The role of solar variability on climate change, despite having a very long scientific tradition, is currently downplayed as a climatic factor within the most popular hypothesis for climate change.
As the root of this neglect lie two fundamental problems. Solar variability is quite small (about 0.1% of total irradiation), and there is no generally accepted mechanism by which the solar variability signal could be amplified by the climate system.
While progress is being made to solve these problems, there is a growing number of scientific paleoclimatology articles published every year that defend a significant role for solar variability in paleoclimate change. The explanation for this contradiction is that evidence always trumps theory, and there is very solid evidence that periods of low solar activity in the past, identified by a higher rate of cosmogenic isotopes production, have a high degree of correlation with periods of climate deterioration manifested as lower temperatures and precipitation changes.
Frequency analysis of solar variability during the Holocene identifies several cycles (McCracken et al., 2013), with the most important being the 11.4-yr Schwabe cycle, the 87-yr Gleissberg cycle, the 208-yr de Vries cycle, the ~ 1000-yr Eddy cycle, and the ~ 2400-yr cycle. Even longer cycles can be identified from 10-Berilium (10Be) records in ice cores, like a 9600-yr cycle (Sánchez-Sesma, 2015). Comparison of climate and solar variability records leads to the important observation that the length of the cycle correlates with the amplitude of the climate effect observed and in general the longer the cycle the more profound effect it appears to have on climate.
For an analysis of solar cycles during the Holocene you can read “Periodicities in solar variability and climate change: A simple model”
For this article I am going to concentrate mainly on the ~ 2400-yr cycle during the Holocene and on its effects both on climate and people. It is important to highlight two things. First, that solar variability, even if an important factor affecting climate change is neither the main one, nor the only one. Temperatures on Earth appear to depend mainly on orbital changes, firstly obliquity, but also precession and eccentricity, and oceanic cycles, and volcanic activity also play an important role at times, and therefore solar variability alone does not explain climate changes. The second is that solar cycles are irregular in nature. The Schwabe cycle is a good example. Although described as an 11-yr cycle it can be anywhere from 8 to 15 years. Also its amplitude is very variable, and during the Maunder minimum between 1620 and 1700 AD even became inconspicuous. Other solar cycles also manifest this irregularity both in periodicity and amplitude, and similarly the ~ 1000-yr Eddy cycle was inconspicuous between 4500 and 1500 yr BP (years before 1950).
The ~ 2400-yr Bray cycle
This cycle was identified by J. Roger Bray in 1968 from a consilience of geophysical, biological, and glaciological evidence contrasted with solar activity reconstructed from sunspot naked eye observations and aurora records. This proposed solar cycle was later confirmed by Sonnet and Damon by spectral analysis of the 14C record when it became available in the late 80’s and named Hallstattzei (later Hallstatt) for a late Bronze-early Iron cultural transition in an Austrian archeological site during the cycle’s previous to last minimum, 2800 years ago. However the name breaks the tradition of naming solar cycles after their discoverer, and refers to a cultural change at a specific time and location, and so I propose the name to be changed to the Bray cycle.
The average length of this cycle has not been very clearly determined due to its irregularity and the difficulty from separating its signal from the ~ 1000-yr Eddy cycle. It was originally described by J. R. Bray as a ~ 2500-yr cycle, and values as low as ~ 2200 years have been published. My own estimate based on known occurrences of its lows during the entire Holocene is around 2450 years, quite similar to Bray’s estimate 50 years ago.
Figure 1. Variation in 14C after removal of the long-term trend. An oscillation of the ~ 2400-year Bray cycle is superposed on the data to indicate times when periods of very low solar activity would be expected to occur (arrows). As with every solar cycle, there is some variability in the spacing that complicates mathematical analysis. Adapted to show correct cycle length from Clilverd et al., 2003.
The ~ 2400-yr cycle in solar activity could be a lower harmonic of the ~ 9600-yr cycle found by Sánchez-Sesma (2015) in 10Be records for the past 135 kyr. Alternatively the ~ 9600-yr cycle could be the result of a constructive interference between the ~ 1000-yr and ~ 2400-yr cycles.
We do not know what causes a ~ 2400-yr cycle in solar variability. Scientists are divided between those that propose solar internal causes, and those that defend a “planetary hypothesis” where the giant planets in the Solar System would cause these cycles by affecting the movements on the Sun around the Baricenter of the Solar System, either through changes in planetary torque (Abreu et al., 2012), or through changes in solar inertial motion (Charvátová & Hejda, 2014).
Ivanka Charvátová’s hypothesis is specially relevant to this article since she finds a 2402 year cycle in changes in solar inertial motion, a period in close agreement with the ~ 2400-yr cycle in solar variability.n essence periods of high solar activity would coincide with planetary and solar configurations in an equilibrated figure known as the trefoil, while periods of low solar activity would be associated with disordered configurations (Charvátová & Hejda, 2014). An obstacle to the acceptance of the planetary hypothesis is that several massive exoplanets have been discovered at orbits around their star much closer than Jupiter without causing a measurable effect on the star’s activity.
Whatever its cause, the observed effect of the ~ 2400-yr Bray cycle is to result in long grand solar minima (GSM) or clusters of GSM at its lows. According to Usoskin et al., 2007:
“the occurrence of grand minima depicts a weak (marginally significant) quasi-periodicity of 2000–2400 years, which is a well-known period in 14C data… no clear periodicities are observed in the occurrence of grand maxima.”
This fits well with the observation that unlike the last glacial period, during the Holocene after it reaches its Hypsithermal or Climatic Optimum there are no warming events, just cooling events followed by recovery, and thus a warming period cannot be separated from the previous cold event without losing context. According to this view, a low in the Bray cycle would increase the probability of a long GSM or a cluster of GSM that would reduce temperatures and cause changes in precipitation patterns bringing about a general worsening of the climate for a few centuries. The end of the low would bring about a return to normal solar activity with a natural increase in temperatures that can also take a few centuries.
The climatic effect of the GSM caused by the lows in the Bray cycle appears to register mainly as a significant reduction in winter temperatures with a smaller effect on summer temperatures, and profound changes in precipitation patterns, usually registered in the best studied North Atlantic region as a very significant increase in precipitations. The combination usually leads to cold winters, increased snow, glacier re-advance, and spring flooding. The effect on human societies can be postulated as higher frequency of food crises, population decrease, increased migration, increased violence, and higher chance of civilization regression or failure. It is so common that new civilizations emerge after climatic crises that some archaeologists have developed the theory that climate caused environmental stress is an engine to societal change (Weninger et al., 2009; Roberts et al., 2011).
Let’s now review what has happened to the planet and people at the lows of the Bray cycle during the Holocene. These have taken place around the following dates (kyr BP. indicates thousands of years before 1950):
B1. 0.4 kyr BP. Little Ice Age (LIA)
B2. 2.8 kyr BP. Sub-Boreal/Sub-Atlantic Minimum
B3. 5.2 kyr BP. Mid-Holocene Transition. Ötzi buried in ice. Start of Neoglacial period
B4. 7.7 kyr BP. Boreal/Atlantic transition and precipitation change
B5. 10.3 kyr BP. Early Holocene Boreal Oscillation
B6. 12.8 kyr BP. Younger Dryas cooling onset
Figure 2. Holocene climate reconstruction. Major palinological subdivisions of the Holocene (names on top) match a 2450-yr regular spacing (grey arches on top). (a) The global temperature reconstruction (black curve; Marcott et al., 2013 by the differencing method with proxy published dates) has been rescaled in temperature anomaly to match biological, glaciological, and marine sedimentary evidence, resulting in the Holocene Climate Optimum being about 1.2°K warmer than LIA (See Appendix). (b) The general temperature trend of the Holocene follows the Earth’s axis obliquity (purple), and significant downside deviations generally match the lows of the ~ 2400-year Bray cycle of solar activity (grey bands labelled B-1 to B-5). (c) Significant negative climate deviations manifest also in global glacier advances (blue bars; Mayewski et al., 2004) and (d) strong increases in iceberg detrital discharges (red curve, inverted; Bond et al., 2001) that generally agree well with the lows in the ~ 2400-year Bray cycle and ~ 1000-year Eddy cycle (not shown) of solar activity.
The 10.3 kyr event. The Boreal Oscillation.
The fifth low in the Bray cycle at about 10.3 kyr BP coincides with the GSM known as Boreal 1 and the climate worsening named Boreal Oscillation. The planet was at that time still warming towards the Holocene Hypsithermal, due to increasing obliquity and northern latitudinal summer insolation and the event appears to not have been a major trend inversion, but a relatively brief cold and wet period for the areas we have the best evidence.
The 10.3 kyr event and its associated Boreal 1 GSM coincide with Bond event 7 of increased iceberg discharge in the North Atlantic (Bond et al., 2001; figure 2), and with the highest concentration of non-sea salt potassium deposition in the Greenland GISP2 ice core for the Holocene (figure 3e). The presence of non-sea salt potassium in Greenland is associated to the expansion of the Siberian High pressure system that brings polar temperatures over a wide area in the northern hemisphere (Mayewski et al., 2004).
The precise dating of the potassium increase around 10250 BP allows to estimate the duration of the Bray cycle. Since the Spører minimum takes place around 450 BP, that gives an average duration of around 2450 years.
The 10.3 kyr event has been studied mainly by Björk et al. (2001) and attributed to decreased solar forcing. The event is recorded at proxies from multiple sites in the Northern Hemisphere, like Norwegian sea surface temperatures (SST) that show a drop of >2°C both in winter and summer that lasted less than 200 years. This Norwegian sea cooling coincides with harsher conditions in the Faroe Islands lacustrine records that show a decrease in birch pollen and increase in grass and herb pollen. German pines show at the time a tree-ring width minimum and the Santa Barbara basin shows a cold related peak of oxygenation. Greenland and Tibetan ice cores display a d 18O isotope minimum (Björk et al., 2001, and references within). The increase in precipitation in the North Atlantic-European region is supported by the increase in lake levels in west-central Europe and central Italy (Magny et al., 2007; figure 3c), and the increased iceberg discharge (Bond et al., 2001). Glacier readvances took place in Norway (the Erdalen event; Dahl et al., 2002) and Tibet (Seong et al., 2009).
Figure 3. Climate change in the Early Holocene. Upper panel shows the correlation between solar activity and cold phases reflected by highstands of mid-European lake levels (c). Higher indicates wetter. Periods of lower solar activity coincide with periods of higher production of the cosmogenic radionuclides 10Be (a) and 14C (b). Source: Magny et al., 2007. Bottom panel (d) d 18O speleothem (‰, inverted) from Qunf Cave (Southern Oman), a proxy for the strength of the Indian monsoon (as represented, weaker towards bottom). (e) GISP2 non-sea salt [K+] (ppb, inverted), a proxy for Siberian High polar conditions. (f) Relative abundance (%, inverted) of the cold-water dinocyst Spiniferites elongatus in the Aegean sea core SL21. Lower indicates colder. Source: Marino et al., 2009. The position of the two lows for the ~ 2400-yr Bray cycle in this period at 10.3 and 7.7 kyr BP is indicated by the two violet bars.
In the Aegean sea, the increase in abundance of the cold-water dinocyst species Spiniferites elongatus indicates a strong biological response to the climatic deterioration and lower SST in the Eastern Mediterranean during the 10.3 kyr event (Marino et al., 2009; figure 3f). Within dating uncertainties a coincident dryer period for the Indian monsoon can be postulated based on an increase in d 18O speleothem in the Qunf cave of Oman (Marino et al., 2009; figure 3d).
The climatic deterioration described must have had an impact on the prehistoric societies, but then most of the world was populated by hunter-gatherer cultures. In the Fertile Crescent, humans were in a pre- or proto-agricultural state, with increased population densities. At the time of the 10.3 kyr event Göbekli Tepe was being constructed, and Jericho, one of the oldest cities in the world was the first city known to have built a wall, dated precisely at 10.3 kyr BP. The proposed roles for the first city wall fit what we would expect from a climatic deterioration: defensive role against raiders in search of stored food, or protective role against flooding, as mud deposits indicate it had become more common.
Whatever the reason, the walls of Jericho do not appear to have spared the city from the societal changes usually associated to bad climatic conditions. For over 700 years Jericho inhabitants were part of the Pre-Pottery Neolithic A (PPNA) culture, a strange culture characterized for living with their dead (home burials) and construction of the first granaries. The 10.3 kyr event marks the end of PPNA in Jericho (Weninberg et al., 2009; figure 4). There is a gap of about 200 years in radiocarbon dates that suggests a hiatus or strong reduction in construction and habitation and afterwards Jericho’s inhabitants belong to the PPNB culture that has Anatolian influence, which suggest northern immigration, and is characterized by domesticated sheep, and a different more advanced flint toolkit. The PPNB culture disappears at the 8.2 kyr event, a climate pessimum that does not belong to the ~ 2400-yr Bray cycle.
Figure 4. Cultural shift at Jericho coinciding with the 10.3 kyr event. Radiocarbon Data from Jericho arranged according to cultural period (Top: Pre-Pottery Neolithic B, PPNB; Middle: PPNA; Lower: Combined PPNA and PPNB), in comparison to (lower graph): Gaussian smoothed (200 yr) and high-resolution GISP2 potassium (non-sea salt [K+]; ppb) ion proxy for the Siberian High. The calibrated 14C-age distribution (radiocarbon periodization) gives reason to assume a hiatus between PPNA and PPNB. Source: Weninger et al., 2009.
The 7.7 kyr event. The Boreal/Atlantic transition.
The low in the Bray cycle at about 7.7 kyr BP coincides with a cluster of four GSM known as Jericho 0 to 3 and a long period of climate worsening between 7.8 and 7.0 kyr BP. The planet had reached maximum North polar insolation over a millennia before but the melting of the ice sheets was completed at about this time so general conditions were still within the Holocene Hypsithermal. The 7.7 kyr event marks a climatic change in Holocene conditions in northern Europe from the warm relatively dry Boreal period to the warm more humid Atlantic period, reflected in a vegetation change in high northern latitudes with a significant arboreal species expansion.
The 7.7 kyr event coincides with Bond event 5a of increased iceberg discharge in the North Atlantic (Bond et al., 2001; figure 2). It is not however a very strong minimum in the ~ 2400-yr Bray cycle and it is completely overshadowed by the strength of the close 8.2 kyr event, the most profound cooling in the Holocene, that does not belong to the Bray series. Due to that, although detected in most proxy series for climate change, it is seldom studied.
As with any Bray low, it is characterized both by winter cooling reflected in Greenland and the Aegean Sea, and an increase in precipitations in the North Atlantic-European region (figures 3c & 5). A detailed study of the hydrology of the Rhone Valley of France during this time by Berger et al. (2016) shows that, very much like the LIA, this long period of climate deterioration can be subdivided into three subperiods (A, B, and C in figure 5): two cold and wet sub-periods separated by a warm and drier interval. During the cold-humid periods the Citelle river changed to a braided fluvial style, greatly increasing the liquid flow and sediment discharges. This fluvial changes coincide with increased hydrological activity elsewhere in Europe, lower temperatures in the Greenland ice core GISP2 and glacier advances in the Alps (Berger et al., 2016; figure 5).
Figure 5. Hydrological and climate indicators during the 6.5-8.5 kyr BP. Hydrological analysis defines seven phases at Lalo site (Rhone valley, France). Four of them correspond to periods of soil formation (pedogenesis), meandering entrenched Citelle river, and normal sediment discharge. Three periods at 8.2, 7.7, and 7.2 kyr BP show braided Citelle river flow, and enhanced flux and sediment discharges. They coincide with periods of low or decreasing temperatures in Greenland, reduced solar activity, increased hydrology elsewhere in Europe and Alps glacier advances. The blue bands correspond to colder periods in the Greenland ice sheet and alpine areas and to moister signals in western/central hydrosystems, defining the known 8.2, 7.7, and 7.2 kyr events. A, B and C letters indicate the tripartite climate division of the 7.7-7.1 period. Source: Berger et al., 2016.
Solar activity cycle timing indicates that the cold 7.7 kyr event belongs to the ~ 2400-yr Bray cycle, while the 7.2 kyr event belongs to the ~ 1000-yr Eddy cycle. The Eddy cycle is modulated by a longer cycle resulting in very strong lows (both in solar activity and climate worsening) during the early Holocene at 11.2, 10.2, 9.2, 8.2, 7.2, 6.2 and 5.2 kyr BP, followed by subdued lows at 4.3, and 2.3 kyr BP, and again increasingly stronger lows at 1.3 and 0.3 kyr BP. Every two ~ 2400-yr Bray cycle lows, nearly five Eddy cycles have taken place and both Bray and Eddy cycle lows fall close enough to produce a longer significantly colder period. This has happened not only at the LIA (0.4 kyr event), but also at the Mid-Holocene transition (5.2 kyr event), and at the Boreal Oscillation 1 (10.3 kyr event).
7.7 kyr ago the agro-pastoral system was being introduced in Central and Southern Europe by two routes: One in Central Europe following the river valleys originating in the Hungarian region of the Danube, by a culture known as the Linear Pottery Culture or LBK, and the other in Southern Europe following a maritime route from the costs of Greece all the way to the Iberian peninsula by the Cardium Pottery Culture. Both group of farmers descended directly from Neolithic people that established in the Aegean from the Fertile Crescent during the 8.2 kyr event.
The effect on human societies in Central Europe of the cold and wet periods that coincide with periods of reduced solar activity is to reduce the length of the growing season for plants, reducing the output of both natural ecosystems and agro-pastoral systems, resulting in food shortages and food crises. Suddenly the population cannot be sustained. Malnourishment often comes accompanied by plagues. Before increased mortality can reduce the population naturally, some societies resort to emigration and the destination places, also experimenting food shortages, are usually subjected to increased violence in the ensuing fight for resources. It is the Four Horsemen scenario. Archeologists are increasingly aware that the pattern of advances of farming in Europe follows stages that coincide within dating uncertainties with periods of climate deterioration often coincident with periods of reduced solar activity. Agro-pastoral societies appear to have expanded faster during periods of climate worsening, pushed by human overshooting conditions caused by climate crises.
During the 8.2 kyr event, the worse climate event of the Holocene produced by the coincidence of several solar and non-solar causes, farmers from the Levant and Southern Anatolia moved to the shores of the Aegean Sea and expanded into the Balkans. This event marked a significant population decline in the hunter-gatherer societies of Europe (Shennan et al., 2013), probably facilitating the invasion. At around 7.7 kyr BP, when the climate deteriorates again, arises the LBK culture that flourishes and expands into hunter-gatherer areas during the 7.7-7.0 kyr period, substituting the human populations that lived in Central Europe (figure 6c, d, & e). According to Dubouloz (2008), the LBK culture was well adapted to cold, wet periods through construction of robust buildings, placement of villages in tertiary drainage networks, well away from flood risk areas, the importance of cattle-herding, a marked reduction of the Balkan early Neolithic range of cultivated plants, and the practice of autumn sowing in intensively cultivated plots.
Figure 6. The effect of 8th millennium BP climate changes on human societies of Central Europe. (a) Solar activity reconstruction by Steinhilber et al., 2012 (in black) shows the cluster of Jericho 0-3 grand solar minima. (b) Bi-decadal Greenland GISP2 temperatures (2-period averaged, in red) highlights a general correspondence between low solar activity and climate cooling during the two periods marked by light blue bands, that correspond also to the same periods in figure 5. (c) Dubouloz, 2008 Linear Pottery LBK culture demographic analysis (blue lines) shows demographic peaks coincident with temperature valleys just at the time of the major dispersal periods of LBK (blue boxes). (e) Shennan et al., 2013 analysis of Central/West Europe population (in purple) shows that the population decline of the 8.2 kyr event did not recover until the arrival of the LBK agro-pastoral culture, and that this population increase took place during the cold 7.7 and 7.2 events, when LBK was expanding. (d) The agro-pastoral expansion was at the expense of the hunter-gatherer population that did not get diluted but disappeared, according to genetic mitochondrial DNA studies (Brandt et al., 2013).
Dubouloz (2008), and Gronenborn et al. (2013), show that LBK expansion follows a climatic rhythm (figures 6c & 7). LBK forms during the increasingly colder 7.7 kyr event and initiates its dispersal around 7.5 kyr BP at the peak of cold conditions. During the period of warmer drier climate that followed the 7.7 kyr event LBK consolidates a wide territory. The next period of dispersal initiates again at the next cold period around 7.3 kyr BP when LBK crosses the Rhine into Alsace and present time Dutch area. It is 200 years later during another cold period around 7.1 Kyr BP when LBK experiments its last dispersal into the Seine basin. Demographic analysis of LBK habitation (Dubouloz, 2008) indicates that periods of dispersal coincide not only with cold, wet, periods but also with periods of maximal population (figure 6c dark blue lines), suggesting that the difficult conditions that gave the LBK its edge over other human groups, also caused the hardship and population decline that usually instigates climate migration. The arrival of better climate conditions after 7.0 kyr BP probably rendered the harsh climate adaptations of LBK disadvantageous and the culture quickly disorganized, losing its vast circulation networks of raw materials, and disappeared.
Figure 7. Geographical and temporal expansion phases of Linear Pottery Culture (LBK), according to Gronenborn et al., 2008. Expansion phases coincide with periods of cooling in Greenland, and in fact the entire LBK culture appears to encompass the long period of climate worsening between 7700 and 6900 yr BP, known as the Cerin phase in the Alpine region and defined by some authors as a Little Ice Age within the Holocene Climatic Optimum.
The expansion of the agro-pastoral system in Europe marked the end of the hunter-gatherers. Analysis of mitochondrial DNA (mtDNA) frequencies in Central Europe human remains shows that hunter-gatherers mtDNA alleles essentially vanished during the 7.7 kyr event with the arrival of the LBK early Neolithic mtDNA alleles (Brandt et al., 2013; figure 6d). This mtDNA genetic shift and population substitution takes place even as the general population of Western/Central Europe experiments a great population boom due to the arrival of the agro-pastoral societies (Shennan et al., 2013; figure 6e). That the two biggest increases in population in Western/Central Europe take place during the 7.7 and 7.2 kyr events further confirms the expansion of the agro-pastoral system in Central Europe during periods of climate worsening, and lends support to similar expansions elsewhere, like the Aegean expansion during the 8.2 kyr event.
The 5.2 kyr event. The Mid-Holocene Transition and the start of the Neoglacial period.
During the fifth millennium BC (~ 7000-6000 yr BP), the climate of the Atlantic period was in general warm, humid, and stable, constituting ideal conditions for Neolithic farming. In the following millennium, however, the entire climate system of the planet changed, driven by orbital changes in precession and producing a reduction in solar forcing while the oceanic/atmospheric forcing increased in importance. Since 10 kyr BP northern summer insolation has been reducing and southern winter insolation increasing. The balance between northern and southern insolation determines the position of the Inter-Tropical Convergence Zone (ITCZ), a low pressure belt around the planet that organizes the wind patterns, separating the hemispheres and determining the location of the monsoons. The southward displacement of the ITCZ and the changes in insolation during the fourth millennium BC completely altered the planet’s climate, putting an end to the Holocene Climatic Optimum in what is called the Mid-Holocene Transition, setting the path for the colder, dryer world of the Neoglacial Period (figure 8). Between its most notorious effects, the southward displacement of the African monsoon brought an end to the African humid period causing the desertification of the Sahara.
Figure 8. Timing of global glacier fluctuations during the Holocene. Horizontal dark blue bars indicate times of glacier advance; vertical light blue bars are periods of glacier advance based on the global data set. The main periods of the Holocene are indicated. Position of the ~ 2400-yr Bray solar cycle lows are indicated as B1-B5. Timing of glacier advances indicates a progressive cooling of the Holocene. Source: J. Koch & J.J. Clague. 2006.
Within this context of climate instability and increasingly difficult conditions takes place the next low in the Bray cycle between 5.6 and 5.2 kyr BP, coinciding with Bond event 4 of increased iceberg discharge in the North Atlantic (Bond et al., 2001; figure 2). A cluster of three GSM known as Sumerian 1 to 3 are responsible for the reduced solar activity of this low (figure 9a).
GISP2 temperature proxy points to an important but not extraordinary cooling centered at around 5400 BP, but other climate proxies do show exceptional climate for this period. Sea and non-sea salts in GISP2 have been used to reconstruct the polar atmospheric circulation (O’Brien et al., 1995; inverted in figure 9d). The estimated orthogonal function (EOF) that represents the fluxes of these salts displays its highest values respect its baseline of the entire Holocene (figure 9d), indicating very strong high latitude winds and a great polar vortex expansion that could have brought a very cold period over the northern hemisphere while reducing the cold near the pole. Further confirmation for an extraordinary atmospheric circulation at the time of the Bray 3 cycle low comes from Iceland, where Jackson et al. (2005) detected the strongest winds and associated cooling of the entire Holocene in eolic loess deposits between 5600-5100 BP (figure 9e). Both polar circulation changes and Icelandic wind intensity proxies also match very well other known periods of strong climate deteriorations, including all of the Bray cycle lows. The final extinction of the mammoth at its St. Paul island (Alaska) refuge has been dated with precision at 5550 BP, coinciding with the beginning of this climatic deterioration, and attributed to climate change (Graham et al., 2016; figure 9c).
Figure 9. Climate indicators of the 5.2 kyr event. From top to bottom: (a) Solar activity reconstruction by Steinhilber et al., 2012 (in black) shows the cluster of Sumerian 1-3 grand solar minima. (b) Bi-decadal Greenland GISP2 temperatures (2-period averaged, in red) displays a significant cooling centered at 5400 BP. (c) Orange box is mammoth extinction at 5550 ± 100 BP at St. Paul Island (Alaska; Graham et al., 2016). (d) Polar Circulation Index determined by sea and non-sea salt fluxes from GISP2 ice core by O’Brien et al., 1995 (inverted, in blue) manifests at this period one of its biggest departures from baseline (blue straight line) of the entire Holocene. (e) Iceland wind strength determined by eolic loess deposit size by Jackson et al., 2005 (inverted, in green) displays the highest values of the entire 8000-years series at this time. (f) Arboreal/non-arboreal pollen ratio (in magenta) in the Austrian Alps highlights periods of forest retraction (below the baseline) due to colder and wetter climate (this proxy and next in Magny et al., 2006, and references within). (g) Black boxes represent periods of Mid-European higher lake levels. (h) Turquoise box is the Alpine glacier advance period known as Rotmoos II in Austria and Piora oscillation in Switzerland. Blue bands highlight periods of climate deterioration.
In Central Europe, pollen analysis in Alpine Austria shows a retraction of warmer loving arboreal species versus cold-resistant grasses and shrubs in three periods coincident with the three GSM, while increasing precipitation reflects in three corresponding periods of higher lake levels (Magny et al., 2006 and references within; figure 9f & g). At the same time glacier advances are recorded in several parts of the world (figure 2), and in the Alpine region this glacier advance receives the names of Rotmoos II, or Piora oscillation (figure 9h).
The 5.2 kyr event was demonstrated to have been a global phenomenon by Lonnie Thompson (2006) through a variety of records that show that at 5200 BP a strong and sudden cooling took place all over the world. Those records include simultaneous freezing of organic remains at glaciers in Tyrol, Peru, and Western Canada, at the same time the Kilimanjaro ice cores display a sudden and profound cooling. The transition from wet to dry conditions is recorded by changes in the water balance in many African lakes and the driest excursion recorded at the Soreq Cave speleothem. Concurrently, global atmospheric CH4 concentrations recorded in both Greenland and Antarctica stopped decreasing and started to increase for the rest of the Holocene. Dendrochronological records from Irish and Lancashire oaks extending back to 7000 BP exhibit some of their most narrow rings during the decade-long 5145 BP (3195 BC) event (Thompson et al., 2006 and references within).
In Mesopotamia the Uruk culture started to develop around 6000-5800 BP as a response to an increasingly arid period as attested by several Near East proxies (Soreq Cave, Lake Van, and Gulf of Oman), and flourished based on a system of high yield cereal irrigation, efficient canal transportation, and slavery and forced labor. Uruk culture first developed mass production and writing. By 5500 BP Uruk had become the first city-state and started to expand through “colony” settlements founded across the dry-farming portions of the Near East. A short but very arid period at around 5200 BP coincides with the abrupt collapse of the Uruk colony system, as the colonies in the North and smaller settlements in the South are abandoned, and formerly cultivated areas are turned into pastures with the changes of river courses (Brooks, 2012). The 5200 BP arid event in the Near East is also reflected in the abrupt cessation of precipitations over the Nile river delta. At that time an Early Bronze culture was flourishing in Egypt based on a very high population density at the Nile Valley boosted by climate refugees from the Sahara aridification over the previous centuries. The sudden 5200 BP dry period must have increased competition over resources resulting in widespread violence that ended with the subjugation of the Lower (Northern) Nile by the Upper Nile and the unification of Egypt under the first pharaoh at that date (Brooks, 2012).
The difficult climatic conditions through the 5.2 kyr event constituted an authentic disaster for Neolithic farmers in Central Europe. There is a widespread record of settlement abandonment at the Late Neolithic/Chalcolithic-Early Bronze transition, as attested by lake dwellings at France and Switzerland (Arbogast et al., 2006; figure 10b), and the almost complete absence of radiocarbon dates for a period of four centuries in Bulgaria (Weninger et al., 2009; figure 10c). At the same time and with an eerie similarity to the solar activity record, the population of West/Central Europe crashed, revealing the true extent of one of the most difficult periods for humankind (Shennan et al., 2013; figure 10e). The population fell so hard that it is believed that diseases must have played an important role in bringing down the debilitated Neolithic farmers. About this time, at 5800 BP the pneumonic plague (Yersinia pestis) is believed to have emerged for the first time between the Kurgan nomadic herders of the Pontic steppe.
Figure 10. The effect of 6th millennium BP climate changes on human societies of Central Europe. From top to bottom: (a) Solar activity reconstruction by Steinhilber et al., 2012 (in black) shows the cluster of Sumerian 1-3 grand solar minima. (b) Number of Neolithic lake villages (in red) in an area comprising East France and West Switzerland by Arbogast et al., 2006. (c) Cultural shift in Northern Greek area (Bulgaria) from Chalcolithic (Copper Age) to Early Bronze based on radiocarbon data (black boxes) from Weninger et al., 2009. The calibrated 14C-age distribution (radiocarbon periodization, in blue) supports a hiatus during the 5.2 kyr event. (d) The burial dating of Ötzi, the Tyrolean iceman (orange box). (e) Shennan et al., 2013 analysis of Central/West Europe population (in purple) reveals a catastrophic decline coincident with the climatic deterioration, with no recovery until the following millennia. (f) The population decline was accompanied by a shift in mtDNA frequencies (Brandt et al., 2013) that supports a recovery of the descendents of the Paleolithic hunter-gatherer population. (g) Turquoise box corresponds to the Globular Amphora Culture, the first Indo-European culture in Central Europe. Blue columns correspond to the same periods of climatic deterioration as in figure 9.
The decline of the Neolithic farmers of Central Europe allowed a return of the Western European hunter-gatherers as attested by the re-appearance of their genetic signature in areas where they had previously disappeared (Brandt et al., 2013; figure 10f). It was the prelude to the second major invasion and last big population turnover of the Holocene in Europe. Starting around 5350 BP the first nomadic herders from the steppes invaded Central Europe establishing the Globular amphora culture (figure 10g), probably pushed by the climate pessimum conditions and taking advantage of the weak state of Neolithic farmers. A few centuries later came the great invasion by the Battle Axe people (Corded Ware culture). The Indo-European nomads had domesticated the horse, developed the war chariot, acquired the bronze culture and had a patriarchal war-like culture. The Late-Neolithic farmers did not stand a chance, and according to genetics the third known major genetic shift in Europe took place, being the first the Neanderthals substitution by Paleolithic hunter-gatherers and the second the replacement of the hunter-gatherers by Neolithic farmers.
Ötzi, the iceman from Tyrol, was a Neolithic farmer closely related to the LBK people that lived in this violent times and met a violent final (figure 10d). Whether he was killed by other waning Neolithic farmers, by resurging hunter-gatherers now pastoralists, or by invading Indo-European nomadic herders is not possible to know. But as he fled his enemies uphill only to be buried in ice for over 5000 years, he was a testimony of both the changing climate of the Mid-Holocene Transition and its devastating effect on human societies.
The 2.8 kyr event. The Sub-Boreal/Sub-Atlantic Minimum.
The second low in the Bray cycle at about 2.8 kyr BP coincides with the Homer grand solar minimum that took place right after one of the worst climate-induced human catastrophes known to history. This crisis that took place around 3160 BP (1210 BC) has been convincingly linked to a severe drought that affected the Black Sea area and the Eastern Mediterranean in what probably constituted a global climatic event known as the 3.2 kyr event. The drought is likely to have triggered a massive migration by land and sea of the people that lived North and West of the Black Sea and the Balkans. They are collectively known by historians as the Sea Peoples. They destroyed everything in their way bringing about the Late Bronze Age collapse. In a period of less than 50 years the Hittite Empire and the Mycenaean Kingdoms of Greece were destroyed while the New Kingdom of Egypt was brought to its knees by the combined effect of the climatic crisis and the invaders. Every city between Pylos in the Peloponnese and Ashkelon in Gaza was destroyed, including the famous Troy at the Dardanelles, most of them never to be rebuilt. The Late Bronze Age collapse was so destructive that nothing similar has taken place later. The Fall of the Roman Empire pales in comparison. The palace cultures were substituted by small villages and writing was lost in most areas. Greece entered the Greek Dark Ages and Egypt the Third Intermediate Period.
The 3.2 kyr event with its two centuries long megadrought in the Eastern Mediterranean was not associated with a reduction in solar activity nor did it display the climate signature of a solar event. Given the date it is probable that it coincided with a low in the ~1500 year oceanic cycle that also occurred around 1700 years ago at the end of the Roman Warm Period and during the LIA.
The 2.8 kyr event and its associated Homer GSM coincide with Bond event 2a of increased iceberg discharge in the North Atlantic (Bond et al., 2001; figure 2), and with an abrupt climate change from relatively warm and continental to cooler and wetter conditions that marks the transition of the Sub-Boreal to the Sub-Atlantic period. This transition is reflected by an abrupt change in moss species in raised bogs of Northwest Europe at this time (van Geel et al., 1998).
The event started with a 50 year reduction in solar activity and temperatures around 2950 yr BP (figure 11a, b), that brought much needed precipitations to the Eastern Mediterranean, briefly interrupting the drought and allowing a temporary recovery of agriculture in the region (Kaniewski et al., 2013, figure 11f, g). After a return to dry warmer conditions for another century, the Homer minimum started at 2800 yr BP coinciding with an abrupt descent in land and sea surface temperatures (figure 11a, b, c). In Europe the event is very well described as a change to colder, wetter conditions from Northwestern Europe to the Eastern Mediterranean (van Geel et al., 1998; figure 11f). In South America, glacier advances, peat changes, vegetation changes, and sediments, indicate also an abrupt transition to colder, wetter conditions dated at 2750 BP. North America was also affected by a general climate change around 3000-2600 BP towards cooler temperatures and increased precipitation. Increased flooding in the Mississippi basin and an hyperactive storm period in the Gulf coast are dated at that time. Pollen and sediment organic contents in Central Asia support also a coincidental increase in precipitation. Lake Pupuke (New Zealand) isotopic levels indicate a 400 year marked increase in precipitations starting at 2800 BP.
Figure 11. Climate indicators of the 2.8 kyr event. From top to bottom: (a) Solar activity reconstruction by Steinhilber et al., 2012 (in black) shows the Homer grand solar minimum. (b) Northern Hemisphere temperature reconstruction by Kobashi et al., 2013 (in red) displays a significant cooling centered at 2700 BP. (c) Iceland summer sea surface temperature diatom-based reconstruction by Jiang et al., 2015 (in blue) displays a similar simultaneous cooling. (d) Asian summer monsoon proxy from Dongge Cave stalagmite DA oxygen isotope ratio (in green) suggests a weakening of the monsoon starting about 2950 BP and reaching its lowest values at 2750 BP. (e) Tree pollen percentage (in brown) in Cameroon indicates the biggest Central African forest retraction in the entire Holocene starting at about 2800 BP (Maley & Brenac, 1998). (f) Pollen-derived proxy of moisture availability (in purple) at Gibala-Tell Tweini, a city in the ancient Ugarit Kingdom, Northwest Syria, plotted as Principal Components Analysis scores, and (g) pollen-derived proxy of agriculture showing the percentage of pollen coming from cultivated species (in orange) at the same location (Kaniewski et al., 2013) indicating that humid periods at this time in the Eastern Mediterranean coincided with periods of reduced solar activity. (h) White boxes show historic periods in the Eastern Mediterranean. Light blue bands highlight colder and wetter periods that correspond to the 2.8 kyr event. Light orange bar indicates the dry period that constituted the 3.2 kyr event.
While in mid-latitude areas of North America, South America, Europe, Central Asia and Australasia there was an important increase in precipitations, analysis of the Dongge Cave stalagmite DA in China shows one of the biggest reductions in oxygen isotopes ratio of the entire series (Wang et al., 2005; figure 11d), indicative of an important weakening of the summer monsoon and an increase in aridity in South Asia. This prominent weakening of the Asian summer monsoon coincided with what has been described as the “dramatic forest decline” in Central Africa (Maley & Brenac, 1998; figure 11e), the biggest forest reduction in the area for the entire Holocene, of which the forests of Central Africa are still recovering 2000 years later. A possible weakening of the West African monsoon is the likely cause, and although drier conditions started at around 3150 BP, it was around 2750 BP that the forests initiated a marked decline accompanied by expansion of grasses and very dry conditions as attested by the complete drying of several lakes such as lake Sinnda in the Niary valley (Congo). The forest cover opened up and fragmented, and enclosed savannas appeared.
In Europe the 2.8 kyr event separates the Late Bronze Age from the Early Iron Age. The impact of this climatic crisis is somewhat diluted by the previous dramatic crisis of the 3.2 kyr dry event from which there had been no recovery. In fact the increase in precipitations, despite the cooling, was very beneficial for agriculture in drier areas (figure 11f, g) and probably was a significant factor contributing to the end of the Greek Dark Ages (figure 11h). In wet marginal areas however the change had a negative impact. In West Friesland (Netherlands), the Late Bronze settlement phase came to an end coincident with rising water tables as houses started to be built on artificial mounds. The rising water and bog expansion caused a loss of agricultural land forcing the migration of the population to coastal salt marshes, richer in food resources, that also started to appear around 2700 BP (van Geel et al., 1998). The end of habitation in West Friesland is also coincident with the end of Late Bronze lakeside village construction in Central Europe for half a millennium from 2.8 kyr BP. In Central and Western Europe the Late Bronze Urnfield culture gave way to the Early Iron Hallstatt culture that expanded during the 8th century BC (2750-2650 BP), amid the dramatic changes in flora and fauna that accompanied the 2.8 kyr event.
In North America the abrupt climate change at 2.8 BP also separates two cultural periods in the Mississippi basin, supporting the theory that abrupt climate change is a motor for cultural change. The Archaic hunter-gatherer period reached an end in the 3000-2600 BP, marked by a hiatus of several hundred years in riverside settlements, suggesting an abandonment of frequently flooded areas, after which the new settlements belong to the Woodland period, characterized by widespread use of pottery and domesticated plants (Kidder, 2006). Meanwhile in Central Africa the opening of the forest allowed the migration of Bantu speaking, metal working people into areas that are now completely forested.
In the Central Asian steppes the increase in precipitations at 2800 BP brought the Scythians into preeminence. They were semi-nomadic herders of the Eastern Iranian language group from the Tuva region at the intersect of Russia, Mongolia, China, and Kazakhstan. With the wetter climate the steppes expanded and could support huge herds of horses, sheep, and goats. The Scythians abandoned any trace of settlement and became nomadic riders. They are credited with developing mounted warfare using composite bows. By 2700 BP they invaded the Northern Black Sea and the Caucasus pushing the Cimmerians southward into conflict with the Assyrians. The Scythians would continue expanding their territory up to Thrace and Eastern Europe, and played a leading role in the destruction of the Assyrian Empire. To the Greeks they were the prototype of savage barbarians, as they were very war-like and practiced human sacrifices. They were also their main providers of slaves from selling their captives. They inspired two well known Greek myths, the centaurs from their riding combat skills, and the amazons because their women also fought, as one in three women was buried with weapons and many sustained war wounds.
Figure 12. The steppe migration climatic hypothesis. From top to bottom: North Atlantic stacked percentage of ice-rafted debris (a, in red), indicative of iceberg activity (Bond et al., 2001). The main peaks have been labeled with their accepted numbers. Pollen-derived aridity index (b, inverted, in green) from a Central Mongolian lake (Fowell et al., 2003). Light orange bars indicate drought periods. Light blue bars mark millennial humidity maxima. Arrows indicate downward trends in humidity from millennial maxima. Main historic migration events are indicated by boxes, and they took place after humidity maxima within an increasing aridity context. Notice that the Bond events pattern does not correspond to the humidity pattern at the Central Asian steppes, however big changes in humidity tend to coincide with Bond events. Background picture: Scythian king and warriors, drawn after figures on an electrum cup from the Kul’Oba kurgan burial.
In the perpetual conflict between nomads and settlers, climate change appears to have played an essential role in setting the stage for numerous conflicts. From the invasion of Central Europe during the 7.7 kyr event by LBK agro-pastoralists, to the Sea Peoples invasion at the 3.2 kyr event, and the periodical invasions of Eurasia from steppe nomads, we find evidence of climate change creating conditions that resulted in migration as a response, and conflict as a consequence. The productivity of the steppes is very dependent on precipitations and the nomads and their herds cannot rely on stored food during bad years. When analyzing precipitations in Central Asia we find a common pattern for nomadic invasions. They don’t take place during arid periods, but following a maximum in humidity (figure 12b), suggesting that the increase in precipitations, like in the 2.8 kyr event, brings the nomad population and their herds to a maximum, and from that point, any decrease in precipitations, even if not pronounced (figure 12b arrows), places the population in overshooting. The result is a high number of steppe nomads migrating to adjacent areas where easy conquests stimulate further advances, pushing other groups into migration. This pattern is detected not only in the case of the Scythians, but also with the Huns in the 2nd century AD, the Turkic peoples in the 11th century and the Mongols in the 13th century (figure 12b). I have termed this pattern the steppe migration climatic mechanism. A similar pattern is observed with lemmings, that don’t mass migrate during bad years that keep the population in check, but after good years that push the population up creating overshooting conditions.
The 0.4 kyr event. The Little Ice Age.
The first low in the Bray cycle at about 0.4 kyr BP coincides with the Wolff/Spører/Maunder/Dalton cluster of GSM that took place during the coldest period of the Holocene that is generally known as the LIA. Due to have taken place during modern historic times, it is also the most well studied and known cold period.
The LIA coincides with Bond event 0 of increased iceberg discharge in the North Atlantic (Bond et al., 2001; figure 2). Different authors choose a different start for the LIA, since the climate started to deteriorate progressively from its previous warm period at about 1150 AD, but did not become significantly colder than the previous four hundred years until after 1250 AD, becoming a serious problem for human societies of the time after 1300 AD. Other authors however wait until after 1500 AD, when a relatively warm interlude in the 15th century ended. I place the start of the LIA at 1258 AD, a year after the Rinjani eruption (Lombok, Indonesia), the strongest since writing was invented. As with other lows in the Bray cycle, in the LIA there is a pattern of colder phases recognizable in many climate proxies that in general matches quite well the pattern of solar activity (figure 13a). Some temperature reconstructions (Christiansen & Ljunqvist, 2012; figure 13b) show good agreement with solar activity except for starting the initial cooling before solar activity declined with the Wolf minimum and showing a very cold period before the Maunder minimum. This general pattern of four cold phases for the LIA can be defended on the basis of decreasing Mediterranean sea surface temperatures (Versteegh et al., 2007; figure 13c), increased Iceland sea ice (Massé et al., 2008; figure 13d), glacier advances in the Alps (Holzhauser et al., 2008; figure 13e) and Venezuela (Polissar et al., 2006; figure 13f), and increased North Atlantic deposition of ice-rafted debris (Bond et al., 2001). Alpine glaciers do not show an advance during the Spører minimum, and this requires some explanation. Unlike during most GSM, the LIA GSM do not show a pattern of increased Central European precipitations, and during the Spører minimum Central Europe experienced a very dry period (Büntgen et al., 2010; figure 13g). It has been reported that in England due to the fields not been covered in snow during severe winters around 1458 AD the seeds in the field were killed by the cold resulting in several years of poor crops and famine (figure 14e, g, h). The reduction in precipitations would have prevented glacier advances in the region (but not in Venezuela) and might have reduced the growth of Iceland sea ice that was lesser during the Spører minimum that at other minima during the LIA (figure 13d).
Figure 13. Climate indicators of the 0.4 kyr event. From top to bottom: (a) Solar activity reconstruction by Steinhilber et al., 2012 (in black) shows the Wolf, Spører, Maunder, and Dalton grand solar minima. (b) Northern Hemisphere temperature reconstruction by Christiansen & Ljungqvist, 2012 (in red) displaying a pattern that generally matches solar activity. (c) Mediterranean sea surface temperature proxy record (in brown) also displays four cooling periods (Versteegh et al., 2007). (d) A biomarker sea ice proxy from Iceland (in medium blue) agrees well with the sea surface temperature (Massé et al., 2008). (e) Glacier retreat in km from maximum extent in the Alps (in dark blue) does not show glacier advances during the Spører minimum (Holzhauser et al., 2005), while (f) Venezuelan glaciers (in purple) show glacier advances at every minima (Polissar et al., 2006). (g) Precipitation in Central Europe (in green) measured from German oak rings (Büntgen et al., 2010) showing a period of very low precipitation during the Spører minimum (light orange box). Blue bars highlight the four periods of climatic deterioration within the LIA as determined by the climate proxies.
Why would there have been dry conditions in West/Central Europe during the Spører minimum, when we have seen a general pattern of increased precipitations in this region during previous GSM? A possible explanation comes from the ~1500 yr oceanic cycle that based on evidence not presented here appears to have had a low around 400 BP. 2,800 years before (two periods), at around 3200 BP, another dry period brought about the Late Bronze Age collapse, and 1,400 years before (one period), at around 1800 BP, the Roman Empire suffered the Third Century Crisis at a time of poor harvests form their North African and Iberian granaries that greatly contributed to produce the crisis.
The Dalton minimum, that is also unusual in some climatic aspects, including precipitations in Central Europe and glacier advances in Venezuela (figure 13f, g), has the same problem as planet/dwarf planet Pluto. Whether it qualifies as a GSM or not is a matter of opinion as it was both brief and barely showed the required reduction in solar activity. Had the Dalton minimum taken place farther from us and isolated it would probably not have been considered a GSM. To make matters worse, most of the climatic effects during the Dalton minimum are of clear volcanic origin.
Why was the LIA so cold? There was a confluence of causes that made the LIA the coldest period in the Holocene. To start, the Holocene has been cooling since the Climatic Optimum due to an accelerating reduction in obliquity (figure 2), so the LIA started from a lower temperature than previous Bray events. In addition the LIA has been a very long cold event (600 years), longer than similar periods during the Holocene, and its cooling phase was also longer and therefore more profound. The probable reason that the LIA was so long is the coincidence of the lows from three long climate cycles, the ~ 2400-yr Bray solar cycle at around 400 BP, the ~ 1500-yr oceanic cycle also at around 400 BP, and the ~ 1000-yr Eddy solar cycle at around 300 BP, and significant volcanic activity both at its beginning from 1150 AD, and at its end at 1815-1840 AD. The third reason that the LIA was so cold was a very significant contribution from very high volcanic activity during this period.
While volcanic activity during the past 2000 years does not correlate with solar activity, the concentration of strong eruptions during the LIA is so high that it has been proposed that the LIA was mainly due to a volcanic effect on climate. Available evidence however indicates that this is not the case. Moberg et al. (2005) Northern Hemisphere temperature reconstruction has very good yearly resolution and allows to investigate this issue (figure 14b). Very strong isolated volcanic eruptions like the 536 and 540 AD eruptions or the 1453 and 1458 AD eruptions (one of them the Kuwae eruption in Vanuatu) have a very clear effect on temperatures that last 1 to 2 decades at most (figure 14 blue bars), while clusters of eruptions, like around 1257 and 1815, can reduce temperatures for about 4-5 decades. But in every case, after the effect of the volcanic aerosol ends, temperatures recover (figure 14 orange bars), and the general temperature trends continue, whether they were going up or down. It is clear therefore that even though volcanism contributed to the cold and misery of the LIA, and can explain why the cooling at the Wolf and Maunder minima started before there was a significant reduction in solar activity (figure 15a, b, c), it cannot have been the driving factor behind the LIA.
Figure 14. The effect of volcanic forcing on temperatures. (a) Reconstruction of the time and aerosol forcing of major volcanic eruptions from sulfate levels in Greenland and Antarctic ice cores (Sigl et al., 2015) for the past 2000 years. (b) Multi-proxy temperature reconstruction (Moberg et al., 2005) AD 1-1979 (grey line) with a 10 year moving average (black line) and its >80-yr slow component (red line). Light blue columns indicate the temperature reduction after the four biggest volcanic eruptions. Orange columns indicate the temperature recovery after the temperature effect of volcanic eruptions ended.
The effect on human societies of the climate deterioration brought about by the LIA is better known, and the case of the Vikings in Greenland has been brought up often. However it was almost the entire population of the planet who suffered the situation. Data for grain production in England shows that yield per acre decreased following a similar pattern to Northern Hemisphere temperatures (figure 15c, e), probably reflecting the shortening of the season. Back to back and even three in a row (highly unusual) years of bad crops took place on this period (figure 15 vertical grey lines) causing a marked increase in wheat prices (figure 15d, inverted) and major famines. The first one at 1315-17 was the worst, affecting most of Northern and Central Europe and initiating the Crisis of the 14th Century. Climatic factors also determined the increase in contact between rodents and humans in Central Asia, giving rise to the bubonic plague, a different manifestation of the plague that had a near 100% mortality. The plague reached Europe in 1347 and in six years killed over one third of the population in the Black Death pandemic. The population decline was so large (figure 15f), that subsequent crop failures had less effect on people’s famine and wheat prices even one century later, like the bad crops of 1459-61. The plague became recurrent in Europe being always present somewhere in the continent until 1750, and causing major epidemics periodically. The spread to other countries of the Hundred Year War between England, France and Burgundy through the Free Companies of mercenary bandits, the start of the peasants revolts, and the Western Schism in the Church completed the Crisis of the 14th Century, that manifested as a complete failure of the institutions to cope with climate-related natural disasters that were seen then as acts of God.
Figure 15. The effect of LIA climate changes on human societies of Europe. From top to bottom: (a) Solar activity reconstruction by Steinhilber et al., 2012 (in black), shows the Wolf, Spører, Maunder, and Dalton grand solar minima. (b) Volcanic activity reconstruction by Sigl et al., 2015, (in magenta), with dates for the three major eruptions. (c) Northern Hemisphere temperature reconstruction (in red), by Christiansen & Ljungqvist, 2012. (d) Wheat price in Dutch guilders per 100 kg (Lamb, 1995; inverted, in blue), for France (continuous), England (dashed) and Germany (dotted). (e) Three main crops of grain net yield per acre in England, with annual data in pink, and long term trend in brown (Campbell & Ó Gráda, 2011). (f) Northern Hemisphere population growth in % (Zhang et al., 2010; in orange). (g) Northern Hemisphere famine index in events per decade (Zhang et al., 2010; in green). (h) Major famine events (green boxes) and major epidemic and pandemic events (brown boxes). Main historical periods of crisis are shown in boxes at the bottom. Grey vertical lines link multiyear crop failures in (e) with major famines in (h). Light blue boxes are periods of climate deterioration defined in figure 13.
The 15th century was a period of recovery and Renaissance in Europe, despite the severe impact of what is believed to have been the coldest decade of the millennium according to both climate and historic reconstructions, the 1430s during the Spører minimum. However one can never underestimate the capacity of humans to make a difficult situation worse, and so while Japan was developing successful strategies to cope with the challenges that the LIA posed on food production, most of Europe and a great part of the world was again engulfed by man-made crises at the time of the Maunder cold period. The General Crisis of the 17th Century was again a period when the Four Horsemen of Apocalypse rode unchallenged, as the world saw the biggest number, and duration of wars, and war casualties in recorded history to that time. Climatic factors contributed to the general worsening, and glaciers advanced destroying farms, houses, and villages. Climate worsening, together with peasants revolts and war destruction, produced a record number of major famines and accompanying epidemics, to the point of producing a collapse in population growth (figure 15f, g, h). It was a period that coincided with major political upheaval, including frequent government replacements and even state failures. One of the biggest countries in Europe, the Polish-Lithuanian Confederation, completely disappeared, together with one third of its population. The Seven Ill Years of Scotland in the 1690s, were caused by a major famine event in Northern Europe that killed half of the population in Finland and 15% in Scotland, and were decisive for its union to England.
The 18th century was again a period of recovery, after which the climatic deterioration and social problems returned. The inability of the Old Regime to respond to the frequent crises was the cause of the French Revolution when bad crops due to a drought in 1788, and resulting high food prices in 1789 affected again the population of France. The French Revolutionary Wars, followed by the Napoleonic Wars engulfed Europe once more.
However the European farmers from the second half of the 18th century had learned to cope with the challenging climatic conditions through a series of adaptations that constituted the Agricultural Revolution, which in turn helped drive the Industrial Revolution. The final disappearance of the plague from Europe around 1750 was followed however by the appearance of the recurrent cholera pandemics of the 19th century. By 1850 the LIA had been left behind and much better climatic conditions have accompanied human societies since then.
Climatic effects of grand solar minima.
The considerable amount of climate information about the climatic effects of GSM essentially points to an atmospheric effect. The observed phenomenology is usually:
– Increased precipitation in mid and high-latitudes
– Decreased precipitation in tropical and subtropical areas
– Weakening of tropical monsoons.
– Increase in mid and high-latitude wind strength
– Increase in polar circulation
– General cooling
– Sea surface cooling
– Glacier advances
– Increased iceberg activity
These effects are consistent with an expansion of the polar cells, a southward displacement of the polar jet, an equatorial shift of the Ferrel cell and subtropical jet, and a similar displacement of the descending parts of the Hadley Cells that contract. The resulting change in wind patterns would be responsible for the alterations in precipitations and temperatures. These atmospheric changes were described by Joanna Haigh in her landmark 1996 article “The impact of solar variability on climate” where she described the changes found in a general circulation model when simulating changes in solar irradiance and stratospheric ozone. Since then Haigh’s hypothesis has received support not only from paleoclimatology, as seen here, but also from meteorological data reanalysis. The hypothesis states that solar variability affects climate through a bottom up mechanism from surface changes in irradiation coupled to a top down mechanism from stratospheric UV and ozone changes, being the second one the main in terms of effect.
Although changes in oceanic circulation have been implicated by some authors in the climatic changes of the Bray cycle cold events, the global nature of these suggests that oceanic changes, although potentially very important, are probably of secondary nature, induced by atmospheric changes in wind patterns.
Analysis of Holocene climate shows a long term cooling trend punctuated by cold events (figure 2; see also Appendix). For the past seven thousand years, every millennia has been colder on average than the previous one, driven by orbital changes. Obliquity is now decreasing at its fastest rate in 40,000 years and Northern summer insolation is at its minimum value in 20,000 years, an orbital configuration that supports a continuation of the multi-millennial cooling trend. Within this background, the Holocene, since reaching the Climate Optimum, does not display warming events, as any significant multi-centennial warming period is preceded by a similarly significant multi-centennial cooling period, and the Current Warming Period is no exception, as it is preceded by the LIA.
Comparing the Holocene major cold events that include the lows in the Bray cycle at 0.4, 2.8, and 5.2 kyr, plus the major 8.2 kyr event (figure 16) shows that both cooling and posterior warming can last from 2 to 4 centuries. Given that solar activity usually returns quite quickly to normal levels after a SGM, the slow recovery in temperatures suggests that the atmospheric reorganization induced by the changes in solar activity is a slow process, and since the entire event usually takes over half a millennia, the climate seems to settle in a different configuration, as the orbital conditions have changed, and this could be the reason why the lows in the Bray cycle broadly mark the separation between the different climatic periods of the palynological Blytt-Sernander series (Boreal, Atlantic, Sub-Boreal, Sub-Atlantic periods).
Figure 16. Global temperature reconstruction during major Holocene cooling events (blue curves; Marcott et al., 2013 by the differencing method with proxy published dates) has been rescaled in temperature anomaly to match biological, glaciological, and marine sedimentary evidence (see Appendix). Black curve instrumental temperature anomaly data from HadCRUT4.
There is paleoclimatological evidence that a poleward atmospheric expansion of the Hadley and Ferrel cells, and associated wind regimes, including the Southern Westerly Winds strengthening and southern displacement associated with persistently positive phases of the Southern Annular Mode (Antarctic Oscillation) has been taking place, as assessed in the Patagonia (Chile), for over 100 years (Moreno et al., 2014). The expansion of the Hadley cells, that is usually attributed to ozone depletion, has been measured since 1979 at about 1-2° in latitude. The continuation of the Hadley cells expansion is an indication that natural recovery from the LIA has not ended, since it is believed that greenhouse gases contribute little to this phenomenon (Allen et al., 2012) and it seems to have been taking place for over 100 years.
To the natural warming caused by the post-LIA recovery we have added the anthropogenic warming. Nowhere is the anthropogenic effect more noticeable than in the status of the cryosphere. Globally glaciers have retreated to a point last seen around 5000 years ago, during the Mid-Holocene Transition at the start of the Neoglacial period. That is the reason why organic remains like Ötzi, the iceman from Tyrol, from 5200 BP are being uncovered. It is possible that the cryosphere is particularly sensitive to greenhouse gas warming, since water vapor content is very low when the air is very cold. Nevertheless we cannot rule out that global average temperatures are approaching values that took place during the 5.2 kyr event as figure 16 suggests.
The closest orbital analogue for the Holocene or Marine Isotope Stage 1 (MIS 1) is not MIS 11, 407 kyr BP, but MIS 19, 777 kyr BP, when obliquity, precession, and eccentricity all align for both interglacials (Pol et al., 2010; figure 17). In about 1000 years from the current analogue position MIS19 likely started its glacial inception as indicated by an accelerated cooling (figure 17 blue arrow). Afterwards during the next 15,000 years of low obliquity it experienced three Antarctic Isotope Maxima (AIM) that probably represent global warming events (figure 17 orange arrows). Two warming events have also been described for MIS 5e/d interglacial at the Eemian to Early Weichselian glacial transition (Boettger et al., 2009) indicating that strong warming events at the end of an interglacial could be a general feature of glacial inception, something that should be taken into account before claiming that the next glaciation has been delayed or abolished by the Current Warming Period.
Figure 17. Alignment of the Holocene (red) and MIS 19 (black). From top to bottom: a) δDeuterium (‰, temperature proxy) of Holocene (red) and the 650 years averaged signal (dark red); b) MIS 19 δDeuterium (‰) data. In panels a) and b) the thin dashed horizontal lines correspond to the present-day (last millennium average) δD levels; e) eccentricity (dashed, right axis) and North Hemispheric June 21st insolation (solid, left axis); f) precession parameter (dashed, right axis), reported on inversed axis to evolve in phase with the NH insolation, and finally obliquity (°, solid, left axis). Source: Pol et al., 2010. Blue arrow proposed time for MIS 19 glacial inception, orange arrows Antarctic Isotope Maxima that probably represent warming events within the cooling trend.
After about 350 years from the low of the 0.4 kyr event, the post-LIA atmospheric and warming recovery could be close to end. Once it is finished there should be a natural pressure to return to a climatic state more adequate to current orbital conditions that should oppose anthropogenic warming. Whether this will result in a reduced rate of warming or even some cooling it is not possible to know. However barring some unforeseeable strong volcanic eruption the climate conditions from the solar variability point of view appear as favorable as during the long Roman Warm Period. The next low in the ~ 1000-yr Eddy cycle is not due until around 2700 AD (figure 18), followed by a low in the ~ 1500-yr oceanic cycle around 3000 AD. GSM outside the lows of the Eddy and Bray cycles have a much lower probability, since most GSM during the Holocene have taken place close to the lows of these two solar cycles (figure 18d).
It is very likely that anthropogenic warming has been overestimated, since solar variability warming is not properly accounted for, according to paleoclimatic data presented here, due to a poor understanding of its mechanisms. An unfavorable orbital configuration and the end of the post-LIA recovery should help determine the true extent of the anthropogenic warming over the next hundred years.
Assuming that half of the warming since 1950 is due to natural causes, that post-LIA recovery is due to end within the 21st century, that human CO2 emissions will stall and start declining within the next decades, and that the post-LIA natural warming is analogous to the global warming event reflected by a peak in Deuterium (AIM C) that took place 772 kyr BP during MIS 19 interglacial under similar orbital conditions (figures 17 & 18e), a temperature projection can be traced (figure 18f) that shows stable or slightly declining temperatures for the next centuries until the lows in the Eddy and ~ 1500-yr oceanic cycles bring them to negative anomaly (below 1960-1991 average) around 3000 AD, with the low in the Bray cycle around 4000 AD pushing temperatures well below LIA levels.
The period between the last Bray low 400 years ago, and the next in 2000 years that has been proposed to be named Anthropocene should be characterized by Pre-Glacial conditions, with declining temperatures, increased precipitations and flooding events in mid and high latitudes, increased El Niño conditions, increased snow, glacier advances and sea ice expansion. While the northern treeline is likely to retreat, forests should advance in mid and high latitudes, while diminishing in the tropics due to weakened monsoons. An increase in the latitudinal thermal gradient is likely to produce increased storminess and extreme weather events, just the opposite of what global warming has produced. But as planetary cooling is slower than warming, humanity will have ample time to adapt and develop successful strategies for the next global cooling.
Figure 18. Solar cycles and temperatures during the Holocene. Major palinological subdivisions of the Holocene (names on top) match a 2500-yr regular spacing (light blue arches on top). (a) The global temperature reconstruction (black curve; Marcott et al., 2013 by the differencing method with proxy published dates) has been rescaled in temperature anomaly to match biological, glaciological, and marine sedimentary evidence, resulting in the Holocene Climate Optimum being about 1.2°K warmer than LIA (see Appendix). (b) The general temperature trend of the Holocene follows the Earth’s axis obliquity (purple), and significant downside deviations generally match the lows of the ~ 2400-year Bray cycle of solar activity (light blue bands labeled B-1 to B-4 that correspond to similar bands in previous figures). (c) Significant negative climate deviations manifest also in strong increases in iceberg detrital discharges (red curve, inverted; Bond et al., 2001) that generally agree well with the lows in the ~ 2400-year Bray cycle and ~ 1000-year Eddy cycle (orange bands) of solar activity. (d) Solar activity reconstruction (Steinhilber et al., 2012) shows that the majority of grand solar minima correspond very well with Bond events and tend to occur at the lows of the Bray (light blue bars) and Eddy (orange bars) cycles. Significant Holocene climate changes tend to occur when Bray and Eddy cycle lows coincide, like at the Mid-Holocene Transition that ended the Holocene Climatic Optimum and started the Neoglacial period, and the LIA that started the Current Warm Period, now proposed to be named Anthropocene. The regular spacing of the ~ 1000-yr Eddy cycle is shown by orange arches at bottom. Solar cycles can be projected into the future, when the situation could be analogous to interglacial MIS 19 (Marine Isotope Stage) AIM C (Antarctic Isotope Maximum) that is likely to represent a natural global warming event at 771 kyr BP (e). Considering all these factors, temperatures can be projected into the future (f) defining a Pre-Glacial period that could end around 4000 AD in the next glacial inception.
Given that the lows in the Bray cycle usually produce long term changes in climatic conditions after a significant period of climate deterioration that can last several centuries, and given the analogy to MIS 19, the next low in the Bray cycle, at around 4000 AD in two thousand years appears to be the first good candidate for the next glacial inception (figure 18f).
- According to cosmogenic isotopes, there is a near 2450 year solar variability cycle, named here the Bray cycle for its discoverer.
- At every low of this cycle for the past 10,300 years there have been significantly long grand solar minima or clusters of grand solar minima, as determined by cosmogenic isotopes.
- At every low of this cycle for the past 12,800 years there has been a significant period of climate deterioration, generally characterized by global cooling in land and sea, increased iceberg activity, glacier advances, atmospheric changes consistent with equatorward expansion of polar circulation, Hadley cells contraction, and changes in wind and precipitation patterns usually increasing at mid and high latitudes and decreasing at low latitudes with a weakening of the equatorial monsoons.
- After every low of this cycle for the past 12,800 years there has been a long lasting change in climatic conditions manifested mainly in high latitude changes in vegetation, and reflected in the palynological subdivisions of the Holocene.
- The lows in this cycle coincide with periods of crisis for human societies while providing also opportunities for adaptation an advancement, and often coincide with important cultural transitions lending support to the hypothesis that climate change acts as an engine for societal progress.
- Despite a clear and intense paleoclimatic effect, changes in solar activity are not properly accounted for in our current understanding of climate forcings due to our ignorance of the underlying physical mechanisms. This underestimation of solar forcing has the inevitable consequence of an overestimation of anthropogenic forcing.
- The paleoclimatic effects of very low solar activity lend support to Haigh’s hypothesis of solar induced atmospheric changes due to changes in irradiation and stratospheric ozone.
- No support for an approaching grand solar minimum comes from the analysis of the ~ 1000-yr Eddy and ~ 2400-yr Bray cycles. Analysis of these two long solar cycles indicates that we are probably in a prolonged warm period likely to last for several more centuries.
- Analogue analysis suggests an increased probability that the next low in the Bray cycle around 4000 AD could mark the start of the next glacial period.
Moderation note: As with all guest posts, please keep your comments civil and relevant.