The ‘Little Ice Age’ in the Southern Hemisphere in the context of the last 3000 years: Peat-based proxy-climate data from Tierra del Fuego

Abstract

The so-called ‘Little Ice Age’ (LIA) of the 15th to 19th centuries ad is well-attested from much of Europe and from some other parts of the Northern Hemisphere. It has been attributed to solar forcing, associated with reduced solar activity, notably during the Spörer, Maunder and Dalton solar minima, although other causes have also been proposed and feature strongly in recent papers. Detection of the LIA in some proxy-climate records from the Southern Hemisphere is less clear, leading to suggestions that the LIA was perhaps not a global phenomenon. Resolving this issue requires more data from the Southern Hemisphere. We present proxy-climate data (plant macrofossils; peat humification) covering the past three millennia from an ombrotrophic mire (peat bog) in Tierra del Fuego, southern South America, but focus our discussion on the period traditionally associated with the LIA. During parts of this time, the mire surface was apparently relatively dry compared with much of its 3000-year record. It was reported earlier that a particularly dry episode in the mire coincided with the 2800 cal. BP ‘solar’ event (since identified as a Grand Solar Minimum), which was attributed to solar-driven changes in atmospheric circulation, and more specifically to a shift in position of the Westerlies. Parts of the LIA record show a similar shift to dryness, and we invoke a similar cause. The shifts to and from dry episodes are abrupt. These new data support the concept of a global LIA, and for at least the intense dry episodes might reinforce the claim for solar forcing of parts of the LIA climate.

Keywords

‘Little Ice Age’peat humification proxy-climate data solar forcing Tierra del Fuego

Introduction

While the timing of its onset varies both geographically and between archives, the so-called ‘Little Ice Age’ (LIA) from the 15th to 19th centuries ad has long been characterised in Europe as a period of advancing glaciers (Le Roy Ladurie, 1971 [1967]), relatively cold winters (Grove, 1988) and, at times, failing harvests (Lamb, 1995). Often characterised as having intensely cold winters in Northwest Europe under anticyclonic conditions, the late autumn/winter weather in the LIA was much more varied than this and included a particularly severe storm event in England in late November ad 1703, classically reported by Defoe (1704).

With glaciers in Southern Norway reaching their maximum Holocene extent variously from the early 18th to the late 19th century ad (Nesje, 2009), the LIA has long been regarded in Europe as the most severe prolonged downturn in climate warmth since the late-glacial. However, it was not as deep as the centennial-scale downturn of the so-called 8.2 kyr event of the early Holocene, which is now regarded as the most severe climate shift in the post-glacial (Daley et al., 2011; Kobashi et al., 2007). A pronounced putative global climate shift in c. 2800 cal. BP (Chambers et al., 2007; Van Geel et al., 1996) and a decade of relative coolness in ad 540s, recorded especially in Europe but also elsewhere (Baillie, 2001), are other candidates for the second-most severe downturn since the late-glacial. In those instances, the inferred climate changes are abrupt, but short-lived. By contrast, the LIA is of much longer duration than any other inferred reduction in climate warmth during the Holocene, although the climate within the LIA was variable and was not uniformly colder, either temporally or geographically.

While proxy-climate data from individual sites show that the period from c. ad 1420 to 1850 was not cold continuously, it can be viewed as a multi-centennial period in which temperatures in the Northern Hemisphere were on average some 0.8°C lower than the 1961–1990 reference period (Bradley and Jones, 1993; Mann, 2002; Mann et al., 1998), although some archival datasets imply far greater temperature reductions in specific locations (e.g. in northern Norway; Lauritzen and Lundberg, 1999). More recent examination of tree-ring data over the last millennium in northern Scandinavia by Esper et al. (2012) suggests some reappraisal of the magnitude of temperature changes is required.

In Europe and North America, a wide range of proxy-climate methods has been used to infer the climate over the LIA. Detection of the LIA elsewhere has been patchy. Mann (2002) reviewed some of this evidence, which was largely terrestrially based, and noted rather subdued cooler overall temperatures in eastern China, more pronounced cooling for a time in the tropical Andes and evidence of glacial advances in New Zealand, but that there were large regional variations (Jones and Mann, 2004). Indeed, subsequently, Schaefer et al. (2009) reported glacier fluctuations in New Zealand that are not fully in accord with those in the Northern Hemisphere.

Whether the LIA can be regarded as a ‘global’ event has been disputed (compare, for example, views expressed in Jones et al., 2009; Rhodes et al., 2012). Part of the reason for caution is the paucity of LIA proxy-climate records from southern latitudes, which contrasts with the far greater concentration of proxy-climate studies of the LIA in the Northern Hemisphere. This is attributable in part to the large expanse of ocean compared with land in the Southern Hemisphere (SH), which restricts the availability of suitable SH terrestrial sites to provide land-based records. However, new data from the Antarctic Peninsula suggest the period of lowest ice melt and coolest temperatures in the last millennium was c. ad 1410 and 1460, ‘when mean temperature was 1.6°C lower than that of 1981–2000’ (Abram et al., 2013: 404). Ocean-margin ice-core records point to significant LIA climate changes in the SH, averaging 1.6 ± 1.4°C cooler than during the last 150 years (Rhodes et al., 2012); these authors emphasise that other SH proxy-climate records are required to enrich the LIA database.

The depth, geographical extent and duration of the LIA have a bearing on expectations of future climate, because both the trajectory of recent climate changes and computer projections for future climate are made against a background of the proxy-climate of the past millennium (Mann, 2007; Mann et al., 1998) or past two millennia (e.g. IPCC, 2013; Mann et al., 1999; 2008) in which both the Medieval Climate Anomaly and LIA feature – more prominently in some reconstructions than in others (Mann et al., 2009; cf. Esper et al., 2012). The importance of being able to reconstruct climates of past recent millennia was stressed by McCarroll (2010), who stated, ‘The wider the range of climate parameters that can be reconstructed, and the wider the range of locations, the more constraint will be provided’ (p. 1661).

Here, in this Research Report, we present proxy-climate data from a peat bog in Tierra del Fuego, southern South America, for the last 3000 years. The methods used to derive a proxy-climate record from this southerly site are identical to those that have been used widely in northern and central Europe and in parts of North America (Chambers et al., 2012) and for which a set of field and laboratory protocols has been established (De Vleeschouwer et al., 2010b). We focus our discussion on the period of time commonly associated with the LIA, and draw comparisons within the 3000-year record, including a previously reported particularly intense but short-lived climate shift in that record, dated to c. 2800 cal. BP (Chambers et al., 2007), which corresponded in time with one of opposite direction, reported from Europe (Van Geel and Renssen, 1998). Our proxy-climate record is based largely on bog surface wetness (sensu Barber et al., 1994), rather than temperature. Although there may be a temperature component in the data, this is unlikely to be a major component at such southerly cool latitudes; indeed, on the basis of changes within the period of meteorological records, it seems more likely that precipitation changes in response to variation in the Westerlies would have the strongest influence on bog surface wetness (Daley et al., 2012).

Our data can be seen in the context of claims made recently by Moreno et al. (2014) based on stratigraphic data from a lake in southwest Patagonia that there is ‘in-phase inter-hemispheric coupling of palaeoclimate over the last 3000 years through atmospheric teleconnections’ (p. 1).

Study site and field methods

The field location for analyses reported here is a Sphagnum-dominated raised mire, located c. 6 km north-east of Ushuaia, in the broad Andorra valley (Figure 1), Southern Argentina. Details of the mire vegetation are provided in Mauquoy et al. (2004) in which results of pollen analysis on a separate core of peat were reported. The 2-m core of peat (labelled AND-2) for the present study was extracted later by DM using a series of 10 cm × 10 cm × 50 cm stainless steel boxes, hammered into a cleaned peat face exposed during the creation of ditches for peatland harvesting (De Vleeschouwer et al., 2010a), and sub-sampled into 1 cm slices, which were bagged and transported back to the United Kingdom, for laboratory analysis.

Figure 1.

Map to show location of Valle de Andorra mire and Ushuaia, Tierra del Fuego.

Laboratory methods

Two standard methods were used for producing a proxy-climate record from peats: plant macrofossil analysis (Mauquoy et al., 2010), conducted by DM in Uppsala, and determination of peat humification (Blackford and Chambers, 1993), the latter method modified (as in Chambers et al., 2011) and conducted by SB at Cheltenham. Radiocarbon dates for the core were requested by FMC from the Natural Environment Research Council, UK, in two submissions: a rangefinder series that focused on major shifts in the peat-humification data, followed by groups of samples that amplified the original rangefinder data (see Table 1), to facilitate possible wiggle-matching (Piotrowska et al., 2011; Van Geel and Mook, 1989). With one exception (an Ericales macrofossil, close to the base of the core), all dates were carried out on Sphagnum leaf macrofossils, separated assiduously from the peat by SB, following the methods used in the Northern Hemisphere ACCROTELM project (Chambers, 2006). An attempt to date the upper part of the core by ²¹⁰Pb was unsuccessful, probably owing to low radon source in this region.

Table 1.

Radiocarbon dates for Core AND-2.

Sample name	Material	Depth (cm)	Date (¹⁴C BP)
SUERC-2468	Sphagnum	37	395 ± 24
AA-54892	Sphagnum	40	468 ± 38
SUERC-2469	Sphagnum	43	394 ± 24
SUERC-2470	Sphagnum	46	448 ± 20
SUERC-2471	Sphagnum	49	402 ± 18
SUERC-2472	Sphagnum	52	565 ± 24
SUERC-2473	Sphagnum	70	952 ± 22
AA-54893	Sphagnum	73	1101 ± 39
SUERC-2476	Sphagnum	76	961 ± 22
SUERC-2477	Sphagnum	79	1050 ± 25
SUERC-2478	Sphagnum	82	1029 ± 20
SUERC-3033	Sphagnum	85	1131 ± 20
SUERC-2481	Sphagnum	118	1721 ± 25
SUERC-2482	Sphagnum	121	1722 ± 22
SUERC-2483	Sphagnum	124	1747 ± 25
SUERC-2486	Sphagnum	127	1813 ± 22
SUERC-2487	Sphagnum	130	1796 ± 25
AA-54894	Sphagnum	133	1810 ± 38
SUERC-3877	Sphagnum	153	2196 ± 25
SUERC-3878	Sphagnum	157	2236 ± 25
SUERC-3881	Sphagnum	161	2256 ± 19
SUERC-3882	Sphagnum	165	2255 ± 25
SUERC-3883	Sphagnum	169	2238 ± 25
SUERC-3884	Sphagnum	173	2241 ± 21
SUERC-3886	Sphagnum	180	2398 ± 23
SUERC-3887	Sphagnum	184	2418 ± 25
SUERC-3888	Sphagnum	188	2465 ± 21
SUERC-3891	Sphagnum	192	2508 ± 26
SUERC-3892	Sphagnum	196	2577 ± 23
SUERC-5570	Sphagnum	198	2717 ± 30
SUERC-6312	Ericales wood	199	2704 ± 22
SUERC-5573	Sphagnum	199	2805 ± 30
AA-54895	Sphagnum	200	2866 ± 41

Initial deposition-rate curves were produced using a precursor of the ‘Bacon’ method of Blaauw and Christen (2011); limitations of this method for extrapolation when the basal age is unknown were exposed by initial lack of dating control beyond the base of the core, and so further samples were dated close to the base, to anchor the basal age by wiggle-matching. Dated samples in this basal-age wiggle match fitted the calibration curve very closely (Figure 2; see also Figure 3 in Chambers et al., 2007), and so permitted precise and accurate dating of the base of the sequence. The closely spaced sets of radiocarbon dated samples at intervals along the core have ensured precision and accuracy in dating the proxy-climate records, except in the uppermost layers, for which the sampling date provides a topmost dating-anchor point.

Figure 2.

Radiocarbon-calibrated deposition-rate curve for the AND-2 core, produced using the Bacon method of Blaauw and Christen (2011), showing groups of samples that assist with ‘wiggle-matching’ (Van Geel and Mook, 1989), resulting in a tight chronology for most of the past 3000 years. Grey-scale plot indicates relative precision of dating; calibration based on SHCal13.

Figure 3.

Percentages of identifiable Sphagnum (showing proportion of Sphagnum magellanicum) leaves and stems, and percentage light transmission through alkali extract of the AND-2 core peat (low values indicate dryness; high values indicate wetness) in contiguous 1 cm samples for the past three millennia, plotted against Empetrum/Ericaceae roots (indicating relative dryness). Age estimates derive from SHCal04, to be consistent with data published in Chambers et al. (2007); there are minor differences from SHCal13 used for Figure 2.

Results

Proxy-climate data for the past 3000 years are presented in a composite diagram (Figure 3), showing peat humification (expressed inversely as percentage light transmission) and the proportions of Sphagnum leaves and Empetrum/Ericaceae roots in plant macrofossils recorded from the mire. These data show that for most of the past 3000 years, the principal peat former has been Sphagnum, normally with over 70% and typically 75–95% of plant macrofossils being Sphagnum remains. Overwhelmingly, the dominant species has been S. magellanicum, with some 80–100% of the Sphagnum leaves attributable to this species. Owing to this dominance, it was not necessary to apply k-value corrections to the humification data (cf. Hughes et al., 2012). While original above-ground macrofossils (e.g. Sphagnum leaves) and below-surface remains (e.g. Empetrum/Ericaceae roots) in the same horizon may not be exactly contemporaneous, the latter are relatively shallow-rooted in ombrotrophic peats.

There are notable exceptions to the Sphagnum dominance: (1) a short-lived but pronounced episode c. 2800 cal. BP in which Sphagnum leaves are below 30% of plant macrofossils, while Empetrum/Ericaceae macrofossils were more abundant at >50%; (2) a short-lived episode c. 750 cal. BP (c. cal. ad 1200) in which total Sphagnum remains dip below 70%; and (3) a longer period towards the top the core, estimated between 275 cal. BP and 200 cal. BP (i.e. cal. ad 1675 to cal. ad 1750) in which Sphagnum remains are below 50%, within which they twice dip below 10%, and over which Empetrum/Ericaceae remains are generally higher than for most of the time period covered by the core. In each of these low points in the Sphagnum curve, the light transmission data also dip.

Peat-humification data from peat bogs often show an overall trend from higher humification values (i.e. lower light transmission) at the base to lower humification (or higher light transmission) towards the surface. Part of the reason for this is undecayed plant material in the upper peat layer (acrotelm), but the overall trend relates to continued slow decomposition over time in the catotelm below (Clymo, 1984). Removal of that overall trend (Figure 4) by linear regression reveals the low points in the light transmission curve to be c. 2800 cal. BP, 1735–1710 cal. BP, c. 950 cal. BP, c. 490 cal. BP and from 250 to 180 cal. BP. These are equivalent to c. 850 cal. bc, c. cal. ad 215–240, c. cal. ad 1000, c. cal. ad 1460 and c. cal. ad 1700–1770.

Figure 4.

Detrended peat-humification data for the AND-2 core, plotted using absorption data, for the past 3000 years (age estimates as per Figure 3). Note the reverse vertical axis used when plotting these data, so as to create the same visual impression as in Figure 3: deep troughs indicate short-lived dry episodes.

The ages cited above are based on the new calibration curve (Figure 2) using the Bacon method of Blaauw and Christen (2011), with the estimated ages then rounded to the nearest 5 years.

Interpretation and discussion

Plant macrofossils

The overall plant macrofossil record for the past 3000 years, shown in Figure 3, indicates a mire dominated by S. magellanicum for most this period, but with some short interludes in which the mire has increased proportions of Empetrum/Ericaceae when Sphagnum is less abundant. This general dominance of S. magellanicum for millennia is reminiscent of northwest European bogs in which Sphagnum imbricatum (syn. Sphagnum austinii) has overwhelming dominance as the major peat former. Interestingly, in parts of Northwest Europe, notably in Britain, S. imbricatum becomes extinct at many sites during the last millennium, possibly related to airborne eutrophication (McClymont et al., 2008); at many sites, it is replaced by S. magellanicum, apparently occupying the same ecological niche, leading to its record being interpreted as having the same climate-indicator value in data analysis, with the data from the two species amalgamated (Barber et al., 1994). As S. imbricatum is a Northern Hemisphere species, it is not native to Patagonia. No major moss species change took place in this core, and so no species amalgamation has been necessary for interpreting the Sphagnum record.

The longest period in which the proportion of Sphagnum is lower, while Empetrum/Ericaceae remains are more abundant, although not as prominent as c. 2800 cal. BP, is that between c. 275 and 200 cal. BP (i.e. c. cal. ad 1675–1750). This period can therefore be regarded as a significant deviation from an overwhelming dominance by S. magellanicum during most of the past three millennia.

Peat humification

There is generally close agreement in direction, if not in magnitude, between the plant macrofossil record of a generally wet site (indicated by high percentages of Sphagnum remains) interrupted by short, dry episodes (shown by lower Sphagnum values and increased representation of Empetrum/Ericaceae remains), and the light transmission data, which indicate a mainly wet bog surface (shown by values >40%) interposed by drier episodes (when values dip below 40%). The most extreme dry episode is short-lived, at c. 2800 cal. BP (reported by Chambers et al., 2007). A short dry episode is suggested subsequently at c. cal. ad 1460, while a longer dry episode is indicated at c. cal. ad 1700–1770.

Combining the proxy-climate records

When the two proxy-climate records are combined, there is evidence mainly from a pronounced low in the light transmission curve for a dry episode at c. cal. ad 1460, while a combination of plant macrofossil and light transmission values suggests a particularly dry episode from c. cal. ad 1675 to 1770. These data compare with those from Rhodes et al. (2012), who inferred from a coastal ice-core record in the Ross Sea, Antarctica, that there were ‘three 12–30 yr intervals of cooler temperatures’ (p. 1223) within the LIA, centred on ad 1690, 1770 and 1840. The time period c. cal. ad 1675–1770 encompasses two of those three intervals, and noticeably has a somewhat less dry period in between. The overall time period corresponds with the most severe climatic episodes within the LIA in terms of cold climate in Northwest Europe, matched only by the early 15th century (Lamb, 1995).

Relationship to solar activity?

Short-term climate changes in the Holocene have been attributed to volcanism (Robock, 2000), changes in solar activity (Beer and Van Geel, 2008) or other extra-terrestrial causes (Baillie, 2007), whereas the LIA has in the past been attributed to reduced solar activity (Grove, 1988), notably around the Maunder (Eddy, 1976) and Dalton minima. However, other factors, including volcanism (Robock, 2000) and ocean circulation, are claimed to play a part (Mann, 2002), while its onset has recently been claimed to be caused by four, large, sulphur-rich explosive eruptions in a 50-year period, and its continuance to be sustained by sea ice/ocean feedback ‘during a hemispheric insolation minimum’ (Miller et al. 2012: 1), rather than requiring a reduction in solar activity. In contrast, Hunt (2006) used only internal forcing in the CSIRO mark 2 coupled global climate model and concluded that both the ‘Medieval Warm Period’ (i.e. the MCA) and LIA could not be explained by natural (i.e. internal) processes, and that ‘external forcing must be involved’ (p. 677). This conclusion was earlier exemplified by Mauquoy et al. (2002; 2008), who used proxy-climate techniques on mires in Denmark and northern England, and identified solar-forcing signals during the LIA.

It is notable that the dry episode we infer from the Valle de Andorra mire at c. cal. ad 1460 is between a putative start of the LIA at ad 1420, according to Lamb (1995), and the lowest point in the solar-activity Spörer Minimum (at ad 1490), while the dry episode of c. cal. ad 1675–1770 is very close to the Maunder Minimum of ad 1645 to ad 1715. Indeed, the period c. cal. ad 1675–1770 presents as the driest in this mire record since the inferred solar-driven ‘global’ climate change c. 2800 cal. BP (Chambers et al., 2007; see also Van Geel et al., 1996).

Suck-in and smear

The deposition-rate curve from this site (Figure 2) contains some of the best examples of wiggle-matched dates from peat sequences, but this tight chronology only serves to emphasise the lesser chronological control in recent centuries. Baillie (1991) cautioned archaeologists about the dangers of both ‘suck-in’ – in which unrelated events, apparently closely dated in time, might be interpreted falsely as presenting evidence for a single event – and ‘smear’ in which the dating of a single event at different sites may cover a wide dating range, owing to inaccuracy and/or imprecision in dating techniques. Interpretations made in the present study are subject to that cautionary advice. For example, although the dry period from c. ad 1675–1770 is close both in time and duration to that of the Maunder Minimum (ad 1645–1715), limitations of the method preclude more precise dating of the mire data, because the dating of the peat that accumulated during recent centuries is compromised by the acknowledged ‘fuzziness’ of the calibration (see Figure 2) and the failure of ²¹⁰Pb dating at this site. Indeed, it is unfortunate that some other methods to help date the top of the core, which can be used in Europe (Van der Plicht et al., 2013), could not be applied here. Nevertheless, the closeness of fit of the driest episodes in the mire with cold episodes noted elsewhere and with solar minima is striking, even allowing for imprecision in calibrated ages (and therefore possible inaccuracy in dating) in this part of the core.

Relationship to previous work

In previous work at the Valle de Andorra bog, using climate proxies on a separate peat core, Mauquoy et al. (2004) did not find clear evidence for the LIA. However, this was under a prevailing mindset that expected the mire to respond in the same way as those in Northwest Europe – to show evidence of cooler, wetter conditions – and these were not evident. In this more recent detailed work, the opposite tendency is shown for the LIA. Mauquoy et al. (2004) identified two notably wet periods: from c. ad 1030–1100 and from c. ad 1800 to 1930. It is interesting to note that the former is close to the peak warmth of the Medieval Climate Anomaly (a time period associated in Europe with warmer drier conditions and a relatively ice-free North Atlantic: Lamb, 1995), while the latter encompasses the global temperature recovery from the LIA and the warming trend in the early 20th century (cf. Folland et al., 2002). Taken at face value, these data might imply that when conditions in Northwest Europe, and arguably also globally, become noticeably warmer, conditions in this South American mire become wetter. This is the counterpoint to the drier-than-normal episodes recorded in the present study during the LIA and from this same core at 2800 cal. BP. Even so, the same cautionary remarks concerning ‘suck-in’ must apply here, and this post hoc interpretation did not feature in the 2004 paper.

Large-scale controls on the Patagonian climate have been explored recently using results from a 30-year numerical simulation: Garreaud et al. (2013) conclude that variation in zonal winds account for much of the rainfall variability in Patagonia ‘at synoptic and interannual timescales’ (p. 215) and note somewhat opposite responses to the west of the Andes (recent drying trend) compared with the southern tip (modest increase). In general, a West–East contrast was noted, with a positive/negative correlation between the 850-hPa zonal wind speed and monthly precipitation for sites located to the west/east of the Andes ridge. It is difficult to relate their findings to our interpretations above, because their data covered only a limited time period; it also would be necessary to decide into which of two regions the mire falls. Geographically, it is East for direct precipitation; but in terms of original moisture source for the river in the valley, then arguably it could be West, because it originates there from glaciers and snowpacks. Hypothetically, during times of stronger Westerlies, and so with a more pronounced föhn effect, there could be more snow melt, causing a higher river discharge, which in turn could impede efflux from the mire; so, local water tables rise and appear to give a ‘West’ response. However, the overall scale of change suggested in the 3000-year Andorra bog record is greater than encompassed in that 30-year simulation, while the abrupt episodes of LIA dryness are pervasive for at least a few decades. This implies a more significant mechanism for parts of the LIA than minor variations in strength of the Westerlies, such as, for example, a major shift in their geographical position (cf. Moreno et al., 2009).

Varma et al. (2011) suggest that during times of reduced solar forcing (as, for example, during parts of the LIA), the Westerlies moved north; whereas, during times of higher solar activity (e.g. Medieval Climate Anomaly; late 20th century), the Westerly belt was more southerly. This contrasts with an earlier view of Moy et al. (2008), who suggested a poleward shift of the Westerlies during the LIA, based on data from a lake site in Chile. Also to the west of the Andes, from Lake Puyehue in Chile, Bertrand et al. (2006) report a humid period from ad 1490 to 1700, which they interpret as coinciding with the LIA, whereas, our record, well to the southeast of the Andes, suggests dryness in the latter part of that period. Note that the dating does not coincide exactly, because after ad 1770, Bertrand et al. (2006) report (in the west) a drying trend; our record shows that a putative countervailing wetter period (to the east) did not start until after c. cal. ad 1770. Our data are, however, seemingly in accord with Neukom et al. (2010) whose reconstructions show dry summers for the 17th century in southern South America, compared with those for the mid-late 20th century, which are noticeably wetter. They have only sparse data points for southernmost South America, whereas our reconstructions extend the evidence for (late) 17th century dryness, further south into Tierra del Fuego. Daley et al. (2012) noted that Ushuaia (see Figure 1) has in recent decades been getting wetter over the instrumental period in response to intensification and southward movement of the Westerlies. In that context, it may be hypothesised that a LIA drying of the surface of the Andorra bog is not inconsistent with the Westerlies moving northward, which itself would be consistent with the notion of reduced Hadley cell circulation that would be expected with reduced solar activity.

Limitations to the proxy-climate measures

We used two proxy-climate measures, rather than a single method, on this core. However, the measures are not completely independent, because there may be a species effect in humification data (Chambers et al., 1997; Yeloff and Mauquoy, 2006). The peat-humification technique has also been criticised for not fully capturing decay processes (Hansson et al., 2013). In a limited comparative study of methods to assess peat decay at a single site in Europe by Biester et al. (2014), techniques other than the colorimetric technique could be considered to have performed better; several of these other methods have been used in three pristine mires elsewhere in Patagonia (Broder et al., 2012). Nevertheless, the datasets reported here indicate that the greatest inferred climatic deviations fit well with climate excursions elsewhere, even though it is assumed that the principal change in the mire relates to bog surface wetness, rather than to temperature. Further work might be directed to investigating deuterium excess, as suggested in Daley et al. (2012), which could help to show whether shifting Westerlies were indeed the mechanism for inferred changes in bog surface wetness at this site.

Conclusion

Proxy-climate data from a mire in Tierra del Fuego, Southern Argentina, using the techniques of plant macrofossil analysis and determination of peat humification, show that the main climate changes recorded during the past 3000 years were (1) at 2800 cal. BP and (2) during parts of the time period conventionally associated with the LIA. These changes manifest as vegetation changes and changes in peat humification, both indicative of changes in bog surface wetness; they present as significantly drier-than-normal interludes. Although there may well be a temperature component in the records, no temperature-change equivalent can at present be calculated from these data.

The 2800 cal. BP shift is sharp, short-lived and coincides with a claimed abrupt ‘global’ climate change attributed to a temporary decrease in solar activity during a ‘Grand Solar Minimum’. That drier episode was interpreted by Chambers et al. (2007) as indicating a shift in the position of the moisture-bearing Westerlies. Subsequent short-lived climate shifts, of lesser magnitude, seemingly occurred at c. cal. ad 215–240 and 1000. Later, a more pronounced short-lived shift occurred at c. cal. ad 1460, close to the putative claimed start of the LIA, while a twin-peak episode of more comparable magnitude to that of c. 2800 cal. BP lasted from c. cal. ad 1675 to 1770, coinciding with the most severe interludes of the LIA in Europe. A similar mechanism to the pronounced shift at c. 2800 cal. BP might be invoked to explain these shifts to dryness in the last millennium: namely, a change in the position of the Westerlies.

While the proximal cause of major changes in bog surface wetness may be shifts in the southerly jet stream, the data from this site suggest (1) that the LIA is recorded in Southern Argentina; (2) at its peak intensities, it is comparable in magnitude to the 2800 cal. BP climate perturbation; (3) the data support Moreno et al.’s (2014) claim of ‘inter-hemispheric [temporal] symmetry’; (4) these data are not inconsistent with the long-standing interpretation of the more severe multi-decadal episodes within the LIA being associated with decreased solar activity; and (5) that it may be hypothesised that decreased solar activity caused an equatorward shift of the Westerlies, resulting in a drier mire surface at the Andorra bog in Tierra del Fuego at c. 2800 cal. BP (which while temporally coincident with a major change to wetness recorded in mires in northwest continental Europe is nevertheless an opposite response of the bog surface here) and again during the more intense phases of the LIA; this hypothesis remains to be tested at other sites.

Footnotes

Acknowledgements

Radiocarbon dating was provided by the Natural Environment Research Council, UK, under Allocation number 1012.1002.

Funding

DM’s research (at Department of Earth Sciences, Uppsala University) was supported through a European Community Marie Curie Fellowship (Contract HPMF-CT-2000-01056).

References

Abram

Mulvaney

Wolff

. (2013) Acceleration of snow melt in an Antarctic Peninsula ice core during the twentieth century. Nature Geoscience 6: 404–411.

Baillie

MGL

(1991) Suck in and smear – Two related chronological problems for the 90s. Journal of Theoretical Archaeology 2: 12–16.

Baillie

MGL

(2001) The ad 540 event. Current Archaeology 15: 266–269.

Baillie

MGL

(2007) The case for significant numbers of extraterrestrial impacts through the late Holocene. Journal of Quaternary Science 22: 101–109.

Barber

Chambers

Maddy

. (1994) A sensitive high-resolution record of late-Holocene climatic change from a raised bog in northern England. The Holocene 4: 200–207.

Beer

Van Geel

(2008) Holocene climate change and the evidence for solar and other forcings. In: Battarbee

Binney

(eds) Natural Climate Variability and Global Warming: A Holocene Perspective. Chichester: Wiley-Blackwell, pp. 138–162.

Bertrand

Boes

Castiaux

. (2006) Temporal evolution of sediment supply in Lago Puyehue (Southern Chile) during the last 600 yr and its climatic significance. Quaternary Research 64: 163–175.

Biester

Knorr

K-H

Schellekens

. (2014) Comparison of different methods to determine the degree of decomposition in peat bogs. Biogeosciences 11: 2691–2707.

Blaauw

Christen

(2011) Flexible palaeoclimate age-depth models using an autoregressive gamma process. Bayesian Analysis 6: 457–474.

10.

Blackford

Chambers

(1993) Determining the degree of peat decomposition for peat-based palaeoclimatic studies. International Peat Journal 5: 7–24.

11.

Bradley

Jones

(1993) ‘Little Ice Age’ summer temperature variations: Their nature and relevance to recent global warming trends. The Holocene 3: 367–376.

12.

Broder

Blodau

Biester

. (2012) Peat decomposition records in three pristine ombrotrophic bogs in southern Patagonia. Biogeosciences 9: 1479–1491.

13.

Chambers

(2006) ACCROTELM: Abrupt climate changes recorded over the European Land Mass. Final Report to European Commission, Contract No. EVK2-CT-2002-00166, 117 pp. Available at: http://cordis.europa.eu/project/rcn/67956_pl.html.

14.

Chambers

Barber

Maddy

. (1997) A 5500-year proxy-climate and vegetational record from blanket mire at Talla Moss, Peebleshire, Scotland. The Holocene 7: 391–399.

15.

Chambers

Bielman

(2011) Methods for determining peat humification and for quantifying peat bulk density, organic matter and carbon content for palaeostudies of climate and peatland carbon dynamics. Mires and Peat 7: Article 07, 1–10.

16.

Chambers

Booth

De Vleeschouwer

. (2012) Development and refinement of proxy-climate indicators from peats. Quaternary International 268: 21–33.

17.

Chambers

Mauquoy

Brain

. (2007) Globally synchronous climate change 2800 years ago: Proxy data from peat in South America. Earth and Planetary Science Letters 253: 439–444.

18.

Clymo

(1984) The limits to peat bog growth. Philosophical Transactions of the Royal Society of London, Series B 303: 605–654.

19.

Daley

Mauquoy

Chambers

. (2012) Investigating late Holocene variations in hydroclimate and the stable isotope composition of precipitation using southern South American peatlands: An hypothesis. Climate of the Past 8: 1457–1471.

20.

Daley

Thomas

Holmes

. (2011) The 8200 yr BP cold event in stable isotope records from the North Atlantic region. Global and Planetary Change 79: 288–302.

21.

Defoe

(1704) The Storm. London: John Nutt.

22.

De Vleeschouwer

Chambers

Swindles

(2010a) Coring and sub-sampling of peatlands for palaeoenvironmental research. Mires and Peat 7: Article 01, 1–10.

23.

De Vleeschouwer

Hughes

PDM

Nichols

. (eds) (2010b) Special volume: A review of protocols in peat palaeoenvironmental studies. Mires and Peat 7: Article 00, 1.

24.

Eddy

(1976) The Maunder minimum. Science 192: 1189–1192.

25.

Esper

Frank

Timonen

. (2012) Orbital forcing of tree-ring data. Nature Climate Change 2: 862–866.

26.

Folland

Rayner

Brown

. (2002) Global temperature change and its uncertainties since 1861. Geophysical Research Letters 28: 2621–2624.

27.

Garreaud

Lopez

Minvielle

. (2013) Large-scale control on the Patagonian climate. Journal of Climate 26: 215–230.

28.

Grove

(1988) The ‘Little Ice Age ’. London: Methuen.

29.

Hansson

Rydberg

Kylander

. (2013) Evaluating paleoproxies for peat decomposition and their relationship to peat geochemistry. The Holocene 23: 1666–1671.

30.

Hughes

PDM

Mallon

Essex

. (2012) The use of k-values to examine plant ‘species signals’ in a peat humification record from Newfoundland. Quaternary International 268: 156–165.

31.

Hunt

(2006) The Medieval Warm Period, ‘Little Ice Age’ and simulated climatic variability. Climate Dynamics 27: 677–694.

32.

IPCC (2013) Climate Change 2013: The Physical Science Basis (Working Group I Contribution to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change). Cambridge: Cambridge University Press.

33.

Jones

Mann

(2004) Climate over past millennia. Reviews of Geophysics 42: RG 2002, 1–42.

34.

Jones

Briffa

Osborn

. (2009) High-resolution palaeoclimatology of the last millennium: A review of current status and future prospects. The Holocene 19: 3–49.

35.

Kobashi

Severinghaus

Brook

. (2007) Precise timing and characterization of abrupt climate change 8200 years ago from air trapped in polar ice. Quaternary Science Reviews 26: 1212–1222.

36.

Lamb

(1995) Climate, History and the Modern World. 2nd Edition. London: Routledge.

37.

Lauritzen

S-E

Lundberg

(1999) Calibration of speleothem delta function: An absolute temperature record for the Holocene in northern Norway. The Holocene 9: 659–669.

38.

Le Roy Ladurie

(1971 [1967]) Histoire du climat depuis l’an mil. Paris: Flammarion (Translated by Barbara Bray as Times of Feast, Times of Famine: A History of Climate since the Year 1000. Revised and updated. Garden City: Doubleday).

39.

McCarroll

(2010) Future climate change and the British Quaternary research community. Quaternary Science Reviews 29: 1661–1672.

40.

McClymont

Mauquoy

Yeloff

. (2008) The disappearance of Sphagnum imbricatum from Butterburn Flow, UK. The Holocene 18: 991–1002.

41.

Mann

(2002) ‘Little Ice Age’. In: MacCracken

Perry

(eds) Encyclopedia of Global Environmental Change. Chichester: Wiley, pp. 504–509.

42.

Mann

(2007) Climate over the past two millennia. Annual Review of Earth and Planetary Sciences 35: 111–136.

43.

Mann

Bradley

Hughes

(1998) Global-scale temperature patterns and climate forcing over the past six centuries. Nature 392: 779–787.

44.

Mann

Bradley

Hughes

(1999) Northern hemisphere temperatures during the past millennium: Inferences, uncertainties, and limitations. Geophysical Research Letters 26: 759–762.

45.

Mann

Zhang

Hughes

. (2008) Proxy-based reconstructions of hemispheric and global surface temperature variations over the past two millennia. Proceedings of the National Academy of Sciences of the United States of America 105: 13252–13257.

46.

Mann

Zhang

Rutherford

. (2009) Global signatures and dynamical origins of the ‘Little Ice Age’ and Medieval Climate Anomaly. Science 326: 1256–1260.

47.

Mauquoy

Blaauw

Van Geel

. (2004) Late Holocene climatic changes in Tierra del Fuego based on multi-proxy analyses of peat deposits. Quaternary Research 61: 148–158.

48.

Mauquoy

Hughes

PDM

van Geel

(2010) A protocol for plant macrofossil analysis of peat deposits. Mires and Peat 7: Article 6, 1–5.

49.

Mauquoy

Van Geel

Blaauw

. (2002) Evidence from North-West European bogs shows ‘Little Ice Age’ climatic changes driven by variations in solar activity. The Holocene 12: 1–6.

50.

Mauquoy

Yeloff

Van Geel

. (2008) Two decadally resolved records from north-west European peat bogs show rapid climate changes associated with solar variability during the mid-late Holocene. Journal of Quaternary Science 23: 745–763.

51.

Miller

Geirsdóttir

Zhong

. (2012) Abrupt onset of the ‘Little Ice Age’ triggered by volcanism and sustained by sea-ice/ocean feedbacks. Geophysical Research Letters 39: L02708.

52.

Moreno

François

Villa-Martínez . (2009) Millennial-scale variability in Southern Hemisphere westerly wind activity over the last 5000 years in SW Patagonia. Quaternary Science Reviews 28: 25–38.

53.

Moreno

Vilanova

Villa-Martínez

. (2014) Southern Annular Mode-like changes in southwestern Patagonia at centennial timescales over the last three millennia. Nature Communications 5: Article 4375.

54.

Moy

Dunbar

Moreno

. (2008) Isotopic evidence for hydrologic change related to the westerlies in SW Patagonia, Chile, during the last millennium. Quaternary Science Reviews 27: 1335–1349.

55.

Nesje

(2009) Latest Pleistocene and Holocene glacier fluctuations in Scandinavia. Quaternary Science Reviews 28: 2119–2136.

56.

Neukom

Luterbacher

Villalber

. (2010) Multi-centennial summer and winter precipitation variability in southern South America. Geophysical Research Letters 37: L14708.

57.

Piotrowska

Blaauw

Mauquoy

. (2011) Constructing deposition chronologies for peat deposits using radiocarbon dating. Mires and Peat 7: Article 10, 1–14.

58.

Rhodes

Bertler

NAN

Baker

. (2012) ‘Little Ice Age’ climate and oceanic conditions of the Ross Sea, Antarctica from a coastal ice core record. Climate of the Past 8: 1223–1238.

59.

Robock

(2000) Volcanic eruptions and climate. Reviews of Geophysics 38: 191–219.

60.

Schaefer

Denton

Kaplan

. (2009) High-Frequency Holocene Glacier Fluctuations in New Zealand Differ from the Northern Signature. Science 324: 622–625.

61.

Van der Plicht

Yeloff

van der Linden

. (2013) Dating recent peat accumulation in European ombrotrophic bogs. Radiocarbon 55: 1763–1778.

62.

Van Geel

Mook

(1989) High-resolution 14C dating of organic deposits using natural atmospheric 14C variations. Radiocarbon 31: 151–155.

63.

Van Geel

Renssen

(1998) Abrupt climate change around 2,650 BP in North-West Europe: Evidence for climatic teleconnections and a tentative explanation. In: Issar

Brown

(eds) Water, Environment and Society in Times of Climatic Change. Dordrecht: Kluwer, pp. 21–41.

64.

Van Geel

Buurman

Waterbolk

(1996) Archaeological and palaeoecological indications for an abrupt climate change in The Netherlands and evidence for climatological teleconnections around 2650 BP. Journal of Quaternary Science 11: 451–460.

65.

Varma

Prange

Lamy

. (2011) Solar-forced shifts of the Southern Hemisphere Westerlies during the late Holocene. Climate of the Past 7: 339–347.

66.

Yeloff

Mauquoy

(2006) The influence of vegetation composition on peat humification: Implications for palaeoclimatic studies. Boreas 35: 662–673.