Causal explanations for the 4.2 ka BP event are based on the amalgamation of seasonal and annual records of climate variability that was manifest across global regions dominated by different climatic regimes. However, instrumental and paleoclimate data indicate that seasonal climate variability is not always sequential in some regions. The present study investigates the spatial manifestation of the 4.2 ka BP event during the boreal winter season in Eurasia, where climate variability is a function of the spatiotemporal dynamics of the westerly winds. We present a multi-proxy reconstruction of winter climate conditions in Europe, west Asia, and northern Africa between 4.3 and 3.8 ka. Our results show that, while winter temperatures were cold throughout the region, precipitation amounts had a heterogeneous distribution, with regionally significant low values in W Asia, SE Europe, and N Europe and local high values in the N Balkan Peninsula, the Carpathian Mountains, and E and NE Europe. Further, strong northerly winds were dominating in the Middle East and E and NE Europe. Analyzing the relationships between these climatic conditions, we hypothesize that in the extratropical Northern Hemisphere, the 4.2 ka BP event was caused by the strengthening and expansion of the Siberian High, which effectively blocked the moisture-carrying westerlies from reaching W Asia and enhanced outbreaks of cold and dry winds in that region. The behavior of the winter and summer monsoons suggests that when parts of Asia and Europe were experiencing winter droughts, SE Asia was experiencing similar summer droughts, resulting from failed and/or reduced monsoons. Thus, while in the extratropical regions of Eurasia the 4.2 ka BP event was a century-scale winter phenomenon, in the monsoon-dominated regions it may have been a feature of summer climate conditions.
The 4.2 ka BP climate event was a ca. 200–300-year period of synchronous abrupt megadrought, cold temperatures, and windiness that were manifest globally (Walker et al., 2018). Coincident societal collapses and habitat tracking, particularly in regions where archaeological data are both extensive and high resolution, have attracted the attention of many paleoclimatologists and archaeologists since the event's first observation (Gasse and van Campo, 1994; Weiss et al., 1993; Dalfes et al., 1997). Numerous attempts have therefore been made to characterize and quantify the event's nature and to identify its causes at several levels of explanation. These studies have first defined the spatial extent and variability of the event. Megadrought developed abruptly at ca. 4.2 ka cal BP across North America, Andean South America, the Mediterranean basin from Spain to Turkey (except for a few records from N Morocco and S Spain that indicate wetter conditions), Iran, India, Tibet, and north China and Australia (Booth et al., 2005; Staubwasser and Weiss, 2006; Arz et al., 2006; Berkelhammer et al., 2013; Cheng et al., 2015; Weiss, 2016; Kathayat et al., 2018). In southern Asia, failure of the monsoon (Wang et al., 2005) caused widespread droughts (Staubwasser et al., 2003; Berkelhammer et al., 2013). Abrupt cold conditions, however, appeared at ca. 4.2 ka cal BP in the northern North Atlantic (Geirsdóttir et al., 2019), the midlatitudes of the northern Eurasia (Hughes et al., 2000; Mayewski et al., 2004; Andresen and Björck, 2005; Mischke and Zhang, 2010; Larsen et al., 2012; Baker et al., 2017), and Antarctica (Peck et al., 2015) and surrounding oceans (Moros et al., 2009).
These descriptive data have encouraged numerous causal hypotheses at both a
regional and, to a lesser extent, global level for the event's spatiotemporal
distribution and qualities. Possible thermohaline circulation weakening or
shutdown due to freshwater release in the North Atlantic (similar to the
8.2 ka event; Alley et al., 1997), changes in the loading of the Earth's
atmosphere with aerosols or
The abrupt century-scale wet event recorded at very high resolution in North America at Mt. Logan, Yukon (Fisher et al., 2008), suggests an interval of massive advection of tropical air to NW North America linked to El Niño emergence at ca. 4.2 ka (Shulmeister and Lees, 1995). A southward shift of the Intertropical Convergence Zone (ITCZ) could result in the observed cooling at high latitudes and stronger westerlies in the Northern Hemisphere and widespread drought in the tropics (Gasse and Van Campo, 1994; Mayewski et al., 2004). However, the widespread droughts both at the northern and southern margins of the ITCZ suggest that rather than migrating, the ITCZ was narrowing, resulting in megadrought affecting the tropics both south and north of the Equator (Weiss, 2016). Combining the above observations, it seems that while some of the climate variability at ca. 4.2 ka cal BP can be attributed to regionally observable causes, explanations do not yet account for the global nature of the event, which includes the disruption of the westerlies and reduction of moisture advection to continents.
Hypothesized causal explanations for the 4.2 ka BP event are based on the amalgamation of winter, summer, and annual records of climate variability that was manifest in regions dominated by different climatic regimes (e.g., westerly dominated vs. monsoon dominated). However, both instrumental (Balling et al., 1998) and paleoclimate data (Perşoiu et al., 2017) indicate that, on scales ranging from annual to millennial, seasonal climate variability was not always sequential; i.e., warm (cold) summers were not always followed by warm (cold) winters. To address this conundrum, we have investigated the spatial manifestation of the 4.2 ka BP event during winter in a region dominated by climate variability induced by the strength and dynamics of westerly winds. We present a reconstruction of winter climate conditions in Europe, the Near East, and northern Africa between 4.3 and 3.8 ka cal BP. From examination of the spatial distribution of temperature and precipitation excursions during this period, we hypothesize that, in the regions around the Eurasian landmass, the 4.2 ka BP event was caused by the strengthening and expansion of the Siberian high-pressure cell centered over western Asia that caused widespread cooling at midlatitudes in the Northern Hemisphere and aridification in the Middle East. We further discuss the possible causes and mechanisms leading to this phenomenon in a global perspective.
For our analysis, we have selected proxy records from Europe, the Middle
East, northern Africa, and the Atlantic Ocean that cumulatively fulfilled a
set of five criteria on interpretation, chronology, resolution, and nature of
climatic variability. We have selected only records of winter climate
variability, either precipitation amount (the vast majority) or air
temperature, as indicated by the authors. Where no season was indicated we
assumed that the proxy is recording annual climatic changes and we excluded
it from our analysis. We have selected records with at least two absolute age
determinations for the millennium encompassing the 4.2 ka BP event and for
which measurement uncertainties were less than 50 years. A few
high-resolution records from the fringes of the core study area (mainland
continental Europe, the Middle East, and the Mediterranean basin) with age
uncertainties up to 80 years were nevertheless used to refine the spatial
interpretation of the results. To allow for chronological uncertainties, we
have selected records that showed the onset of the local event within
The response of European temperatures and precipitation to the variability of
the Siberian High (SH) (Fig. 1) is based on the Climatic Research Unit
Timeseries (CRU TS) 4.01 dataset (Harris et al., 2014). The relationship
between the SH intensity, sea level pressure (SLP), and 10 m wind has been
analyzed within composite maps for the years when the SH index was greater
(high) and lower (low) than a value of 1 standard deviation. We have
computed composite maps instead of correlation maps because the former
considers the nonlinearities included in the analyzed data. The SH index has
been obtained by averaging the SLP over the key regions between 40 and
65
Climatic conditions at 4.2 ka cal BP in Europe and western Asia.
The background map in
The list of records with information on the type of proxy used and its climatic
interpretation, chronology, and resolution is presented in Table 1
and plotted in Fig. 1. Of the 30 selected proxies, 11 register winter (or
cold season) temperature and 19 register winter precipitation amount. The
temperature-sensitive proxies are from central and northern Europe and SW
Asia, while the precipitation-sensitive proxies cover the entire study area
(between 30
List of proxies used and their interpretation. Numbers in the first column correspond to numbers in Fig. 1.
The 4.2 ka BP event appears generally as cold during winter throughout
Europe, from the Urals to the Atlantic Ocean (Fig. 1a). The highest amplitude
of cooling is seen in the Ural Mountains (Baker et al., 2017) and at high
altitude in the Alps (Fohlmeister et al., 2013), both recorded by speleothem
The composite map of the winter (DJF) sea level pressure (SLP) and
wind at 10 m for the years when the SH index
Cold winters in Europe are associated with either blocking conditions over
central Europe or westward expansion of the high-pressure cell – the
Siberian High – centered over Asia (Cohen et al., 2001; Rîmbu et al.,
2014; Ionita et al., 2018). In the Northern Hemisphere (NH), during the
winter season, three semipermanent and quasi-stationary systems prevail over
the middle to high latitudes: the Icelandic Low (over the Atlantic Ocean), the
Aleutian Low (over the Pacific Ocean), and the Siberian High (SH). The SH is a
semipermanent anticyclone centered over Eurasia and is associated with cold
and dense air masses in the NH and extreme cold winters over Europe and Asia
(Cohen et al., 2001). The composite maps of the SH index, SLP, and 10 m
wind are shown in Fig. 2. As expected, in the case of a positive SH index (HIGH
years, Fig. 2a) an extensive area of strong and positive SLP anomalies
prevails over the whole Eurasian landmass, with the highest anomalies over
Siberia. The positive anomalies in Fig. 2 were found to be statistically
significant at the 5 % level using a two-sample
The robust association between the instrumental-based response of
European and Asian temperatures to a strong SH (base map in Fig. 1) and the
proxy-based reconstructions of winter air temperatures (blue dots in Fig. 1a)
supports the hypothesis that a strengthened SH was active at the time of the
4.2 ka BP event (the possible mechanisms are described below). The
seasonality of the SH implies its onset in mid-autumn, likely linked to
diabatic heating anomalies initiated by snow cover development in NE Siberia
(Foster et al., 1983; Cohen et al., 2001). The cooling resulting from the
expanding snow cover leads to anomalously high SLP in NE Asia, which in turn
results in more snowfall and further strengthening of the SLP anomaly. The
rapidly developing high-pressure and cold anomaly extends westwards, being
limited towards the north and east by the warm ocean SSTs (Cohen et al., 2001).
The end result of an enhanced SH is a westward-rolling high-pressure system
that also brings cold air, heavy snowfall, and strong winds towards both
Europe and central Asia (Ding and Krishnamurti, 1987; Gong and Ho, 2002;
Panagiotopoulos et al., 2005). The development of the SH also leads to
strengthening of the subtropical jet stream over SE China (Panagiotopoulous
et al., 2005), a characteristic feature of the East Asian winter monsoon
(EAWM; Cheang, 1987), and instrumental data (Wu and Wang, 2002; Jhun and Lee,
2004) show that strengthening of the SH results in a stronger than average
EAWM. Paleoclimate data from Asia further indicate the strengthening of the
EAWM at 4.2 ka cal BP (e.g., Hao et al., 2017; Giosan et al., 2018),
likely linked to stronger and more frequent outbreaks of cold air from the
core of the SH. Similarly, paleoclimate records from the outer limits of the
region impacted by the SH have documented significant increases in the
strength of the local winds, frequently a local diagnostic signature of the
4.2 ka BP event. Various proxies in different sedimentary archives across
west Asia have documented strong northerly winds at 4.2 ka cal BP: soil
micromorphology at Tell Leilan (NE Syria; Weiss et al., 1993), detrital
dolomite and calcite in Gulf of Oman (Cullen et al., 2000) and Red Sea (Arz
et al., 2006) marine cores, high Ti counts in Lake Neor on the Iranian plateau
(Sharifi et al., 2015),
Inferred winter climatic conditions between
The strengthened EAWM and high windiness in SW Asia are consistent with the climatology of the SH, with a strong clockwise flow of anomalously cold air from its center of action, located in north–central Asia (Fig. 2a). Paleoclimate records from Europe also document 4.2 ka BP-related increases in wind strength and/or storminess, such as at the raised bogs in SW Sweden (linked to cold temperatures and possible increased sea ice; Björck and Clemmensen, 2004), aeolian sandbanks in coastal Denmark (Clemmensen et al., 2003; Goslin et al., 2018), and Gotland, Baltic Sea (Muschitiello et al., 2013) (Fig. 3), where strong winter winds and high precipitation, the product of Baltic Sea moisture delivered by intense easterly winds, indicate the reinforcement and westwards expansion of the Siberian High. These data suggest that a belt of strong winds extended around the core region of the SH, from East Asia through west Asia and SE Europe up to the Baltic and North Sea (Fig. 3).
Summarizing the above information, at ca. 4.2 ka a cold temperature anomaly settled over most of Europe from the Ural Mountains to the Atlantic Ocean, including Scandinavia, and extending to the region south and east of the Caspian Sea, likely the result of a deeper than average Siberian High. Further, anomalously high SLP over this region resulted in the strengthening of winter winds in eastern, southern, and southwestern Asia and eastern and northeastern Europe, linked to the clockwise and outward movement of cold air from the core of the SH-impacted region.
Data from winter precipitation records at the time of the 4.2 ka BP event
suggest a far more complex image of precipitation distribution across our
study area (Fig. 1b) compared with the simpler temperature distribution
dipole (Fig. 1a). The SE Mediterranean and the wider Middle East were dry
(Bini et al., 2018), with most of
the droughts occurring rather abruptly (Cheng et al., 2015; Sharifi et al.,
2015; Dean et al., 2018). In the wider Mediterranean basin, winter drought
was also recorded in S Greece (Finné et al., 2017), north–central Italy
(Drysdale et al., 2006; Regattieri et al., 2014; Isola et al., 2019), N
Algeria (Ruan et al., 2016), and central Spain (Smith et al., 2016), with all
records pointing towards an abrupt onset and a ca. 150–200-year duration.
Against this background of generalized megadrought in the Mediterranean, in two regions an increase in winter
precipitation amounts was registered (Fig. 1b), most notably in NW Africa and
SW Europe (Walczak et al., 2015; Wassenburg et al., 2016; Zielhofer et al.,
2017) as well as in the central Balkans and Carpathian Mountains (Zanchetta
et al., 2012; Panait et al., 2017; Perşoiu et al., 2017). Multiple
records and different proxies (speleothem and lake sediment
Apart from SW Europe, the Balkans, and the Carpathian Mountains, high precipitation at 4.2 ka in Europe was also registered in a lake at the foothills of the Alps (Cartier et al., 2019) and in Gotland, the Baltic Sea (Muschitiello et al., 2013). In the Alps, high flooding activity at 4.2 ka was linked to increased autumn precipitation (Cartier et al., 2019), while in the Baltic, high winter precipitation is consistent with strong easterly winds picking up local moisture form the Baltic Sea (Muschitiello et al., 2013, as well as the discussion in Sect. 3.1 above).
The winter precipitation record in Europe and the Middle East can now be
summarized as follows (Fig. 1b).
Regionally significant dry conditions occurred during winter in the
Middle East, southern Europe (Italy and Greece), northern Africa, and
on a band stretching from the Atlantic Ocean, through the north European
plains, towards eastern Europe, including Scandinavia. Regionally significant wet conditions occurred during winter around the
Strait of Gibraltar (northern Morocco and southern Spain) and in the northern
Balkan Peninsula (including the Carpathian Mountains).
The distribution of precipitation minima and maxima on the western (Atlantic)
side of Europe is similar to that occurring during the negative phase of the
North Atlantic Oscillation (NAO), one of the main modes of climate
variability in Europe (Hurrell et al., 2013) that is mainly active during winter.
The NAO is defined as the difference in atmospheric pressure between the
Icelandic Low and the Azores High. A below average difference between the two
pressure systems (negative NAO, or NAO
The paleoclimate evidence we have compiled collectively suggests cold winter
conditions in N Asia and Europe, likely induced by cold air outbreaks from
high-pressure fields located over Siberia, conditions that in modern climates
are associated with a strong Siberian High. The sole reconstruction of the
past behavior of the Siberian High is based on an analysis of the
continental-sourced nss K
Previous studies based on instrumental, tree ring, and ice core impurity content have shown a clear link between a strong SH and a cold and dry climate in Europe (Meeker and Mayewski, 2002; D'Arrigo et al., 2005). The close match between the impact of the SH on temperature and precipitation amounts and the reconstructed climate (Fig. 1) suggests that at 4.2 ka there was a stronger than usual SH, leading to cooling in Asia and Europe, disruption of the westerlies, and drought in the Middle East (Fig. 3). The possible causes of this chain of events remains, however, elusive. Some possible forcings behind climate changes do not appear abruptly at 4.2 ka. Orbital forcing resulted in low winter insolation in the Northern Hemisphere and comparably high, but decreasing, summer insulation, while radiative forcing was going through a remarkably long state of stable, albeit high, values (Steinhilber et al., 2009). Volcanic and greenhouse forcing were both low and stable at 4.2 ka, with no abrupt changes (e.g., Wanner et al., 2011). The high contrast between summer and winter insolation would have resulted in a weak polar vortex (Orme et al., 2017) and thus more meridional polar vortex and associated southward-displaced storm tracks in the Atlantic. The same meridionally displaced polar vortex could have led to cold air advection to N Asia and the early onset of winter, with earlier formation of snow cover.
The early presence and persistence of snow in NE Asia is one of the most
important triggers of a strong SH (Cohen et al., 2001; Wu and Wang, 2002).
The causes and mechanisms by which snow accumulates in early winter in NE
Asia are elusive; possible causes include a positive feedback from the NAO,
with NAO
The above inferences suggest that at ca. 4.2 ka, orbital and solar forcing led to a chain of atmospheric changes, transmitted and amplified by ocean circulation, which caused abrupt cold and dry climatic conditions in northern Eurasia. These atmospheric changes included the weakening of the polar vortex and southward advection of cold air over N Asia. The enhanced meridional transport generated earlier and more persistent autumn snow cover. In turn, this led to the onset of a stronger than usual Siberian High that lowered Eurasian surface temperatures with strong outbreaks of cold and dry northerly winds in a belt stretching from eastern Asia through portions of west Asia and central and northern Europe. The above average SLP associated with the strengthened SH resulted in the blocking of the moisture-bearing westerlies in Europe. Megadrought across the Mediterranean and west Asia may have also been enhanced by the weak and southward-displaced Atlantic storm track that resulted from lower than average NAO conditions. The conditions associated with a weak polar vortex strengthened sea ice towards the Nordic Seas, further contributing to the weakening of the thermohaline circulation and reduction in the strength of the NAO and the westerlies.
We have gathered records of changes in winter temperature, precipitation amount, and associated climatic conditions in the wider Eurasian region during the 4.2 ka BP event. The data show that 4200 years ago cold winter temperature anomalies dominated western Asia and most of Europe. The strength of winter winds in eastern and southern Asia was strongly enhanced, while those in western Europe weakened. Regionally significant droughts settled over the Middle East, southern and northern Europe, and western Asia, while locally significant increases in precipitation were reconstructed in the Balkan Peninsula, the Carpathian Mountains, around the Baltic Sea, and in NW Africa and southern Spain.
We propose a multi-causal hypothesis of partially mutual reinforcing vectors and mechanisms to explain the regionally coherent north Eurasian and adjacent region 4.2 ka phenomena. Thus, we hypothesize that before and at 4.2 ka, the orbitally induced high insolation gradient between summer and winter in the high latitudes of the Northern Hemisphere led to a weakening of the polar vortex, resulting in a meandering jet that promoted an early onset of winter in NE Siberia. In turn, this resulted in decreasing temperatures and an early and stronger Siberian High that expanded south and westwards, bringing cold and dry conditions across Eurasia. The same circulation pattern led to more sea ice export in the North Atlantic and weakening of the subpolar gyre, resulting in the slowdown of the thermohaline circulation and a decrease in sea level pressure around Iceland, thus possibly leading to a shift towards a negative phase of the North Atlantic Oscillation. In turn, these changes resulted in weaker and southward-displaced westerly winds across Europe. However, the high-pressure systems in Europe effectively blocked these weakened westerlies, causing reduced winter precipitation and drought conditions across the eastern Mediterranean and western Asia. Clockwise circulation around the Asia-centered high-pressure field induced strong northerly winds in southern and western Asia and in eastern Europe. Further, the strong thermal pressure gradient between central and northern Asia and the Indian and Pacific oceans determined the strengthening of the East Asian and Indian winter monsoons. However, given the drought in the source regions of the winter monsoon, these strengthened winds did not result in increased moisture advection. Nevertheless, several regions experienced a slight increase in winter precipitation due to strong winds picking up moisture from local sources (NW Africa, N Balkan Peninsula and the Carpathian Mountains, the Baltic region).
In the context of the above data and description, we suggest that, in the extratropical regions of Eurasia, the 4.2 ka BP event was an abrupt century-scale boreal winter phenomenon. While not the subject of our study, we note that a clear antiphase behavior of the winter and summer monsoons has been evidenced (Kang et al., 2018), suggesting that at the times when parts of Asia and Europe were experiencing winter droughts related to strong, dry, winter monsoons, SE Asia was experiencing similar abrupt summer megadroughts resulting from failed and/or reduced monsoons. Whether these were caused by the same orbitally induced changes and/or teleconnections transmitted via the weakened AMOC are questions to be investigated within future proxy-based and modeling studies. Especially important would be winter precipitation records from western Asia and eastern Europe, as well winter temperature records from southern Europe and the wider Middle East, where such data are scarce. Further, most of the winter records are of low resolution and/or with poor chronological control such that improvements in these fields are required to further test our hypothesis.
All data in this study have been obtained from the cited references.
AP designed the hypothesis, AP and HW collected, reviewed, and analyzed the paleoclimate data, AP and MI discussed the climatology of the SH, and AP synthesized the evidence and wrote the text with input from HW and MI. MI drew the base maps in Fig. 1 and created Fig. 2.
The authors declare that they have no conflict of interest.
This article is part of the special issue “The 4.2 ka BP climatic event”. It is a result of “The 4.2 ka BP Event: An International Workshop”, Pisa, Italy, 10–12 January 2018.
The Scărişoara ice core analyses in Romania were partially supported by UEFISCDI Romania through grant nos. PN-III-P1-1.1-TE-2016-2210 and PNII-RU-TE-2014-4-1993 awarded to Aurel Perşoiu, ELAC2014/DCC-0178/FP7, and contract 18PFE/16.10.2018 funded by the Ministry of Research and Innovation in Romania within Program 1 – Development of national research and development system, Subprogram 1.2 – Institutional Performance-RDI excellence funding projects. Aurel Perşoiu further acknowledges support from SP-PANA-W1010. The Associazione Italiana per lo studio del Quaternario and the organizers of the “4.2 ka BP Event: An International Workshop” (Pisa, Italy) financially supported Aurel Perşoiu for attendance at the workshop where some of the ideas presented here were born. Monica Ionita was funded by the Helmholtz Climate Initiative REKLIM and by the Polar Regions and Coasts in the Changing Earth System (PACES) program of the AWI. We thank the editor, Giovanni Zanchetta, and two anonymous referees for their comments.
This paper was edited by Giovanni Zanchetta and reviewed by two anonymous referees.