University of Birmingham Holocene changes in marine productivity

University of Birmingham
Holocene changes in marine productivity and
terrestrial organic carbon inputs into an Icelandic
fjord: Application of molecular and bulk organic
proxies
Moossen, Heiko; Abell, R.; Quillmann, U.; Bendle, James
DOI:
10.1177/0959683613505346
License:
None: All rights reserved
Document Version
Publisher's PDF, also known as Version of record
Citation for published version (Harvard):
Moossen, H, Abell, R, Quillmann, U & Bendle, J 2013, 'Holocene changes in marine productivity and terrestrial
organic carbon inputs into an Icelandic fjord: Application of molecular and bulk organic proxies' The Holocene,
vol 23, no. 12, pp. 1699-1710. DOI: 10.1177/0959683613505346
Link to publication on Research at Birmingham portal
General rights
When referring to this publication, please cite the published version. Copyright and associated moral rights for publications accessible in the
public portal are retained by the authors and/or other copyright owners. It is a condition of accessing this publication that users abide by the
legal requirements associated with these rights.
• You may freely distribute the URL that is used to identify this publication.
• Users may download and print one copy of the publication from the public portal for the purpose of private study or non-commercial
research.
• If a Creative Commons licence is associated with this publication, please consult the terms and conditions cited therein.
• Unless otherwise stated, you may not further distribute the material nor use it for the purposes of commercial gain.
505346
13505346The HoloceneMoossen et al.
2013
HOL0010.1177/09596836
Research paper
Holocene changes in marine productivity
and terrestrial organic carbon inputs into
an Icelandic fjord: Application of
molecular and bulk organic proxies
The Holocene
0(0) 1­–12
© The Author(s) 2013
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0959683613505346
hol.sagepub.com
Heiko Moossen,1,2 Richard Abell,3 Ursula Quillmann4
and James Bendle2
Abstract
This study examines the dynamics of organic carbon contributions from different sources to the sediments of a ~39 m core from Ísafjarðardjúp Fjord,
Northwest Iceland, throughout the Holocene. Furthermore, it shows that the variability of terrestrial organic carbon (OCterr) and marine organic carbon
(OCmar) is linked to palaeoclimatic change throughout the Holocene. glycerol-dialkyl-glycerol-tetraether (GDGT), alkenone, n-alkane, total OC and total
nitrogen analyses were conducted on 326 samples to yield high-resolution branched versus isoprenoid tetraether index (BIT-index), n-alkane/alkenone
index and C/N ratio records from ~10,800 to ~300 cal. a BP. These records were used to estimate the OCterr and the OCmar contributions to the
sediments. Three different approaches of estimating the OCterr contribution yield different relative amounts, but similar long-term trends. These results
indicate that the combination of biomarker records is a good approach to reconstruct OCterr contributions but also highlight the strengths and weaknesses
of the individual biomarkers. The OCterr contribution to the total OC inventory continually increases throughout much of the Holocene but does not
rise above 30%. It seems to have been driven by changing climate rather than changing sedimentation rates, and during the late Holocene, anthropogenic
activity may have been an influence. The reconstructed OCmar contribution to the sediment was used to model changes in palaeoproductivity throughout
the Holocene. These changes were likely forced by changes in nutrients supplied both by the catchment area and the Irminger Current.
Keywords
branched versus isoprenoid tetraether index, C/N ratio, Holocene, Iceland, n-alkane/alkenone index, palaeoproductivity
Received 21 May 2013; revised manuscript accepted 24 July 2013
Introduction
Fjords play a major role in the global carbon cycle as storage reservoirs of organic carbon (OC), containing at least 12% of the
total OC buried in the continental margin over the past 100,000
years (Nuwer and Keil, 2005). The generally high sedimentation
rates prevalent in fjords are conducive to preserving OC (Smittenberg
et al., 2004). Additionally, the seasonal or permanent anoxic conditions prevalent in the water-columns of many fjords further contribute to the preservation of OC (Howe et al., 2010; Paetzel and
Schrader, 1992). These attributes make fjords prime locations to
study the carbon cycle but also for high-resolution palaeoreconstructions of terrestrial and marine climate change.
The deposited OC is made up of terrestrial organic carbon
(OCterr) and marine organic carbon (OCmar); however, there is a
general lack of data showing the relative abundance of OCterr and
OCmar in fjord basins (Skei, 1983). Researchers have only recently
started to elucidate the relative contributions of OCterr and OCmar
to fjordic sediments (Nuwer and Keil, 2005; Smittenberg et al.,
2004; Walsh et al., 2008). Variations in the relative abundance of
carbon from different carbon pools can offer insights into variable
climatic conditions in response to, for example, river discharge
(Weijers et al., 2009) and erosion driven by catchment precipitation (Smittenberg et al., 2004), or changes in oceanography and
sea level (Knies, 2005).
Beyond studying climate change, the relative contributions
of different carbon pools to fjordic sediments also offer the
opportunity to model changes in marine primary productivity
through geological time (Knies, 2005; Knies and Mann, 2002).
Global satellite observations and microcosm experiments show
that changing climate affects marine primary productivity
(Behrenfeld et al., 2006; Gao et al., 2012). Since primarily produced organic matter (OM) forms the bases of the marine food
chain and carbon (marine and terrestrial) sequestered in marginal basins can act as a sink for CO2 (Borges et al., 2005), it is
imperative to better understand climatic controls on palaeoproductivity (PP) over geological timescales.
Here, we show the changing OCterr and OCmar contributions to
the sediment of Ísafjarðardjúp Fjord. Furthermore, we use the
sedimentary OCmar content to model changes in Holocene PP.
Recent studies have highlighted the need for a multiproxy approach
to elucidate changing OCterr contributions to fjordic sediments
1
University of Glasgow, UK
University of Birmingham, UK
3
The Scottish Marine Institute, UK
4
University of Colorado, USA
2
Corresponding author:
Heiko Moossen, School of Geography, Earth and Environmental
Sciences, University of Birmingham, Birmingham B15 2TT, UK.
Email: [email protected]
2
The Holocene 0(0)
Figure 1. Core location and main surface water masses in the North Atlantic (modified after Hansen and Østerhus, 2000; source of North
Atlantic map: Schlitzer, 2010). Arrows indicate warm, saline MNAW, the NAW and the cold PWs. The CSC and the NWAC originate from the
NAW. The NAC, originating from the MNAW branches into the FC and the IC. Part of the IC flows through the Denmark Strait and forms the
NIIC on the North Icelandic Shelf, while another part turns southwards. The EIC originates from the PW. The EIC, a branch of the PW, splits off
north of Iceland and meets the NIIC. Inset:Vestfirdir Peninsula and Denmark Strait Bathymetry. The MD99-2266 core location is marked with a
black dot in the mouth of Ísafjarðardjúp Fjord (modified after Quillmann et al., 2010).
MNAW: Modified North Atlantic Water; NAW: North Atlantic Water; PW: polar water; CSC: continental slope current; NWAC: Norwegian Atlantic
Current; NAC: North Atlantic Current; FC: Faroe Current; IC: Irminger Current; NIIC: North Icelandic Irminger Current; EGC: East Greenland Current;
EIC: East Icelandic Current.
(Belicka and Harvey, 2009; Huguet et al., 2007; Walsh et al.,
2008). Therefore, we employ the n-alkane/alkenone-index (Marret
et al., 2001), the branched versus isoprenoid tetraether (BIT)index (Hopmans et al., 2004) and the C/N ratio (Meyers, 1997) in
three binary mixing models to reconstruct OCterr contributions
throughout the Holocene. The alkenones are derived from haptophyte algae and constitute the marine component of the n-alkane/
alkenone-index (Volkman et al., 1995, 1998), while the longchained, odd-carbon numbered n-alkanes are derived from terrestrial higher plants (Eglinton et al., 1962). The BIT-index compares
the relative amounts of predominantly terrestrial-derived branched
glycerol-dialkyl-glycerol-tetraether (GDGTs) with the relative
amount of crenarchaeol produced by marine Thaumarchaeota
(Hopmans et al., 2004). Finally, terrestrial and marine OM has distinct C/N ratios allowing for the differentiation between these two
sources (Lamb et al., 2006).
Methods
Site description and sampling
The 3890 cm piston core (MD99-2266) was retrieved on Leg III
of the 1999 IMAGES V cruise aboard the R/V Marion Dusfresne
from 106 m water depth from Ísafjarðardjúp fjord in Northwest
Iceland (66° 13′77″ N, 23° 15′93″ W; Figure 1; Quillmann et al.,
2010, and references therein).
The age model published by Quillmann et al. (2010) is used
to calculate sediment accumulation rates (Figure 2), assuming
linear sedimentary deposition between each 14C-accelerator mass
spectrometry (AMS)-dated horizon. Sample ages are given in
calibrated (kilo)ages before present (cal (k)a BP; present: ad
1950; Stuiver et al., 1998). The age of each sample was calculated using the sediment accumulation rate determined by the
14
C-AMS dates bracketing the section of sediment from which
the sample was taken. The ages of seven samples below the
youngest 14C-AMS dated sediment horizon were extrapolated,
assuming that the sedimentation rate between the core top and
the youngest 14C-AMS date is the same as that of the sediment
interval bracketed by the two youngest 14C-AMS dates.
Elemental analysis
The total organic carbon (TOC) and total organic nitrogen contents
of 156 samples were analysed following the method of Verardo
et al. (1990). A range of 0.1–1.0 mg acetanilide standards were
analysed to correct instrumental drift, and all standards and samples were blank corrected. The mean standard deviation from 17
triplicate analyses was ±3.87% for nitrogen and ± 2.47% for carbon. The mean standard deviation of the molar C/N ratio is ±0.14.
The C/N ratio shows changes in the nitrogen concentration,
while the N/C ratio reflects the amount of OC (Perdue and
Koprivnjak, 2007). Since the C/N ratio is more commonly used in
3
Moossen et al.
respectively. The GC columns and temperature programme were
identical to the GC-FID analysis. The ionisation energy was 70
eV, and the scan width was 50–800 mass units.
The n-alkanes and alkenones were quantified by comparing
their peak areas with the peak areas of the internal standard. The
mean analytical quantification errors (mean standard deviation)
of 10 n-alkane and 34 alkenone-containing fractions run in duplicate were 5.5% and 3.3%, respectively.
The n-alkane-/alkenone-index proposed by (Marret et al.,
2001) and modified by Weijers et al. (2009) was calculated using
the concentrations of the n-alkanes and alkenones (Figure 3).
n − alkane/alkenone-index =
odd-chained (C 25 − C35 )n-alkanes
odd-chained (C 25 − C35 )n-alkanes  + 3 ( C37 alkenones )
Figure 2. Age model of core MD99-2266. It is based on 19
14
C-AMS-dated sediment horizons and the depth horizon of
the Saksunarvatn tephra (dashed line), which is dated at 10,180
± 120 cal. a BP (Gronvold et al., 1995; Quillmann et al., 2010).
Sedimentation rates are calculated using the calibrated ages of the
dated horizons.
the scientific literature, we follow Weijers et al. (2009) in giving
the C/N ratio in the text, but using the N/C ratio to model the sedimentary OCterr contribution.
Sample preparation
Freeze-dried and powdered samples were extracted by ultrasonication using a dichloromethane/methanol mixture (3:1 v/v). An
internal standard consisting of squalane, 2-nonadecanone,
1-nonadecanol and erucic acid was added to each sample. An aliquot of each sample was separated into four fractions using silica
gel column chromatography after Bendle et al. (2007). Aliphatic
and alicyclic hydrocarbons were eluted using 4 mL of n-hexane.
Aromatic hydrocarbons were eluted using 2 mL of n-hexane/
dichloromethane (2:1 v/v). Ketones, n-alcohols and aldehydes
were eluted with 4 mL of dichloromethane, and acids, sterols,
tetraether-lipids and diether-lipids were eluted using 5 mL of
methanol/dichloromethane (95:5 v/v). The tetraether containing
fraction of each sample was re-dissolved in 200 µL of n-hexane/ipropanol (99:1 v/v) and filtered using a 0.45 µm polytetrafluoroethylene (PTFE) syringe filter prior to analysis.
(1)
The relative tetraether abundances in 299 MD99-2266 sediment samples were analysed using high-performance liquid chromatography–atmospheric pressure chemical ionisation–mass
spectrometry (HPLC-APCI-MS) at the Organic Geochemistry
Unit at the University of Bristol. The analyses were conducted on
a Thermo Scientific TSQ Quantum Access equipped with an
Acella Autosampler, Acella pump and Xcalibur software. The LC
was equipped with an Alltech Prevail Cyano column (150 mm ×
2.1 mm; film thickness: 3 µm). Two mobile phases, n-hexane (A)
and i-propanol (B) were used at a flow rate of 0.2 mL/min. Initially, 1% of B v/v was held for 7 min. Then the concentration of
B was increased on a linear gradient over 43 min to 1.6%. The
concentration of B was increased to 10% v/v at 51 min and held
for 2 min. Finally, the concentration of B was decreased to 1% v/v
and held for 10 min. Single ion mode (SIM) was used to monitor
the abundance of the [M+H]+ (molecular ion + proton) ion. The
peak areas were used to calculate the relative abundance of each
GDGT. The BIT-index was calculated using the equation published by Hopmans et al. (2004; Figure 3c). Roman numerals
indicate relative abundances and structures of GDGTs as described
by Hopmans et al. (2004; see also supplementary material S3,
Figure S2).
BIT = ( I + II + III ) ( I + II + III ) + ( IV ) 
(2)
The mean standard deviation of BIT-index values is ±0.01
determined by the analyses of nine samples in triplicate and two
in duplicate.
Biomarker analysis
Statistical analyses
The alkenones and n-alkanes in 326 samples were analysed
using a gas chromatograph (GC; Shimadzu 2010) with a flame
ionisation detector (FID). The carrier gas was hydrogen (constant pressure; 190 kPa). The separation of the different compounds was achieved using one of two identical columns, either
a BP1 (SGE Analytical Science) or a TG-1MS (Thermo Scientific) column (length: 60 m, diameter: 0.25 mm, film thickness:
0.25 µm, coating: 100% dimethyl-polysiloxane). The GC oven
was held at 60°C for 2 min, then the temperature was ramped
up to 120°C at 30°C/min and then to 350°C at 3°C/min, where
the temperature was held for 20 min. An injection standard consisting of methyl behenate was added to each sample prior to
analysis.
The alkenones and n-alkanes were identified by comparing the
retention time of the substances in the samples to the retention
time of standard substances and by using a Shimadzu OP2010Plus Mass Spectrometer (MS) interfaced with a Shimadzu 2010
GC. The carrier gas was helium (constant pressure: 230 kPa). The
ion source and interface temperatures were 200°C and 300°C,
Statistical analyses were performed using SigmaPlot 11.0 (Systat Software, Inc.). Data were smoothed using the Arand time
series software (Howell et al., 2006). The analytical error (standard deviation) of the measured variables has been propagated
through the equations where appropriate (see supplementary
material S1).
Mass accumulation rates
Mass accumulation rates (MARs) for TOC (mg cm/a) and individual biomarkers (ng cm/a) were calculated using Eq. (3)
(Rommerskirchen et al., 2003, and references therein), where X is
the TOC or specific biomarker concentration in milligrams or
nanograms per gram of dry sediment (mg/g Sed or ng/g Sed),
respectively, ρ is dry bulk density (DBD; g/cm3) and LSR is the
linear sedimentation rate (LSR; cm/a; Figure 4). The DBD was
obtained from the project collaborator Ursula Quillmann.
MAR = X * ρ * LSR
(3)
4
The Holocene 0(0)
Figure 3. Biomarker and bulk parameter variability of core MD99-2266. (a) TOC content, black triangles indicate the 14C-AMS-dated
sediment horizons, (b) C/N ratio, (c) BIT-index, (d) odd-chained (C25–C35) n-alkane/alkenone index, (e) concentration of odd-chained (C25–C35)
n-alkanes and (f) concentration of C37:2 and C37:3-alkenones. Grey dots and analytical error bars signify values of individual samples and the
black line indicates the moving average.
TOC: total organic carbon; AMS: accelerator mass spectrometry; BIT: branched versus isoprenoid tetraether.
Figure 4. Sedimentation and MARs of core MD99-2266. (a) Linear sedimentation rate of core MD99-2266, black triangles indicate the
14
C-AMS dated sediment horizons; (b) MAR of odd-chained (C25–C35) n-alkanes; (c) MAR of TOC; (d) MAR of the sum of C37:2- and C37:3alkenones.
TOC: total organic carbon; AMS: accelerator mass spectrometry; MAR: mass accumulation rate.
5
Moossen et al.
Figure 5. Estimates of the OCterr contribution to the sedimentary TOC pool using a binary mixing model approach. (a) OCterr estimate using
the n-alkane/alkenone index, black triangles indicate the 14C-dated sediment horizons; (b) OCterr estimate using the BIT-index; (c) the C/N
ratios, grey dots with propagated standard deviations represent individual sample values, the black line indicates the moving average and (d)
mean OCterr estimate (black line) and standard deviation of the mean (grey-shaded area). Only samples where the n-alkane/alkenone index, BITindex and C/N ratio data were available were used to compile the mean OCterr estimate (n = 137).
TOC: total organic carbon; OC: organic carbon; BIT: branched versus isoprenoid tetraether.
Modelling OCterr contributions
The molar C/N ratio, n-alkane/alkenone-index and the BIT-index
were employed in a binary mixing model (Eq. (4)) previously
used by Weijers et al. (2009) to assess the percentile contribution
of OCterr (fterr) to the TOC pool of Ísafjarðardjúp Fjord sediment
(Figure 5).
− X Mar
X
f terr = sample
⋅100%
(4)
X Terr − X Mar
Xsample, XMar and XTerr are the values of the sample, the marine
end member value and the terrestrial end member value of the
C/N ratio, the n-alkane/alkenone-index and the BIT-index,
respectively.
Marine sourced OC and modelling marine paleoproductivity
The sedimentary TOC comprises two carbon pools, OCterr and
OCmar. The OCmar content of the fjordic sediment is calculated
using Eq. (5).
(5)
OC mar = TOC − OC terr
marine OC content (%), DBD is dry bulk density (g/cm3), LSR is
the linear sedimentation rate (cm/ka) and D is the water-column
depth (m).




 OC mar ⋅ 0.378 ⋅ DBD ⋅ LSR ⋅ D0.63 
PP = 

1
 
 1 − 




1.5
   0.037 ⋅ LSR + 1   
0.71
(6)
Early Holocene sea-level change in the Ísafjarðardjúp Fjord as
a response to final deglaciation and the isostatic rebound of Iceland is considered in the PP estimates. Quillmann et al. (2010)
show that the relative sea level decreased by ~30 m between
~10,700 and ~8900 cal. a BP. Subsequently, relative sea level
increased again and reached its contemporary level at ~5700 cal.
a BP. Assuming a linear sea level decrease and subsequent
increase, the relative sea-level change was calculated using
today’s sea level as a reference point.
Results
Bulk parameters
The marine PP (gC·m2/a) was calculated using the model published by Knies and Mann (2002; Eq. (6)), where OCmar is the
The LSR varies from 0.09 to 4.09 cm/a (Figures 2 and 4a). The
highest LSR coincides with the Saksunarvatn tephra at 10,180 ±
6
120 cal. a BP (Gronvold et al., 1995; Quillmann et al., 2010). A
second peak in sedimentation rate at ~8250 cal. a BP is likely an
artefact caused by the close proximity of the adjacent 14C-AMS
dates, as there is no visual disturbance in sedimentation in that
interval (Quillmann, personal communication, 2013). The sedimentation rates are highest in the early Holocene and decrease
towards the late Holocene.
The TOC content of the sediment core varies between 0.7% and
2.2% throughout the Holocene (Figure 3a). The mean TOC content
is 1.2%. Over the first ~2400 years, the TOC content increases
before decreasing from ~8400 to ~8100 cal. a BP. The TOC content
fluctuates between 1% and 1.5% from ~8000 and ~3000 cal. a BP
with the exception of a TOC spike at ~4600 cal. a BP. Throughout
the last ~2000 years of the record, TOC values increase.
The MAR of TOC follows the sedimentation rate closely
(Figure 4b). The highest values are recorded during the early
Holocene with the highest value of 34 mg/cm2/a at ~10,100 cal.
a BP.
The molar C/N ratio increases throughout the first ~7800 years
of the record from values of 3.6 to values of nearly 7 (Figure 3b).
The variability of the C/N ratio increases throughout the last
~3500 years of the record compared with the previous ~7000
years, and the clear tendency towards increasing C/N values is not
observed. Excursions to C/N values of >6.5 occur at ~3300 and
from ~2700 to ~2300, and at ~820 cal. a BP.
Biomarkers
Throughout the record, the BIT-index does not rise above 0.15,
and the lowest values are seen in the early Holocene, with a mean
value of 0.04 between ~10,800 and ~8300 cal. a BP (Figure 3c).
BIT-index values increase from ~8300 cal. a BP, and the highest
values of 0.13 and 0.14 are reached between ~4000 and ~3000
cal. a BP.
A total of 16 n-alkane-containing fractions were discarded due
to contamination (see supplementary material S2; Figure S1). The
average concentration of the odd-chained (C25–C35) n-alkanes is
537 ng/g Sed (Figure 3e). The n-alkane concentration varies
between 200 and 800 ng/g Sed throughout most of the Holocene
(~10,800 to ~1200 cal. a BP) with one exception at ~6900 cal. a
BP, where the concentration increases to just over 1000 ng/g Sed,
and three exceptions between ~7600 and ~8100 cal. a BP, where
the concentration falls to ~150 ng/g Sed. The odd-chained
n-alkane concentrations rise after ~1200 cal. a BP to their highest
values of nearly 1400 ng/g Sed.
The MAR of the odd-chained n-alkanes follows the sedimentation rate (Figure 4b). The highest value of just over 1000 ng/
cm2/a is recorded at ~10,100 corresponding to the high LSR.
The mean combined concentration of the C37:2- and C37:3alkenones is 1134 ng/g Sed throughout the Holocene (Figure 3f).
The highest concentration of alkenones of 3000 ng/g Sed is
recorded at ~9000 cal. a BP, and the lowest concentration of 120
ng/g Sed is recorded at 8200 cal. a BP.
The MAR of the sum of the C37:3 and C37:2 alkenones follows
variations of LSRs (Figure 4d). During the early Holocene, from
~10,800 to ~7000 cal. a BP, the alkenone MAR exhibits the
highest variability. Alkenone MARs of 2210 and 1600 ng/cm2/a
are the highest, recorded at ~10,100 and ~9000 cal. a BP,
respectively.
High n-alkane-/alkenone-index values indicate increased
OCterr input, while lower values suggest a higher proportion of
OCmar input (Figure 3d). The average value of the index throughout the Holocene is 0.15. The highest fluctuation in the index is
seen between ~10,800 and ~8000 cal. a BP, owing to the high
variability of the C37-alkenone concentration during that interval.
Throughout the rest of the Holocene, the n-alkane/alkenone index
values are below 0.3.
The Holocene 0(0)
Modelling the terrestrial OC contribution to the
sediments of Ísafjarðardjúp Fjord
The percentage of terrestrial OC contributing to the sedimentary
TOC pool is estimated using three different proxies in a binary
mixing model (Figure 5). The choice of terrestrial and marine end
member values is discussed below.
The mean sedimentary OCterr contribution estimated by the
n-alkane/alkenone index is 14%. The OCterr contribution estimated by this proxy is highest in the oldest part of the record,
from ~10,800 to ~8000 cal. a BP. This proxy is the only one showing OCterr estimates higher than 30% in the early Holocene,
between ~9000 and ~8000 cal. a BP.
The mean estimated OCterr contribution to the fjordic sediment
is 8% when using the BIT-index in the binary mixing model. The
OCterr contributions remain low during the early Holocene until
~8500 cal. a BP, at which point the OCterr estimates sharply
increase. The OCterr contributions continually increase until
~3000 cal. a BP, before slowly decreasing again.
The C/N ratio–inferred OCterr estimates show the highest variability. The mean estimated OCterr contribution to the sediment is
23%. In the early Holocene, some of the estimated OCterr values
are negative. From ~8000 to ~4000 cal. a BP, the C/N ratio estimated OCterr increases from below 10% to above 40%. In the latter part of the record, the estimated OCterr remains high.
Sedimentary OCmar content and modelled paleoproductivity
The mean estimated OCterr values derived from the three binary
mixing models were used to calculate the OCmar content of MD992266. The concentrations and MARs of TOC, OCterr and OCmar
are shown in Figure 6. Throughout the Holocene, the TOC MAR
correlates well with OCmar (Spearman rank order correlation coefficient: 0.99; p < 0.05), but less so with OCterr (Spearman rank
order correlation coefficient 0.76; p < 0.05).
The marine PP of Ísafjarðardjúp Fjord was estimated using the
estimated OCmar content and the model by Knies and Mann (2002;
Figure 7c). The model estimates primary productivity values of
around 300 gC/m2/a between ~10,000 and ~8300 cal. a BP and a
PP peak of 500 gC/m2/a at ~8200 cal. a BP followed by a sharp
decrease. After 8000 cal. a BP, PP values gradually decline and
reach their minimum between ~2300 and ~2500 cal. a BP.
Throughout the youngest part of the record, PP values rise again
sharply and vary around ~200 gC/m2/a.
Discussion
Factors influencing estimates of OCterr contribution to
the fjordic sediments
The estimated percentile terrestrial OC fraction of the TOC
depends on the end member values employed in the binary mixing model. For the n-alkane/alkenone index, XTerr is 1 and XMar is
0, assuming that the n-alkanes are produced by terrestrial higher
plants and alkenones are produced in situ by marine haptophytes
only, and that there is no, or negligible, input of alkenones from
lakes (Castañeda and Schouten, 2011). Furthermore, it is assumed
that n-alkanes and alkenones are representative fractions of the
OCterr and OCmar contributions into the sediment, and that their
concentrations vary in step with changes of OCterr and OCmar
contributions.
The XTerr value used to evaluate the terrestrial OC input via the
BIT-index is 0.91 and not 1, as crenarchaeol is found in soils and
in marine environments (Weijers et al., 2006, 2009). Assuming
that branched GDGTs are produced by terrestrial bacteria, 0 is
used as the marine end member value. However, branched GDGTs
likely have an aquatic source as well (Fietz et al., 2011, 2012),
therefore the marine end member value is likely higher than 0.
7
Moossen et al.
Northwest Iceland (Table 1). Most soils in Iceland have C/N
ratios between 11 and 17 (Pitty, 1979). Due to the large variety of
C/N values from different terrestrial carbon sources, the choice of
a ‘true’ C/N end member value, representing 100% of terrestrial
OC input is difficult. Here, we choose the C/N value of the sedge
Carex cf. rariflora (C/N = 48) that dominates the wetland environment north of Ísafjarðardjúp Fjord today (Skrzypek et al.,
2008). Pollen analysis show that the vegetation on Vestfirdir Peninsula was made up of sedge and shrub type plants from ~10,100
cal. a BP and that sedge-dominated wetlands were a significant
and highly resilient part of the landscape (Caseldine et al., 2003),
supporting the choice of the XTerr value. If a C/N value typical for
Icelandic soils (C/N = 15) is chosen as the terrestrial end member
value, then fTerr increases by 20%. Therefore, the choice of endmembers for the binary mixing model has a profound effect on the
result, indicating that they need to be interpreted with caution.
OCterr contribution to the fjordic sediments
Figure 6. Concentration and MARs of total, marine and terrestrial
OC. The error bars indicate the propagated analytical error (mean
standard deviation) of the mean OCterr estimates and the analytical
error of the TOC measurements. Only those samples where the
mean OCterr estimate was calculated were used to estimate OCmar
(n = 137).
TOC: total organic carbon; OC: organic carbon; MAR: mass accumulation rate.
Furthermore, Fietz et al. (2012) have shown that the concentration of crenarchaeol in marine sediments co-varies with the concentration of the branched GDGTs used in the BIT-index. The
linear regression (r2 = 0.76; supplementary material S3, Figure
S3) between the relative abundance of crenarchaeol and the combined relative abundances of the branched GDGTs confirms the
co-variance in the sediment core studied here. Fietz et al. (2012)
argue that Thaumarchaeota may be more productive during periods where more terrestrial matter is washed into the watercolumn, as indicated by high amounts of branched GDGTs.
However, the increase of the BIT-index throughout the Holocene
broadly agrees with the other two proxies, suggesting that the
BIT-index does reflect OCterr inputs.
The terrestrial organisms producing the branched GDGTs
used to calculate the BIT-index are specific to soil and peat environments (Hopmans et al., 2004; Weijers et al., 2006). Thus,
results of the binary mixing model employing the BIT-index
likely underestimate the amount of terrestrially derived OC contributing to the TOC pool by exclusion of fresh higher plant material (as represented by the n-alkane plant waxes).
Different OM pools have widely varying C/N ratios (Lamb
et al., 2006), with typical C/N values for algae between 4 and 10,
while vascular land plants have C/N values of ≥20 (Meyers,
1994). In order to be sure, that the marine end member of the C/N
ratio exclusively represents a marine source, 4 is used as the representative value of XMar, assuming that algae are the sole or at
least main contributor of marine sourced OC to the sediment.
Possible XTerr end member values for use in the binary mixing
model employing the C/N ratio exhibit considerable variability in
The n-alkane/alkenone index indicates the highest OCterr contribution between ~8000 and ~9000 cal. a BP (Figure 5a). During
that period, concentrations of C37-alkenones in the sediment are
among the lowest of the Holocene, while n-alkane concentrations
range around the mean concentrations during Holocene. Thus, the
estimated high OCterr contribution between ~8000 and ~9000 cal.
a BP is not caused by an actual high terrigenous input into sediments, but rather by relatively low C37-alkenone concentrations,
suggesting adverse growing conditions for alkenone-producing
haptophytes. This explanation is supported by reduced salinity of
Ísafjarðardjúp Fjord waters caused by meltwater events that had a
strong influence on the fjordic environment before ~8000 cal. a
BP (Quillmann et al., 2010, 2012). The influx of meltwater may
have impaired the growth of the resident alkenone-producing
algae.
The estimated OCterr contribution between ~8000 and ~1200
cal. a BP fluctuates between ~10% and ~20%. During the last
~1000 years of the record, the concentration of the n-alkanes
increases, and this increase of terrestrially derived matter is
reflected in the rising n-alkane/alkenone-index OCterr contribution
reaching 30% at ~300 cal. a BP.
The two periods of relatively high estimated OCterr values due
to low alkenone concentration in the early Holocene and high
n-alkane concentrations in the late Holocene highlight the necessity to cautiously approach the sedimentary OCterr estimates. Furthermore, the n-alkane/alkenone-index may underestimate the
OCmar contribution, if the assumption is made that alkenones are
preferentially degraded compared to n-alkanes (Hoefs et al.,
2002; Sinninghe Damsté et al., 2002).
Compared to the other two proxies, the BIT-index provides the
lowest percentile OCterr estimates throughout the record. This is
expected, as the BIT-index primarily reflects changing contributions of soil OM and excludes higher plant OM (Hopmans et al.,
2004; Walsh et al., 2008; Weijers et al., 2006). The observed correlation between the concentrations of the crenarchaeol and the
branched GDGTs also suppresses the BIT-index and suggests a
potential contribution of aquatic branched GDGTs (e.g. Fietz
et al., 2012).
OCterr estimates show the highest variability when using C/N
ratios. The C/N values of the samples which produce negative
OCterr values are lower than the XMar end member value, explaining the negative OCterr estimates in the early Holocene. Lamb
et al. (2006) show that bacterial C/N values can be lower than 4.
Therefore, very low C/N values due to the contribution of bacterial OM to the total OM pool could, in conjunction with very low
terrestrially derived organic matter, cause very low C/N values.
The OCterr increase between ~8000 and ~4000 cal. a BP, based
on changing C/N ratios, may indicate a deteriorating climate. In a
8
The Holocene 0(0)
Figure 7. Holocene palaeoclimate records compared with the OCterr and palaeoproductivity variability of Ísafjarðardjúp Fjord. (a) Summer
insolation at 60°N (Laskar et al., 2004). The black triangles indicate the 14C-ΑΜS-dated sediment horizons of core MD99-2266. (b) Mean OCterr
estimate (black line) and standard deviation of the mean (grey-shaded area). (c) Marine palaeoproductivity of Ísafjarðardjúp Fjord modelled after
Knies and Mann (2002). The error bars indicate the propagated analytical error of the percentile OCmar concentration. Only those samples where
the mean OCterr estimate was calculated were used to estimate marine palaeoproductivity (n = 137). (d) Reconstructed NAO (NAOms) variability
(Trouet et al., 2009). (e) Variations in the speed of the ISOW south of Iceland linked to the speed of the meridional overturning circulation (50 year
running average; Hall et al., 2004). (f) August sea surface temperatures based on diatom assemblages (50 year running average; Justwan et al., 2008).
(g) GISP 2 oxygen isotope inferred temperature variations of the North Atlantic sector (Grootes and Stuiver, 1997). The MCA, the neoglaciation
and the Holocene thermal maximum are indicated by dashed lines, the 8.2 event is indicated through a grey bar.
ISOW: Iceland-Scotland-Overflow-Water; NAO: North Atlantic Oscillation; MCA: Medieval Climate Anomaly; GISP 2: Greenland Ice Sheet Project 2; AMS:
accelerator mass spectrometry.
Table 1. C/N values of different samples from terrestrial sources in northwest Iceland. The C/N value of the sedge Carex cf. rariflora (*) is used
as the terrestrial end member value when converting C/N values into percentile OCterr values.
Sample type
No. of samples
Location
C/N ratio
Literature
Terrestrial plants
Moss Warnstorfia exannulata
Segde C. cf. rariflora*
Peat (upper 10 cm)
Soil
Carex peat
6
1
1
1
3
5
64°32′ N; 22°31′ W
66°06′ N; 22°23′ W
66°06’ N; 22°23′ W
64°32’ N; 22°31′ W
64°32’ N; 22°31′ W
66°06′ N; 22°23′ W
41.3–72.6
62
48*
15.4
16.5–22
33–44
Langdon et al. (2010)
Skrzypek et al. (2008)
Skrzypek et al. (2008)
Langdon et al. (2010)
Langdon et al. (2010)
Skrzypek et al. (2008)
lacustrine environment, C/N values have been shown to increase
with decreasing temperatures (Axford et al., 2009). Axford et al.
(2009) attribute the changing C/N values to either changes in primary production or changes in the flux of terrestrial material into
9
Moossen et al.
the lacustrine sediments. Thus, increasing C/N values throughout
the Holocene are associated with deteriorating climatic conditions, as discussed in the following section.
Variations of OCterr in response to environmental
and/or anthropogenic influences
The three model approaches used to estimate the amount of OCterr
show divergent results highlighting the variability of the proxies
applied to the problem of quantifying sedimentary OCterr input.
By combining the OCterr estimates by the three different proxies
into one record showing the mean OCterr content of the fjordic
sediment, the most plausible approximation of broad Holocene
OCterr trends is derived (Figures 5d and 7b). However, this results
in an inevitable loss of finer detail that is observed in individual
records (see supplementary material S4).
The combined record of OCterr estimates shows two periods
where the OCterr content in the sediment rises. Throughout the
early Holocene, from ~10,700 to ~8800 cal. a BP, the OCterr content in the sediment increases from 2% to 18%. The early Holocene was characterised by high Northern Hemisphere summer
insolation, and glaciers on Iceland were continually retreating
(Geirsdottir et al., 2009), freeing up land area for soil formation
and subsequent increase in vegetation cover. This interpretation is
supported by increasing pollen concentration in lacustrine sediments on Vestfirdir Peninsula and North Iceland (Caseldine et al.,
2003; Langdon et al., 2010) and explains the OCterr increase.
From ~8600 to ~8000 cal. a BP, the contribution of OCterr
tends to decrease (Figure 7b). Several marine archives from the
Denmark Strait and the North Icelandic Shelf have recorded
either one cooling event centred around 8200 cal. a BP (Giraudeau
et al., 2000; Knudsen et al., 2004; Ólafsdóttir et al., 2010;
Quillmann et al., 2012), or a number of shorter cooling events
(Jennings et al., 2011). Based on chironomid assemblages, temperatures in Northwest Iceland decreased slightly at ~ 8200
(Caseldine et al., 2003) and at ~8500 cal. a BP (Langdon et al.,
2010). This downturn of climate could have caused a reversal or
halting in the development of soil reservoirs and vegetation, thus
decreasing the amount of OCterr contribution to the sediment.
Indeed, Hallsdóttir (1995) notes that the development towards
subalpine birch woodland suddenly halted at ~7500 14C years
(~8300 cal a BP). Ólafsdóttir et al. (2001) have modelled a dramatic decrease in vegetation cover at ~8000 cal a BP. The sharp
increase in soil OM contribution to the sediments as indicated by
the BIT-index (Figure 3) at 8.2 cal. a BP supports this interpretation. Decreased vegetation cover, related to the 8.2 event (Figure
7; Alley and Ágústsdóttir, 2005), would have led to more soil erosion and thus to higher input of soil OC into the sediments.
Despite the short, cold interval causing a decrease of the OCterr
contribution as previously discussed, the broader period from
~8600 to ~5400 cal. a BP was regarded as the warmest episode of
the Holocene (Figure 7; Kaufman et al., 2004; Knudsen et al.,
2008). This warm climate was conducive to extensive soil and
vegetation development in Iceland (Caseldine et al., 2003;
Hallsdóttir, 1995; Wastl et al., 2001) and likely explains both the
initial OCterr increase in the fjordic sediments after ~8000 cal. a
BP, during a recovery phase, and then a period of relative stability
from ~7500 to ~5000 cal. a BP as carbon was stored up in soil
reservoirs.
The marked increase of OCterr input to the sediments after
~5000 cal. a BP is attributed to the climatic deterioration following the Holocene Thermal Maximum that is apparent in marine as
well as terrestrial records (Figure 7f; Jennings et al., 2002; Justwan et al., 2008; Principato, 2008; Wastl et al., 2001). The deteriorating climate possibly caused accelerated soil degradation due
to periodic freezing and thawing making soils and vegetation
cover more susceptible to erosion by high velocity
winds (Jackson et al., 2005; Ólafsdóttir and Gudmundsson, 2002).
Furthermore, birch woodland was in retreat throughout the neoglaciation shown by decreasing Betula sp. pollen concentrations
(Hallsdóttir, 1995; Hallsdóttir et al., 2005). Heath land and mires
expanded, and in the lowlands, woods were replaced and buried
by peat (Hallsdóttir et al., 2005). Changing soil and vegetation
types as well as increased amounts of peat could have caused
increased amounts of OCterr to be transported into the fjord, compounding the effects of increased soil erosion as discussed above.
The sharp increase in the estimated contribution of OCterr to
the fjordic sediments between ~1200 and ~600 cal. a BP may
have been caused by anthropogenic actions and/or climatic fluctuations. The period from ~1100 to ~550 cal. a BP, known as the
Medieval Climate Anomaly (MCA; Hughes and Diaz, 1994), was
characterised by warm climate throughout Europe (Buntgen
et al., 2011; Graham et al., 2011). The Norse established settlements in Iceland at ~ad 870 (~1080 cal. a BP; Andrews et al.,
2001). The first generation of settlers cleared woodland to make
space for farmland (Hallsdóttir et al., 2005) likely causing
increased soil erosion (Dugmore et al., 2000). Human-driven vegetation change is probably reflected by a drastic rise in the oddchained n-alkane concentration at ~1200 cal. a BP (Figure 3).
Increased soil erosion, along with possible burning of woodland
to make space for agricultural land, could therefore explain
increased amounts of OCterr in the sediments.
A second explanation is provided by the change in climate
itself. The ameliorated climate of the MCA has been associated
with a prevailing positive North Atlantic Oscillation (NAO) phase
(Figure 7d; Trouet et al., 2009) that is associated with increased
precipitation in Iceland (Hurrell, 1995). Increased precipitation
would have caused more terrestrial runoff resulting in a higher
amount of OCterr to be washed into the fjord.
Marine paleoproductivity
Today’s average annual primary production around Vestfirdir
Peninsula lies between 200 and 400 gC/m2/a (Astthorsson et al.,
2007; Longhurst et al., 1995; Zhai et al., 2012). The Knies and
Mann (2002) model produces average PP values of 170 gC/m2/a
in the most recent 100 years of the record, suggesting that it may
underestimate PP. If the sedimentary C37-alkenone concentrations
are used in the model instead of the sedimentary OCmar content,
the PP values are on average 80% lower throughout the Holocene
(see supplementary material S5 and Figure S4).
The primary production in Ísafjarðardjúp Fjord may have
reached 350 gC/m2/a between ~10,000 and ~8000 cal. a BP as
indicated by the model (Figure 7c). Similar high primary production values are found in the Arctic today in areas with a high supply of nutrients due to upwelling or riverine input (Wollenburg
et al., 2004, and references therein). The apparent peak in primary
production at ~8250 cal. a BP is likely an artefact, as it coincides
with a sedimentation rate anomaly and is not reflected in the alkenone data (Figure 3). However, we suggest the boarder period of
high productivity between ~10,000 and ~8000 cal. a BP is linked
to the amount of nutrients being washed into the fjord. Quillmann
et al. (2010) show that the benthic foraminifera species Fursenkoina fusiformis, that is associated with high nutrient inputs, was
most abundant during the early Holocene while sedimentation
rates were high. Hence, the relatively high PP in the early Holocene was likely driven by high amounts of terrigenous material
being washed into the fjord as shown by the close correlation
between OCmar and the sedimentation rate (Figure 6).
Besides the terrestrial supply of nutrients from the catchment,
an allochthonous supply of advected marine nutrients may also
have played a major role in affecting PP throughout the Holocene.
A maximum in primary production from 8000 to 6000 cal. a BP
on the northwest Norwegian coast has been attributed to a strong
10
inflow of nutrient-rich Atlantic water (Knies, 2005), and periods
of high primary productivity in the Fram Strait have also been
associated with an increased nutrient supply (Wollenburg et al.,
2004). The primary production in Ísafjarðardjúp Fjord rises dramatically from below 200 to 300 gC/m2/a at ~10,200 cal. a BP
coincidental with the Irminger Current, which supplies Atlantic
water to the Denmark Strait fully penetrating the area (Ólafsdóttir
et al., 2010). After ~8000 cal. a BP, the primary production
decreases and reaches values varying around 100 gC/m2/a at
~3000 cal. a BP, coinciding with a diminishing influence of the
Irminger Current in the waters surrounding Northwest Iceland
(Giraudeau et al., 2004; Koc et al., 1993; Ólafsdóttir et al., 2010).
Today, periods with increased influx of Atlantic water show significantly increased primary production on the Northwest Icelandic Shelf (Jónsson and Valdimarsson, 2012, and references
therein). This is compelling evidence that the long-term changes
in primary production in Ísafjarðardjúp Fjord have been forced by
variations in the nutrient supply brought about by either Arctic or
Atlantic water masses dominating the area. The influence of
Atlantic water masses in the Denmark Strait have been controlled
by changing climate drivers such as changes in Northern Hemisphere insolation (Koc and Jansen, 1994), and changes in the
strength of the meridional overturning circulation (Figure 7e;
Bianchi and McCave, 1999; Hall et al., 2004; Hoogakker et al.,
2011) but also by variations in atmospheric circulation patterns in
the Denmark Strait (Blindheim and Malmberg, 2005). Therefore,
changes in the amount of primary production throughout the
Holocene may have at least indirectly been forced by climatic
change.
Conclusion
In this study, the contributions of OC derived from terrestrial and
marine sources to Ísafjarðardjúp Fjord sediments have been estimated. Three biomarker and bulk proxy data sets in binary mixing
models were used to calculate the contribution of OCterr to the
TOC pool. Subsequently, the results of the three models were
combined to produce the most conservative and accurate OCterr
estimate possible.
The estimated OCterr contribution to the sediment increases
throughout the Holocene to a maximum of 25%. The OCterr influx
into the sediments is not strongly correlated with sedimentation
rate but rather controlled by climatic changes. The multiproxy
approach to estimating OCterr contribution highlights the large
uncertainties that are still associated with the different biomarkers
and bulk approaches. Hence, further studies are needed to model
OCterr contributions to marine sediments more accurately.
Changes in Holocene PP were estimated using the model published by Knies and Mann (2002) and the reconstructed OCmar
contribution to the fjordic sediments. It appears that changes in
the PP throughout the Holocene were forced by changing amounts
of terrestrial catchment and advected marine nutrient supplies.
Northern Hemisphere insolation change and changes in the
strength of the meridional overturning circulation caused different current systems to dominate throughout the Holocene and
thus drove the advected marine nutrient supply and marine primary productivity at the coring site.
Acknowledgements
We would like to thank Professor Philip Meyers and Dr Jochen
Knies for their useful and insightful comments. This research used
samples extracted from sediment core MD99-2266, collected during the 1999 IMAGES cruise programme, by CALYPSO piston
coring deployed from the research vessel Marion Dusfresne II.
We especially thank the crew and staff of the Marion Dufresne
II, the CALYPSO coring team, Professor John Andrews and the
The Holocene 0(0)
technical staff at the Institute of Arctic and Alpine Research, Dr
Richard Pancost and technical staff at the Organic Geochemistry
Unit, Bristol, and the Glasgow Molecular Organic Geochemistry laboratory. We thank Dr Ellen Roosen from the Woods Hole
Oceanographic Institution who sent additional u-channels of the
topmost 3.5 m of the sediment core archive half.
Funding
We thank the Scottish Alliance for Geoscience, Environment and Society (SAGES) who funded the PhD to Dr Heiko
Moossen.
References
Alley RB and Ágústsdóttir AM (2005) The 8k event: Cause and consequences
of a major Holocene abrupt climate change. Quaternary Science Reviews
24: 1123–1149.
Andrews JT, Caseldine C, Weiner NJ et al. (2001) Late Holocene (ca. 4 ka)
marine and terrestrial environmental change in Reykjarfjordur, North
Iceland: Climate and/or settlement? Journal of Quaternary Science 16:
133–143.
Astthorsson OS, Gislason A and Jonsson S (2007) Climate variability and the
Icelandic marine ecosystem. Deep Sea Research Part II: Topical Studies
in Oceanography 54: 2456–2477.
Axford Y, Geirsdóttir Á, Miller G et al. (2009) Climate of the Little Ice Age and
the past 2000 years in Northeast Iceland inferred from chironomids and
other lake sediment proxies. Journal of Paleolimnology 41: 7–24.
Behrenfeld MJ, O’Malley RT, Siegel DA et al. (2006) Climate-driven trends in
contemporary ocean productivity. Nature 444: 752–755.
Belicka LL and Harvey HR (2009) The sequestration of terrestrial organic carbon in Arctic Ocean sediments: A comparison of methods and implications for regional carbon budgets. Geochimica et Cosmochimica Acta 73:
6231–6248.
Bendle J, Kawamura K, Yamazaki K et al. (2007) Latitudinal distribution of
terrestrial lipid biomarkers and n-alkane compound-specific stable carbon
isotope ratios in the atmosphere over the Western Pacific and Southern
Ocean. Geochimica et Cosmochimica Acta 71: 5934–5955.
Bianchi GG and McCave IN (1999) Holocene periodicity in North Atlantic
climate and deep-ocean flow south of Iceland. Nature 397: 515–517.
Blindheim J and Malmberg S-A (2005) The mean sea level pressure gradient
across the Denmark strait as an indicator of conditions in the North Icelandic Irminger Current. In: Drange H, Dokkken T, Furevik T and et al.
(eds) The Nordic Seas: An Integrated Perspective. Washington, DC: AGU,
pp. 65–71.
Borges AV, Delille B and Frankignoulle M (2005) Budgeting sinks and sources
of CO2 in the coastal ocean: Diversity of ecosystems counts. Geophysical
Research Letters 32: L14601.
Buntgen U, Tegel W, Nicolussi K et al. (2011) 2500 years of European climate
variability and human susceptibility. Science 331: 578–582.
Caseldine C, Geirsdottir A and Langdon P (2003) Efstadalsvatn – A multiproxy study of a Holocene lacustrine sequence from NW Iceland. Journal
of Paleolimnology 30: 55–73.
Castañeda IS and Schouten S (2011) A review of molecular organic proxies
for examining modern and ancient lacustrine environments. Quaternary
Science Reviews 30: 2851–2891.
Dugmore A, Newton A, Larsen G et al. (2000) Tephrochronology, environmental change and the Norse settlement of Iceland. Environmental Archaeology 5: 21–34.
Eglinton G, Gonzalez AG, Hamilton RJ et al. (1962) Hydrocarbon constituents
of the wax coatings of plant leaves: A taxonomic survey. Phytochemistry
1: 89–102.
Fietz S, Huguet C, Bendle J et al. (2012) Co-variation of crenarchaeol and
branched GDGTs in globally-distributed marine and freshwater sedimentary archives. Global and Planetary Change 92–93: 275–285.
Fietz S, Martínez-Garcia A, Huguet C et al. (2011) Constraints in the application of the branched and isoprenoid tetraether index as a terrestrial input
proxy. Journal of Geophysical Research 116: C10032.
Gao K, Xu J, Gao G et al. (2012) Rising CO2 and increased light exposure synergistically reduce marine primary productivity. Nature Climate Change
2: 519–523.
Geirsdottir A, Miller GH, Axford Y et al. (2009) Holocene and latest Pleistocene climate and glacier fluctuations in Iceland. Quaternary Science
Reviews 28: 2107–2118.
Giraudeau J, Cremer M, Manthé S et al. (2000) Coccolith evidence for instabilities in surface circulation south of Iceland during Holocene times. Earth
and Planetary Science Letters 179: 257–268.
Moossen et al.
Giraudeau J, Jennings AE and Andrews JT (2004) Timing and mechanisms
of surface and intermediate water circulation changes in the Nordic Seas
over the last 10,000 cal years: A view from the North Iceland Shelf. Quaternary Science Reviews 23: 2127–2139.
Graham NE, Ammann CM, Fleitmann D et al. (2011) Support for global climate reorganization during the ‘Medieval Climate Anomaly’. Climate
Dynamics 37: 1217–1245.
Gronvold K, Oskarsson N, Johnsen SJ et al. (1995) Ash layers from Iceland in
the Greenland GRIP ice core correlated with oceanic and land sediments.
Earth and Planetary Science Letters 135: 149–155.
Grootes PM and Stuiver M (1997) Oxygen 18/16 variability in Greenland
snow and ice with 10-3-10-5 year time resolution. Journal of Geophysical
Research 102: 26455–26470.
Hall IR, Bianchi GG and Evans JR (2004) Centennial to millennial scale Holocene climate-deep water linkage in the North Atlantic. Quaternary Science Reviews 23: 1529–1536.
Hallsdóttir M (1995) On the pre-settlement history of Icelandic vegetation.
Buvisindi 9: 17–29.
Hallsdóttir M, Caseldine CJ, Caseldine ARJH et al. (2005) 14. The Holocene
vegetation history of Iceland, state-of-the-art and future research. Developments in Quaternary Sciences 5: 319–334.
Hansen B and Østerhus S (2000) North Atlantic-Nordic Seas exchanges. Progress in Oceanography 45: 109–208.
Hoefs MJL, Rijpstra WIC and Damste JSS (2002) The influence of oxic degradation on the sedimentary biomarker record I: Evidence from Madeira
Abyssal Plain turbidites. Geochimica et Cosmochimica Acta 66: 2719–
2735.
Hoogakker BAA, Chapman MR, McCave IN et al. (2011) Dynamics of North
Atlantic deep water masses during the Holocene. Paleoceanography 26:
PA4214.
Hopmans EC, Weijers JWH, Schefuss E et al. (2004) A novel proxy for terrestrial organic matter in sediments based on branched and isoprenoid tetraether lipids. Earth and Planetary Science Letters 224: 107–116.
Howe JA, Austin WEN, Forwick M et al. (2010) Fjord Systems and Archives:
A Review, vol. 344. London: Geological Society (special publications),
pp. 5–15.
Howell P, Pisias N, Ballance J et al. (2006) ARAND Time-Series Analysis Software. Providence, RI: Brown University.
Hughes MK and Diaz HF (1994) Was there a medieval warm period, and if so,
where and when? Climatic Change 26: 109–142.
Huguet C, Smittenberg RH, Boer W et al. (2007) Twentieth century proxy
records of temperature and soil organic matter input in the Drammensfjord, Southern Norway. Organic Geochemistry 38: 1838–1849.
Hurrell JW (1995) Decadal trends in the North Atlantic Oscillation – Regional
temperatures and precipitation. Science 269: 676–679.
Jackson MG, Oskarsson N, Trønnes RG et al. (2005) Holocene loess deposition in Iceland: Evidence for millennial-scale atmosphere-ocean coupling
in the North Atlantic. Geology 33: 509–512.
Jennings A, Andrews J and Wilson L (2011) Holocene environmental evolution of the SE Greenland Shelf North and South of the Denmark Strait:
Irminger and East Greenland current interactions. Quaternary Science
Reviews 30: 980–998.
Jennings AE, Knudsen KL, Hald M et al. (2002) A mid-Holocene shift in Arctic
sea-ice variability on the East Greenland Shelf. The Holocene 12: 49–58.
Jónsson S and Valdimarsson H (2012) Water mass transport variability to the
North Icelandic Shelf, 1994–2010. ICES Journal of Marine Science: Journal du Conseil. Epub ahead of print 26 February 2012. DOI: 10.1093/
icesjms/fss024.
Justwan A, Koç N and Jennings AE (2008) Evolution of the Irminger and
East Icelandic Current systems through the Holocene, revealed by diatom-based sea surface temperature reconstructions. Quaternary Science
Reviews 27: 1571–1582.
Kaufman DS, Ager TA, Anderson NJ et al. (2004) Holocene thermal maximum in the Western Arctic (0-180°W). Quaternary Science Reviews 23:
529–560.
Knies J (2005) Climate-induced changes in sedimentary regimes for organic
matter supply on the continental shelf off Northern Norway. Geochimica
et Cosmochimica Acta 69: 4631–4647.
Knies J and Mann U (2002) Depositional environment and source rock potential of Miocene strata from the central Fram Strait: Introduction of a new
computing tool for simulating organic facies variations. Marine and
Petroleum Geology 19: 811–828.
Knudsen KL, Jiang H, Jansen E et al. (2004) Environmental changes off North
Iceland during the deglaciation and the Holocene: Foraminifera, diatoms
and stable isotopes. Marine Micropaleontology 50: 273–305.
Knudsen KL, Søndergaard MKB, Eiriksson J et al. (2008) Holocene thermal
maximum off North Iceland: Evidence from benthic and planktonic
11
foraminifera in the 8600-5200 cal year BP time slice. Marine Micropaleontology 67: 120–142.
Koc N and Jansen E (1994) Response to the high-latitude northern-hemisphere
to orbital climate forcing – Evidence from the Nordic Seas. Geology 22:
523–526.
Koc N, Jansen E and Haflidason H (1993) Paleoceanographic reconstructions
of surface ocean conditions in the Greenland, Iceland and Norwegian Seas
throughout the last 14-ka based on diatoms. Quaternary Science Reviews
12: 115–140.
Lamb AL, Wilson GP and Leng MJ (2006) A review of coastal palaeoclimate
and relative sea-level reconstructions using δ13C and C/N ratios in organic
material. Earth-Science Reviews 75: 29–57.
Langdon PG, Leng MJ, Holmes N et al. (2010) Lacustrine evidence of earlyHolocene environmental change in Northern Iceland: A multiproxy palaeoecology and stable isotope study. The Holocene 20: 205–214.
Laskar J, Robutel P, Joutel F et al. (2004) A long-term numerical solution for
the insolation quantities of the Earth. Astronomy & Astrophysics 428:
261–285.
Longhurst A, Sathyendranath S, Platt T et al. (1995) An estimate of global
primary production in the ocean from satellite radiometer data. Journal of
Plankton Research 17: 1245–1271.
Marret F, Scourse JD, Versteegh G et al. (2001) Integrated marine and terrestrial evidence for abrupt Congo River palaeodischarge fluctuations during
the last deglaciation. Journal of Quaternary Science 16: 761–766.
Meyers PA (1994) Preservation of elemental and isotopic source identification
of sedimentary organic matter. Chemical Geology 114: 289–302.
Meyers PA (1997) Organic geochemical proxies of paleoceanographic, paleolimnologic, and paleoclimatic processes. Organic Geochemistry 27:
213–250.
Nuwer JM and Keil RG (2005) Sedimentary organic matter geochemistry of
Clayoquot Sound, Vancouver Island, British Columbia. Limnology and
Oceanography 50: 1119–1128.
Ólafsdóttir R and Gudmundsson HJ (2002) Holocene land degradation and
climatic change in northeastern Iceland. The Holocene 12: 159–167.
Ólafsdóttir R, Schlyter P and Haraldsson HV (2001) Simulating Icelandic
vegetation cover during the Holocene implications for long-term land
degradation. Geografiska Annaler: Series A, Physical Geography 83:
203–215.
Ólafsdóttir S, Jennings AE, Geirsdóttir Á et al. (2010) Holocene variability of
the North Atlantic Irminger Current on the south- and northwest shelf of
Iceland. Marine Micropaleontology 77: 101–118.
Paetzel M and Schrader H (1992) Recent environmental changes recorded in
anoxic Barsnesfjord sediments: Western Norway. Marine Geology 105:
23–36.
Perdue EM and Koprivnjak J-F (2007) Using the C/N ratio to estimate terrigenous inputs of organic matter to aquatic environments. Estuarine, Coastal
and Shelf Science 73: 65–72.
Pitty AF (1979) Geography and Soil Properties. London: Methuen & Co
Ltd.
Principato SM (2008) Geomorphic evidence for Holocene glacial advances
and sea level fluctuations on Eastern Vestfiroir, Northwest Iceland. Boreas
37: 132–145.
Quillmann U, Jennings A and Andrews J (2010) Reconstructing Holocene palaeoclimate and palaeoceanography in Isafjaroardjup, Northwest Iceland,
from two fjord records overprinted by relative sea-level and local hydrographic changes. Journal of Quaternary Science 25: 1144–1159.
Quillmann U, Marchitto TM, Jennings AE et al. (2012) Cooling and freshening
at 8.2 ka on the NW Iceland Shelf recorded in paired δ18O and Mg/Ca
measurements of the benthic foraminifer Cibicides lobatulus. Quaternary
Research 78: 528–539.
Rommerskirchen F, Eglinton G, Dupont L et al. (2003) A north to south transect of Holocene Southeast Atlantic continental margin sediments: Relationship between aerosol transport and compound-specific δ13C land plant
biomarker and pollen records. Geochemistry, Geophysics, Geosystems 4:
1101.
Schlitzer R (2010) Ocean Data View. Available at: http://odv.awi.de.
Sinninghe Damsté JS, Rijpstra WIC and Reichart GJ (2002) The influence of
oxic degradation on the sedimentary biomarker record II. Evidence from
Arabian Sea sediments. Geochimica et Cosmochimica Acta 66: 2737–
2754.
Skei J (1983) Why sedimentologists are interested in fjords. Sedimentary Geology 36: 75–80.
Skrzypek G, Paul D and Wojtun B (2008) Stable isotope composition of plants
and peat from Arctic mire and geothermal area in Iceland. Polish Polar
Research 29: 365–376.
Smittenberg RH, Pancost RD, Hopmans EC et al. (2004) A 400-year record of
environmental change in an euxinic fjord as revealed by the sedimentary
12
biomarker record. Palaeogeography, Palaeoclimatology, Palaeoecology
202: 331–351.
Stuiver M, Reimer PJ and Braziunas TF (1998) High-precision radiocarbon
age calibration for terrestrial and marine samples. Radiocarbon 40: 1127–
1151.
Trouet V, Esper J, Graham NE et al. (2009) Persistent positive North Atlantic
Oscillation mode dominated the Medieval climate anomaly. Science 324:
78–80.
Verardo DJ, Froelich PN and McIntyre A (1990) Determination of organic
carbon and nitrogen in marine sediments using the Carlo Erba NA-1500
analyzer. Deep Sea Research Part A: Oceanographic Research Papers
37: 157–165.
Volkman JK, Barrerr SM, Blackburn SI et al. (1995) Alkenones in Gephyrocapsa oceanica: Implications for studies of paleoclimate. Geochimica et
Cosmochimica Acta 59: 513–520.
Volkman JK, Barrett SM, Blackburn SI et al. (1998) Microalgal biomarkers:
A review of recent research developments. Organic Geochemistry 29:
1163–1179.
Walsh EM, Ingalls AE and Keil RG (2008) Sources and transport of terrestrial
organic matter in Vancouver Island fjords and the Vancouver-Washington
The Holocene 0(0)
Margin: A multiproxy approach using δ13Corg, lignin phenols, and the ether
lipid BIT index. Limnology and Oceanography 53: 1054–1063.
Wastl M, Stötter J and Caseldine C (2001) Reconstruction of Holocene variations of the upper limit of tree or shrub birch growth in Northern Iceland
based on evidence from Vesturardalur-Skidadalur, Trölaskagi. Arctic, Antarctic, and Alpine Research 33: 191–203.
Weijers JWH, Schouten S, Schefuss E et al. (2009) Disentangling marine, soil
and plant organic carbon contributions to continental margin sediments: A
multi-proxy approach in a 20,000 year sediment record from the Congo
deep-sea fan. Geochimica et Cosmochimica Acta 73: 119–132.
Weijers JWH, Schouten S, Spaargaren OC et al. (2006) Occurrence and distribution of tetraether membrane lipids in soils: Implications for the use of the
TEX86 proxy and the BIT index. Organic Geochemistry 37: 1680–1693.
Wollenburg JE, Knies J and Mackensen A (2004) High-resolution paleoproductivity fluctuations during the past 24 kyr as indicated by benthic
foraminifera in the marginal Arctic Ocean. Palaeogeography, Palaeoclimatology, Palaeoecology 204: 209–238.
Zhai L, Gudmundsson K, Miller P et al. (2012) Phytoplankton phenology and
production around Iceland and Faroes. Continental Shelf Research 37:
15–25.