Reconstruction of Holocene patterns of change in a High Arctic coastal
landscape, Southern Sassenfjorden, Svalbard
Evangeline G. Sessford
⁎, Mateusz C. Strzelecki
, Anne Hormes
Department of Arctic Geology, The University Centre in Svalbard, P.O. Box 156, N-9171 Longyearbyen, Norway
Department of Geomorphology, University of Wroclaw, pl. Uniwersytecki 1, 50-137 Wroclaw, Poland
Received 15 March 2014
Received in revised form 6 December 2014
Accepted 10 December 2014
Available online 29 January 2015
Coastal evolutionsea-level change
Abrupt shifts in sediment supply, relative sea level, permafrost regime, glacier state, snow cover and sea ice
conditions associated with Holocene climate changes control processes operating on High Arctic coasts and
make reconstructions of their past evolution a signiﬁcant research challenge. This study attempts to describe
the development of the coastal zone insouthern Sassenfjorden,Svalbard, throughout the Holocene focusing on
the styles of adjustment to major landscape changes. Five marine terraces (MT1-5) are identiﬁed and assessed.
Spatial and chronological analysis suggests that the highest terrace, MT5, is pre-LGM (Last Glacial Maximum)
and that MT4-3 underwent rapid uplift (151 and 11.4 mm/year, respectively) shortly prior to 11 061 ±
174 cal. yr BP and became fully terrestrial by 9100 years ago (as indicated by emergence rates) during the
Holocene Thermal Maximum (HTM). Uplift rates for MT2-1 slowed to 5 and 2 mm/year, respectively, with
suggestedemergence between7200 and 6800 cal. yr BP. A ﬁnal 2 m uplift of therelict alluvial plain probablyhap-
pened during the Medieval Warm Period (1200–950 cal. yr BP). Most recent coastal development (AD 1912–
2012) is characterised by episodes of coastal erosion on the cliff and progradation of the Nøiselva delta.
Interactions between sea ice, snow cover, permafrost, wind and wave regimes are assessed to understand their
implications on future coastal development in a warming climate.
© 2015 Elsevier B.V. All rights reserved.
It has long been acknowledged that uplifted marine terraces are of
primary importance in regard to glacial reconstruction, and are con-
sidered isostatic ﬁngerprints of past ice volume expansions in Svalbard
(Birkenmajer, 1960;Feyling-Hanssen, 1965; Boulton, 1979; Forman and
Miller, 1984; Salvigsen, 1984; Landvik et al., 1987; Salvigsen et al., 1990;
Ziaja and Salvigsen, 1995; Forman et al., 2004; Salvigsen and Høgvard,
2005; Ingólfsson, 2011; Long et al., 2012; Ingólfsson and Landvik,
2013). Extensive investigations of maximum elevations and marineter-
race locations in the Svalbard and Barents Sea region have led to the
consensus that the Svalbard–Barents Sea Ice-Sheet (SBSIS) of the Last
Glacial Maximum (LGM) has experienced differential loading on the
land surface (Salvigsen and Slettemark, 1995; Ziaja and Salvigsen,
1995; Forman et al., 2004; Ingólfsson, 2011). Modern theory revolves
around a multi-domed SBSIS of varying ice thicknessesand thereby var-
iations in differential loading and isostatic rebound (Hogan et al., 2010;
Hormes et al., 2011; Long et al., 2012; Ingólfsson and Landvik, 2013).
Therefore, the marine limit of one area in Svalbard, for example, may
vary from one location to another within the same fjord.
Although the application of fossil shoreline records to estimate past
sea levels has recently been challenged by Pedoja et al. (2011) by
emphasising the key role of plate tectonics in controlling world-wide
Quaternary coastal uplift, careful investigations of Arctic uplifted
beaches combining sedimentological analyses and isotopic dating of
well-preserved material (e.g. whale bones, shells and driftwood) can
be used for reconstructions of the timing of deglaciation and the rate
of isostatic uplift for spatial and temporal comparison between shore-
lines. Collection of fossils from uplifted sediments of marine terraces,
provided they are contemporaneous with the geomorphology and
have remained in situ, may correspond with relative sea level (RSL)syn-
chronous with the time datable material was deposited. However, the
age of material can easily be over- or under-estimated due to reworking
and re-deposition by waves, rivers, glaciations and periglacial processes,
and must be taken into account in the analysis. Shoreline displacement
curves assist in the reconstruction of themagnitude and pattern of post-
glacial isostatic uplift such as that for Svalbard. They are created from
the mapping of marine terraces in association with dated material.
The Holocene shoreline displacement curve as described by Salvigsen
(1984) gives inner Isfjorden a displacement from approximately 70 m
a.s.l. to 15 m a.s.l. between 10,000 and 6000 cal. yr BP, suggesting an
Geomorphology 234 (2015) 98–107
⁎Corresponding authorat: P.O. Box 412, N-9171,Longyearbyen,Norway. Tel.: +47 400
E-mail addresses: firstname.lastname@example.org (E.G. Sessford), email@example.com
(M.C. Strzelecki), firstname.lastname@example.org (A. Hormes).
University of Gothenburg, Department of Earth Sciences, Box 460, SE-405 30
0169-555X/© 2015 Elsevier B.V. All rights reserved.
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/geomorph
isostatic rebound of about 55 m during the Holocene Thermal Maxi-
mum (HTM). Isostatic rebound was of the order of 8 m during
Neoglacial cooling (ca.2–5kyrBP)(Werner et al., 2013), and 2 m during
the Medieval Warm Period (MWP) (Salvigsen, 1984; Miller et al. , 2010).
Therefore, it is apparent that glacier growth following the LGM was not
extensive enough to depress thesurface of Svalbard signiﬁcantly. It also
suggests that any terraces at elevations higher than 70 m a.s.l. should be
a result of glaciations prior to the LGM. This is supported by Feyling-
Hanssen (1965), where an 84.5 m terrace at Kapp Ekholm was dated
to be 24,945 ± 734 cal. yr BP.
By using ages of marine terraces in combination with surface topog-
raphy, it is possible to hypothesize about climate variations at the time
of uplift. Furthermore, changes in climatic forcing may be associated
with the development of the shoreline. This is a branch of study which
has only brieﬂy been touched upon (Møller et al., 2002; Nichol, 2002).
Uplifted marine terraces should then be useful tools not only as indica-
tors of isostasy due to differential loading and unloading of ice sheets
and glaciations but also as clues into past shoreline development and
This study aims to improve our understanding of how High Arctic
coastal zones in Svalbard developed in response to Holocene climatic
changes and the interplay between global sea level and local glacio-
isostasy at Fredheim, southern Sassenfjorden. Understanding the
evolution of coastal systems in connection with climate change, rel-
ative sea-level rise, shifts in sediment supply and glacio -isostasy pro-
vides a crucial basis for constructing possible future scenarios for
High Arctic coastal environments (Overduin et al., 2014).
2. Study site
The study was conducted along the coast of Fredheim, a small cape
located on the southern coast of Sassenfjorden in central Spitsbergen
(Fig. 1). The study area included the present-day coastal zone in con-
nection with ﬁve palaeo-coastlines, as delineated by uplifted marine
terraces. The tidal range in the area is ca. 1.5 m and the modern storm
ridges are formed at an altitude of between 1 and 0.7 m above mean
tide level (MTL). The total analysed area covered approximately
and incorporated a number of geomorphic features inﬂuential
on coastal development including; the prograding Nøiselva delta, an
unconsolidated coastal cliff, active layer interﬂow (sensu (Ballantyne,
1978)), snow-fed streams, rivers, and large snow drift accumulations
along the coastline which are often connected to the ice-foot during
winter. Snow cover around Fredheim is varied in distribution and thick-
ness. Alongthe coastal escarpment it tends to build up to between one
and two metres, however, higher up on the terraces snow depth ranges
from 0 to 45 cm (and possibly more), as has been observed during this
Unlike most other Arctic regions which have a short open water
season of 3–4 months (Lantuit et al., 2012), Sassenfjorden and much
of western Spitsbergen have a longer season of approximately 5–6
months lasting from June through November and sometimes into
December (data collected between 1986 and 2012), produced by
Eliassen (2013). This is due to the warm Atlantic Waters that ﬂow
past the western edge of Spitsbergen and enter into the fjord systems,
causing signiﬁcant inter-annual variability in fjord water temperatures
and sea ice content (Ådlandsvik and Loeng, 1991; Nilsen et al., 2008).
There have even been years, such as the case in 2011–12, where sea
ice was not present at Fredheim at all (Eliassen, 2013, and personal ob-
servation). There was, however, an ice-foot as well as large snowdrifts
that built up along the coastal escarpment.
Fjord water circulation in frontof Fredheim along the coast appears
to be toward the west, as seen from aerial images since 1977 (Fig. 2).
Dominant winter wind directions in Svalbard are from the southeast,
though local winds may vary from this, and measured wind data for
the region are lacking. An overall warming trend is seen where the av-
erage mean annual air temperature MAAT measurements at Svalbard
Airport rose from −6.7
C for the periods 1961–1990 and
1981–2010, respectively (Førland et al., 2012).
Permafrost in Svalbard is approximately 100 m thick, with an ac-
tive layer thickness varying between 74 and 110 cm (Humlum et al.,
2003; Christiansen et al., 2010) in valleys close to sea level, such as is
thecaseinFredheim.Recentpermafrost temperatures at depth of
zero annual amplitude vary between 2.3 and 5.6 °C, however, a sig-
niﬁcant warming of the order of 0.04 −0.07 °C yr
has been ob-
served in Svalbard permafrost over the last decade (Christiansen
et al., 2010). The study by Christiansen and Etzelmüller (2010)
shows that Svalbard has the warmest permafrost so far north in the
Fig. 1. A1) Thelocation of Svalbard within a global context, as indicatedby the white circle. A) Map of Svalbard indicatingthe study area in Isfjorden,central Spitsbergen. B) The ﬁeld site
location, Fredheim (1), as well as the locations for radiocarbon data used in the construction of the shoreline displacement curve: Erdmannﬂya (2), Bohemanﬂya (3), Gåsodden
(4), Mytilusbekken (5), Phantomvika (6), Kapp Ekholm (7), Ekholmvika (8), Teltfjellbekken (9) and Ebbadalen (10).
99E.G. Sessford et al. / Geomorphology 234 (2015) 98–107
Fig. 2. Fredheim over the years from aerialphotographs from 1936, 1977, 1990 and 2009. Note the coastline retreat directly in front of Fredheim and the growth of the Nøiselva delta.
Fig. 3. Quaternary geological and geomorphological map of Fredheim indicating sample sites and geomorphological landforms mentioned throughout the text (after Sessford and
Hormes, 2013; reproduction with permission from the Norwegian Polar Institute Report Series).
100 E.G. Sessford et al. / Geomorphology 234 (2015) 98–107
Spatial observations and mapping (Sessford and Hormes, 2013)of
the geomorphological beach characteristics of uplifted marine terraces
(MT1-5) were conducted during the summers of 2011 and 2012
(Fig. 3). Present-day coastal processes were studied using year-long
observations with time-lapse imagery on site and data modelled from
meteorological observations at the Longyearbyen Airport.
The sediments and their stratigraphy were logged at two sites on
marine terraces MT4 and MT3 (Fig. 4A, B). Sedimentary interpretations
are used in order to conﬁrm and constrain glacial isostasy. An absolute
chronology has been established through collection of Mya truncata
shell halves and shell fragments found in marine terraces and dated
by radiocarbon techniques. Mollusc samples were collected 75 cm and
100 cm below the terrace surface (64.25 and 64 m a.s.l., respectively)
in MT4 (Fig. 4A). On MT3 mollusc samples were collected at 0.50 and
2.57 m below surface (50.5 and 48.40 m a.s.l.) (Fig. 4B).
Both mollusc fragments and shell halves were identiﬁed before
chemical preparation following techniques of Feyling-Hanssen (1965).
The radiocarbon analyses were determined using the Uppsala EN-
tandem accelerator after ultrasonic bathing and leaching of the samples
(Possnert, 1990). Radiocarbon ages are reported in Table 1 as conven-
tional ages with 1σstandard deviation and as calibrated ages. In the
text all radiocarbon ages are calibrated given as cal. yr BP(before pres-
ent: 1950 AD). The radiocarbon dates were calibrated to calendar ages
using a reservoir age of 440 ± 52 years. This is based on two different
recommendations using a marine reservoir effect of 450 ± 52 years
(Mangerud recommendation) and 438 ± 52 years (Bondevik and
Guliksen recommendation) for molluscs and foraminifera in Spitsber-
gen (Mangerud et al., 2006). The calibration is based on the Fairbanks
‘0107’calibration curve with the online calibration software (http://
et al., 2005) as this curve uses only coral U/Th dates.
The shoreline displacement curves for the locations other than
Fredheim are inferred from recalibrated, previously published radiocar-
bon dates as indicated in Table 1 (Feyling-Hanssen, 1965; Salvigsen,
1984; Salvigsen et al., 1990; Long et al., 2012) that they may be compa-
rable to thenew curve for Fredheimwhich is plotted on the same ﬁgure
(Fig. 5). Radiocarbondates are plotted against time inan idealized curve
with time on the x-axis showing zero asAD 1950. Sedimentological de-
scriptions from sections on the uplifted marine terraces corresponding
to the same locations asdated material are used to indicate minortrans-
gressions (locations marked in Fig. 3). Elevations are in metres above
mean tide level (m MTL) and used for all elevations, including the
map as shown in Fig. 3. Field observations showed that the crest of
modern storm ridges has an altitude of between 1 and 0.7 m above
mean tide level. Therefore in our palaeo-shoreline interpretations we
assumed that the elevation of a surface of a relict beach (storm ridge)
is within a metre above mean tide level at the time of ridge formation.
Uplift rates are inferred by the difference between the highest and
lowest elevations of each terrace in combination with age constraints
determined from the radiocarbon dates (Fig. 5).
Fig. 4. A) Sedimentological log of MT4 indicatingradiocarbon sample depths and changes in depositional environment. B) Sedimentological log of MT3 with radiocarbon sample depths
and indications of a transgression between Unit 2 and Unit 4.
101E.G. Sessford et al. / Geomorphology 234 (2015) 98–107
4. Results and interpretation
4.1. Spatial observations
The relative chronology of Fredheim is established through observa-
tions of the Holocene landforms and sediments. By analysing how land-
forms have developed in relation to each other, the relative order of
events is determined (Fig. 3). Five uplifted marine terraces have been
identiﬁed, where four are of Holocene age and the ﬁfth may be pre-
LGM (Last Glacial Maximum) due to its high elevation between 70
and 80 m MTL (Forman et al., 2004; Ingólfsson and Landvik, 2013).
Post-glacial alluvial fans, i.e. alluvial fans that are not currently active,
deposited during the Holocene (marked as ‘Fluvial material, pre-
recent’,inFig. 3) have been deposited subsequently to beach develop-
ment and therefore overlie beach ridges on top of each terrace. Hum-
mocky sections and beach deformation by ice are also identiﬁed.
Landscape development is interpreted based on the Quaternary map
(Fig. 3) alongside shoreline displacement combined with radiocarbon
dates (Fig. 5) and sedimentary logs (Fig. 4).
4.2. Sediment description and interpretation of marine terrace four (MT4)
4.2.1. Unit 1
The lowermost unit is a horizontally bedded and imbricated deposit
with rounded and sub-rounded pebbles. It is less sorted than units 3 and
4 and is interpreted as a storm beach deposit.
4.2.2. Unit 2
A gradational, conformable boundary separates thestorm beach de-
posit from a 10-cm thick sand layer that changes height over the proﬁle.
This sand layer with shell fragments is interpreted to represent a pe-
riod of formation of a shallow embayment formed after RSL rise and
supplied by sandy sediments from ﬂuvial and slope sources. It con-
tains shell fragments (MT4 shell-1, Ua-44107) and was deposited
between 10,960 and 10,574 cal. yr BP (Table 1). The other dated
shell fragments indicate only a slightly older age of terrace MT4
that was dated to between 11,235 and 10,887 cal. yr BP (MT4 shell-
2, Ua-44108). These age overlaps can be interpreted to represent
rapid isostatic uplift and or reworking of older shell fragments by
4.2.3. Units 3 and 4
The uppermost unit consists of imbricated, well sorted, sub-rounded
and rounded pebbles and represents a littoral environment duringuplift
of the site, and is indicative of a fair weather berm beach. The upper
20 cm of pebbles are frost shattered and carbonate precipitation pene-
trates to the same depth (Unit 4).
4.3. Sediment description and interpretation of marine terrace three (MT3)
4.3.1. Unit 1
The lowermost unit 1 of terrace MT3 is very compact, unsorted and
massive diamict with a yellowish silty clay matrix including smaller and
bigger clasts. The clasts are sub-rounded and sub-angular, and one clast
showing very clear striations was found. Therefore, this unit is
interpreted as a subglacial till. Another indicator for the glacial origin
of this unit is a rip-up clast of reddish clay that is incorporated within
Radiocarbon ages and associated information used in this study. Radiocarbon ages are reported as conventional dates with 1 standard deviation and as calibrated ages.
Reference Location Material ID m a.s.l
C yrs ±1σcal BP ±
Feyling-Hanssen (1965) Teltfjellbekken Mya truncata No. 358 Ua-132 56.0 9965 160 10,830 315
Feyling-Hanssen (1965) Phantomvika Mya truncata No. 349 Ua-128 50.7 10,105 150 11,032 294
Feyling-Hanssen (1965) Ekholmvika Astarte No. 350 Ua-124 42.0 9435 200 10,117 339
Feyling-Hanssen (1965) Ekholmvika Astarte U-203 9.7 4500 90 4553 204
Feyling-Hanssen (1965) Mytilusbekken Mytilus No. 343 Ua-126 5.2 3935 100 3770 196
Salvigsen (1984) Kapp Ekholm Larix occidentalis driftwood unknown 65.0 10,030 140 11,633 423
Salvigsen (1984) Gåsodden Mytilus edulis T-4628 18.1 6440 80 6841 167
Salvigsen et al. (1990) Erdmannﬂya Shell fragments T-6287 47.0 10,160 110 11,113 225
Salvigsen et al. (1990) Erdmannﬂya Hiatella arctica T-6282 26 (29) 10,120 110 11,060 234
Salvigsen et al. (1990) Erdmannﬂya Mya t. and Hiatella a. T-6286 32 (41) 9940 100 7406 133
Salvigsen et al. (1990) Erdmannﬂya Mytilus edulis T-6285 16.0 9410 110 10,098 228
Salvigsen et al. (1990) Erdmannﬂya Modiolus modiolus T-6535 6.5 9110 90 9659 189
Salvigsen et al. (1990) Erdmannﬂya Mytilus edulis T-6284 5.0 8650 90 9171 195
Salvigsen et al. (1990) Erdmannﬂya Mytilus edulis T-6288 9.8 (11.6) 8500 100 8962 223
Salvigsen et al. (1990) Erdmannﬂya Mya truncata T-8629 7.0 8370 100 8775 222
Salvigsen et al. (1990) Erdmannﬂya Mytilus edulis T-6283 8.0 8120 90 8474 138
Salvigsen et al. (1990) Erdmannﬂya Whale cranium T-6289 3.5 5590 80 5902 151
Salvigsen et al. (1990) Bohemanﬂya Hiatella arctica Lu-2364 20.0 9950 90 10,808 235
Salvigsen et al. (1990) Bohemanﬂya Mya truncata Lu-2363 0–1 9650 90 10,384 177
Salvigsen et al. (1990) Bohemanﬂya Mya truncata Lu-2138 18–20 9630 90 10,360 173
Salvigsen et al. (1990) Bohemanﬂya Mya truncata Lu-2136 (in till) 9440 80 10,147 173
Salvigsen et al. (1990) Bohemanﬂya Mytilus edulis Lu-2137 10.0 8130 80 8481 128
Salvigsen et al. (1990) Bohemanﬂya Hiatella arctica Lu-2139 (in till) 4620 60 4722 157
This study Fredheim Mya truncata fragments Ua-44108 64.3 10,106 57 11,061 174
This study Fredheim Mya truncata fragments Ua-44107 64.0 9927 60 10,767 193
This study Fredheim Mya truncata half Ua-44106 50.5 9867 63 10,674 181
This study Fredheim Mya truncata fragments Ua-44105 48.4 9878 64 10,690 186
This study Fredheim Mya truncata fragments Ua-44104 48.4 9842 60 10,636 170
Long et al. (2012) Ebbadalen Astarte borealis, paired, juvenile SUERC-35206 1.24 3411 37 3144 135
Long et al. (2012) Ebbadalen Astarte borealis, paired, juvenile SUERC-33589 3.71 4188 37 4107 131
Long et al. (2012) Ebbadalen Astarte borealis, paired, juvenile SUERC-35209 5.01 4545 37 4618 145
Long et al. (2012) Ebbadalen Astarte borealis, paired, juvenile SUERC-35210 5.97 4930 37 5141 161
Long et al. (2012) Ebbadalen Astarte borealis, paired, juvenile SUERC-35211 11.73 6397 38 6784 110
Long et al. (2012) Ebbadalen Astarte borealis, paired, juvenile SUERC-35212 16.26 7292 39 7684 81
Long et al. (2012) Ebbadalen Astarte borealis, paired, juvenile SUERC-35213 20.5 8248 39 8591 112
Long et al. (2012) Ebbadalen Astarte borealis, paired, juvenile SUERC-35214 27.9 9125 39 9646 128
102 E.G. Sessford et al. / Geomorphology 234 (2015) 98–107
4.3.2. Unit 2
This unit is composed of sub-rounded/rounded gravel and sand. A
bullet-shaped boulder was found at the bottom of this deposit that
might have been reworked from the underlying subglacial till. Half
a shell of M. truncata and shell fragments were found close to the
lowermost sharp boundary of the unit (MT3 shell-1, UA-44104;
MT3 shell-2, Ua-44105, respectively). The coarse sand and pebbles
are horizontally bedded and interpreted as beach deposit. According
to the radiocarbon ages, beach deposition started between 10,876
and 10,466 cal. yr BP (Table 1). The erosive boundary of the beach de-
posit on top of the diamict indicates a rather rapid sea-level rise at
4.3.3. Unit 3
A gradual boundary separates this medium to coarse sand con-
taining pebbles of local carbonate lithologies and quartzite, from the
beach deposit above. It is suggested that the gradual boundary can be
interpreted as a transgressional phase.
4.3.4. Unit 4
The beach deposits of Unit 4 are very loose and continued to cover
the lower part during logging; therefore, the boundary of Unit 4 to
Unit 3 could not be described. The top unit of the logging site shows
horizontally bedded, well sorted, well-rounded pebbles and sand. The
radiocarbon ages of the shell fragments (MT3 shell5, Ua-44106;
10,855–10,493 cal. yr BP) that were found in this beach deposit overlap
with one standard deviation with the beach deposit of Unit 2.
4.4. Shoreline displacement
Results from sedimentological descriptions and radiocarbon dates
are used to produce the shoreline displacement curve for Fredheim
shown in Fig. 5. Deglaciation of Sassendalen began approximately
Fig. 5. Inferred shorelinedisplacement curvesas indicated by recalibrated radiocarbon datesfrom Feyling-Hanssen (1965),Salvigsen (1984), Salvigsen et al. (1990)Long et al. (2012) and
this study. Original datesand their recalibrations can be found in Table 1. Solid lines join sample dates; dashed lines are inferred extensions of the shoreline displacement. The squares
made up of dashedlines indicate the time range and elevation range used for emergence rates. Note: samples from Salvigsen et al. (1990) that had more than one sample depth were
not used in the curve.
103E.G. Sessford et al. / Geomorphology 234 (2015) 98–107
11,300–11,200 cal. yr BP (Forwick et al., 2010) and glacier retreat was
signiﬁcant to allow marine ﬂooding over all of Fredheim coastal plain.
The onset of uplift for upper marine terrace four (MT4) (70–65 m
MTL.) began before lower MT4 (65–60 m MTL.), as indicated by
M. truncata fragments sampled at 64 and 64.25 m MTL producing ages
of 11,061 ± 174 cal. yr BP and 10,767 ± 193 cal. yr BP, respectively.
MT3 underwent rapid uplift during the Holocene Thermal Maximum
(HTM) and became fully terrestrial by 9100 cal. yr BP. Uplift of MT2-1
were completed during a cooler period following the HTM and perhaps
into Neoglaciation.Suggested emergence for MT2 is between 7200 and
6800 cal. yrBP. Uplift of MT1 was completed by 3770 cal. yr BP, as sug-
gested by radiocarbon ages and curves from Feyling-Hanssen (1965),
Salvigsen (1984)andSalvigsen et al. (1990). Final uplift of the non-
active alluvial plain probably happened during the Medieval Warm Pe-
riod (MWP) (1200–950 cal. yr BP). Other than the Fredheim curve, the
Holocene shoreline displacement curves shown in Fig. 5 are inferred
through recalibrated dates from Feyling-Hanssen (1965),Salvigsen
(1984), Salvigsen et al. (1990) and Long et al. (2012). Combined, the
curves suggest that in inner Isfjorden relative sea level (RSL) fell very
rapidly from ca. 65 m MTL to ca. 34 m MTL between 11,300 and
9000 cal. yr BP, by 3770 cal. yr BP it had reached 5.8 m MTL and by
3100 cal. yr BP it fell close to the elevation of the non-active alluvial
plain (ca. 1.2 m MTL.), suggesting an isostatic rebound of over 60 m dur-
ing the early and mid-Holocene.
With few radiocarbon data points at a single site or within a smallre-
gion, minor transgressions during uplift are often missed (Salvigsen,
1984). The sedimentary composition at Fredheim sheds light on these
minor changes and shows a gradual boundary between units 2 and 3
on MT3 (Fig. 4B), which is interpreted to be a small transgression. The
gradual boundary is between 48 and 50 m MTL and therefore occurred
approximately 10,500 cal. yr BP when plotted on the shoreline displace-
ment curve in relation to its elevation above current mean tide level. A
much larger transgression is noted by Salvigsen et al. (1990) on the
western coast of Isfjorden at 5–10 m MTL approximately 9000–
8000 cal. yr BP. This is not seen in any of theother locations but cannot
be ruled out as existing. It is the only curve from the western side of
Isfjorden, and therefore may show differences to those further east
that are closer to the SBSIS domes. The results from Long et al. (2012)
show a similar stable, steady trend as those of Feyling-Hanssen (1965)
except that uplift occurs earlier. In northern Billefjorden region (eastern
coast of Petuniabukta), located 40 km from our study site, Long et al.
(2012) found the highest Holocene beaches, deposited shortly after
local deglaciation at ca. 10,000 cal. yr BP, at ca. 40–45 m a.s.l. According
to Long et al. (2012), during the mid-Holocene RSL probably fell below
present level, and around 3100 cal. yr BP roseto reach within a metre of
present sea level.
5.1. Landform development and uplift
The surface topography of marine terraces, i.e. beach ridge ampli-
tude and wavelength, and shore gradient, serves as indicators of pro-
cesses acting upon the shoreline during emergence and thereby
record vertical and horizontal movement of the shoreline, presence of
sea-ice cover as well as shifts in sediment supply to the coast. Due to
isostatic uplift the resulting beach plains will slope so that forelands
prograde and older beach ridges will be more elevated than those that
are younger. The slope gradient therefore depends upon the rate of
emergence and progradation of the shoreline, resulting in rapid (~N10
mm/year scale) or slow (~1–10 mm/year scale) displacement (inferred
from, Feyling-Hanssen and Olsson, 1960). At Fredheim, marine terrace
four (MT4) displays themost rapid uplift, as inferred by thedistance be-
tween the highest and lowest elevations of each terracein combination
with age constraints from the shell fragments, showing progradation of
approximately 15.1 mm/year between 11,235 and 10,574 cal. yr BP
(Table 1 and Fig. 5). The younger age overlaps at 1σwith the ages that
were yielded from terrace MT3. This overlap may suggest that the uplift
of MT4 to MT3 was fairly continuous and that the pause in uplift be-
tween the two was not long. It could also suggest that shell fragments
have been reworked after deposition, thereby showing older ages
than they ought to be on MT3. However,as we did not observe any car-
bonate precipitation on MT3, as we did on MT4, this observation might
be used as an additional argument that terrace MT4 has undergone
weathering processes for longer than MT3 and must therefore be rela-
tively older (Forman and Miller, 1984).
MT3 emerged quite rapidly, as is noted by the erosional boundary
between the glacial diamict and beach sediments, in addition to lack
of ﬁnes. Radiocarbon ages for MT3 suggest beach deposition started
shortly prior to 10,636 ± 170 and 10,690 ± 186 cal. yr BP (Table 1).
An emergence rate of 11.4 mm/year determined using the shoreline
emergence curves and surface topography, i.e. beach ridges, (between
10,855 and 9100 cal. yr BP) is suggested for MT3.
Emergence rates for marine terraces at Fredheim are supported by
those suggested by Salvigsen (1984) where average uplift rate between
10,000 and 8000 cal. yr BP is 19 mm/year and slowed down to 4.5mm/
year between 8000 and 4000 cal. yr BP. Uplift rates coincide with re-
constructions of deglaciation by Forwick et al. (2010) which suggest
step-wise retreat of the Sassendalen glacier, where glacial proximal
conditions were present during the mid-Holocene. It is likely that re-
bound also occurred in a step-wise fashion, as indicated by the steep
slopes between each terrace that may indicate periods of coastal erosion
in former terrace sediments during the slower sea-level fall.
Subsequent terraces MT2 and MT1 have not beendated as no mate-
rial suitable for dating was found during ﬁeldwork.Based on their eleva-
tions in comparison tothe local Isfjorden shoreline displacement curve
(Feyling-Hanssen, 1965; Salvigsen, 1984; Long et al., 2012), it is as-
sumed that MT2 and MT1 were most likely uplifted sometime between
7200 and 3574 cal. yr BP. Between 3200 cal. yr BP and during the Medi-
eval Warm Period (MWP), when sea-level was over 1 m higher than at
present, a large alluvial plain formed (Fig. 3). MT1 and MT2 appear to
have undergone much slower progradation of only 5 mm/year between
7200 and 6800 cal. yr BP, and 2.3 mm/year between 4200 and
3770 cal. yr BP, respectively. Forwick et al. (2010) propose that general
glacier growth in the area occurred between 4000 and 5000 cal. yr BP
(Neoglacial advance), which coincides well with the decrease in uplift
rate of MT1.
Aﬁnal period of uplift of the post-glacial alluvial plain, on which the
buildings at Fredheim have been built, likely occurred sometime during
the MWP between 1200 and 950 cal. yr BP, as indicated by the shoreline
displacement curve by Salvigsen (1984).
Since the uplift of the alluvial plain, both erosion and deposition
have been taking place at the coast. Aerial images show that between
1977 and 2009, the average rate of erosion on the coastal escarpment
made up of the alluvial plain has been 0.33 m/year and that average
growth of the modern delta is 3.6 m/year (Sessford, 2013).
Marine terraces 1–4(Fig. 5) support the rebound models for the
Svalbard–Barents Sea Ice Sheet (SBSIS) by Ingólfsson and Landvik
(2013) who suggest that in central Spitsbergen (in the area of Fredheim
and Kapp Ekholm) an isobase for relative crustal rebound since around
7600 cal. yr BP is approximately 15 m above sea level, and that since
around 12,000 cal. yr BP isostatic rebound caused by isostatic depres-
sion has been approximately 50–60 m (both of which correspond well
with the shoreline displacement curve in this paper) (Fig. 5). However,
MT5 at 70–80 m a.s.l., which has not been thoroughly discussed, is too
high for the Holocene. Following previous models, MT5 may belong to
a pre-LGM (Last GlacialMaximum) terrace system. There are otherloca-
tions in Svalbard that are accepted to have been covered bynon-erosive
cold-based ice during LGM (Forman et al., 2004; Hormes et al., 2011;
Ingólfsson and Landvik, 2013), and therefore did not get reworked dur-
ing the LGM, and pre-existing terraces were left intact after retreat.
Although none of the central Isfjorden higher elevation marine terraces
104 E.G. Sessford et al. / Geomorphology 234 (2015) 98–107
(above 70 m a.s.l.) have beenthoroughly examined, evidence appearsto
support the presence of pre-LGM marine terraces. MT5 at Fredheim
needs to be more thoroughly examined so as to clarify if indeed it is
older than the LGM. If it is not, and belongs to the Holocene, it may be
that isostatic uplift curves need to be re-examined.
5.2. Geomorphic processes and climate
Site-speciﬁc factors such as exposure, wind direction, fjord circula-
tion, coastal plain gradient and sediment supply affect the formation
of Arctic beach accumulation and erosion processes (Møller et al.,
2002). Beach ridges are formed along the top of the forelands due to
dominant longshore drift, and represent either storm ridges (Feyling-
Hanssen, 1965) or fair weather berm ridges (Mason, 2010). A fair
weather beach might be characterised by low amplitude (b0.5 m)
beach crests with wavelengthsof approximately5 m and are ﬂat crested
(Mason, 2010; Long et al., 2012). Møller et al. (2002) identiﬁes ridges
having a steep seaward scarp and a marked landward swale to beindic-
ative of erosional stormy events in comparison to a gentle seaward
slope associated with fair weather. Although beach ridges were not
measured directly at Fredheim, one can see in Fig. 6A that, from marine
terrace three (MT3),beach ridges are relatively ﬂat crested, have an ap-
proximate amplitude of 30–50 cm and wavelengths approximately 5–
6 m. The same trend is exempliﬁed on MT4 and MT1. Ridges on MT2
are however, difﬁcult to assess due to geliﬂuction and active layer
Modern surface water circulation in front of Fredheim is along
the coast toward the west, as indicated by aerial photographs since
1977 (Fig. 2). However, the delta is prograding eastward and wave-
generated beach ridges grow parallel with progradation. Wave activity
at Fredheim is largely limited to local wind waves as fetch distance is
minimal due to its location with the fjord system and the now present
Nøiselva delta protecting the shoreline. However, this is not the case
when winds are from the northeast through Sassenfjorden where
fetch distance is of the order of 14 km. Nonetheless, by broadening the
observation area, insight into dominant wind direction during the
Holocene is toward the southeast and beach ridges formed from sedi-
ment supplied from deltaic systems coming from Nøisdalen (Figs. 2,
3). Similar processes are described in Mason (2010) from northern
Greenland. Ridges imply a long-term easterly direction of littoral
sediment transport at Fredheim. Modern beach ridges to the west of
Fredheim and at the western edge of the Nøis delta follow this pattern,
and it is only directly in front of Fredheim that local winds seemto have
the most impact on beach development.
The surfaces of beach ridges in Arctic environments are often
disrupted and covered by hummocks and pits forming characteristic
pitted beach micro-relief (Nichols, 1961; Urdea, 2007; St-Hilaire-
Gravel et al.,2010). There appear to be three differentprocesses forming
pitted beach morphology at Fredheim: (i) melt-out of remnants of ice-
foots, ice ﬂoes, and icebergs buried in beach sediments which are
often referred to as sea-ice-kettles (found on the present beach, MT1
and MT2); (ii) ice deformation ridges and swales (found on the edges
of MT1 and MT3); and (iii) accumulationsof sediments around boulders
driven by wind wave action (found on the presentbeach and MT3).
The morphology of hummocks on MT1 and MT2 shows higher
ridges up-slope, a curved back-scarp and tend to run parallel to the
shore. Pits on these terraces tend to be behind the ridges (upslope)
and are elongated along the palaeo-shorelines. According to Lindner
and Marks (1989) larger pits found on raised beaches mark places of
iceberg stranding and may indicate periods of intensiﬁed calving and
iceberg production. When observing the present-day beach, hummocks
and pits appear in sites where the ice-foot forms during the winter sea-
son and show similar characteristics to those on MT1 and MT2 (Fig. 6C),
(Nichols, 1961; Urdea, 2007; Rodzik and Zagórski, 2009; St-Hilaire-
Gravel et al., 2010).
An increase in ice rafted debris in the area at 7930–5470 cal. yr BP
is attributed to enhanced sea ice formation (Baeten et al., 2010;
Rasmussen et al., 2013). Therefore, it is possible that these sea-ice-
kettles on MT1–2 have been produced by sea ice as an ice-foot and/or
grounded ice. These terraces were likely a product of uplift during cooler
times prior to or duringNeoglaciation,and it is possible that some ridges
may have formed from iceberg ride-up or pushing due to strong winds
(Møller et al., 2002; Urdea, 2007). A modern example of anice deforma-
tion ridge can be seen (Fig. 7A) where an ice ﬂoe had been pushed up
onshore, over the modern beach, thereby disrupting sediments and
crushing two boats on MT1 during an extreme event in 1999. It is un-
clear whether this was caused by ice ride-up, thrusting or by high
winds during a major storm event. They are also found just on the
steep erosional edge of MT3, suggesting that during the formation of
MT2 beach ice was thrust all the way up the escarpment onto MT3
The type iii hummocks are foundin the east on the modern beach as
well as on MT3. These hummocks and pits are present closer to the ex-
posed bedrock which has been affected by chemical and mechanical
weathering by water, though subsequently affected by periglacial and
Aeolian processes (Fig. 6B). It is possible that these hummocks are cre-
ated because of large boulders falling onto the shoreline through ero-
sion of the bedrock, followed by redistribution of sediments through
wind and wave action. Sediments would then not build smoothly as
ridges but around the boulders in hummocks. As MT3 was developed
during the Holocene Thermal Maximum (HTM) it seems unlikely that
sea ice was present due to higher sea level, air temperatures and
Fig. 6. A) Fair weather beach ridges on MT3. The black backpack dimensions are
approximately 60 by 20 cm. B) Hummocky beach on MT3 indicated within the
black dashed line area, note the rounded boulders from mechanical weathering
through wave action, and holes from chemical weathering. C) Modern beach ridges
as delineated by the solid line and hummocks outlines by the dashed line.
Photos: Sessford, June 2012.
105E.G. Sessford et al. / Geomorphology 234 (2015) 98–107
increased inﬂuence of Atlantic Water masses (Miller et al., 2010;
Rasmussen et al., 2013). This suggests that the hummocks are likely
formed through sediment redistribution around boulders that were
undergoing wave erosion.
This paper describes changes that occurred over the Holocene in the
coastal landscape of Fredheim, Southern Sassenfjorden. Our study sug-
gests that the analysed coastal system dynamically adjusted to shifts
in climate conditions, sediment supply and sea level over the Holocene.
Based upon the rates of uplift for marine terraces, development of
Holocene stratigraphy and detailed geomorphological mapping, we
conclude the following:
a) Spatial and chronological analysis suggests that uplifted marine ter-
races MT4 and MT3 underwent rapid uplift during the Holocene
Thermal Maximum starting shortly prior to 11,061 ±
174 cal. yr BP and becoming fully terrestrialby 9100 years ago. Uplift
of MT2 and MT1 were completed during the cooling period follow-
ing HTM and maybe into Neoglaciation, with suggested emergence
between 7200 and 6800 cal. yr BP for MT2 and 4200 and
3770 cal. yrBP for MT1. The last uplift of the relict alluvial plain prob-
ably happened during the MWP (1200–950 cal. yr BP). Present
coastal erosion andprogradation of theNøiselva delta are recognised
as having taken place within the last 100 years.
b) MT5 may belong to a pre-LGM terrace system and needs to be more
thoroughly examined so as to clarify if indeed it is older than LGM.
c) The uplifted marine terraces are interpreted as having formed
during predominantly calm periods where sediment availability
and accommodation space were in ample supply, thereby causing
d) Dominant wind direction during the Holocene was toward the
southeast, as indicated by the direction of beach ridge formation.
e) Hummocky and pitted beach sections on MT3 are interpreted to
have developed due to the presence of boulders disrupting clear
beach formation. However, on MT1 and MT2 sea-ice-kettles were
produced by the ice-foot and or ice-push as they were deposited
during cooler times.
The coastal zone of Fredheim provided a good opportunity to inves-
tigate High Arctic coastal response to Holocene sea-level and climate
changes and to observe various types of coastal adjustment to ongoing
environmental change. Our study suggests that Svalbard coastal sys-
tems responded abruptly to the Holocene landscape transformation.
We thank Jens Morten Hansen and one anonymous reviewer of this
paper for helpful suggestions. Thanks are also attributed to Andrew
Plater for the ﬁnal editorial review.
Matt Strzelecki is supported by National Science Centre Postdoctoral
Fellowship (Project: Model of the interaction of paraglacial and periglacial
processes in the coastal zone and their inﬂuence on the development of
Arctic littoral relief' award no. 2013/08/S/ST10/00585) and Foundation
for Polish Science HOMING PLUS (grant no. 2013-8/12) and START
This paper is a contribution to the Palaeo-Arctic Spatial and
Temporal Gateways Programme and the IAG Sediment Budgets in
Cold Environments Working Group.
Ådlandsvik, B., Loeng, H., 1991. A study of the climatic system in the Barents Sea. Polar
Res. 10, 45–59.
Baeten, N.L., Forwick, M., Vogt, C., Vorren, T.O., 2010. Late Weichselian and Holocene sed-
imentary environments and ice rafting in Isfjorden, Spitsbergen. Geol. Soc. Lond.
Spec. Publ. 344 (207-223).
Ballantyne, C.K., 1978. The hydrologic signiﬁcance of nivation features in permafrost
areas. Geogr. Ann. Ser. A, Phys. Geogr. 60 (1/2), 51–54.
Birkenmajer, K., 1960. Raised marine features of the Hornsund area, Vestspitsbergen.
Stud. Geol. Pol. 5, 1–95.
Boulton, G.S., 1979. Glacial history of the Spitsbergen archipelago and the problem of a
Barents Shelf ice sheet. Boreas 8, 31–57.
Christiansen, H.H., Etzelmüller, B., 2010. Report from the International Permafrost Associ-
ation: THIRD European Conference on Permafrost (EUCOP III) in Longyearbyen, Sval-
bard. Permafr. Periglac. Process. 21 (4), 366–369.
Christiansen, H.H., Etzelmüller, B., Isaksen, K., Juliussen, H., Farbrot, H., Humlum, O.,
Johansson, M., Ingeman-Nielsen, T., Kristensen, L., Hjort, J., Holmlund, P., Sannel,
A.B.K., Sigsgaard, C., Åkerman, H.J., Foged, N., Blikra, L.H., Pernosky, M.A., Ødegård,
R.S., 2010. The thermal state of permafrost in the Nordic area duringthe International
Polar Year 2007–2009. Permafr. Periglac. Process. 21 (2), 156–181.
Eliassen, A., 2013. Været som var, Longyearbyen, Svalbard. Norwegian Meteorological In-
Fairbanks, R.G., Mortlock, R.A., Chiu, T.C., Cao, L., Kaplan,A., Guilerson, T.P., Fairbanks, T.W.,
Bloom, A.L., 2005. Marine radiocarbon calibration curve spanning 0 to
50,000 yearsB.P. based on paired 230Th/234U/238U and14C dates on pristine corals.
Quat. Sci. Rev. 24.
Feyling-Hanssen, R.W., 1965. Shoreline displacement in central Vestspitsbergen and a
marine section from the Holocene of Talavere on Barentsøya in Spitsbergen. Nor.
Polarinst. Skr. 93, 1–5.
Feyling-Hanssen, R.W., Olsson, I., 1960. Fiveradiocarbon datings ofpost glacial shorelines
in central Spistsbergen. Nor. Geogr. Tidsskr. 86, 121–131.
Førland,E.J., Benestad, R.,Hanssen-Bauer, I.,Haugen, J.E., Skaugen, T.E., 2012. Temperature
and precipitation development at Svalbard 1900–2100. Adv. Meteorol. 2011 (Article
ID 893790), 1–14.
Forman, S.L., Miller, G.H., 1984. Time-dependent soil morphologies and pedogenic pro-
cesses on raised beaches, Bröggerhalvöya, Spitsbergen, Svalbard Archipelago. Arct.
Alp. Res. 16 (4), 381–394.
Forman, S.L., Lubinski, D.J., Ingólfsson, Ó., Zeeberg, J.J., Snyder, J.A., Siegert, M.J., Matishov,
G.G., 2004. A review of postglacial emergence on Svalbard, Franz Josef Land and
Novaya Zemlya, northern Eurasia. Quat. Sci. Rev. 23 (11–13), 1391–1434.
Forwick, M., Vorren, T.O., Hald, M., Korsun, S., Roh, Y., Vogt, C., Yoo, K., 2010. Spatial and
temporal inﬂuence of glaciers and rivers on the sedimentary environment in
Sassenfjorden and Templefjorden, Spitsbergen. In: Howe, J., Austin, W., Forwick, M.,
Paetzel, M. (Eds.), Fjord Systems and Archives. Geochem. Soc. Spec. Publ. 344,
Fig. 7. A) Modernsediment deformation as a ridge by ice on the edgeof MT1 (also noted in
the map in Fig. 3). B) Ice ridgedeformation on the edgeof MT3 during deposition of MT2 in
the Late Holocene.
Photos: Sessford, June 2012.
106 E.G. Sessford et al. / Geomorphology 234 (2015) 98–107
Hogan, K.A., Dowdeswell, J.A., Noormets, R., Evans, J., Cofaigh, Ó.C., 2010. Evidencefor full-
glacial ﬂow and retreat of the Late Weichselian Ice Sheet from the waters around
Kong Karls Land, eastern Svalbard. Quat. Sci. Rev. 29, 3563–3582.
Hormes, A., Akçar, N., Kubik, P.W., 2011. Cosmogenic radionuclide dating indicates ice-
sheet conﬁguration during MIS 2 on Nordaustlandet, Svalbard. Boreas 40, 636–649.
Humlum, O., Instanes, A., Sollid, J.L., 2003. Permafrost in Svalbard: a review of research
history, climatic background and engineering challenges. Polar Res. 22 (2), 191–215.
Ingólfsson, Ó., 2011. Fingerprints of Quaternary glaciations on Svalbard. Geochem. Soc.
Spec. Publ. 354, 15–31.
Ingólfsson, Ó., Landvik, J.Y., 2013. The Svalbard–Barents Sea ice-sheet —historical, cur rent
and future perspectives. Quat. Sci. Rev. 64, 33–60.
Landvik, J.Y., Mangerud, J., Salvigsen, O., 1987. The Late Weichselian and Holocene shore-
line displacement on the west-central coast of Svalbard. Polar Res. 5, 29–44.
W., Rachold, V., Sedenko, S., Solomon, S., Steenhuisen, F., Streletskaya, I., Vasilie v,
A., 2012. The Arctic Coastal Dynamics database. A new classiﬁcation scheme and
statistics on Arctic permafrost coastlines. Estuar. Coasts 35 (2), 383–400.
Lindner,L., Marks, L., 1989. Impact of icebergs on relief development of marine beaches in
Spitsbergen. Quaest. Geographicae 2 (Special Issue), 111–119.
Long, A.J., Strzelecki, M.C., Lloyd, J.M., Bryant, C.L., 2012. Dating high Arctic Holocene rel-
ative sea level changes using juvenile articulated marine shells in raised beaches.
Quat. Sci. Rev. 48, 61–66.
Mangerud,J., Bondevik, S., Gulliksen, S.,Hufthammer, K.A.,Hoisaeter, T., 2006.Marine 14C
resevoirages for 19th century whales and molluscs from theNorth Atlantic. Quat. Sci.
Rev. 25, 3228–3245.
Mason, O.K., 2010. Beach ridge geomorphology at Cape Grinnell, northern Greenland: A
less icy Arctic in the mid-Holocene. Geogr. Tidsskr.-Dan. J. Geogr. 110 (2), 337–355.
Miller, G.H., Brigham-Grette, J., Alley, R.B., Anderson, L., Bauch, H.A., Douglas, M.S.V.,
Edwards, M.E., Elias, S.A., Finney, B.P., Fitzpatrick, J.J., Funder, S.V., Herbert, T.D.,
Hinzman, L.D., Kaufman, D.S., MacDonald, G.M., Polyak, L., Robock, A., Serreze, M.C.,
Smol, J.P., Spielhagen, R., White, J.W.C., Wolfe, A.P., Wolff, E.W., 2010. Temperature
and precipitation history of the Arctic. Quat. Sci. Rev. 29 (15–16), 1679–1715.
Møller, J.J., Yevzerov, V.Y., Kolka, V.V., Corner, G.D., 2002. Holocene raised-beach ridges
and sea-ice-pushed boulders on the Kola Peninsula, northwest Russia: indicators of
climatic change. The Holocene 12 (2), 169–176.
Nichol, S.L., 2002. Morphology, stratigraphy and origin of last interglacial beach ridges at
Bream Bay, New Zealand. J. Coast. Res. 18 (1), 149–159.
Nichols, R.L., 1961. Characteristics of beaches formed in polar climates. Am. J. Sci. 259,
Nilsen, F., Cottier, F., Skogseth, R.,Mattsson, S., 2008. Fjord-shelf exchanges controlled by
ice and brine production: the interannual variation of Atlantic Water in Isfjorden,
Svalbard. Cont. Shelf Res. 28, 1838–1853.
Overduin, P.P., Strzelecki, M.C., Grigoriev, M.N., Couture, N., Lantuit, H., St-Hilaire-Gravel,
D., Günther, F., Wetterich, S., 2014. Coastal changes in the Arctic. In: Martini, I.P.,
Wanless,H.R. (Eds.), Sedimentary CoastalZones From High to Low Latitudes: Similar-
ities and Differences. Geological Society, Special Publications 388, pp. 103–130.
Pedoja, K., Husson, L., Regard, V., Cobbold, P.R., Ostanciaux, E., Johnson, M.E., Kershaw, S.,
Saillard, M., Martinod, J., Furgerot, L., Weill, P., Delcaillau, B., 2011. Relative sea-level
fall since the last interglacial stage: are coasts uplifting worldwide? Earth-Sci. Rev.
Possnert, G., 1990. Radiocarbon dating by the accelerator technique. Nor. Archaeol. Rev.
Rasmussen, T.L., Forwick, M., Mackensen, A., 2013. Reconstruction of inﬂow of Atlantic
Water to Isfjorden, Svalbard during the Holocene: correlation to climate and season-
ality. Mar. Micropaleontol. 99, 18–28.
Rodzik, J., Zagórski, P., 2009. Shore ice and its inﬂuence on development of the shores of
southwestern Spitsbergen. Oceanol.Hydrobiol. Stud. 38 (1), 163–180.
Salvigsen, O., 1984. Occurrence of pumice on raised beaches and Holocene shoreline dis-
placement in the inner Isfjorden area, Svalbard. Polar Res. 2 (1), 107–113.
Salvigsen, O., Høgvard, K., 2005. Glacial history, Holocene shoreline displacement and
palaeoclimate based on radiocarbon ages in the area of Bockfjorden, north-western
Spitsbergen, Svalbard. Polar Res. 25 (1), 15–24.
Salvigsen, O., Slettemark, Ø., 1995. Past glaciation and sea levels on Bjørnøya, Svalbard.
Polar Res. 14 (2), 245–251.
Salvigsen, O., Elgersma, A., Hjort, C., Lagerlund, E., Liestøl, O., Svensson, N., 1990. Glacial
history and shoreline displacement on Erdmannﬂya and Bohemanﬂya, Spitsbergen,
Svalbard. Polar Res. 8, 262–273.
Sessford,E.G., 2013. Spatial andTemporal Analysis ofHolocene Coastal Development:Ap-
plications to Erosion Assessment and Cultural heritage Mitigation in Svalbard. (MSc.
Thesis MSc thesis). The University of Oslo, Oslo.
Sessford, E.G., Hormes, A., 2013. Quaternary geological and geomorphological maps of
Fredheim and Skansbukta. Reportserie; 142. The Norwegian Polar Institute, Tromsø.
St-Hilaire-Gravel,D., Bell, T.J., Forbes, D.L., 2010.Raised gravel beaches as proxy indicators
of past sea-ice and wave conditions, Lowther Island, Canadian archipelago. Arctic 63
Urdea, P., 2007. About some geomorphological aspects of the polar beaches. Rev.
Geomorfologie 9, 5–16.
Werner,K., Spielhagen, R.F.,Bauch, D., Hass, H.C., Kandiano, E., 2013.Atlantic water advec-
tion versus sea-ice advances in the eastern Fram Strait during the last 9 ka:
multiproxy evidence for a two-phase Holocene. Paleoceanography 28, 283–295.
Ziaja, W., Salvigsen, O., 1995. Holocene shoreline displacement in southernmost Spitsber-
gen. Polar Res. 14 (3), 339–340.
107E.G. Sessford et al. / Geomorphology 234 (2015) 98–107