New Study Reveals Potential Causes Of The Little Ice Age (LIA), Including Low Solar Activity, Disturbances To The Jet Stream, Volcanic Activity, And High-Variability Of Arctic Sea Ice Cover

A new paper entitled Split Westerlies Over Europe In The Early Little Ice Age was published in Nature on Aug 20, 2022. The study, authored by 15 researchers, lead by Hsun-Ming Hu, delves into the true, sometimes seemingly contradictory causes of the LIA.

Below is my breakdown of the paper using only the scientists’ words–unless otherwise stated; and while I may not necessarily agree with all of the findings and conclusions, the study is most certainly intriguing and furthers the dialogue.


The Little Ice Age (LIA; ca. 1450–1850 C.E.) is the best documented cold period of the past millennium, characterized by high-frequency volcanism, low solar activity, and high variability of Arctic sea-ice cover.

Past studies of LIA Atlantic circulation changes have referenced the North Atlantic Oscillation (NAO), but recent studies have noted that LIA climate patterns appear to possess complexity not captured by an NAO analogue.

Here, we present a new precipitation-sensitive stalagmite record from northern Italy that covers the past 800 years.

We show that in the early LIA (1470–1610 C.E.), increased atmospheric ridging over northern Europe split the climatological westerlies away from central and northern Europe, possibly caused by concurrent Arctic sea-ice reduction.

With ongoing ice melting in the northern high latitudes and decreasing solar irradiance in the coming years, the early LIA may potentially serve as an analogue for European hydroclimatic conditions in the coming decades.


The westerlies over the North Atlantic sector are the primary source of moisture transport to Europe, especially in winter half-year (October-March).

A well-known influence on their path and strength is the pressure difference between the Icelandic Low and the Azores High; variations in this pressure difference give rise to the North Atlantic Oscillation.

NAO; Supplementary Fig. 1A) and affect the precipitation patterns in Europe (Supplementary Fig. 1B):

Fig. 1: Climate and Atlantic sea-level pressure (SLP) variability

a Correlation between SLP and (i) average precipitation at Genoa, Milan and Nice stations (G/M/N PP) (shades); (ii) Climate Prediction Center (CPC) Scandinavia index (SCAND) (contours) during September–February in 1950–2008 C.E. b Vectors: climatological winds at 200-mb level plus regression of 200-mb winds on SCAND index multiplied by two standard deviations of SCAND index during September-February in 1950–2008 C.E., indicating a positive SCAND condition. Shades: correlation between SCAND and ground precipitation during September-February in 1950–2008 C.E. The shades and contours indicate the correlation coefficient(s) above 90% confident level. Climate data are from 20th century reanalysis v3.

Another feature that gives rise to variation in the westerlies are atmospheric blocking events, i.e., persistent and stationary high-pressure systems that block the regular westerly flow for several days to weeks.

Atmospheric blocking over the North Atlantic sector and in particular Scandinavia plays an important role in extreme winter weather over Europe by modulating the trajectory of the westerlies and associated storm tracks.

Their presence prevents the transport of warm and moist air masses, leading to cold spells in northern and central Europe such as the winter of 2010 C.E.

To better understand the variability of the westerlies and the occurrence of atmospheric blockings on different timescales, natural archives that extend back beyond the instrumental era are needed.

Several proxy records have recorded variability in the North Atlantic westerlies over the past millennia and especially during the Little Ice Age (LIA; ca. 1450–1850 C.E.). This was the coldest episode of the past millennium and featured low solar irradiance, high variability in sea ice extent, and frequent volcanic eruptions.

An early reconstruction of the NAO over the last millennium suggested a persistent positive NAO during the Medieval Climate Anomaly prior to a shift to negative NAO conditions during the LIA, but this conclusion has since been questioned by other NAO reconstructions of the LIA.

A more recent multiproxy reconstruction suggests instead that the NAO was neutral to weak positive during the LIA. There have been attempts to reconcile differing proxy reconstructions within the NAO framework, but another possibility is that diverse patterns of large-scale atmospheric circulation may instead be at play.

Here, we present a new autumn-winter precipitation-sensitive stalagmite-based record from northern Italy that spans the past 800 years. Our record documents enhanced atmospheric ridging (anticyclone) over northern Europe accompanied by a split in the climatological westerlies, with the main branch extending towards the Mediterranean and a weaker branch northward towards Greenland. This circulation is characteristic of a pronounced positive phase of the Scandinavian teleconnection pattern, a mode of wintertime large-scale atmospheric circulation variability over the North Atlantic and Europe that is dynamically distinct from the NAO.

The Little Ice Age

During the LIA, the climate in the North Atlantic/European realm was mostly cold, but proxy records show considerable variability on multi-decadal to centennial time-scales within this period.

On centennial time-scales, the Bàsura record [NW Italy] reveals a distinct wet interval in northern Italy during the early LIA (1470–1610 C.E.), suggesting possible strong and persistent positive SCAND[Scandinavia pattern]-like conditions.

Within this period, however, the Bàsura record (Fig. 2f–below) shows strong multi-decadal variability: precipitation increased at the beginning of the LIA, peaked in ~1550 C.E., and then decreased until 1620 C.E. Precipitation reached a second peak in the late 1700 s, followed by a decreasing trend until the end of the LIA.

Fig. 2: Climate records from Europe and northern Africa for the last 800 years

a Na+ concentration in ice core GISP2 as an indicator of wind strength. b Reconstructed North Atlantic Oscillation (NAO) index. c Green: Stalagmite growth rate from Roaring cave (Scotland) as a proxy of precipitation amount. Low growth rate indicates positive NAO phase. Mustard: Bromine concentration from aeolian sediments as a proxy of wind strength. High values indicate strong winds. d Bunker (dark green) and Bleßberg (light green) δ18O record from Germany. Low values reflect a warm and wet climate. e Spannagel δ18O record from Austria. Low values denote a warm climate. f Bàsura Sr/Ca record from northern Italy. g Ifoulki δ18O record from Morocco. Grey vertical band marks the Little Ice Age. The yellow vertical bar highlights the period 1470-1610 C.E. Colored-coded dots and bars are U-Th ages with 2-sigma uncertainties. h Map showing the climate configuration and location of the cited records. The red-blue shades show the correlation coefficient(s) between Scandinavia index and ground precipitation during September-February in 1950–2008 C.E. Climate data are from 20th century reanalysis v3. Triangles and circles mark westerly-affected sites with a wet/warm (blue) or dry/cold (red) climate. The Bàsura cave is highlighted by a dark blue edge. 1: GISP2. 2: Korallgrottan cave. 3: Neflon. 4: Roaring cave. 5: Outer Hebrides. 6: Bunker cave. 7: Bleßberg cave 8: Spannagel cave. 9. Bàsura cave. 10: Kaite cave. 11: Buraca Gloriosa cave. 12: Sofular cave. 13: Ifoulki cave. 14: Chaara cave.

The early LIA wet interval in northern Italy corresponds to a period of strong westerlies over the North Atlantic (1470–1610 C.E.), as suggested by a sodium ion record in a Greenland ice core (Fig. 2a), concurrent with a neutral-positive NAO mode as indicated by a composite NAO reconstruction (Fig. 2a).

The British Isles, however, experienced a relatively dry and less windy climate during this interval, as recorded in a stalagmite-based reconstruction of Roaring cave (Fig. 2c, green) and the aeolian sediment Bromine record from Scotland (Fig. 2c; mustard).

Dry and cold conditions are also inferred for Germany, based on stalagmite δ18O data from Bunker cave (Fig. 2d, dark green) and Bleßberg cave (Fig. 2d, light green), and for high-latitude Sweden (Supplementary Fig. 7a), based on stalagmite δ18O data and lacustrine records.

The precipitation and wind minima in these regions thus do not reflect a neutral-positive NAO mode. Our Sr/Ca record from Bàsura cave (Fig. 2f), on the other hand, suggests that southern Europe experienced warm and humid conditions, as also documented by a stalagmite δ18O record from Spannagel cave, Austria (Fig. 2e).

Similarly, warm and wet conditions are registered by stalagmite records from Portugal (Supplementary Fig. 7b), Spain (Supplementary Fig. 7c), and Turkey (Supplementary Fig. 7d), whereas stalagmite records from Morocco record a dry interval (Fig. 2g and Supplementary Fig. 7f).

Atmospheric ridging over Scandinavia during 1470–1610 C.E.

The dry/cold climate in northern Europe during 1470–1610 C.E. can be reconciled with a neutral-positive NAO phase during this period as suggested by Ortega et al. (2015) through increased atmospheric ridging and hence increased frequency of atmospheric blocking events over Scandinavia.

Comas-Bru and McDermott (2013) have argued that the additional influence of the SCAND pattern on top of an NAO can explain the nonstationarity in the relationship between European winter climate and the NAO.

The combination of the Greenland record and our Bàsura record suggests that such ridging split the westerlies during the early LIA, with a northern branch directed towards the Arctic, consistent with the windy/warm/humid climate over southeastern Greenland, and a southern branch directed towards the Mediterranean, consistent with increased rainfall at Bàsura cave (Fig. 2h).

This climatic setting is similar to the anomalously wet conditions in southeastern Greenland and the northwestern Mediterranean during positive SCAND phases (Fig. 1b and Supplementary Fig. 1d). The positive SCAND phase does not explain the early LIA drying over North Africa, but this could be attributed to an enhanced Azores High during a neutral-positive NAO phase (Fig. 2a), which prevented moisture transport into Morocco (Fig. 2b).

Now onto the ‘meat and potatoes’:

Connections to reduced Arctic sea ice and solar forcing

Fig. 3: Comparison of volcanic forcing, solar activity, sea-ice variability, and Bàsura record

a Global volcanic aerosol forcing. b Violet: concentration of benthic foraminifera from the North Greenland shelf (PS2641-4; Supplementary Fig. 2) as an indicator of sea-ice cover. Pink: ice-rafted debris (IRD) (MSM5/5; Supplementary Fig. 2) from the Fram Strait. High values of these two records denote large sea-ice cover. c Five-point averaged diatom concentration (Thalassiosira nordenskioeldii) from the west Greenland shelf (GA306-4; Supplementary Fig. 2). High value denotes large sea-ice cover. d Red: 40-year smoothed reconstructed late summer Arctic sea-ice extent. e Bàsura Sr/Ca record. f Reconstructed NAO index. g Orange: Atlantic Multidecadal Variability index. Black: total solar irradiance. The intervals of the Spörer Minimum (1388–1558 C.E.) and decreased sea-ice event (1450–1620 C.E.) are marked. The grey vertical bar denotes the Little Ice Age. The yellow vertical bar highlights the period 1470–1610 C.E.

Ice-rafted debris (IRD) records from the Fram Strait (Fig. 3b), foraminiferal-inferred sea-ice records on the North Greenland shelf (Fig. 3b), and diatom-based sea-ice reconstructions from the west Greenland shelf (Fig. 3c) all show extensive sea-ice cover in the North Atlantic during the middle to late 1300 s.

This extensive sea-ice cover in the North Atlantic was presumably induced by intense volcanism (Fig. 3a) and low solar irradiance (Fig. 3g) in the late 1200 s to early 1300 s and in turn it triggered a change in ocean circulation and cooling in Europe starting around 1400 C.E.

From the 1400 s C.E. onwards, sea ice extent decreased significantly (Fig. 3b, c) due to an intrusion of warm Atlantic Water into the North Atlantic, as reflected by a positive phase of the Atlantic Multidecadal Variability (Fig. 3g).

Considering the dating uncertainties of major Bàsura Sr/Ca decreases between 1466 ± 20 and 1559 ± 40 C.E., the period with Scandinavian ridging we inferred falls in the interval with Arctic sea-ice maximum at 1450 C.E. to minimum at 1586 C.E., as this interval was identified as a “decreased sea ice extent” event. Such Scandinavian ridging can be triggered by sea ice loss in the Barent-Kara seas. While direct sea-ice proxy records from Barent-Kara seas are not available, coupled model simulations constrained by the assimilation of available global proxy data indicate a reduction of Barent-Kara sea ice during 1470–1520 C.E.

Reduced sea ice results in increased heat fluxes into the atmosphere that, in turn, excite a stationary Rossby wave propagating towards the southeast and can increase atmospheric ridging over northern Europe [aka ‘Meridional’ jet stream flow]. This ocean-atmosphere feedback could thus provide a mechanism for the link between reduced sea-ice extent and the Scandinavian ridging during 1470–1610 C.E. suggested by our results (Fig. 2g).

The atmospheric ridging in the early LIA could have been further amplified by low solar irradiation (Fig. 3g).

Model simulations and proxy records suggest that solar irradiation changes can have a significant effect on ozone chemistry in the stratosphere that disturbs the polar vortex and thus influences the tropospheric jet stream and atmospheric circulation.

Our Bàsura stalagmite Sr/Ca data, decreasing from 0.055 mmol/mol at 1460 C.E. to 0.035 mmol/mol at 1550 C.E., suggest that Scandinavian ridging progressively strengthened during this interval (Fig. 3e). This 90-yr interval falls during the Spörer Minimum (1388–1558 C.E., Fig. 3g), supporting this linkage between atmospheric changes and solar variability.

The decreased sea ice extent from 1450–1620 C.E. arguably shows a long duration over the past 1400 years and is more pronounced than that during the Medieval Climate Anomaly (ca. 800–1300 C.E.).

Our results present proxy-based evidence of enhanced atmospheric ridging over northern Europe during this multi-decadal interval at the early LIA, possibly in response to the sea ice reduction and solar minimum. Our results thus potentially provide an analogue for the coming decades, when the sun could enter a grand minimum and the Arctic is projected to be ice free by 2030 C.E.

For a full read of the paper with extensive citations and additional graphics click HERE.

See also:

The COLD TIMES are returning, the mid-latitudes are REFREEZING in line with historically low solar activitycloud-nucleating Cosmic Rays, and a meridional jet stream flow (among many other forcings, including the impending release of the Beaufort Gyre).

Social Media channels are restricting Electroverse’s reach: Twitter are purging followers, while Facebook are labeling posts as “false” and have slapped-on crippling page restrictions. And most recently, the CCDH stripped the website of its ability to advertise with Google.

So, be sure to subscribe to receive new post notifications by email. And also consider becoming a Patron or donating via Paypal (button located in the sidebar >>> or scroll down if on mobile). The site receives ZERO funding, and never has.

Any way you can, help me spread the message so others can survive and thrive in the coming times.