warmed from 1976 to 1998 (22 years), flatlined to 2006 (8 years), warming at the same rate as the previous cycle, and has now cooled slightly for 4 years (36 years overall). Could it be ready for a 0.3 degree C plunge during the next 7 years?
The 1400-1500 year Bond Event is the wild card. Chiefio (http://chiefio.wordpress.com/2010/09/13/an-interesting-view-of-temperatures/) has already mused on the idea that the last Bond event (event number one) occurred during the Dark Ages or Migration Period cooling (about 400 – 900 A.D). It appears in several climate proxy records at about 550 A.D. as a brief downward spike, but shows up as a severe cooling event in Greenland ice cores. 1470 years later is 2020! Oops!
Solar activity – timing looks like a deep grand minimum for cycle 24/25. We can probably expect a cooling at least as severe as the Dalton minimum, and possibly similar to the L I A. However, Wilson (http://ozwx.plasmaresources.com/wilson/Syzygy.pdf) has raised the possibility, even while casting doubt on it, that we might be back to something like the Oort Minimum, which occurred during the MWP and left no real mark on the climate. Since the Oort was not a DGM, I would share Wilson’s doubts.
To use the cycles enumerated above to backcast, I have to make assumptions about past phasing. My assumptions following seem to work pretty well.
• The coolest part of a DGM occurs 25 to 40 years after onset
• The 950 – 1100 year cycle is close to 1050 years long and last peaked during the MWP at about 1150 and bottomed last during the LIA AD 1650-1700
• The Jose cycle last peaked about 1940-1945
• There was a 60 year cycle peak at 1940.
• Leaving out Gleissberg and DeVreis because too variable
For prior warm/cold periods we then find:
• Roman warm period: 1000 year peak about 200 AD, Jose peak 156 AD, 60 year cycle peaks at 140, 200, 260 AD. DGM at 182 AD. 3 warm, 1 cool.
• Dark Ages cold period: 1000 year bottom at 670 AD, Jose mins at 424 and 603 AD, 60 year mins at 530 and 590 AD, and a Bond event about 550 AD. 4 cools with an extended spread.
• MWP warm 1000 year peak at about 1150, Jose peak at 1050, 60 year peaks at 1040 and 1100 and 1160, SGM at 1060. 3 warm, 1 weak cool. Oort swamped by warm peaks.
• LIA cold: 1000 year bottom at 1650-1700, Jose bottom about 1680, 60 year bottoms at 1610m and 1670, Maunder DGM coolest about 1670-1680. 4 cool, very closely grouped.
• Dalton cool: 1000 year 30% up from bottom, Jose bottom 1840, 60 year peak 1820, SGM weak bottom 1810-1820. 2and 1/3 cool, 1 warm.
• 20th Century warm: 1000 year ½ way+ to peak, Jose peak 1940-1945, 60 year peak 1940, mid cycle between grand minima. 2 and ½ warm. Like Roman warm period, cooler than MWP?
• Cycle 24/25 cool: 1000 year ¾ way to peak, Jose bottom 2030-2040, 60 year bottom 2030, DGM coolest about 2030-2040. Bond event 2020?? 3 cool and ¾ of a warm with Bond event wild card. Cooler than Dalton, not as cold as LIA.
• Looking a lot farther forward we would have the next peak of the 1000 year cycle at about 2200, corresponding to a Jose bottom at near 2215, and about half way between 2 DGMs. Not as warm as the MWP, consistent with the long trend down since at least the Minoan Optimum. The next 1000 year cycle bottom would be about 2725 and would correspond closely with a Jose bottom near 2755 and a DGM near 2773, and 60 year bottoms near 2690, 2750 and 2810. This would give us 5 coolings fairly closely grouped, cooler than the LIA. Could that be the start of the next ice age? Happily it does not align with a Bond event.
Questions for the amateur experts
• Are my phasing assumptions supported by planetary alignments, or no?
• Why would grand mimima repeat in an alternating 20 and 13 sunpot cycle pattern?
• Why would 4 DGM periods equal one D-O period?
• What is the significance of Wilson’s 224 sidereal year “strange attractor”?
• Why would the phase change period every 2400 years last one DGM period?
• What mechanism could drive Landscheidt’s phase reversals?
• What are your thoughts looking forward?
Additional References: - far from exhaustive
http://astroclimateconnection.blogspot.com/ I.R.G. Wilson- also Google to find several papers – Probably the easiest to understand and most informative
http://www.ann-geophys.net/18/399/2000/angeo-18-399-2000.pdf - Charvatova
http://www.jupitersdance.com/ a lot of info, perhaps marred by Mayan speculation
http://www.appinsys.com/GlobalWarming/SixtyYearCycle.htm the outstanding reference on the 60 year cycle
but the prediction has been blown.
http://tallbloke.wordpress.com/ A lot of good material and discussions, but you have to hunt for it.
http://www.vukcevic.talktalk.net/ Lots of interesting analyses, but lousy labeling so hard to understand.
http://www.landscheidt.info/ excellent source
http://scienceandpublicpolicy.org/images/stories/papers/originals/climate_change_cause.pdf A most informative argument for solar forcing and against AGW. See the appendices
http://www.fel.duke.edu/~scafetta/pdf/scafetta-JSTP2.pdf More from Scafetta
http://climaterealists.com/index.php?tid=37 Stephen Wilde makes a lot of contributions to the discussion
http://agbjarn.blog.is/users/fa/agbjarn/files/ljungquist_temperature_reconstructions_2009.pdf Many proxy reconstructions for last 2 milennia.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC18099/ 1800 year tidal cycle, note connection to 360 year cycle that seems like DGM, plus Jose cycle and probable 18 year lunar nodal cycle.
http://icesjms.oxfordjournals.org/content/63/3/401.full Lunar influence in Arctic proxies
http://spaceweb.oulu.fi/~kalevi/publications/non-refereed2/ESA_SP477_lostcycle.pdf Usoskin et al.