10.1016 j.quascirev.2018.07.017

Quaternary Science Reviews 197 (2018) 246-256

ELSEVIER

Contents lists available at ScienceDirect

Quaternary Science Reviews

journal homepage: www.elsevier.com/locate/quascirev

An Icelandic terrestrial record of North Atlantic cooling c. 8800-8100
cal. yr BP

Check for
updates

Sigrun Dogg Eddudottir • , Egill Erlendsson , Gudrun Gisladottir a> b

a Institute of Life and Environmental Sciences, University of Iceland, Sturlugata 7, 101 Reykjavik, Iceland
b Institute of Earth Sciences, University of Iceland, Sturlugata 7, 101 Reykjavik, Iceland

ARTICLE INFO

Article history:

Received 4 May 2018
Received in revised form
6 July 2018
Accepted 9 July 2018
Available online 25 August 2018

Keywords:

Climate change
Vegetation dynamics
Pollen

8.2 ka event
Palaeoclimate
North Atlantic
Iceland
Holocene
Lake sediment

1. Introduction

Iceland‘s location in the North Atlantic makes the terrestrial
environment of the island sensitive to changes in atmospheric and
oceanic systems in the region. The terrestrial ecosystem is partic¬
ularly responsive to changes in ocean currents since the island is
located at the junction of the relatively warm Irminger Current and
the colder East Greenland Current (e.g. Stefansson, 1962; Hansen
and 0sterhus, 2000). Palaeoenvironmental reconstructions from
Icelandic lake sediments have revealed centennial and millennial
scale environmental responses to climate change during the Ho¬
locene (Hallsdottir, 1995; Rundgren, 1998; Hallsdottir and
Caseldine, 2005; Geirsdottir et al., 2009, 2013; Larsen et al., 2011,
2012; Striberger et al., 2012; Blair et al., 2015; Eddudottir et al.,
2015, 2016; Schomacker et al., 2016). A prominent anomaly in
some Icelandic palaeoenvironmental reconstructions is a cold
period between c. 8700-7900 cal. yr BP, interrupting a trend of
warming climate during the early Holocene (Geirsdottir et al., 2009,

* Corresponding author.

E-mail address: sde6@hi.is (S.D. Eddudottir).

https://doi.Org/10.1016/j.quascirev.2018.07.017
0277-3791 /© 2018 Elsevier Ltd. All rights reserved.

2013; Larsen et al., 2012; Eddudottir et al., 2015). The cooling during
this period is evident from proxies sensitive to changes in spring
and summer temperatures and is reflected in changes in both
terrestrial vegetation (Eddudottir et al., 2015) and lake productivity
(Larsen et al., 2012). The cooling observed in Iceland falls within the
period c. 9000-8000 cal. yr BP, which is characterised by cold
conditions in the Northern Hemisphere (Mayewski et al., 2004),
and coincides with a cold anomaly beginning c. 8600 cal.yr BP
recorded around the North Atlantic (cf. Rohling and Palike, 2005).
This longer cold period is distinct from the cold anomalies related
to the 8.2 ka cold event (e.g. Alley et al., 1997; Baldini et al., 2002;
Alley and Agustsdottir, 2005; Rohling and Palike, 2005; Bamberg
et al., 2010), which has been linked to the flux of meltwater at
the final drainage of the glacial lakes Agassiz and Ojibway into the
North Atlantic c. 8430 ± 300 cal. yr BP (Barber et al., 1999).

The timing and nature of terrestrial ecosystem changes in Ice¬
land can highlight the nature of centennial scale climate fluctua¬
tions in the North Atlantic and provide information to complement
marine records. The aim of this study was to create a detailed re¬
cord of environmental changes in northwestern Iceland in response
to early Holocene climate change. Here we present a palynological
reconstruction of the period c. 10,100-7000 cal.yr BP. To establish
whether changes in environmental stability occurred during this
period the multi-decadal resolution palynological dataset is sup¬
plemented by plant macrofossils and measurements of organic and
physical properties of the sediment, including organic matter (OM),
total carbon (TC), total nitrogen (TN), carbon-to-nitrogen ratio (C/
N), stable isotope ratios of carbon (5 13 C) and nitrogen (6 15 N),
magnetic susceptibility (MS) and dry bulk density (DBD).

2. Regional setting

The site chosen for this study is Kagadarholl palaeolake (KAGA;
65°35' 16" N, 20°07' 58" W, 114 m a.s.l.) located in northwest Ice¬
land, about 10 km south of the town of Blonduos (Fig. 1). The
palaeolake is today a fen and is about 0.45 km 2 in area, with a
drainage basin of about 0.6 km 2 . The vegetation is dominated by
sedges (Cyperaceae spp.), especially Eriophorum angustifolium
(common cotton grass), the dwarf shrubs Betula nana (dwarf birch),
Salix phylicifolia (tea-leaved willow) and Salix lanata (woolly

S.D. Eddudottir et al. / Quaternary Science Reviews 197 (2018) 246-256

247

20°30'0"W

20°0'0"W

19°30'0"W

65°50 , 0 ,, N-

es^o'CN-

65°30'0"N-

65°20'0"N-

▲ Weather st
★ Coring
— 1000
m 400
0

Blonduos

▲

★

Kagad

Hunafloi

Skagafjordur

Fig. 1 . The Kagadarholl study site, a) Map showing the location of the coring site in northwest Iceland, b) map of Iceland showing the location of Kagadarholl and, c) overview of the
palaeolake.

willow), and the herb Cardamine nymanii (lady's smock). The
overgrown lake is surrounded by Betula nana -dominated dwarf-
shrub heath, hayfields and eroded, gravelly hills. The closest
weather station is Blonduos (Table 1 ), which is located by the coast
and may therefore represent a more maritime climate than
Kagadarholl. For more detailed site information see (Eddudottir
et al., 2015, 2017).

3. Material and methods

Sediment cores were retrieved from the centre of the palaeolake
using a Livingstone piston corer and a Bolivia adaptor fitted with 75
mm diameter polycarbonate tubes. A series of overlapping cores
was used to construct a continuous Holocene sequence with
0 cm at the top of the core (see Eddudottir et al., 2015). The period c.
10,100-7000 cal. yr BP is covered by three overlapping core seg¬
ments connected by changes in magnetic susceptibility (MS) and
tephra geochemistry (Fig. SI). A description of sediment charac¬
teristics according to the Troels-Smith system (Aaby and Berglund,
1986) is presented in Table SI.

The pollen dataset consists of 46 samples, covering the period c.
10,100-7000 cal. yr BP, comprising 31 samples previously analysed

as part of a dataset published in Eddudottir et al. (2015) and 15 new
samples. Subsamples for pollen analysis (2 cm 3 ) were collected at
2-8 cm intervals. Pollen samples were prepared using 10% HC1,10%
NaOH, acetolysis (Faegri et al., 1989; Moore et al., 1991 ) and dense-
media separation (Bjorck et al., 1978; Nakagawa et al., 1998) using
LST Fastfloat (a sodium heteropolytungstate solution, density 1.9
gem -3 ). A tablet containing spores of Lycopodium clavatum
(Stockmarr, 1971; batch no. 177745) was added to each sample to
calculate pollen accumulation rates (PARs; grains cm -2 year -1 ).
Samples were mounted on glass microscope slides in silicone oil of
12,500 cSt viscosity. A minimum of 300 indigenous terrestrial
pollen grains were counted for each sample. Identification of pollen
grains and spores was based on Moore et al. (1991) and a pollen
type slide collection at the University of Iceland. Pollen sums and
categories followed Hallsdottir (1987) and Caseldine et al. (2006).
Pollen and spore taxonomy followed Bennett (2016) and amend¬
ments specific to the Icelandic flora by Erlendsson (2007). The
sediments were analysed for plant macrofossils at contiguous 5 cm
intervals and sample volume varied between 35 and 50 mL (for
further detail see Eddudottir et al. (2015)). Pollen and macrofossil
diagrams were constructed using TILIA version 1.7.16 (Grimm,
2011 ).

Table 1

Observations from the weather station Blonduos. Unpublished data from the Icelandic Met Office.

Elevation 8 m a.s.l.

Period 1961-1990

Mean July temperature 9.4 ° C

Mean January temperature -2.5 °C

Mean tritherm temperature (mean June, July, August temperature) 8.7 °C

Mean annual precipitation 458 mm

248

S.D. Eddudottir et al. / Quaternary Science Reviews 197 (2018) 246-256

Fig. 2. Age-depth model for the Kagadarholl core.

DBD and OM were measured at 1 cm contiguous intervals. OM
was measured by loss on ignition (LOI) by combustion for 5 h at
550 °C in a muffle furnace (Bengtsson and Enell, 1986) and DBD was
calculated by dividing the dry weight of a sample with the volume
of the wet sample (Brady and Weil, 1996). Magnetic susceptibility
(MS) measurements were performed using a Bartington MS2 meter
and Bartington MS2F probe (Dearing, 1994) at 1 cm intervals on
split core segments.

The core was subsampled for measurements of total carbon
(TC), total nitrogen (TN) and 5 13 C and 5 15 N stable isotope ratios of
bulk sediments at the same depths as the pollen samples, where
possible. The samples were dried at 50 °C, homogenized using a
ball mill and sieved through a 150 pm mesh. A total of 46 samples
were analysed. The carbon, nitrogen and isotope measurements
were performed on a Thermo Delta V isotope ratio mass spec¬
trometer (IRMS) interfaced to a NC2500 elemental analyser at the
Cornell University Isotope Laboratory (COIL).

The chronology for the Kagadarholl core (Fig. 2) was constructed
using a combination of tephrochronology and radiocarbon dates
(Eddudottir et al., 2015). The age-depth model for the core (Fig. 2) is
based on the previously dated tephra layers Hekla 6 (6060 cal. yr
BP; Gudmundsdottir et al., 2011), Katla S (6600 cal.yr BP;
Eddudottir et al., 2015), Hekla 5 (7050 cal.yr BP; Thorarinsson,
1971) and Saksunarvatn (10,300 cal.yr BP; Rasmussen et al.,
2006) and radiocarbon dated macrofossils (Table 2). A smooth
spline age-depth model (Fig. 2) was constructed in R using the
package Clam (Blaauw, 2010). Ages are given in calibrated years
before present (cal. yr BP). According to the 95% confidence of the
age model the age uncertainty is ±100-120 yrs for most of the
record; lowest c. 8600-7400 cal.yr BP (<± 100 yr) and highest

(±120-130 yrs) c. 10,100-9700 cal. yr BP.

Ordination analyses were performed on the pollen dataset to
detect patterns in terrestrial vegetation development in the vicinity
of the lake. Detrended correspondence analysis (DCA) was initially
used. As the first axis length of 1.3527 suggests a linear response in
the dataset, a principal component analysis (PCA) was preferred.
The PCA was performed on a Hellinger transformed dataset that
included terrestrial pollen taxa with abundances >1%. PCA analysis
was also performed on a standardised dataset of %TC, %TN, %OM,
C/N, MS, DBD, 5 13 C and 5 15 N along with the most important pollen
groups and taxa ( Betula , Juniperus communis , shrubs, herbs and
aquatics) to determine the relationships between vegetation and
sediment properties. This analysis includes 42 samples analysed for
both pollen and organic and physical properties. All ordination
analyses were made in R using the package vegan (Oksanen et al.,
2016).

4. Results

4.1. Pollen assemblage zones (PAZs)

The pollen record covers the period c. 10,100-7000 cal.yr BP
and is divided into five pollen assemblage zones (PAZs) based on
CONISS performed in TILIA (Fig. 3).

PAZ 1 (10,100-9600 cal. yr BP)

Juniperus communis pollen is -15—32% of total land pollen (TLP),
Betula is -14-32% of TLP and Cyperaceae is -16-24% of TLP. Other
important pollen types are Empetrum nigrum (-1-10% of TLP), Salix
(-3-11% of TLP), Galium (-0.3-6% of TLP) and Thalictrum alpinum
(-3-9% of TLP) (Fig. 3). The Juniperus communis PAR increases to
>600 grains cm -2 yr -1 c. 9800 cal. yr BP (Fig. 4). Mean Betula pollen
diameters are -17 ± 1 pm within this PAZ (Fig. 5). Potamogeton
natans- type pollen (-18-82% of total pollen and spores) is the most
important aquatic pollen taxon (Fig. 3). Selaginella selaginoides
megaspores and Potamogeton natans seeds along with Luzula and
Carex nigra seeds are the most common plant macrofossils recorded
within this PAZ (Fig. 6). Non-triporate Betula pollen range between
2 and 9% (Fig. 4).

PAZ 2 (9600-9200 cal. yr BP)

Juniperus communis pollen increases to -38-58% of TLP and
Cyperaceae decreases to -4-13% of TLP. Betula ranges between 14
and 30% of TLP. Galium and Thalictrum alpinum pollen decrease
within this PAZ, along with Empetrum nigrum and Potamogeton
natans- type pollen. Iso'etes echinospora spores appear c. 9500 cal. yr
BP (Fig. 3). The relative abundance of Juniperus communis pollen
increases after c. 9600 cal. yr BP and the PAR is -1300-3200 grains
cnrT 2 yr -1 between c. 9600 and 9200 cal. yr BP. This period delivers
the highest TLP PAR in the record (-3200-6400 grains cm -2 yr -1 ),
as well as high TLP concentration (-26,000-50,000 grains cm -3 ;
Fig. 4). Mean Betula pollen diameters are -18 ± 1 pm within this PAZ
(Fig. 5). Betula nana seeds and leaves and Selaginella selaginoides
megaspores are the most common plant macrofossils recorded.

Table 2

Radiocarbon dates from the Lake Kagadarholl sediment.

Lab code

Depth (cm)

14 C date ± la

5 13 C

Calibrated age range

Material

ETH-56125

385

7003 ±36

-20.6

7934-7751

Potamogeton leaf fragments

ETH-56126

401

7170±35

-35.0

8033-7937

Empetrum nigrum and Betula pubescens various items

ETH-56135

461

8209 ±38

-19.1

9280-9031

Potamogeton leaf fragments

ETH-56129

524

8695 ±38

-11.7

9761-9546

Potamogeton fruits and leaf fragments

ETH-56130

583

8997 ±39

-11

10,240-9942

Potamogeton fruits and leaf fragments

S.D. Eddudottir et al. / Quaternary Science Reviews 197 (2018) 246-256

249

Fig. 3. Pollen percentage diagram for the Kagadarholl core.

Betula pubescens macro fossils appear at the top of this PAZ (Fig. 6).
Non-triporate Betula pollen range between 2 and 5% (Fig. 4).

PAZ 3 (9200-8800 cal.yr BP)

Juniperus communis pollen decreases within this PAZ to
-22-35% of TLP and Betula pollen increases to -34-49% of TLP.
Pollen of Oxyria digyna and Potentilla-type appear within this PAZ
but remain <1% (Fig. 3). The Juniperus communis PAR decreases to
<900 grains cm -2 yr -1 c. 9100 cal.yr BP, and the Betula PAR in¬
creases to >1000 grains cm -2 yr -1 after c. 9200 cal.yr BP (Fig. 4).
Mean Betula pollen diameters increase to -20 ± 1 pm (Fig. 5). Betula
pubescens macrofossils are abundant within this PAZ (Fig. 6). Non-
triporate Betula pollen range between 2 and 7% (Fig. 4).

PAZ 4 (8800-8100 cal. yr BP)

Betula pollen decreases after c. 8700 cal.yr BP to <32% of TLP
and Juniperus communis pollen continues to decrease to -11-26% of
TLP. Poaceae pollen increases to >20% of TLP c. 8800 cal.yr BP and
Cyperaceae to >16% of TLP c. 8700 cal. yr BP (Fig. 3). The Betula PAR
decreases to <600 grains cm -2 yr -1 after c. 8800 cal.yr BP and
<200 grains cm -2 yr -1 after c. 8500 cal.yr BP. The TLP PAR de¬
creases to <1400 grains cm -2 yr -1 c. 8800 cal.yr BP and <900
grains cm -2 yr -1 c. 8500 cal.yr BP (Fig. 4). Mean Betula pollen di¬
ameters are -20 ± lpm within this PAZ (Fig. 5). Myriophyllum
alterniflorum pollen increases while Potamogeton natans- type pol¬
len decreases (Fig. 3). Myriophyllum alterniflorum seeds are the
most common plant macrofossils within this PAZ (Fig. 6). Non-
triporate Betula pollen range between 1 and 10% (Fig. 4).

PAZ 5 (8100-7000 cal. yr BP)

Betula pollen increases rapidly between c. 8100 and 8000 cal. yr
BP from -37% TLP c. 8100 cal. yr BP to -79% TLP c. 8000 cal. yr BP.
Juniperus communis pollen decreases dramatically within this PAZ
to -1-9% of TLP. Other pollen taxa such as Poaceae, Cyperaceae,
Salix, Thalictrum alpinum and Empetrum nigrum decrease. Galium
pollen disappears along with Selaginella selaginoides spores (Fig. 3).
The Betula PAR is high from c. 8000 cal.yr BP (-1500—2500 grains
cm -2 yr -1 ) and the TLP PAR is -1800-3200 grains cm -2 yr -1
(Fig. 4). Mean Betula pollen diameters are -21 ± 1 pm (Fig. 5). Betula
pubescens catkin scales and fruits and Myriophyllum alterniflorum
seeds are the most abundant plant macrofossils in this PAZ (Fig. 6).
Non-triporate Betula pollen range between 2 and 5% (Fig. 4).

4.2. Principal component analysis of pollen data

The first PCA axis accounts for 66.5% of the variance in the
dataset and the second axis for 12.6%. The terrestrial pollen
assemblage in PAZ 1 (c. 10,100-9600 cal.yr BP) is influenced by
Empetrum nigrum, Poaceae, Cyperaceae, Thalictrum alpinum and
Galium. Juniperus communis is the dominant pollen type in PAZ 2 (c.
9600-9200 cal. yr BP), however the influence of Betula pollen in¬
creases in PAZ 3 (c. 9200-8800 cal.yr BP). Empetrum nigrum,
Poaceae, Cyperaceae, Thalictrum alpinum and Galium, as well as
Oxyria digyna show a strong influence on the pollen assemblages
within PAZ 4 (c. 8800-8100 cal.yr BP). Betula becomes the domi¬
nant pollen taxon within PAZ 5 (c. 8100-7000 cal.yr BP) (Fig. 7).

4.3. Organic matter and physical properties

MS and DBD values are high (DBD 0.16-0.34 g cnrr 3 , excluding
tephra layers) between c. 10,100 and 9300 cal. yr BP (PAZ 1 and 2).
Values decrease after c. 9300 cal. yr BP and the lowest values for the

250

5.D. Eddudottir et al. / Quaternary Science Reviews 197 (2018) 246-256

ts\

/>

///

7000

7500

8000

8500

9000

9500

10000

340

360

380

400

420

440

460

480

500

520

540

560

580

0

/

/ #

100020003000

&

2000 4000

/

✓

#

Jr

500 10001500

,<s*

iV

✓

.o\^

/

4000 8000

PAZ

5

Fig. 4. Pollen accumulation rates of selected pollen taxa and TLP, concentration of TLP and percentage non-triporate Betula pollen.

Fig. 5. Boxplot of Betula pollen grain diameters for each PAZ in the Kagadarholl record.

two proxies are recorded after c. 8500 cal. yr BP (DBD 0.09-0.19
gem -3 ). Organic matter proxies (%TC, %TN, C/N and %OM) increase
with time. %TC increases from -10-13% in PAZ 1 to >20% in PAZ 5.
The C/N ratio increases from -13% in PAZ 1 to -15% in PAZ 5. The
5 13 C isotope record shows decreased values with time, from — 18%o
at c. 10,100 cal.yr BP to<— 21%o c. 9100 cal.yr BP. This trend is
reversed after c. 8700 cal. yr BP and higher values (>— 20%o) are
recorded between c. 8400 and 8100 cal.yr BP. Values decrease
to — 21 and -23%o after c. 8100 cal. yr BP (Fig. 8).

Although dissolved inorganic carbon (DIC) is found in rivers in
Iceland (Kardjilov et al., 2006), the contribution of inorganic carbon
toTC in Icelandic lake sediments is considered small due to the lack
of carbonate bedrock (e.g. Langdon et al., 2010; Larsen et al., 2012;
Johannesson, 2014; Harning et al., 2018) and changes in TC are
considered to reflect changes in total organic carbon (TOC).

4.4. Principal Component Analysis of physical and organic matter
properties and pollen

The first PCA axis accounts for 61.8% of the variance in the
dataset and the second axis for 14.5%. High MS and DBD influence

the samples in PAZs 1 and 2. Herb and shrub pollen, 5 13 C, as well as
aquatics are the dominant features in the samples in PAZ 1. In PAZ 2,
Juniperus communis pollen and 5 15 N are the most important factors
influencing the samples. PAZ 3 is characterised by increased in¬
fluence of Betula pollen and organic proxies (%OM, %TC and %TN),
while the importance of the C/N ratio increases in PAZ 4. The
dominant factors influencing the samples in PAZ 5 are Betula pollen
and organic proxies (%OM, %TC, %TN and C/N) (Fig. 9).

5. Discussion

5.1. Early Holocene (c. 10,100-9200 cal. yr BP)

The pollen assemblage from Kagadarholl suggests the presence
of Juniperus communis-Betula nana-Salix shrub heath for the first
four centuries of the record (Fig. 3). This is supported by the
dominance of pollen of Juniperus communis and Betula within PAZ 1.
Other important pollen taxa are Empetrum nigrum and Salix, grasses
(Poaceae), sedges (Cyperaceae), Thalictrum alpinum and Galium
(Figs. 3 and 7). The presence of Betula nana is evidenced by small
Betula pollen diameters -17 ± 1 pm (Fig. 5) concurrent with modern
Betula nana pollen (Makela, 1996), as well as the presence of Betula
nana macro fossils (Fig. 6). Macro fossils of heath taxa such as
Empetrum nigrum, Salix spp., Selaginella selaginoides and grami-
noids (e.g. Carex nigra and Luzula spp.) support the presence of
heath around the lake (Fig. 6).

The shrub heath was replaced by Juniperus communis scrub c.
9600 cal. yr BP, evidenced by a relative increase in Juniperus com¬
munis pollen to >38% of TLP (Fig. 3) and PARs of >1300 grains cm -2
yr -1 (Fig. 4). Juniperus communis is dominant in the TLP assemblage
in PAZ 2. The high PAR of Juniperus communis pollen in the lake
during this period indicates the presence of the upright growth
form of the species around the lake (Eddudottir et al., 2015). Juni¬
perus communis produces large quantities of pollen (Huntley and
Birks, 1983), but low amounts of pollen are found in lakes despite
strong local presence (Birks and Bjune, 2010). However, high per¬
centages and PARs of Juniperus communis can be produced by an
upright growth form of dense scrub (Birks, 1973). Similar peaks in
Juniperus communis pollen are also seen in other records from
northern Iceland during this period (Hallsdottir, 1995; Rundgren,
1998; Caseldine et al., 2003, 2006; Hallsdottir and Caseldine,
2005), when climate was relatively warm (Langdon et al., 2010)

S.D. Eddudottir et al. / Quaternary Science Reviews 197 (2018) 246-256

251

Trees and shrubs Herbs Pteridophytes Aquatics Sphagnum

Fig. 6. Plant macrofossil diagram for Kagadarholl. Concentrations in 50 mL of sediment.

and environmental conditions probably favoured Juniperus com¬
munis over Betula pubescens (Eddudottir et al M 2015). A decrease in
pollen from dwarf shrubs, graminoids and other herbs (Figs. 3, 4
and 7) also reflects the transition to a new vegetation commu¬
nity. This period is characterised by environmental instability,
indicated by relatively high sedimentation rate (Fig. 2) and large
minerogenic input into the lake, as suggested by relatively high and
variable DBD (>0.16 gem -3 ) and MS. Organic matter (<30%), %TC
(-8-12%) and %TN (~0.8-1.2%) are relatively low and 5 13 C values
decrease from -18.6 to -20.6%o. Although the lowest C/N ratios in
the record occur during this period, values of 14.2-15.9 (Fig. 8)
show that terrestrial organic matter contributed a significant
component to the sediment (Meyers and Lallier-Verges, 1999). The
TLP PAR (-3200-6400 grains cm -2 yr -1 ) is the highest during this
period, and the TLP concentration is also high (-26,000-50,000
grains cm -3 ) (Fig. 4), suggesting warm conditions favourable for
pollen production.

Small mean Betula pollen diameters (-18 ± 1 pm) c. 9600-9200
cal.yr BP (Fig. 5) indicate that Betula nana was still the dominant
Betula species, although Betula pubescens macrofossils appear in
the record c. 9300 cal. yr BP (Fig. 6). The percentage of non-triporate
Betula pollen ranges between 2 and 9% during this period (Fig. 4)
and is above 3% in most of the pollen samples. A high proportion of
non-triporate pollen indicates the presence of hybrid individuals of
Betula nana and Betula pubescens. Modern individuals of Betula
nana and Betula pubescens in Iceland produce on average 2.4% and
0.7% of non-triporate pollen, respectively. In contrast, hybrid in¬
dividuals of the two species produce about 12.2% non-triporate
pollen (Karlsdottir et al., 2008). The relatively high percentages of
non-triporate Betula pollen indicate some hybridisation between
the two species during the early Holocene as has been recorded at
other sites in Iceland (Karlsdottir et al., 2009, 2012; Karlsdottir,
2014; Eddudottir et al., 2015, 2016).

5.2. Early birch woodland development (c. 9200-8800 cal.yr BP)

Birch woodland development is indicated by increased impor¬
tance of Betula in the pollen assemblage c. 9200 cal.yr BP (PAZ 3;
Fig. 7). The Betula PAR increases to >1000 grains cm -2 yr -1 after c.
9200 cal. yr BP (Fig. 4) and the relative abundance to >40% of TLP c.
9100 cal.yr BP (Fig. 3). This corresponds to a decrease in Juniperus
communis pollen to <35% of TLP, probably reflecting competition
for space with Betula pubescens and the effects of an increasingly
closed canopy upon the shade intolerant Juniperus communis
(Thomas et al., 2007). Mean Betula diameters increase to
-20 ±1 pm (Fig. 5), indicating an increased importance of Betula
pubescens pollen (Makela, 1996). As woodland developed, envi¬
ronmental stability increased, resulting in less minerogenic mate¬
rial deposited in the lake, which is reflected in lower values of DBD
(0.13-0.19 g cm -3 ) and MS. This change in sediment characteristics
is reflected in the sedimentation rate that decreases c. 9200 cal. yr
BP (Fig. 2). Organic matter proxies show an increase in %TC
(14.7-20.9%) and C/N (13.3-14.7), while 5 13 C decreases (-20.3
to — 23.7%o) (Fig. 8). Terrestrial plants in Icelandic lakes have 5 13 C
values that are significantly lower (-24.4 to -30.9%o) than those of
aquatic plants (-11,5 to -15,l%o; Wang and Wooller, 2006). Hence,
the decrease in 5 13 C, accompanied by relatively high C/N ratios may

252

5.D. Eddudottir et al. / Quaternary Science Reviews 197 (2018) 246-256

C\J

E

£

o' ^

< CM

1 £ I

TO o

•2 w
a> .g
QQ co

z

un

To

O —

(da ja "i bo) 06v

*

CM

CL

if)

§ "co CO
o c E

8 s

CM O
1 CO

CM

CD O

co

Fig. 8. Organic and physical properties of the KagaSarholl sediments compared with
selected proxies from Greenland ice core records, a) Organic matter (OM), b) total
carbon (%TC), c) total nitrogen (%TN), d) C/N ratio, e) nitrogen isotope ratio (5 15 N), f)

Fig. 9. Principal component analysis of organic and physical properties and pollen
from the KagaSarholl sediments.

represent an increase in terrestrial input into the lake as woodland
encroached upon the area. This is supported by an increase in
Betula pubescens macrofossils in the sediment during this period
(Fig. 6). This stage of woodland development is associated with a
period of warm climate in northern Iceland (e.g. Caseldine et al.,
2006; Langdon et al., 2010; Eddudottir et al., 2015).

5.3. Cooling climate (c. 8800-8100 cal.yr BP)

The progression from Juniperus communis scrub to Betula
pubescens woodland was stalled by external factors c. 8800 cal. yr
BP, prompting a change towards more open landscape for c. 700
years. A change in the pollen assemblage is interpreted as a change
in vegetation community c. 8800 cal. yr BP (Fig. 7), characterised by
an increase in Poaceae and Cyperaceae and a further decrease in
Juniperus communis pollen (Fig. 3). The Betula PAR drops below 600
grains cm -2 yr -1 (Fig. 4) and the relative abundance of Betula pollen
decreases to <45% of TLP c. 8800 cal.yr BP and <32% after c. 8700
cal.yr BP (Fig. 3). The drop in Betula pollen indicates a decrease in
pollen deposition from Betula species. Despite a decrease in Betula
pollen accumulation, mean Betula pollen diameters remain
~20 ± 1 pm (Fig. 5), suggesting that Betula pubescens remained a
part of the vegetation community and was not replaced by Betula
nana. The increase in grass and sedge pollen and decline in Betula
pollen suggests a more open landscape around the lake as wood¬
land retreated (Figs. 3 and 7). These changes in the pollen assem¬
blage are also recorded in a previous (lower resolution)
palynological reconstruction from the same lake (Vasari, 1972;
Vasari and Vasari, 1990).

The low concentration and PAR of terrestrial pollen (Figs. 4 and
8) during this period suggests a decrease in pollen production of
most terrestrial plants. This is reflected in an increase in the relative
abundance of Poaceae and Cyperaceae pollen (Fig. 3), while PARs of
the two pollen taxa do not show significant increases (Fig. 4). The

carbon isotope ratio (5 13 C), g) Betula PAR, h) total land pollen (TLP) concentration, i)
Magnetic susceptibility (MS) showing values below 40 for clarity, j) dry bulk density
(DBD) showing values below 0.4 gem -3 for clarity, k) Volcanic forcing based on the
GISP2 sulfate record (Kobashi et al., 2017), 1) GISP2 I< + , showing values <2.5 ppb for
clarity, black line represents 50 year mean values (Mayewski et al., 1997), and m)
Renland ice-core 5 ls O record, 20 year average, uplift corrected (Vinther et al, 2009).

S.D. Eddudottir et al. / Quaternary Science Reviews 197 (2018) 246-256

253

changes in the pollen assemblage may have been caused by lower
spring and summer temperatures hindering the reproduction of
Betula pubescens (Pichugina, 1971) and possibly other taxa. Colder
conditions may have caused a shift towards a period of vegetative
reproduction by plants. For instance, following damage or stress,
Betula pubescens can reproduce by sprouting from basal buds or
regenerate from stems (Kauppi et al, 1987; Verwijst, 1988). Prior to
c. 8800 cal.yr BP, most of the plant macrofossils found in the
sediment are Betula pubescens catkin scales and fruits, however
between c. 8800 and 8100 cal.yr BP very few terrestrial plant
macrofossils are recorded (Fig. 6). This, along with the decrease in
pollen deposition, may indicate a drop in seed production by
terrestrial plants. Fligher 5 13 C values are recorded c. 8700 cal. yr BP,
with the highest values between c. 8400 and 8100 cal. yr BP (-18.9
to —20.5%o). The C/N ratio increases slightly during the same period
from 14.4 to 15.1 (Fig. 8). This coincides with the lowest input of
land pollen (TLP) into the lake (Fig. 4), and an increase in macro¬
fossils of the aquatic taxon Myriophyllum alterniflorum (Fig. 6). The
increase in aquatic plant macrofossils and decreased deposition of
terrestrial plant pollen and macrofossils in the sediment would lead
to higher 5 13 C values in the sediment closer to the 5 13 C of aquatic
plants (-11.5 to -15.1%o), than terrestrial plants (-24,4 to -30,9%o;
Wang and Wooller, 2006).

Environmental stability at Kagadarholl increased from c. 8800 to
8100 cal. yr BP, despite a response by the vegetation community
suggesting colder climate. Minerogenic input into the lake
decreased during the period, indicated by declining DBD (<0.14
gem -3 ) and MS values, with the exception of a tephra layer
deposited c. 8550 cal.yr BP (Fig. 8). The data show a progression
from more minerogenic to more organic sediment with time,
without a regression to more unstable conditions (Fig. 9). This is
contrary to several other studies in Iceland that have found evi¬
dence of increased soil erosion in the wake of cooling climate c.
8700 cal.yr BP at highland sites (Larsen et al., 2012; Geirsdottir
et al., 2013; Gunnarsson, 2016; Harning et al., 2018). The tall
vegetation cover around the Kagadarholl lake and the position of
the lake in the lowlands may be the reason for the different
response. This may indicate that environmental instability
increased in the highlands, while stable conditions prevailed in the
lowlands during this period.

Indications of colder spring/summer temperatures and
possible glacier advances, corresponding with the changes in
vegetation at Kagadarholl, are recorded in Lake Hvitarvatn in the
western highlands of Iceland beginning c. 8700 cal.yr BP, with
two rapid cooling episodes at c. 8500 and 8200 cal.yr BP (Larsen
et al., 2012). Chironomid-inferred July temperatures (July CI-T)
from nearby Trollaskagi peninsula show a gradual cooling be¬
tween c. 9000 and 8100 cal. yr BP, with a drop in July CI-T c. 8500
cal.yr BP. However, cooling is not evident in the pollen record
from the sites in Trollaskagi. On the contrary, the relative abun¬
dance of Betula increases after c. 8400 cal. yr BP (Caseldine et al.,
2006). Changes in pollen assemblages similar to those at
Kagadarholl are recorded at Vatnskotsvatn in Skagafjordur, east of
Hunafloi between c. 8300 and 8000 cal. yr BP, where development
of birch woodland was interrupted by reoccurrence of Juniperus-
Salix-Betula nana heath (Hallsdottir, 1995). Records that show a
response to long-term cooling c. 8800-7900 cal.yr BP are
concentrated in northwestern and western Iceland, although it
should be noted that most available palaeoecolgical and envi¬
ronmental records covering the period are from northern Iceland.
A short, sharp cooling with a change to cooler winter and spring
temperatures is observed in the only record available from eastern
Iceland c. 8200-8000 cal.yr BP (Striberger et al., 2012), which
shows more similarities to responses observed in northern Europe
(Rohling and Palike, 2005; Seppa et al., 2007, 2009) than those in

northwestern and western Iceland. Marine records northwest and
west of Iceland show a long period of cooling, beginning c.
8750-8600 cal. yr BP, and intensifying c. 8200 cal. yr BP (Knudsen
et al., 2004; Olafsdottir et al., 2010). A shorter cold period,
accompanied by an increase in sea-ice c. 8300-8100 cal.yr BP,
indicating decreased strength of the Irminger current, is recorded
north of Iceland (Ran et al., 2006). Other marine records around
Iceland show a relatively minor cooling associated with the 8.2 ka
event (e.g. Andersen et al., 2004; Castaneda et al., 2004; Ran et al.,
2006; Jiang et al., 2015).

5.4. Woodland recovery (c. 8100-7000 cal.yr BP)

Betula becomes the dominant pollen taxon in the terrestrial
pollen assemblage after c. 8100 cal.yr BP when Betula pubescens
woodland begins to recover (Figs. 3, 4 and 7). A sharp increase in
Betula pollen is recorded after c. 8100 cal. yr BP, as the PAR increases
from -200 to >2000 grains cm -2 yr -1 between c. 8100-8000 cal. yr
BP. This change is confirmed by a relative increase of Betula pollen
from -30 to 80% of TLP, showing relative changes in pollen taxa
abundance, independent of changes in sedimentation rates. Rela¬
tive abundances and PARs of other pollen taxa decrease, including
Juniperus communis , Salix, Poaceae, Cyperaceae and Thalictrum
alpinum (Figs. 3 and 4). The change in pollen assemblage occurs
within less than a century, indicating a rapid response of the
vegetation community to warming climate. A relatively stable
woodland community is reflected in the pollen assemblage until
the end of the record c. 7000 cal. yr BP (Figs. 3, 4 and 7). The most
stable period in the record is characterised by an increase in %OM to
>37% and %TC to >19%. The 5 13 C decreases to < —21 %o and the C/N
ratio is 14.6-16.2, probably indicating a greater input of organic
terrestrial material into the lake (Figs. 8 and 9). This may be in part
due to the reappearance of Betula pubescens catkin scales and fruits
in the sediment (Fig. 6). This period corresponds to the Holocene
Thermal Maximum (HTM), the most stable period of the Holocene
in Iceland, lasting until c. 6000-5500 cal. yr BP (Caseldine et al.,
2006; Larsen et al., 2012; Striberger et al., 2012; Geirsdottir et al.,
2013; Blair et al., 2015; Eddudottir et al., 2015, 2016; Eddudottir,
2016).

5.5. Climate anomalies c. 8800-8000 cal yr BP

The changes to colder springs/summers documented in the
palaeoecological reconstruction from Kagadarholl occur c.
8800±110 cal.yr BP (Figs. 3, 4 and 7). Responses to cooling are
recorded in other lakes in northwestern and western Iceland c.
8700cal.yr BP (e.g. Larsen et al., 2012; Geirsdottir et al., 2013;
Harning et al., 2018). The responses manifested in proxies sensitive
to spring and summer temperatures in Iceland, such as the Betula
PAR at Kagadarholl (Fig. 8), are similar to proxies that record global
climate anomalies c. 8600 and 8000 cal. yr BP (e.g. Rohling and
Palike, 2005; Wanner et al., 2011). A cold temperature anomaly is
observed in some records from the Greenland Ice Sheet beginning a
few centuries before the more severe 8.2 ka event (Fig. 8; Vinther
et al., 2009; Kobashi et al., 2017). The period is characterised by a
complex pattern of climate change in the North Atlantic region and
beyond when remnants of large ice sheets were still present in the
Northern Hemisphere (Mayewski et al., 2004). Rohling and Palike
(2005) identified a longer period of climate anomalies recorded
in summer proxies from c. 8600 cal. yr BP, distinct from a shorter
event affecting winter and early spring temperatures beginning c.
8300 cal. yr BP. The shorter event is recorded in pollen based
temperature reconstructions in Northern Europe and indicates
colder winters and early springs between c. 8300 and 8000 cal. yr
BP in response to freshwater input into the North Atlantic.

254

S.D. Eddudottir et al. / Quaternary Science Reviews 197 (2018) 246-256

However, pollen records from northern Fennoscandia show little or
no evidence of colder conditions in May and June (Seppa et al.,
2007, 2009).

Atmospheric circulation changes are observed in records c.
8650-8000 cal. yr BP, such as expansion of the polar vortex,
suggested by records of potassium ions (Fig. 8) and dust in the
GISP2 core from Greenland (Taylor et al., 1992, 1993; Mayewski
et al., 1997), weakening of the African, Indian and Asian mon¬
soons between c. 8500 and 8000 cal.yr BP (Gasse, 2000; Hong
et al., 2003; Yuan et al., 2004), and migration of the Intertrop-
ical Convergence Zone (ITCZ) (Haug et al., 2001). Several drivers
have been proposed for cooling climate during this period; i.e.
changes in solar output, meltwater discharge from the Laurentide
Ice Sheet and volcanism. Although a solar output minima is
recorded c. 8400 cal. yr BP (Steinhilber et al., 2009) it is not
considered enough to explain the observed changes in climate
(Wanner et al., 2011). Episodic cooling of the Irminger Current
occurred c. 9500—8100 cal. yr BP due to fluxes of freshwater from
the Laurentide Ice Sheet into the North Atlantic (Jennings et al.,

2011) . Three pulses of freshwater from the Laurentide Ice Sheet
are recorded, peaking c. 8636, 8565 and 8147 cal.yr BP (Jennings
et al., 2015). This coincides with three events of relative sea level
increase recorded in Scotland between c. 8760 to 8218 cal. yr BP,
each lasting 105-130 years (Lawrence et al., 2016). The first two
events are linked to the opening of the Tyrrell Sea and the last
one with the drainage of lakes Agassiz and Ojibway. All events
are part of a series of freshwater events originating from Hudson
Bay and Hudson Strait between c. 11,500 and 8000 cal.yr BP
(Jennings et al., 2015). The period between c. 9000-8000 cal.yr
BP is also characterised by large volcanic eruptions, with the
highest occurrence of volcanic activity during the Holocene
recorded in the Greenland Ice Sheet c. 8600-8000 cal.yr BP
(Fig. 8; Kobashi et al., 2017). There is evidence of intensified
volcanic activity in the Kamchatka Peninsula during this period
(Braitseva et al., 1995, 1997), including the Karymsky caldera
eruption c. 8770±210 cal.yr BP (7889±67 14 C yr) and Kuril
Lake-Iliinsky eruption c. 8460±60 cal.yr BP (7666±19 14 C yr;
Braitseva et al., 1997), an extremely large eruption that produced
-120-140 km 3 of pyroclastic material (Braitseva et al., 1997).
Furthermore the largest Holocene lava on Earth, the hjorsarhraun
lava was formed c. 8590±170 cal.yr BP (7800±60 14 C yr;
Hjartarson, 1988). The minimum size of the eruptive material
produced is estimated to be -25 km 3 DRE (dense rock equivalent)
(Hjartarson, 2011). The recent, much smaller (1.2 km 3 DRE;
Bonny et al., 2018), Holuhraun eruption originating from the
same volcanic system, has demonstrated that high amounts of
S0 2 and aerosol particulate matter are emitted from effusive
eruptions from the system (Ilyinskaya et al., 2017). Model sim¬
ulations of volcanic forcing have suggested that a large volcanic
eruption can cause climate cooling and sea-ice expansion that
can last for over a century (Kobashi et al., 2017; Miller et al.,

2012) , resulting in prolonged periods of cold summers (Miller
et al., 2012). A succession of large eruptions can therefore
potentially cause centuries-long cold periods (Kobashi et al.,
2017), such as the one observed in the Kagadarholl record.

The pollen record from Kagadarholl suggests relatively stable
conditions at the site between c. 8800 and 8100 cal. yr BP, despite
a change in vegetation community. The early response of the
palaeoenvironmental proxies in northwestern and western Ice¬
land to colder spring and summer temperatures is probably due
to northwestern Iceland's marginal location and the region's
sensitivity to ocean currents, showing more similarities to re¬
sponses in some marine records (Knudsen et al., 2004;
Olafsdottir et al., 2010) and Greenland ice core records (Fig. 8;
Vinther et al., 2009; Kobashi et al., 2017) than terrestrial records

from Northern Europe (Seppa et al., 2007, 2009). This demon¬
strates the ability of Icelandic palaeoenvironmental re¬
constructions to capture signals of North Atlantic climate
variability.

6. Conclusions

The pollen record from Kagadarholl clearly demonstrates a
change in vegetation community beginning c. 8800 cal.yr BP.
During the period c. 8800-8100 cal.yr BP, cold climate probably
interrupted the vegetation succession from Juniperus communis
scrub to Betula pubescens woodland that had taken place during a
relatively warm early Holocene c. 10.100—8800 cal. yr BP. The re¬
cord shows a significant decrease in pollen production from
terrestrial plants, especially Betula pubescens c. 8800-8100 cal. yr
BP, probably reflecting colder spring and/or summer temperatures.
Despite a colder climate, proxies of organic and physical properties
of the sediment do not indicate increased instability, i.e. soil erosion
in the area around the lake, showing the stability of the established
Betula pubescens woodland and the resilience of the more open
vegetation community in response to colder climate. The birch
woodland shows quick recovery with warming climate between c.
8100 and 8000 cal. yr BP. The results show the potential of
terrestrial multi-proxy records from Iceland for reconstructions of
climate variability in the North Atlantic.

Acknowledgements

The authors would like to thank Olafur Eggertsson, Fri3t>6r Sofus
Sigurmundsson and Porsteinn Jonsson for assistance in the field.
Jessica Lynn Till is thanked for proofreading the manuscript. We
would like to thank Olga Kolbrun Vilmundardottir and Scott John
Riddell for discussions on vegetation communities. We would like
to thank two anonymous reviewers for their valuable comments
and suggestions. The Blonduvirkjun hydropower plant kindly
hosted us during fieldwork. The research was funded by the
Landsvirkjun Energy Research Fund, the University of Iceland
Eimskip Fund, the University of Iceland Research Fund, and the
Icelandic Research Fund (grant no. 141842-051).

Appendix A. Supplementary data

Supplementary data related to this article can be found at
https://doi.Org/10.1016/j.quascirev.2018.07.017.

References

Aaby, B., Berglund, B.E., 1986. Characterization of Peat and lake Deposits. Handbook
of Holocene palaeoecology and palaeohydrologyjohn Wiley & Sons, Chichester,
pp. 231—246.

Alley, R.B., Agustsdottir, A.M., 2005. The 8k event: cause and consequences of a
major Holocene abrupt climate change. Quat. Sci. Rev. 24,1123—1149. https://
doi.org/10.1016/j.quascirev.2004.12.004.

Alley, R.B., Mayewski, P.A., Sowers, T., Stuiver, M., Taylor, K.C., Clark, P.U., 1997.
Holocene climatic instability: a prominent, widespread event 8200 yr ago.
Geology 25, 483-486. https://doi.org/10.1130/0091-7613(1997)025<0483:
hciapw>2.3.co;2.

Andersen, C., Ko^, N., Moros, M., 2004. A highly unstable Holocene climate in the
subpolar North Atlantic: evidence from diatoms. Quat. Sci. Rev. 23, 2155—2166.
Baldini, J.U., McDermott, F., Fairchild, I.J., 2002. Structure of the 8200-year cold
event revealed by a speleothem trace element record. Science 296, 2203—2206.
Bamberg, A., Rosenthal, Y., Paul, A., Heslop, D., Mulitza, S., Riihlemann, C., Schulz, M.,
2010. Reduced North Atlantic central water formation in response to early
Holocene ice-sheet melting. Geophys. Res. Lett. 37.

Barber, D.C., Dyke, A., Hillaire-Marcel, C., Jennings, A.E., Andrews, J.T., Kerwin, M.W.,
Bilodeau, G., McNeely, R., Southon, J., Morehead, M.D., 1999. Forcing of the cold
event of 8,200 years ago by catastrophic drainage of Laurentide lakes. Nature
400, 344-348.

Bengtsson, L., Enell, M., 1986. Chemical analysis. In: Berglund, B.E. (Ed.), Handbook
of Holocene Palaeoecology and Palaeohydrology. John Wiley & Sons, Chichester,

S.D. Eddudottir et al. / Quaternary Science Reviews 197 (2018) 246-256

255

pp. 423-451.

Bennett, K.D., 2016. Catalogue of Pollen Types, http://www.chrono.qub.ac.uk/
pollen/pc-intro.html.

Birks, H.J.B., 1973. Modern pollen studies in some arctic and alpine environments.
In: Birks, H.J.B., West, R.G. (Eds.), Quaternary Plant Ecology. Blackwell, Oxford,
pp. 143-168.

Birks, H.H., Bjune, A.E., 2010. Can we detect a west Norwegian tree line from
modern samples of plant remains and pollen? Results from the DOORMAT
project. Veg. Hist. Archaeobotany 19, 325-340.

Bjorck, S., Persson, T., Kristersson, I., 1978. Comparison of two concentration
methods for pollen in minerogenic sediments. GFF 100,107—111.

Blaauw, M., 2010. Methods and code for ‘classical’ age-modelling of radiocarbon
sequences. Quat. Geochronol. 5, 512-518. https://doi.Org/10.1016/j.quageo.2010.
01 . 002 .

Blair, C.L., Geirsdottir, A., Miller, G.H., 2015. A high-resolution multi-proxy lake re¬
cord of Holocene environmental change in southern Iceland. J. Quat. Sci. 30,
281-292.

Bonny, E., Thordarson, T., Wright, R., Hoskuldsson, A., Jonsdottir, I., 2018. The vol¬
ume of lava erupted during the 2014 to 2015 eruption at Holuhraun, Iceland: a
comparison between satellite- and ground-based measurements. J. Geophys.
Res.: Solid Earth. https://doi.org/10.1029/2017JB015008.

Brady, N.C., Weil, R.R., 1996. Elements of the Nature and Properties of Soils, second
ed. Pearson Prentice-Hall, Upper Saddle River.

Braitseva, O.A., Melekestsev, I.V., Ponomareva, V.V., Sulerzhitsky, L., 1995. Ages of
calderas, large explosive craters and active volcanoes in the Kuril-Kamchatka
region, Russia. Bull. Volcanol. 57, 383-402.

Braitseva, O.A., Ponomareva, V.V., Sulerzhitsky, L.D., Melekestsev, I.V., Bailey, J., 1997.
Holocene key-marker tephra layers in Kamchatka, Russia. Quat. Res. 47,
125-139.

Caseldine, C., Geirsdottir, A., Langdon, P., 2003. Efstadalsvatn—a multi-proxy study
of a Holocene lacustrine sequence from NW Iceland. J. Paleolimnol. 30, 55—73.

Caseldine, C., Langdon, P., Holmes, N., 2006. Early Holocene climate variability and
the timing and extent of the Holocene thermal maximum (HTM) in northern
Iceland. Quat. Sci. Rev. 25, 2314-2331.

Castaneda, I.S., Smith, L.M., Kristjansdottir, G.B., Andrews, J.T., 2004. Temporal
changes in Holocene 5180 records from the northwest and central North Ice¬
land Shelf. J. Quat. Sci. 19, 321-334.

Dearing, J., 1994. Environmental Magnetic Susceptibility: Using the Bartington MS2
System. Chi Publishing, Kenilworth.

Eddudottir, S.D., 2016. Holocene Environmental Change in Northwest Iceland. PhD
dissertation. University of Iceland.

Eddudottir, S.D., Erlendsson, E., Gisladottir, G., 2015. Life on the periphery is tough:
vegetation in Northwest Iceland and its responses to early-Holocene warmth
and later climate fluctuations. Holocene 25,1437—1453.

Eddudottir, S.D., Erlendsson, E., Gisladottir, G., 2017. Effects of the Hekla 4 tephra on
vegetation in northwest Iceland. Veg. Hist. Archaeobotany 26 (4), 389—402.
https://doi.org/10.1007/s00334-017-0603-5.

Eddudottir, S.D., Erlendsson, E., Tinganelli, L., Gisladottir, G., 2016. Climate change
and human impact in a sensitive ecosystem: the Holocene environment of the
Northwest Icelandic highland margin. Boreas 45, 715—728.

Erlendsson, E., 2007. Environmental Change Around the Time of the Norse Settle¬
ment of Iceland. PhD dissertation. University of Aberdeen.

Faegri, K., Kaland, P.E., Krzywinski, K., 1989. Textbook of Pollen Analysis, fourth ed.
John Wiley & Sons, New York.

Gasse, F., 2000. Hydrological changes in the african tropics since the last glacial
maximum. Quat. Sci. Rev. 19,189-211.

Geirsdottir, A., Miller, G.H., Axford, Y., Olafsdottir, S., 2009. Holocene and latest
Pleistocene climate and glacier fluctuations in Iceland. Quat. Sci. Rev. 28,
2107-2118.

Geirsdottir, A., Miller, G.H., Larsen, D.J., Olafsdottir, S., 2013. Abrupt Holocene
climate transitions in the northern North Atlantic region recorded by syn¬
chronized lacustrine records in Iceland. Quat. Sci. Rev. 70, 48—62. https://doi.
org/10.1016/j.quascirev.2013.03.010.

Grimm, E.C., 2011. TILIA 1.7.16. Illinois state museum, Springfield.

Gudmundsdottir, E.R., Larsen, G., Eiriksson, J., 2011. Two new Icelandic tephra
markers: the Hekla 6 tephra layer, 6060 cal. yr BP, and Hekla DH tephra layer,
-6650 cal. yr BP. Land-sea correlation of mid-Holocene tephra markers. Holo¬
cene 21, 629-639.

Gunnarsson, S., 2016. Holocene Climate and Landscape Evolution in the West
Central Highlands, Iceland. MSc dissertation. University of Iceland.

Hallsdottir, M., 1987. Pollen Analytical Studies of Human Influence on Vegetation in
Relation to the Landnam Tephra Layer in Southwest Iceland. PhD Thesis. Lund
University.

Hallsdottir, M., 1995. On the pre-settlement history of Icelandic vegetation. Buvi-
sindi 9,19-29.

Hallsdottir, M., Caseldine, C.J., 2005. The Holocene vegetation history of Iceland,
state-of-the-art and future research. In: Caseldine, C., Russel, A., Hardardottir, J.,
Knudsen, 6 (Eds.), Iceland — Modern Processes and Past Environments. Elsevier,
Amsterdam, pp. 319-334.

Hansen, B., Osterhus, S., 2000. North Atlantic—Nordic seas exchanges. Prog. Oce-
anogr. 45,109-208.

Harning, D.J., Geirsdottir, A., Miller, G.H., 2018. Punctuated Holocene climate of
Vestfirdir, Iceland, linked to internal/external variables and oceanographic
conditions. Quat. Sci. Rev. 189, 31-42. https://doi.Org/10.1016/j.quascirev.2018.
04.009.

Haug, G.H., Hughen, K.A., Sigman, D.M., Peterson, L.C., Rohl, U., 2001. Southward
migration of the intertropical convergence zone through the Holocene. Science
293,1304—1308.

Hjartarson, A., 1988. jDjorsarhraunid mikla, staersta nutimahraun jardar.
Natturufraedingurinn 58,1—16.

Hjartarson, A., 2011. Vidattumesta hraun Islands. Natturufraedingurinn 81, 37—49.

Hong, Y.T., Hong, B., Lin, Q.H., Zhu, Y.X., Shibata, Y., Hirota, M., Uchida, M., Leng, X.T.,
Jiang, H.B., Xu, H., 2003. Correlation between indian ocean summer monsoon
and North Atlantic climate during the Holocene. Earth Planet Sci. Lett. 211,
371-380.

Huntley, B., Birks, H.J.B., 1983. An Atlas of Past and Present Pollen Maps for Europe:
0—13000 Years Ago. Cambridge University Press, Cambridge.

Ilyinskaya, E., Schmidt, A., Mather, T.A., Pope, F.D., Witham, C., Baxter, P.,
Johannsson, T., Pfeffer, M., Barsotti, S., Singh, A., Sanderson, P., Bergsson, B.,
McCormick Kilbride, B., Donovan, A., Peters, N., Oppenheimer, C., Edmonds, M.,
2017. Understanding the environmental impacts of large fissure eruptions:
aerosol and gas emissions from the 2014—2015 Holuhraun eruption (Iceland).
Earth Planet Sci. Lett. 472, 309-322.

Jennings, A., Andrews, J., Wilson, L., 2011. Holocene environmental evolution of the
se Greenland shelf north and south of the Denmark strait: irminger and east
Greenland current interactions. Quat. Sci. Rev. 30, 980—998.

Jennings, A., Andrews, J., Pearce, C., Wilson, L., Olafsdottir, S., 2015. Detrital car¬
bonate peaks on the Labrador shelf, a 13—7ka template for freshwater forcing
from the Hudson Strait outlet of the Laurentide Ice Sheet into the subpolar gyre.
Quat. Sci. Rev. 107, 62-80.

Jiang, H., Muscheler, R., Bjorck, S., Seidenkrantz, M.-S., Olsen, J., Sha, L., Sjolte, J.,
Eiriksson, J., Ran, L., Knudsen, K.-L., 2015. Solar forcing of Holocene summer sea-
surface temperatures in the northern North Atlantic. Geology 43, 203—206.

Johannesson, H., 2014. Jardfraedikort af Islandi 1:600 000 Berggrunnur (Geological
Map of Iceland 1:600 000 Bedrock Geology). Natturufrsdistofnun Islands,
Gardabaer.

Kardjilov, M., Gisladottir, G., Gislason, S., 2006. Land degradation in northeastern
Iceland: present and past carbon fluxes. Land Degrad. Dev. 17, 401^417.

Karlsdottir, L., 2014. Hybridisation of Icelandic Birch in the Holocene Reflected in
Pollen. PhD dissertation. University of Iceland.

Karlsdottir, L., Hallsdottir, M., Thorsson, A.T., Anamthawat-Jonsson, K., 2008. Char¬
acteristics of pollen from natural triploid Betula hybrids. Grana 47, 52—59.

Karlsdottir, L., Hallsdottir, M., Thorsson, 7E.T., Anamthawat-Jonsson, K., 2009. Evi¬
dence of hybridisation between Betula pubescens and B. nana in Iceland during
the early Holocene. Rev. Palaeobot. Palynol. 156, 350-357.

Karlsdottir, L., Hallsdottir, M., Thorsson, /E.T., Anamthawat-Jonsson, K., 2012. Early
Holocene hybridisation between Betula pubescens and B. nana in relation to
birch vegetation in Southwest Iceland. Rev. Palaeobot. Palynol. 181,1—10.

Kauppi, A., Rinne, P., Ferm, A., 1987. Initiation, structure and sprouting of dormant
basal buds in Betula pubescens. Flora 179.

Knudsen, K., Jiang, H., Jansen, E., Eiriksson, J., Heinemeier, J., Seidenkrantz, M.-S.,
2004. Environmental changes off North Iceland during the deglaciation and the
Holocene: foraminifera, diatoms and stable isotopes. Mar. Micropaleontol. 50,
273-305.

Kobashi, T., Menviel, L., Jeltsch-Thommes, A., Vinther, B.M., Muscheler, R., Nakae-
gawa Pfister, P.L., Doring, M., Leuenberger, M., Wanner, H., 2017. Volcanic in¬
fluence on centennial to millennial Holocene Greenland temperature change.
Sci. Rep. 7,1441.

Langdon, P.G., Leng, M., Holmes, N., Caseldine, C.J., 2010. Lacustrine evidence of
early-Holocene environmental change in northern Iceland: a multiproxy
palaeoecology and stable isotope study. Holocene 20, 205—214.

Larsen, D.J., Miller, G.H., Geirsdottir, A., Olafsdottir, S., 2012. Non-linear Holocene
climate evolution in the North Atlantic: a high-resolution, multi-proxy record of
glacier activity and environmental change from Hvitarvatn, central Iceland.
Quat. Sci. Rev. 39,14-25.

Larsen, D.J., Miller, G.H., Geirsdottir, A., Thordarson, T., 2011. A 3000-year varved
record of glacier activity and climate change from the proglacial lake Hvitar¬
vatn, Iceland. Quat. Sci. Rev. 30, 2715-2731.

Lawrence, T., Long, A.J., Gehrels, W.R., Jackson, L.P., Smith, D.E., 2016. Relative sea-
level data from southwest Scotland constrain meltwater-driven sea-level
jumps prior to the 8.2 kyr BP event. Quat. Sci. Rev. 151, 292-308.

Makela, E.M., 1996. Size distinctions between Betula pollen types—a review. Grana
35, 248-256.

Mayewski, P.A., Meeker, L.D., Twickler, M.S., Whitlow, S., Yang, Q., Lyons, W.B.,
Prentice, M., 1997. Major features and forcing of high-latitude northern hemi¬
sphere atmospheric circulation using a 110,000-year-long glaciochemical series.
J. Geophys. Res.: Oceans 102, 26345-26366.

Mayewski, P.A., Rohling, E.E., Stager, J.C., Karlen, W., Maasch, K.A., Meeker, L.D.,
Meyerson, E.A., Gasse, F., van Kreveld, S., Holmgren, K., 2004. Holocene climate
variability. Quat. Res. 62, 243-255.

Meyers, P.A., Lallier-Verges, E., 1999. Lacustrine sedimentary organic matter records
of Late Quaternary paleoclimates. J. Paleolimnol. 21, 345—372.

Miller, G.H., Geirsdottir, A., Zhong, Y., Larsen, D.J., Otto-Bliesner, B.L., Holland, M.M.,
Bailey, D.A., Refsnider, K.A., Lehman, S.J., Southon, J.R., 2012. Abrupt onset of the
Little Ice Age triggered by volcanism and sustained by sea-ice/ocean feedbacks.
Geophys. Res. Lett. 39.

Moore, P.D., Webb, J.A., Collison, M.E., 1991. Pollen Analysis. Blackwell scientific
publications, Oxford.

Nakagawa, T., Brugiapaglia, E., Digerfeldt, G., Reille, M., Beaulieu, J.-L.D., Yasuda, Y.,
1998. Dense-media separation as a more efficient pollen extraction method for

256

S.D. Eddudottir et al. / Quaternary Science Reviews 197 (2018) 246-256

use with organic sediment/deposit samples: comparison with the conventional
method. Boreas 27,15-24. https://doi.Org/10.llll/j.1502-3885.1998.tb00864.x.

Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, RR., O'Hara, R.B.,
Simpson, G.L., Solymos, R, Stevens, M.H.H., Wagner, H., 2016. Vegan: Commu¬
nity Ecology Package, R Package Version 2.3-3 Edn.

Olafsdottir, S., Jennings, A.E., Geirsdottir, A., Andrews, J., Miller, G.H., 2010. Holocene
variability of the North Atlantic Irminger current on the south-and northwest
shelf of Iceland. Mar. Micropaleontol. 77,101-118.

Pichugina, N.P., 1971. Effect of ecological factors on seed formation in certain
interspecific birch hybrids. Ecologiya: Sov. J. Ecol. 6, 89—91.

Ran, L., Jiang, H., Knudsen, K.L., Eiriksson, J., Gu, Z., 2006. Diatom response to the
Holocene climatic optimum on the North Icelandic shelf. Mar. Micropaleontol.
60, 226-241.

Rasmussen, S.O., Andersen, K.K., Svensson, A.M., Steffensen, J.P., Vinther, B.M.,
Clausen, H.B., Siggaard-Andersen, M.L., Johnsen, S.J., Larsen, L.B., Dahl-
Jensen, D., Bigler, M., Rothlisberger, R., Fischer, H., Goto-Azuma, Hansson ME.,
Ruth, U., 2006. A new Greenland ice core chronology for the last glacial
termination. J. Geophys. Res.: Atmosphere 111. D06102.

Rohling, E.J., Palike, H., 2005. Centennial-scale climate cooling with a sudden cold
event around 8,200 years ago. Nature 434, 975.

Rundgren, M., 1998. Early-Holocene vegetation of northern Iceland: pollen and
plant macrofossil evidence from the Skagi peninsula. Holocene 8, 553—564.

Schomacker, A., Brynjolfsson, S., Andreassen, J.M., Gudmundsdottir, E.R., Olsen, J.,
Odgaard, B.V., Hakansson, L., Ingolfsson, 6., Larsen, N.K., 2016. The Drangajokull
ice cap, northwest Iceland, persisted into the early-mid Holocene. Quat. Sci. Rev.
148, 68-84.

Seppa, H., Birks, H.J.B., Giesecke, T., Hammarlund, D., Alenius, T., Antonsson, K.,
Bjune, A.E., Heikkila, M., MacDonald, G.M., Ojala, A.E.K., 2007. Spatial structure
of the 8200 cal yr BP event in northern Europe. Clim. Past Discuss 3,165-195.

Seppa, H., Bjune, A.E., Telford, R.J., Birks, H., Veski, S., 2009. Last nine-thousand years
of temperature variability in Northern Europe. Clim. Past 5, 523—535.

Stefansson, U., 1962. North Icelandic Waters, Reykjavik.

Steinhilber, F., Beer, J., Frohlich, C., 2009. Total solar irradiance during the Holocene.
Geophys. Res. Lett. 36.

Stockmarr, J., 1971. Tablets with spores used in absolute pollen analysis. Pollen

Spores 13, 614-621.

Striberger, J., Bjorck, S., Holmgren, S., Hamerlik, L., 2012. The sediments of Lake
Logurinn—A unique proxy record of Holocene glacial meltwater variability in
eastern Iceland. Quat. Sci. Rev. 38, 76-88.

Taylor, K., Alley, R., Fiacco, J., Grootes, P., Lamorey, G., Mayewski, P., Spencer, M.J.,
1992. Ice-core dating and chemistry by direct-current electrical conductivity.
J. Glaciol. 38, 325-332.

Taylor, K.C., Lamorey, G.W., Doyle, G.A., Alley, R.B., Grootes, P.M., Mayewski, P.A.,
White, J.W.C., Barlow, L.K., 1993. The ‘flickering switch’of late Pleistocene
climate change. Nature 361, 432.

Thomas, P.A., El-Barghathi, M., Polwart, A., 2007. Biological flora of the British isles:
Juniperus communis L. J. Ecol. 95, 1404-1440. https://doi.org/10.llll/j.1365-
2745.2007.01308.x.

Thorarinsson, S., 1971. The age of the light Hekla tephra layers according to cor¬
rected C14—datings. Natturufraedingurinn 41, 99—105.

Vasari, Y., 1972. The history of the vegetation in Iceland during the Holocene. In:
Vasari, Y., Hyvarinen, H., Hicks, S. (Eds.), Climatic Changes in Arctic Areas during
the Last Ten-thousand Years. University of Oulu, Oulu, pp. 239-252.

Vasari, Y., Vasari, A., 1990. L’histoire Holocene des lacs Islandais. In: Malaurie, J.
(Ed.), 102 temoignages en hommage a quarante ans d’etudes arctiques. Editions
Plon, Paris, pp. 279-293.

Verwijst, T., 1988. Environmental correlates of multiple-stem formation in Betula
pubescens ssp. Tortuosa. Vegetatio 76, 29-36. https://doi.org/10.1007/
BF00047385.

Vinther, B.M., Buchardt, S.L., Clausen, H.B., Dahl-Jensen, D., Johnsen, S.J., Fisher, D.A.,
Koerner, R.M., Raynaud, D., Lipenkov, V., Andersen, K.K., 2009. Holocene thin¬
ning of the Greenland ice sheet. Nature 461, 385-388.

Wang, Y., Wooller, M.J., 2006. The stable isotopic (C and N) composition of modern
plants and lichens from northern Iceland: with ecological and paleoenvir-
onmental implications. Jokull 56, 27—38.

Wanner, H., Solomina, O., Grosjean, M., Ritz, S.P., Jetel, M., 2011. Structure and origin
of Holocene cold events. Quat. Sci. Rev. 30, 3109-3123.

Yuan, D., Cheng, H., Edwards, R.L., Dykoski, C.A., Kelly, M.J., Zhang, M., Qing, J., Lin, Y.,
Wang, Y., Wu, J., 2004. Timing, duration, and transitions of the last interglacial
Asian monsoon. Science 304, 575—578.