Reconstruction of mid- to late-Holocene winter temperatures in the Skagerrak region using benthic foraminiferal Mg/Ca and δ 18 O

Abstract

Reconstruction of Skagerrak deep-water renewal is used to assess regional changes in winter thermal conditions over the past 6800 years. Changes in winter climate conditions from the Skagerrak region are in turn linked to shifts in Holocene large-scale atmospheric circulation patterns prevailing over northern Europe. We use Melonis barleeanus Mg/Ca from two sediment cores in the central Skagerrak to reconstruct temperature of Skagerrak intermediate water, representing the warm season temperature variability, and deep water, for monitoring Skagerrak deep-water renewal, reflecting the winter temperature variability. In addition, M. barleeanus δ¹⁸O is used from the deeper core to reconstruct salinity, also monitoring the deep-water renewal. Our results show that the Skagerrak deep-water experienced phases of particularly enhanced renewal during the mid-Holocene reflecting severe winter conditions, followed by a general shift to reduced renewal as a consequence of milder winter conditions over the North Sea around 3500 cal. yr BP. The late-Holocene shift was most likely related to the onset of a regime with intensified winter westerly winds directed toward northern Europe and an increased inflow of North Atlantic water into the Skagerrak–North Sea reflecting more maritime climate conditions. On millennial scale, cold phases in our deep-water records match with low winter precipitation phases in western Norway. They are associated with distinct increases in ice rafted debris (IRD) in North Atlantic sediments, suggesting that phases of iceberg discharge in the Atlantic were associated with cold and dry winter conditions over northern Europe. Interestingly, the cold event centered around 5900 cal. yr BP appears to be only associated with winter variability, while the following one at 4200 cal. yr BP is documented in our winter record, as well as in records related to warmer seasons.

Keywords

benthic foraminifera Bond events late-Holocene mid-Holocene Skagerrak winter

Introduction

In northwestern Europe, climate conditions during the mid-Holocene period were characterized by generally high summer and mean annual temperatures, lower winter precipitation rates, and reduced or lacking continental glaciers (Bakke et al., 2008; Bjune et al., 2005; Davis et al., 2003; Seppä et al., 2009). The end of the mid-Holocene ‘climate optimum’ occurred in northwestern Europe around ~4200 cal. yr BP (Davis et al., 2003), followed by a transition toward the late-Holocene that was associated with a decrease in summer and mean annual temperatures, increased winter precipitation, and re-advancing continental glaciers (Bakke et al., 2008; Bjune et al., 2005; Davis et al., 2003; Seppä et al., 2009).

Until now, most paleotemperature reconstructions from northwestern Europe are based on biological proxies that are biased toward the warm season as productivity mainly occurs between spring and autumn in the North Sea region, and biological production during winter is normally low (Andersen et al., 2004; Blanz et al., 2005; Joint and Pomroy, 1993; Leduc et al., 2010; Liu et al., 2014; Moros et al., 2004). However, many evidences point to the stronger importance of winter variability compared with summer variability in Europe over the Holocene (Mangini et al., 2005, 2007; O’Brien et al., 1995; Oppo et al., 2003; Xoplaki et al., 2005). On the short term, modern climate in Europe is governed by large-scale atmospheric pressure cells responsible for changes in strength and direction of North Atlantic wind fields (Hurrell, 1995; Hurrell et al., 2003). Because the vigor of the flow is related to meridional pressure gradient, the westerly winds (or Westerlies) carrying warmth and moisture across the North Atlantic are the strongest during winter months. However, it is still under debate whether or not this phenomenon, also known as the North Atlantic Oscillation (NAO), has prevailed during the Holocene (Jones et al., 2014; Moreno et al., 2012; Olsen et al., 2012; Spangehl et al., 2010; Trouet et al., 2009, 2012). On a millennial scale, cold events during the early-Holocene are expected to have mainly influenced winter variability as a result of high-latitude remnant ice sheets and low winter insolation (Denton et al., 2005; Wanner and Bütikofer, 2008; Wanner et al., 2014). The mid-Holocene, however, starts with the melting of the last remnant Laurentide and Scandinavian ice sheets and a change in the insolation regime, and the contribution of winter variability to the annual variability during the mid-Holocene might have a change. Most reconstructions of northwestern European temperatures spanning the mid- to late-Holocene are, however, biased toward the warm season, and, therefore, underestimate the potential temperature variability occurring during winter months.

The Skagerrak, deepest basin of the North Sea, is known to be sensitive to regional atmospheric circulation. In particular, Skagerrak deep-water renewal is influenced by changes in winter conditions over the North Sea for the last 1200 years at least (Brückner and Mackensen, 2006; Ljøen and Svansson, 1972). Here, we will argue that the renewal of the deep water has been occurring at least since the onset of the modern circulation about 8500 cal. yr BP and has been reflecting winter atmospheric temperatures over the region. Subsequently, Skagerrak deep archives can be used to trace past winter variability since the onset of the modern circulation. Using Melonis barleeanus Mg/Ca and δ¹⁸O from a core in the deep basin (IOW225514), we have reconstructed deep-water temperature (DWT) and, therefore, traced winter atmospheric temperatures for the last 6800 cal. yr BP. We will also argue that intermediate-water temperature (IWT) reconstructions based on benthic foraminifera Mg/Ca from a core at intermediate depth (IOW225517) are biased toward the warm season.

To complete our understanding of past climate variability in the Skagerrak region, we compare our records with alkenone-based sea surface temperature (SST; Krossa personal communication, 2015) from core IOW225514, interpreted as a proxy for regional mean annual temperatures. Changes in seasonal thermal gradient are discussed in terms of regional atmospheric circulation, as well as the seasonal imprint of mid- and late-Holocene Bond events on northwestern Europe climate.

Regional setting

The Skagerrak is a marginal basin in the northern part of the North Sea and forms a connection between the North Atlantic and the Baltic Sea (Figure 1). It has a fjord-like shape with maximum depths of around 700 m at its northeastern rim (Rodhe, 1987). A sill separates the Skagerrak from the rest of the North Sea at a water depth of ~270 m and creates a boundary between upper and lower water masses (Ljøen and Svansson, 1972; Figure 1).

Figure 1.

General circulation and bathymetry in the eastern North Sea and Skagerrak–Kattegat (modified from Krossa et al., 2015). Dots mark the location of interest for this study (marine cores and regional record location). The main transport is located around 200–300 m water depth (Rodhe, 1987; Svansson, 1975). The sill is represented with a dotted line.

The modern circulation pattern has been established around 8500 cal. yr BP as a consequence of the opening of the English Channel and the Danish Strait (Gyllencreutz, 2005). It is governed by the inflow of North Atlantic water and the outflow of Baltic Sea water (Otto et al., 1990; Rodhe, 1987; Svansson, 1975). North Atlantic water enters the Skagerrak from the northwest, forming the Norwegian Atlantic Current and the central North Sea water, and from the southwest through the English Channel, forming the South Jutland current (Figure 1). Those currents subsequently mix to form the North Jutland current. As this water mass passes Skagen thereby entering the Skagerrak–Kattegat border, it mixes with fresher and colder Baltic Sea water. It subsequently slows down, allowing fine-grained sediments to accumulate at high rates in the central and northeastern parts of the Skagerrak (Rodhe and Holt, 1996). The resulting water mass exits the Skagerrak flowing along the Norwegian coast as the Norwegian Coastal Current, forming an anticlockwise circulation pattern (Figure 1).

The surface layer is permanently stratified mainly because of the outflow of fresher Baltic Sea water, in addition to a thermal stratification component during summer (Otto et al., 1990). Throughout the complete annual cycle, SSTs range from 4°C to 15°C (Locarnini et al., 2010). The intermediate water in the western Skagerrak is strongly influenced by the inflow of a branch of North Atlantic water, with temperature range between 4°C and 8°C, showing intra-annual variations (Ljøen and Svansson, 1972; Figure 2). Finally, the deep waters are dissociated from the main circulation process. They remain stagnant for a few years until they are replaced by cold and dense water coming from the central North Sea. This process is known as Skagerrak deep-water renewal and is strongly driven by temperature and to a lesser extent by salinity of the central North Sea surface water mass (Ljøen and Svansson, 1972). The DWTs do not possess an intra-annual variability but present a multi-annual variability with temperatures ranging between 4°C and 8°C (Ljøen and Svansson, 1972; Figure 2).

Figure 2.

Evolution of instrumental temperatures at different depths in the Skagerrak basin, depicting the difference between the intermediate and the deep waters annual variability between 1947 and 1969 CE (from Ljøen and Svansson, 1972). The cores used in this study are represented at their respective depth on a profile of the Skagerrak basin.

Deep-water renewal and regional climate

During severe winters, slow currents and strong vertical mixing of the water mass of the central North Sea enable cooling of the whole water body in this part of the North Sea. The resulting cold and, therefore, dense water mass subsequently cascades down into the Skagerrak basin and replaces less dense waters prevailing in the deep basin. Trapped in the deeper basins for several years, the deep-water mass preserves the winter temperature signal (Ljøen and Svansson, 1972). The present-day process of deep-water renewal is characterized by a cooling phase of a few months in the deep water, followed by a period of a slight temperature increase that might last a few years. Temperature variability of the deep waters has been positively correlated to winter atmospheric temperatures over the region. It has been documented for the period between 1947 and 1969 (Ljøen and Svansson, 1972), extended here until 2010 (Figure 3), and for the last 1200 years (Brückner and Mackensen, 2006). Finally, the frequency of the deep-water renewal has been positively correlated to the NAO phases for the last decades (Brückner and Mackensen, 2006; Hagberg and Tunberg, 2000; Tunberg and Nelson, 1998).

Figure 3.

Winter air temperatures from central England (mean of December, January, and February, from the HadCET dataset) and Skagerrak deep-water temperatures (bottle data from the Ocean Data View, measurements below 400 m) between 1950 and 1960 CE. The thicker lines represent a 5-year average.

Today, winter climate conditions affecting the North Sea are largely governed by large-scaled atmospheric pressure systems that drive variations in strength and direction of wind fields. This phenomenon of atmospheric circulation is also referred to as the NAO (Hurrell, 1995; Hurrell et al., 2003). The corresponding index, which describes the winter NAO pattern, is defined by differences in atmospheric pressure cells over Iceland and the Azores (NAO index; Hurrell, 1995). During periods of a positive NAO index, the difference between the pressure cells is high, leading to Westerlies shifted toward Northern Europe. The Westerlies bring mild and moist air masses from the North Atlantic over Northern Europe (Hurrell, 1995; Hurrell et al., 2003). During periods of a negative winter NAO index, the difference is low, and consequently, Westerlies are weaker, preventing mild and moist air masses to influence the climate in northwestern Europe. A positive NAO is associated with moist and mild winters over Northern Europe, while a negative NAO is associated with severe and dry winters. The winter NAO index varies from year to year. However, it also exhibits the tendency to remain on one phase for several years (Hurrell, 1995).

Materials and methods

Sample material

We used two gravity cores from the central Skagerrak (Figure 1, Table 1) to reconstruct Skagerrak DWT and IWT. Gravity core IOW225514 is located at 420 m water depth, thus in the range of the deep waters. IOW225517 was cored around sill depth, at 293 m water depth, consequently in the range of the intermediate waters. Both gravity cores consist of homogeneous olive-green silty clay, and an array of paleoceanographic data is available for both stations on these specific cores (Brückner and Mackensen, 2006, 2008; Emeis et al., 2003; Hass, 1996; Krossa et al., 2015). We sampled material of 1-cm slices and wet sieved the sediment fractions between 125 µm and 2 mm. Benthic foraminifera were obtained from the >150-µm coarse fraction, and monospecific samples (M. barleeanus) were prepared for δ¹⁸O and Mg/Ca analyses (see below). We chose M. barleeanus to reconstruct past changes in Skagerrak deep-water conditions, as it is abundant continuously during the Holocene in the Skagerrak and is one of the species reported to be only slightly affected by carbonate dissolution and re-precipitation (Brückner, 2008).

Table 1.

Radiocarbon dates of cores IOW225514 and IOW225517. Depths in core IOW225517 are corrected for sediment loss at the top core by adding 41.5 cm to the original depth (Emeis et al., 2003). The material dated is mixed benthic foraminifera except for samples KIA14030 and KIA14032 where it is bivalves. Calendar years are recalculated for all data using Calib7.0, with Marine13. The discarded dates are marked with a *.

Sample	Depth (cm)	¹⁴C (yr BP)	Median calibrated age (cal. yr BP)	1σ range (cal. yr BP)	From
Core IOW225514 (57″50.279′N, 842.226′E), 420 m water depth
KIA14028	10,5	810 ± 30	454	429–483	Emeis et al. (2003)
KIA14029	105,5	1780 ± 35	1323	1284–1352	Emeis et al. (2003)
KIA14030^*	105,5	1165 ± 30	707	670–731	Emeis et al. (2003)
KIA14031^*	135,5	2710 ± 35	2407	2342–2451	Emeis et al. (2003)
KIA14032	135,5	1980 ± 30	1540	1502–1587	Emeis et al. (2003)
KIA14033	155,5	2755 ± 45	2481	2368–2545	Emeis et al. (2003)
KIA43703	218,5	3545 ± 35	3431	3377–3472	This study
KIA14034^*	250,5	4280 ± 45	4397	4337–4480	Emeis et al. (2003)
KIA43704	300,5	4250 ± 60	4353	4257–4430	This study
KIA14035^*	310,5	6400 ± 45	6876	6810–6939	Emeis et al. (2003)
KIA14036	375,5	5680 ± 45	6086	6022–6160	Emeis et al. (2003)
Core IOW225517 (57″40.002′N, 7′05.465′E), 293 m water depth
KIA15116	82	2185 ± 30	1781	1728–1823	Emeis et al. (2003)
KIA15117	122	3065 ± 35	2828	2772–2868	Emeis et al. (2003)
KIA47690	137	3250 ± 60	3070	2980–3157	This study
KIA47691	196	4150 ± 60	4222	4133–4322	This study
KIA15118	202	4305 ± 30	4434	4396–4487	Emeis et al. (2003)
KIA15119	232	4610 ± 30	4830	4800–4854	Emeis et al. (2003)
KIA43705	253	5015 ± 30	5369	5314–5409	This study
KIA15120	272	5165 ± 35	5528	5505–5571	Emeis et al. (2003)
KIA47693	292	5295 ± 30	5647	5601–5681	This study
KIA15121	352	6485 ± 40	6988	6919–7053	Emeis et al. (2003)
KIA43706	404	7850 ± 40	8320	8282–8369	This study
KIA15122	452	9825 ± 50	10,751	10,656–10,831	Emeis et al. (2003)

Mg/Ca analyses and temperature reconstructions

A total of 10–25 specimens of M. barleeanus were weighed using an ultraprecision scale (Sartorius ME5 OCE) and gently crushed to reveal the inner chamber walls. Subsequently, they were prepared for Mg/Ca analyses following the protocol originally developed by Martin and Lea (2002). The samples were successively cleaned with methanol to remove clays; with hydrazine, ammonium hydroxide, and ammonium citrate to remove metal oxides (reduction step); and with sodium hydroxide to remove organic matter (oxidation step). Finally, the samples were dissolved and diluted in order to achieve a final solution containing about 50 (25–75) mg L⁻¹ of Ca.

The samples were analyzed on a radial viewing simultaneous ICP-AES (SPECTRO Ciros^CCD SOP) at the Institute for Geosciences, Kiel. Standards for foraminifera were used to calibrate the results, following De Villiers et al. (2002). Carbonate reference materials (ECRM 752–1, BAM RS3) were analyzed for monitoring analytical accuracy (Greaves et al., 2008). The typical external error is 0.1% rel. (1σ) for Mg/Ca. According to replicate samples, the standard deviation is about 0.1 mmol mol⁻¹ Mg/Ca for M. barleeanus. Samples with a recovery in Ca concentration of <20% were rejected. To check for potential contamination from metal oxides or ashes, trace elements were additionally monitored, and samples showing a significant correlation between Fe/Ca, Al/Ca, Mn/Ca, and Mg/Ca values were excluded (Schmidt et al., 2004).

Temperatures were derived from Mg/Ca ratios using the following calibration for M. barleeanus (Kristjánsdóttir et al., 2007):

Mg / Ca = (0.658 \pm 0.07) \exp ((0.137 \pm 0.020) T)

Mg/Ca as a DWT proxy in the Skagerrak

Foraminiferal Mg/Ca is largely used in paleoceanography as a proxy for reconstructing past temperatures. However, studies argue that carbon ion concentration at lower temperatures (<3–4°C) and also salinity might affect Mg incorporation in the foraminifera shells (Elderfield et al., 2006; Raitzsch et al., 2008), consequently influencing the temperature reconstructions. Both modern observations (Ljøen and Svansson, 1972) and temperature reconstructions over the past 1200 cal. yr BP (Brückner and Mackensen, 2006) show that Skagerrak DWTs are generally higher than 3–4°C. Additionally, as deep-water masses in the Skagerrak originate from North Atlantic waters and are not affected by fresher Baltic Sea water (Ljøen and Svansson, 1972), the deep-water salinity does not vary enough to influence the Mg incorporation. Therefore, we do not expect the carbonate ion or salinity to bias the Mg incorporation and subsequently our Mg/Ca-derived temperature record. Benthic foraminifera Mg/Ca has also been successfully applied in the northeastern Skagerrak to reconstruct past IWTs, supporting other paleotemperature proxies such as benthic foraminifera transfer functions or stable isotopes (Erbs-Hansen et al., 2011a).

δ¹⁸O analyses

A total of 4–15 individuals of M. barleeanus were crushed into large fragments, stirred in ethanol in an ultrasonic bath for about 20 s and then dried at 40°C in a stove. We measured stable carbon and oxygen isotopes using a Finnigan MAT 251 mass spectrometer at the Leibniz Laboratory at the University of Kiel. The system is coupled online to the Carbo-Kiel device (type I) for automated CO₂ preparation from carbonate samples (for stable isotopic analyses). Samples were reacted by individual acid addition (99% H₃PO₄ at 73°C). Standard external error is lower than ±0.07‰, as documented by the performance of international and laboratory-internal carbonate standard materials.

Values were corrected for a vital effect of 0.276 on M. barleeanus δ¹⁸O values (Kristjánsdóttir et al., 2007). After correction of the vital effect, δ¹⁸O of calcite in the shells of benthic foraminifera can be used to reconstruct past changes in temperature when knowing the isotopic composition of water (δw; Shackleton, 1974):

T (° C) = 16.9 - 4 ((δ^{18} Oc - δ w)

Consequently, as we know the temperature reconstructed using Mg/Ca (see above), we can infer past changes in δw, which is dependent on the salinity of the water mass and also ice volume. Prior to 6000 cal. yr BP, δ¹⁸O values were corrected for an ice volume effect of 0.11‰ per 10 m sea-level change (Fairbanks, 1989). After 6000 cal. yr BP, the general isotopic composition of the world ocean did not change significantly (Fairbanks, 1989), and therefore, no correction was applied. Therefore, after correction for ice volume effect, past changes in salinity in the Skagerrak can be estimated.

In the Skagerrak, Brückner and Mackensen (2006) reconstructed past DWTs using a mixing-line from Sognefjorden in western Norway (Mikalsen and Sejrup, 2000). They found no significant changes in salinity and consequently δw over the past 1200 cal. yr BP. However, when applying Mikalsen and Sejrup (2000) mixing-line or the more recently proposed Skagerrak mixing-lines (Austin et al., 2006; Harwood et al., 2008), we document a relationship of salinity and δw, which results in a much larger influence of salinity on the δ¹⁸Oc signal. Additionally, as the Skagerrak mixing-lines were proposed for a present-day situation of deep-water renewal, it is possible that under different deep-water renewal conditions – for instance, no renewal at all for a period longer than recorded nowadays – the relationship between the salinity and the δw was different in the past ~6800 cal. yr BP. To avoid complications regarding mixing-lines, we used paired Mg/Ca and δ¹⁸O from benthic foraminifera to reconstruct δw, thereby using δw only to infer qualitative changes in past salinity. Plus, δ¹⁸O and δw will only be used in this study in parallel to DWTs, to test the robustness of the latter.

Age model

The gravity cores were dated using the AMS¹⁴C method on specific foraminiferal species (Table 1). The ¹⁴C data, including those previously published by Emeis et al. (2003), were calibrated using the online software CALIB (Calib7.0, ©1986–2014: M. Stuiver and P. Reimer; http://calib.qub.ac.uk/calib/calib.html) and the Marine13 curve (Reimer et al., 2013), with a standard reservoir age of 400 years as generally used in the Skagerrak (e.g. Erbs-Hansen et al., 2011a, 2011b; Krossa et al., 2015). Our results are presented on a calibrated age scale before present (cal. yr BP, AD 1950). Age–depth models were established using linear interpolation between each calibrated radiocarbon date.

No inverted ages occur for gravity core IOW225517, while four inverted dates are found for core IOW225514 (Figure 4), which can either be explained by bioturbation, redeposition, or changes in habitat depth of benthic foraminifera.

Figure 4.

Age–depth relation of radiocarbon calibrated age model for cores IOW225514 and IOW225517 (top). Ca/K ratio from XRF measurements for cores IOW225514 and IOW225517 (bottom). XRF measurements were performed on XRF Avaatech core scanner from Kiel University, with the following settings: 15 s measurement time in a 1.2 × 1-cm rectangle, operated at 10 and 30 kV.

High-resolution geochemical records were produced using x-ray fluorescence (XRF), and analyses of the results were performed following the recommendation from Weltje and Tjallingii (2008). Using XRF Ca/Al records of both gravity cores, we were able to find a good correlation (0.7) between both records. This was done by considering the ¹⁴C too old as outlier when an inversion was occurring (Figure 4), which suggests that outlier dates in core IOW225514 are because of punctual reworking such as bioturbation or redeposition. For the youngest dates, not covered by the XRF records, we used the age model established by Emeis et al. (2003). After removing the outliers in core IOW225514, the good agreement between both XRF curves from both cores (0.7) suggests that the age models are robust. Additional comparisons with regional terrestrial records were done using, for instance, pollen data and tephra chronology (not presented here) and showed that the chronostratigraphy of both cores IOW225514 and IOW225517 were robust.

Results

Over the period covered by the records, between 6800 and 700 cal. yr BP, the general trends in DWT and δw show an increase (+1°C and +0.3‰, respectively). In contrast, IWT seems to have had higher values in the early part of the record compared with the latest part, although the resolution is too low after 4200 cal. yr BP to confidently document a trend.

Several distinct shifts superimposed on the general trend are observed in the DWT and δw records (Figure 5). The interval between 6800 and 3500 cal. yr BP is characterized by strong variability in the DWT with high frequency and amplitude (±4.7°C). The two coldest episodes of deep waters are found during that interval, between 6300 and 5700 cal. yr BP (minimum of 2.8°C) and 4300 and 3500 cal. yr BP (minimum of 4°C). Episodes of low δw occur at the same time.

Figure 5.

M. barleeanus Mg/Ca on intermediate (IOW225517) and deep (IOW225514) cores and M. barleeanus δ¹⁸O of the deep core. M. barleeanus δ¹⁸O measurements from Brückner and Mackensen (2006) are presented by a continued line. The stable isotope records are corrected for vital effect and ice volume (see references in the methods). The bar represents analytical errors.

A strong increase in DWT (+~1.5°C) and δw (+1.3‰) occurs at 3500 cal. yr BP, followed by a period of prevailing high DWT and δw and lower amplitude in DWT variability (±2.9°C) until the end of the record at 700 cal. yr BP. During the late-Holocene, we observe periods of decreased DWTs and in particular δw, centered around 2800 and 1400 cal. yr BP, respectively, but of less amplitude than prior events.

Between 6400 and 4300–4200 cal. yr BP, the IWT varied around 6.8°C in average, with high frequency and amplitude variability where values range between 4.6°C and 10.2°C. Between 4300–4200 and 3500 cal. yr BP, the IWT decreased to reach around 5.5°C. The IWT has low resolution afterward.

Towards the end of the deep-water records, at 1500 cal. yr BP, a rapid increase in temperatures of ~2°C is documented within a few centuries that culminated around 1200 cal. yr BP, followed by a rapid decrease of about the same amplitude over a few centuries.

Discussion

Approach to assess DWT and IWT signals in the Skagerrak

In the North Sea region, the main planktonic blooms occur from spring to autumn (Blanz et al., 2005; Joint and Pomroy, 1993), resulting in organic flux to the sea floor with a strong seasonality influence on the growth of benthic foraminifera (Gooday, 2003).

Temperature measurements from the surface and intermediate water (up to 350 m) in the Skagerrak reflect seasonal-like atmospheric temperature variability (Ljøen and Svansson, 1972). Benthic foraminifera dwelling at intermediate depths will integrate into their shell the thermal conditions during their growth season which is mainly between spring and autumn as discussed previously. As a consequence, we expect low (high) Mg/Ca-based temperatures from benthic foraminifera of core IOW22517 (292 m water depth) to reflect mild (warm) conditions during the warm season (spring to autumn).

However, DWT measurements and reconstructions over the past 1200 cal. yr BP reflect the thermal conditions of the winter months (Brückner and Mackensen, 2006; Ljøen and Svansson, 1972). This is the result of the renewal of Skagerrak deep waters, occurring after particularly severe winters (Ljøen and Svansson, 1972). During severe winters, slow currents and strong vertical mixing of the water mass enable the cooling, and therefore increase in density, of the whole water body of the central North Sea. This dense water mass subsequently flows downward into the central Skagerrak basin where it is trapped for several years until a new episode of deep-water renewal occurs, thus preserving the winter temperature signal (Ljøen and Svansson, 1972). Benthic foraminifera in the deep Skagerrak basin, although precipitating their shells from spring to autumn, will, therefore, record temperatures reflecting winter conditions.

The deep-water renewal is a particular process controlled by the topography and circulation in the North Sea. The shallowness of the central North Sea enables the formation of a cold dense water mass during severe winters on the whole water column, and the circulation brings this water mass into the Skagerrak where it cascades down into the deep basin and is trapped until a new renewal (Ljøen and Svansson, 1972). Since the topography has not considerably changed since the onset of the modern circulation at 8500 cal. yr BP (Gyllencreutz, 2005), it is reasonable to expect that the conditions that allows the process of Skagerrak deep-water renewal were established at least since the establishment of the modern circulation at 8500 cal. yr BP. The frequency of the renewal will itself depends on winter thermal conditions. As a consequence, low (high) Mg/Ca-based temperatures from benthic foraminifera of the deep core IOW22514 should reflect strong (weak) renewal of the deep water following a period of severe (mild) winters, at least for the last 8500 years.

In summary, using Mg/Ca-based temperatures from benthic foraminifera at both deep and intermediate depths give an insight into past atmospheric temperature variability during both warmer season and winter, here, for the last 6800 cal. yr. Additional information about SST variability in the Skagerrak, representing mean annual temperatures, was given by V. Krossa (personal communication, 2015).

Long-term temperature trends in the Skagerrak region

The records in Figure 6 show that the long-term trends in the SST and IWT are inversed compared with the long-term trend of the DWT. Indeed, SST and IWT, representing the mean annual and warm season atmospheric temperatures, respectively, are decreasing over the course of the record, while DWTs, representing winter atmospheric temperatures, are increasing. This is consistent with a decrease in mean annual–summer insolation and increase in winter insolation at northern high latitudes (Laskar et al., 2004).

Figure 6.

(a) Winter and summer insolation at 58°N (Laskar et al., 2004), (b) Skagerrak sea surface temperatures from core IOW225514 (Krossa, personal communication, 2015), (c) intermediate and deep-water temperatures, (d) deep-water salinity (represented by δw), and (e) winter precipitations documented from Norway (Nesje et al., 2001). Note that the solid lines in b, c, and d represent a five-point running mean. Periods of relatively cold and less saline deep waters are highlighted by gray areas.

A strong contrast is observed in Skagerrak water temperatures at different depths prior to 4400–4200 cal. yr BP, with high SST and IWT and low DWT. This contrast suggests a strong seasonal gradient, with warm summers and severe winters, indicating prevalent continental conditions prior to 4200–4300 cal. yr BP. Most likely, the Westerlies that transport warm and moist air from the Atlantic toward NW Europe were weakened between 6800 and ~4400–4200 cal. yr BP. Consequently, cold and dry continental air masses from the north or the east influenced the climate conditions in northern Europe, probably resembling the spatial atmospheric configuration of a negative NAO phase (Hurrell, 1995). In contrast, we document lower differences in temperatures between the SST–IWT and the DWT after 4400–4200 cal. yr BP. We note a decrease in all records but of stronger importance in the SST and IWT records at ~4400–4200 cal. yr BP, followed by an abrupt increase in DWT at 3500 cal. yr BP. This suggests a transition toward lower seasonal gradient, probably related to a stronger dominance of maritime conditions reflective of a strengthening in the Westerlies that in turn bring warm and moist air over northern Europe, resembling the spatial configuration of a positive NAO phase at present (Hurrell, 1995). The transition from a period of strong seasonal gradient to a period of low seasonal gradient at 4200 cal. yr BP is associated with a period of instability in the temperature records of the different seasons between ~4400–4200 and 3500 cal. yr BP.

This is consistent with previous climate reconstructions from the region and the North Atlantic (Bakke et al., 2008; Bjune et al., 2005; Erbs-Hansen et al., 2011b; Krossa et al., 2015; Moros et al., 2004; Nesje et al., 2001; Risebrobakken et al., 2003; Snowball et al., 1999), where a change from a period of strong seasonal gradient to a period of low seasonal gradient has been documented between ~4000 and 3500 cal. yr BP. However, the timing of the transition is not clearly established in regional records. In western Norway, retreating or even lacking glaciers related to reduced winter precipitation rates and warm summers are documented prior to ~4000 cal. yr BP, followed by a period of re-advancing glaciers (Bakke et al., 2008; Bjune et al., 2005; Nesje et al., 2001). Paleoclimate records from the open North Atlantic that are expected to reflect a winter signal suggest a colder mid-Holocene and a warmer late-Holocene with a transition at around 3700 cal. yr BP (Moros et al., 2004). In the Skagerrak, an increased influence of Atlantic Inflow in the eastern basin, discussed to be associated with an increased strength of the Westerlies, is documented starting around 4350 (Erbs-Hansen et al., 2011b), while an increase in Baltic outflow is documented starting at ~3500 cal. yr BP most likely associated with an increase in (winter) precipitation over the catchment area under maritime conditions (Krossa et al., 2015). The period of instability between 4400–4200 and 3500 cal. yr BP evidenced from our Skagerrak records could explain the different timing for the transition from dominant continental to dominant marine climate conditions, documented in the regional records.

According to those results, the transition from mid- to late-Holocene in the region is placed in this study at ~4200 yr BP, at a time when changes occurred in regional temperature records of both the cold and the warm seasons. The choice of this date has been proposed by Walker et al. (2012) in a recent attempt to formally fix the appellations ‘early-’, ‘mid-’, and ‘late’-Holocene.

Abrupt temperature variations in the Skagerrak

The long trends described above are punctuated in our records by multi-centennial fluctuations, particularly distinct during the mid-Holocene in the evolution of the deep-water renewal. These fluctuations are evidenced in both the Mg/Ca-based temperature records and in the δ¹⁸O record from the deep core, suggesting a robust signal. We observe periods of relatively low DWT between ~6300 and 5800, 4800 and 4600, 4300 and 3600 cal. yr BP, and two last centered around 2800 and 1400 cal. yr BP, concomitant with periods of relatively warm deep water. These periods of decreased DWT suggest strong deep-water renewal, reflecting particularly cold winter conditions.

Moreover, the different cold and warm winter phases recorded in the DWT match with winter precipitation reconstruction from Norway (Nesje et al., 2001; Figure 6). Accordingly, low (high) DWT suggesting cold (mild) winter conditions are associated with dry (wet) winters in Norway. The cold and dry winter conditions are most probably reflecting a decreased strength of the Westerlies, while mild and wet winter conditions are most probably reflecting a strengthening of the Westerlies.

Interestingly, the coupled cold and dry winter conditions documented from our records and precipitation reconstructions from Norway (Nesje et al., 2001) also match with the Bond events represented by increases in ice rafted debris (IRD) in northern North Atlantic sediments (Bond et al., 2001; Figure 6). The IRD occurred in the North Atlantic roughly at 1500-year intervals and supposedly are forced by changes in solar intensity and ocean-atmosphere modulation (Bianchi and McCave, 1999; Bond et al., 1997, 2001; Debret et al., 2007; Thornalley et al., 2009). During these events, a general cooling and a shift toward continental-dominated climate conditions are reported in northern Europe (Alley et al., 1997).

Bond event 4, centered around 5900 cal. yr BP, is particularly marked in our DWT record, as well as in the winter precipitation records from Norway (Nesje et al., 2001). However, Bond event 4 is seldom found in the Skagerrak climate records (Emeis et al., 2003; Erbs-Hansen et al., 2011b; Krossa et al., 2015). Moreover, it is not marked in either the IWT or the SST records (this study; Krossa, personal communication, 2015). This suggests that Bond event 4 mainly affected winter climate conditions in Northern Europe. Accordingly, stacked atmospheric temperature records based on pollen from southern Sweden coherently document a cooling event in the annual temperature reconstruction between ~6300 and 5500 yr BP, which is only slightly visible in the summer temperature reconstructions (Seppä et al., 2009). The cooling during winter can be linked to a reduction in Atlantic meridional overturning circulation (AMOC) associated with the Bond event (Wanner and Bütikofer, 2008), most likely resulting in the extension of Artic ice cover during winters and a continental climate over the mid- and high latitudes, as documented for previous Bond events (Denton et al., 2005).

However, Bond event 3 (~4200 cal. yr BP, coinciding with the 4.2 event) is associated in our records with both a period of intense renewal of the deep water (low DWT) and a strong cooling in IWT and SST between about 4200 and 3500 yr BP. This suggests that the cooling event at that time did not only prevail during winter time in the Skagerrak region but also clearly affected the whole year. The difference with the previous period is maybe related to the lower summer–annual insolation at that time, possibly delaying the melting of the extended ice cover.

The following Bond events (2 and 1), centered around 2800 and 1400 cal. yr BP, respectively, are also evidenced in our deep-water records by a decrease in temperature and in particular in salinity but with less significance than prior cold events. Our intermediate water record does not have the resolution to comment on such short-term events, but the SST record does not show any particular cooling (Figure 6). This suggests that those events did not have a strong impact on the climate of the Skagerrak region.

In summary, our records suggest that the 5.9 event was associated with particularly cold and dry winter conditions in the Skagerrak region, and relatively warm summers, implying continental conditions. In contrast, Bond event 3 appears to have affected the climate in the Skagerrak region not only during winter but also during warmer seasons as documented in the intermediate and SST records. Afterward, during the late-Holocene, cold events appear moderate in all seasons, except maybe for the ‘Little Ice Age’ (LIA) which is not entirely covered in our records. The reason for the different seasonal sensitivity of the region to the Bond events may be explained by different insolation at the time of those cold events.

The ‘Medieval Warm Period’

Over the past two millennia, the ‘Medieval Warm Period’ (MWP) and the LIA are the most prominent features in Skagerrak climate records (Hebbeln et al., 2006). The MWP is evidenced in the Skagerrak region between ~1300 and 900 cal. yr BP (Hebbeln et al., 2006) and is covered in our deep-water record. Hebbeln et al. (2006) suggested that an increased advection of saline North Atlantic water reaching the deeper basins in the Skagerrak occurred during the MWP. In our records, this period is characterized by the most pronounced deep-water warming within the complete record, indicating an interval of weakest deep-water renewal. Therefore, our data support this hypothesis and may also suggest that the North Atlantic saline inflow did not only reached deeper levels but probably filled the whole deep basin of the Skagerrak during a period of particularly weak deep-water renewal.

There is an ongoing debate about whether the climate conditions during the MWP were driven by a persistent positive NAO mode (Jones et al., 2014; Moreno et al., 2012; Olsen et al., 2012; Trouet et al., 2009, 2012). Such predominance in a phase of the NAO is attributed to an oceanic modulation by the AMOC that is enhanced or weakened after a positive or negative phase of the NAO (Delworth and Greatbatch, 2000; Robson et al., 2012), which in turn results in a prolonged phase of a locked ocean-atmosphere system (Trouet et al., 2009, 2012). Previous observational and paleoclimate studies have indicated that deep-water renewal in the Skagerrak correlates highly to the NAO (Brückner and Mackensen, 2006; Hagberg and Tunberg, 2000; Tunberg and Nelson, 1998). The exceptional conditions in the Skagerrak during the MWP, with potentially no renewal of the deep water as suggested by our dataset, would support such prevailing positive-like NAO conditions during the MWP (Moreno et al., 2012; Trouet et al., 2009, 2012).

Conclusion

We used benthic Mg/Ca and stable isotopes from a gravity core located at ~420 m water depth in the Skagerrak to reconstruct changes in past DWTs over the last 6800 years that are linked to deep-water renewal during cold winter conditions, and benthic Mg/Ca from a gravity core at intermediate water depth (~270 m) to infer past changes in warm season temperatures in order to comment on seasonal variability in temperature. Overall, the mid-Holocene period was characterized by cold and low-saline deep water and generally warm intermediate waters, suggesting a continental-dominated atmospheric circulation pattern marked by cold and dry winters and also a strong seasonal contrast in temperature. At ~3500 yr BP, warmer DWTs and increased salinity occurred associated with a cooling in the SSTs, indicating a shift toward a more maritime-dominated atmospheric circulation pattern over the late-Holocene period. Superimposed on the general temperature trends, the deep-water records documented periods of relatively frequent or intense deep-water renewal concomitant with dry conditions in Norway and increases in IRD in North Atlantic sediments. The 5.9 event seems to be only associated with winter variability, while younger events appear to be associated with variability during winter and warmer months. During the MWP, there was apparently no renewal of the deep waters which could indicate a period of long-lasting positive-like NAO conditions supporting observations from previous studies. A higher resolution record of the deep-water renewal during the late-Holocene would add to the discussion about winter thermal conditions and seasonality over northwestern Europe for that period.

Footnotes

Acknowledgements

This project is part of the SPP 1400 ‘Frühe Monumentalität und soziale Differenzierung’ and the Graduate School ‘Human Development in Landscape’ (GSHDL) of Kiel University. As such, we would like to express our gratitude to the Deutsche Forschungsgemeinschaft (DFG) and the GSHDL for their financial support. We would also like to thank Dieter Garbe-Schönberg and Marcus Regenberg for their support with the Mg/Ca analyses, Nils Andersen for his help on the stable isotopes analyses, and Marie-José Nadeau for her support on the radiocarbon analyses, together with Walter Dörfler and Christel Van den Bogaard for their contribution to the age model discussion. We would also like to express our thanks toward Alte Nesje who provided us with the winter precipitation data. We thank Thomas Blanz and Sylvia Koch for their work on the alkenone analyses on core IOW225514, following the method described in Leduc et al., 2010. Finally, we want to thank the reviewers and the editor for their time and the constructive criticisms they provided to improve the present manuscript.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Alley

Mayewski

Sowers

. (1997) Holocene climatic instability: A prominent, widespread event 8200 yr ago. Geology 25(6): 483–486.

Andersen

Koç

Moros

(2004) A highly unstable Holocene climate in the subpolar North Atlantic: Evidence from diatoms. Quaternary Science Reviews 23(20–22): 2155–2166.

Austin

WEN

Cage

Scourse

(2006) Mid-latitude shelf seas: A NW European perspective on the seasonal dynamics of temperature, salinity and oxygen isotopes. The Holocene 16(7): 937–947.

Bakke

Lie

Dahl

. (2008) Strength and spatial patterns of the Holocene wintertime westerlies in the NE Atlantic region. Global and Planetary Change 60(1–2): 28–41.

Bianchi

McCave

(1999) Holocene periodicity in the North Atlantic climate and deep-ocean ow south of Iceland. Nature 397(6719): 515–517.

Bjune

Bakke

Nesje

. (2005) Holocene mean July temperature and winter precipitation in western Norway inferred from palynological and glaciological lake-sediment proxies. The Holocene 15(2): 177–189.

Blanz

Emeis

K-C

Siegel

(2005) Controls on alkenone unsaturation ratios along the salinity gradient between the open ocean and the Baltic Sea. Geochimica et Cosmochimica Acta 69(14): 3589–3600.

Bond

Kromer

Beer

. (2001) Persistent solar influence on North Atlantic climate during the Holocene. Science 294(5549): 2130–2136.

Bond

Showers

Chesseby

. (1997) A pervasive millennial-scale cycle in North Atlantic Holocene and glacial climates. Science 278(5341): 1257–1266.

10.

Brückner

(2008) Climatic and hydrographic variability in the late-Holocene Skagerrak as deducted from benthic foraminiferal proxies. Alfred Wegener Institut für Polar- und Meeresforschung. Bremerhaven:

11.

Brückner

Mackensen

(2006) Deep-water renewal in the Skagerrak during the last 1200 years triggered by the North Atlantic Oscillation: Evidence from benthic foraminiferal d¹⁸O. The Holocene 16(3): 331–340.

12.

Brückner

Mackensen

(2008) Organic matter rain rates, oxygen availability, and vital effects from benthic foraminiferal δ 13 C in the historic Skagerrak, North Sea. Marine Micropaleontology 66(3–4): 192–207.

13.

Davis

Brewer

Stevenson

. (2003) The temperature of Europe during the Holocene reconstructed from pollen data. Quaternary Science Reviews 22: 1701–1716.

14.

Debret

Bout-Roumazeilles

Grousset

. (2007) The origin of the 1500-year climate cycles in Holocene North-Atlantic records. Climate of the Past 3(4): 569–575.

15.

Delworth

Greatbatch

(2000) Multidecadal thermohaline circulation variability driven by atmospheric surface flux forcing. Journal of Climate 13(9): 1481–1495.

16.

Denton

Alley

Comer

. (2005) The role of seasonality in abrupt climate change. Quaternary Science Reviews 24(10–11): 1159–1182.

17.

De Villiers

Greaves

Elderfield

(2002) An intensity ratio calibration method for the accurate determination of Mg/Ca and Sr/Ca of marine carbonates by ICP-AES. Geochemistry, Geophysics, Geosystems 3(1): 1001.

18.

Elderfield

Anand

. (2006) Calibrations for benthic foraminiferal Mg/Ca paleothermometry and the carbonate ion hypothesis. Earth and Planetary Science Letters 250(3–4): 633–649.

19.

Emeis

K-C

Struck

Blanz

. (2003) Salinity changes in the central Baltic Sea (NW Europe) over the last 10 000 years. The Holocene 13(3): 411–421.

20.

Erbs-Hansen

Knudsen

Gary

. (2011a) Holocene climatic development in Skagerrak, eastern North Atlantic: Foraminiferal and stable isotopic evidence. The Holocene 22(3): 301–312.

21.

Erbs-Hansen

Knudsen

Gary

. (2011b) Late Younger Dryas and early Holocene palaeoenvironments in the Skagerrak, eastern North Atlantic: A multiproxy study. Boreas 40(4): 660–680.

22.

Fairbanks

(1989) A 17, 000-year glacio-eustatic sea level record: Influence of glacial melting rates on the Younger Dryas event and deep-ocean circulation. Nature 342: 637–642.

23.

Gooday

(2003) Benthic foraminifera (protista) as tools in deep-water palaeoceanography: Environmental influences on faunal characteristics. Advances in Marine Biology 46: 1–90.

24.

Greaves

Caillon

Rebaubier

. (2008) Interlaboratory comparison study of calibration standards for foraminiferal Mg/Ca thermometry. Geochemistry, Geophysics, Geosystems 9: Q0810.

25.

Gyllencreutz

(2005) Late Glacial and Holocene paleoceanography in the Skagerrak from high-resolution grain size records. Palaeogeography, Palaeoclimatology, Palaeoecology 222(3–4): 344–369.

26.

Hagberg

Tunberg

(2000) Studies on the covariation between physical factors and the long-term variation of the marine soft bottom macrofauna in Western Sweden. Estuarine, Coastal and Shelf Science 50: 373–385.

27.

Harwood

AJP

Dennis

Marca

. (2008) The oxygen isotope composition of water masses within the North Sea. Estuarine, Coastal and Shelf Science 78(2): 353–359.

28.

Hass

(1996) Northern Europe climate variations during late-Holocene: Evidence from marine Skagerrak. Palaeogeography, Palaeoclimatology, Palaeoecology 123(1–4): 121–145.

29.

Hebbeln

Knudsen

K-L

Gyllencreutz

. (2006) Late-Holocene coastal hydrographic and climate changes in the eastern North Sea. The Holocene 16(7): 987–1001.

30.

Hurrell

(1995) Decadal trends in North Atlantic Oscillation: Regional Temperatures and Precipitation. Science 269(5224): 676–679.

31.

Hurrell

Kushnir

Ottersen

. (2003) An overview of the North Atlantic Oscillation. The North Atlantic Oscillation: Climatic Significance and Environmental Impact. Epub ahead of print 19 March. DOI: 10.1029/134GM01.

32.

Joint

Pomroy

(1993) Phytoplankton biomass and production in the southern North Sea. Available at: http://www.int-res.com/articles/meps/99/m099p169.pdf.

33.

Jones

Harpham

(2014) Winter responding proxy temperature reconstructions and the North Atlantic Oscillation. Journal of Geophysical Research: Atmospheres 119(11): 6497–6505.

34.

Kristjánsdóttir

Lea

Jennings

. (2007) New spatial Mg/Ca-temperature calibrations for three Arctic, benthic foraminifera and reconstruction of north Iceland shelf temperature for the past 4000 years. Geochemistry Geophysics Geosystems 8(3): 1–27.

35.

Krossa

Moros

Blanz

. (2015) Late-Holocene Baltic Sea outflow changes reconstructed using C37: 4 content from marine cores. Boreas 44(1): 81–93.

36.

Laskar

Robutel

Joutel

. (2004) A long-term numerical solution for the insolation quantities of the Earth. Astronomy & Astrophysics 428: 261–285. Available at: http://www.aanda.org/articles/aa/abs/2004/46/aa1335/aa1335.html (accessed 7 March 2016).

37.

Leduc

Schneider

Kim

. (2010) Holocene and Eemian sea surface temperature trends as revealed by alkenone and Mg/Ca paleothermometry. Quaternary Science Reviews 29: 989–1004.

38.

Liu

Zhu

Rosenthal

. (2014) The Holocene temperature conundrum. Proceedings of the National Academy of Sciences 111(34): E3501–E3505.

39.

Ljøen

Svansson

(1972) Long-term variations of subsurface temperatures in the Skagerrak. Deep Sea Research 19: 277–288.

40.

Locarnini

Mishonov

Antonov

. (2010) World Ocean Atlas 2009, Volume 1: Temperature. In: Levitus

(ed.) NOAA Atlas NESDIS 68. Washington, DC: Government Printing Office, 184 pp.

41.

Mangini

Spötl

C and Verdes P

(2005) Reconstruction of temperature in the Central Alps during the past 2000 yr from a δ¹⁸O stalagmite record. Earth and Planetary Science Letters 235(3–4): 741–751. Available at: http://linkinghub.elsevier.com/retrieve/pii/S0012821X05003195 (accessed 4 March 2016).

42.

Mangini

Verdes

Spötl

. (2007) Persistent influence of the North Atlantic hydrography on central European winter temperature during the last 9000 years. Geophysical Research Letters 34(2): L02704. Available at: http://doi.wiley.com/10.1029/2006GL028600 (accessed 4 March 2016).

43.

Martin

Lea

(2002) A simple evaluation of cleaning procedures on fossil benthic foraminiferal Mg/Ca. Geochemistry, Geophysics, Geosystems 3(10): 8401.

44.

Mikalsen

Sejrup

(2000) Oxygen isotope composition of fjord and river water in the Sognefjorden drainage area, Western Norway. Implications for Paleoclimate studies. Estuarine, Coastal and Shelf Science 50(4): 441–448.

45.

Moreno

Pérez

Frigola

. (2012) The Medieval Climate Anomaly in the Iberian Peninsula reconstructed from marine and lake records. Quaternary Science Reviews 43: 16–32.

46.

Moros

Emeis

Risebrobakken

. (2004) Sea surface temperatures and ice rafting in the Holocene North Atlantic: Climate influences on northern Europe and Greenland. Quaternary Science Reviews 23(20–22): 2113–2126.

47.

Nesje

Matthews

Dahl

. (2001) Holocene glacier fluctuations of Flatebreen and winter-precipitation changes in the Jostedalsbreen region, western Norway, based on glaciolacustrine sediment records. The Holocene 11(3): 267–280.

48.

O’Brien

Mayewski

Meeker

. (1995) Complexity of Holocene climate as reconstructed from a Greenland ice core. Science 270(21): 1–3.

49.

Olsen

Anderson

Knudsen

(2012) Variability of the North Atlantic Oscillation over the past 5,200 years. Nature Geoscience, Nature Publishing Group 5(11): 808–812.

50.

Oppo

McManus

Cullen

(2003) Deepwater variability in the Holocene Epoch. Nature Communications 422: 277–278.

51.

Otto

Zimmerman

Furnes

. (1990) Review of the physical oceanography of the North Sea. Netherlands Journal of Sea Research 26(2–4): 161–238.

52.

Raitzsch

Kuhnert

Groeneveld

. (2008) Benthic foraminifer Mg/Ca anomalies in South Atlantic core top sediments and their implications for paleothermometry. Geochemistry Geophysics Geosystems 9: Q05010.

53.

Reimer

Bard

Bayliss

. (2013) IntCal13 and Marine13 radiocarbon age calibration curves 0–50,000 years cal BP. Radiocarbon 55(4): 1869–1887.

54.

Risebrobakken

Jansen

Andersson

. (2003) A high-resolution study of Holocene paleoclimatic and paleoceanographic changes in the Nordic Seas. Palaeoceanography 18(1): 1017–1031.

55.

Robson

Sutton

Lohmann

. (2012) Causes of the rapid warming of the North Atlantic Ocean in the mid-1990s. Journal of Climate 25(12): 4116–4134.

56.

Rodhe

(1987) The large-scale circulation in the Skagerrak; interpretation of some observations. Tellus A 39(1–3): 245–253.

57.

Rodhe

Holt

(1996) Observations of the transport of suspended matter into the Skagerrak along the western and northern coast of Jutland. Journal of Sea Research 35(1–3): 91–98.

58.

Schmidt

Spero

Lea

(2004) Links between salinity variation in the Caribbean and North Atlantic thermohaline circulation. Nature 428: 160–163.

59.

Seppä

Bjune

Telford

. (2009) Last nine-thousand years of temperature variability in Northern Europe. Climate of the Past 5: 523–535.

60.

Shackleton

(1974) Attainment of isotopic equilibrium between ocean water and the benthonic foraminifera genus Uvigerina: Isotopic changes in the ocean during the last glacial. Colloques Internationaux du C.N.R.S 219: 203–209.

61.

Snowball

Sandgren

Petterson

(1999) The mineral magnetic properties of an annually laminated Holocene lake-sediment sequence in northern Sweden. The Holocene 9(3): 353–362.

62.

Spangehl

Cubasch

Raible

. (2010) Transient climate simulations from the Maunder Minimum to present day: Role of the stratosphere. Journal of Geophysical Research 115. Available at: http://onlinelibrary.wiley.com/doi/10.1029/2009JD012358/full (accessed 8 August 2015).

63.

Svansson

(1975) Physical and chemical oceanography of the Skagerrak and the Kattegat. I. Open Sea Conditions. Institute of Marine Research Report No. 1. Lysekil: Fishery Board of Sweden, pp. 1–88.

64.

Thornalley

DJR

Elderfield

McCave

(2009) Holocene oscillations in temperature and salinity of the surface subpolar North Atlantic. Nature 457(7230): 711–714.

65.

Trouet

Esper

Graham

. (2009) Persistent positive North Atlantic oscillation mode dominated the Medieval Climate Anomaly. Science 324(5923): 78–80.

66.

Trouet

Scourse

Raible

(2012) North Atlantic storminess and Atlantic Meridional Overturning Circulation during the last Millennium: Reconciling contradictory proxy records of NAO variability. Global and Planetary Change 84–85: 48–55.

67.

Tunberg

Nelson

(1998) Do climatic oscillations influence cyclical patterns of soft bottom macrobenthic communities on the Swedish west coast? Marine Ecology Progress Series 170: 85–94.

68.

Walker

Berkelhammer

Björck

. (2012) Formal subdivision of the Holocene Series/Epoch: A Discussion Paper by a Working Group of INTIMATE (Integration of ice-core, marine and terrestrial records) and the Subcommission on Quaternary Stratigraphy (International Commission on Stratigraphy). Journal of Quaternary Science 27(7): 649–659. Available at: http://onlinelibrary.wiley.com/doi/10.1002/jqs.2565/full (accessed 18 July 2015).

69.

Wanner

Bütikofer

(2008) Holocene Bond Cycles: Real or imaginary. Geografie 4(113): 338–349.

70.

Wanner

Mercolli

Grosjean

. (2014) Holocene climate variability and change; a data-based review. Journal of the Geological Society 172(2): 254–263.

71.

Weltje

Tjallingii

(2008) Calibration of XRF core scanners for quantitative geochemical logging of sediment cores: Theory and application. Earth and Planetary Science Letters 274: 423–438.

72.

Xoplaki

Luterbacher

Paeth

. (2005) European spring and autumn temperature variability and change of extremes over the last half millennium. Geophysical Research Letters 32(15): L15713. Available at: http://doi.wiley.com/10.1029/2005GL023424 (accessed 4 March 2016).

Reconstruction of mid- to late-Holocene winter temperatures in the Skagerrak region using benthic foraminiferal Mg/Ca and δ 18 O

Abstract

Keywords

Introduction

Regional setting

Deep-water renewal and regional climate

Materials and methods

Sample material

Mg/Ca analyses and temperature reconstructions

Mg/Ca as a DWT proxy in the Skagerrak

δ18O analyses

Age model

Results

Discussion

Approach to assess DWT and IWT signals in the Skagerrak

Long-term temperature trends in the Skagerrak region

Abrupt temperature variations in the Skagerrak

The ‘Medieval Warm Period’

Conclusion

Footnotes

Acknowledgements

Funding

References

δ¹⁸O analyses