Quaternary International 465 (2018) 117e129

Contents lists available at ScienceDirect

Quaternary International
journal homepage: www.elsevier.com/locate/quaint

MIS 5e1 dinoflagellate cyst analyses and morphometric evaluation of

Galeacysta etrusca and Spiniferites cruciformis in southwestern Black
Sea
Shannon Ferguson a, b, Sophie Warny a, b, *, Gilles Escarguel c, Petra J. Mudie d
a
Department of Geology and Geophysics, E-235 Howe-Russell, Louisiana State University, Baton Rouge, LA 70803, USA
b
Museum of Natural Science, E-235 Howe-Russell, Louisiana State University, Baton Rouge, LA 70803, USA
c
Laboratoire d'

ecologie des hydrosyst
emes naturels et anthropis

es, UMR CNRS 5023, Universit

e Claude Bernard Lyon 1, Boulevard du 11 novembre 1918,
F69622 Villeurbanne Cedex, France
d
Geological Survey Canada Atlantic, Dartmouth, Nova Scotia B2Y 4A2, Canada

a r t i c l e i n f o a b s t r a c t

Article history: Here we evaluate the changes in dinoflagellate cyst assemblages and intraspecies morphometric vari-
Available online 21 August 2016 ability in dominant Black Sea species during MIS 5
1 in sediment recovered from DSDP Site 380.
Twenty-three taxa represent the majority of the 16,000 cysts tabulated. Qualitative assessment of
Keywords: assemblage composition indicates that the taxa are distributed in three ecological assemblages; marine
Dinoflagellate cysts stenohaline, marine euryhaline, and brackish Ponto-Caspian endemic species; the fluctuation of these
DSDP Site 380
assemblages provide evidence for changes in environmental conditions from the end of MIS 5e to MIS 1.
Brackish
Although some fluctuations in sea-surface salinity are highlighted by changes in dinoflagellate assem-
Paleosalinity
Sea level
blages, the continuous presence of Pyxidinopsis psilata, Galeacysta etrusca and Spiniferites cruciformis
likely indicate brackish conditions for most of the interval sampled at this site. Statistical analysis of
measured morphometric changes in Galeacysta etrusca endocyst:ectocyst ratios indicate that the spec-
imens recovered mostly belong to a smaller morphotype forming group ‘b’ of Popescu et al. (2009),
interpreted as indicative of brackishemarine conditions after migration to the Mediterranean from the
Paratethys. Although not fully conclusive because of sampling constraints and limited chronological
control, the lack of sudden, large abundance changes in marine salinity-tolerant species the relatively
low species diversity, and the sequence of sapropelic layers relative to banded clastic muds provides
evidence against the occurrence of sustained marine flooding between ca. 0.07e0.007 Ma and it is not
clear that the last interglacial marine highstand during MIS 5e was fully sampled at this site.
© 2016 Elsevier Ltd and INQUA. All rights reserved.

1. Introduction
The salinity of the Black Sea is controlled by the inflow of marine
water through Marmara Sea and its shallow straits (Fig. 1) during
marine highstands, and by precipitation and the volume of runoff
from vast rivers draining Europe and Western Russia during and
immediately after glacial marine regressions (Major et al., 2006)
The unusual configuration and unique oceanographic features of
the semi-isolated Black Sea makes it a fascinating biome to study
unusual dinoflagellate cysts (Mudie et al., 2007; Ivanova et al.,
* Corresponding author. Department of Geology and Geophysics, E-235 Howe-
Russell, Louisiana State University, Baton Rouge, LA 70803, USA.
E-mail address: swarny@lsu.edu (S. Warny).
2012; Yanko-Hombach et al., 2013, 2014) that have been able to
adapt to large-scale salinity fluctuations and can tell us about these
past environmental changes. Some of these cysts are relics of the
Pliocene Paratethyan Seas, while others are more recent Mediter-
ranean immigrants (Marret et al., 2015). Here we provide new data
on late Pleistocene to early Holocene dinoflagellate cyst assem-
blages recovered from DSDP Site 380 in deep water at the base of
the western continental slope apron, 2107 m water depth.
Morphological changes seen in two of the most abundant relic
Paratethyan endemic dinoflagellate cyst species present, Galeacysta
etrusca and Spiniferites cruciformis, are combined with dinoflagel-
late assemblage data to evaluate a possible morphological response
of the cysts to changes in sea-surface salinity and to examine the
impact of sea-level changes and the history of the connections.

http://dx.doi.org/10.1016/j.quaint.2016.07.035
1040-6182/© 2016 Elsevier Ltd and INQUA. All rights reserved.
118 S. Ferguson et al. / Quaternary International 465 (2018) 117e129
Fig. 1. Location of the study area showing DSDP Site 380 in proximity to the connections between the Eastern Mediterranean Sea and the Marmara Sea.
Furthermore, discriminant analysis is performed to verify the
morphometric variations between these two different dinoflagel-
late species. DSDP Site 380 (Fig. 1) was selected for its proximity to
the narrow Bosphorus Strait as drops in eustatic sea level during
the glacials disconnected the Black Sea from Marmara and the
Aegean Sea, creating a change from higher salinities to more
brackish water conditions that should be evidenced at the studied
site. These salinity changes were further overprinted by fluctua-
tions in evaporation (E) relative to total freshwater input (P)
derived from runoff, melt water and rainfall and will be discussed
as well (Major et al., 2006; Wegwerth et al., 2016).
Debate exists on the exact timing of connections between the
low salinity Black Sea and more saline Mediterranean between the
last interglacial maximum (high salinity phase, MIS 5e) and the
Holocene interglacial (MIS 1). Two main scenarios exist for the Late
PleistoceneeHolocene marine reconnection: 1) a rapid (Cata-
strophic Flood model (Ryan et al., 1997, 2003; Major et al., 2006;
Lericolais et al., 2007; Soulet et al., 2011) around 8.4 ka BP, and 2)
the Progressive or Gradual Flood model starting around 11 ka BP
(Hiscott et al., 2007; Aksu et al., 2002a). In addition to debates over
the timing of the Holocene marine transgression in the Black Sea,
there are significant disagreements regarding the timing and
manner of marine reconnections between the Aegean, Marmara
and Black Seas during MIS 3 and regarding the salinity of the Black
Sea after MIS 5e (Karangatian highstand 130e115 ka BP). Benthic
foraminifera (Yanko-Hombach, 2007) and some mollusk data (e.g.
Aksu et al., 2002b; Panin and Popescu, 2007) indicate a marine
incursion in the southwestern Black Sea during MIS 3 (Tarkanku-
tian beds, dated 40e27 ka BP; Surozhian interstade 40e25 ka BP)
and show that during the Karangatian, salinity was higher, being
comparable to that of the present Mediterranean, ranging from ca.
30e37 psu (practical salinity unit). Dinoflagellate cyst assemblages
in the southeastern Black Sea (Shumilovskikh et al., 2013) also
indicate a higher salinity and temperature for MIS 5e. Oxygen
isotope records from Sofular Cave (Badertscher et al., 2011) indicate
marine connections only during MIS 5e and the Holocene, as also
evident in the absence of an Aegean Sea sapropel between the end
_
of MIS 5 (71 ka BP) and early MIS 1 (Işler et al., 2016). In contrast,
there are three MIS 5 sapropel layers in the northern Aegean Sea
(S3, S4 and S5), with well constrained median termination ages at
_
70.8, 96.2 and 121.0 ka BP (Işler et al., 2016).
2. Regional setting
The Black Sea is a large, deep semi-enclosed basin located be-
tween southeastern Europe and Asia (Fig. 1). The Black Sea is
thought to have evolved from a freshwater lake to a brack-
ishemarine basin around the Oligocene e Miocene (Muratov et al.,
1978). From the Miocene onwards, the basin has been characterized
by successive fluctuations between brackish to fresh water condi-
tions (Muratov et al., 1978). The present marine connection to the
Mediterranean Sea is complex as the shallow Bosphorus Strait first
connects the Black Sea to the Marmara Sea, which then connects to
the Aegean and Mediterranean seas through the deeper Darda-
nelles strait (Fig. 1). The Bosphorus strait is relatively shallow
(average 40 m), with a minimum depth of about 30 m near Istanbul,
Turkey. The depth of this connection limits the salinity, stratifica-
tion and oxygen exchange within the Black Sea (Kerey et al., 2004).
In addition to this marine connection, the Black Sea is connected to
the almost freshwater (<11 psu) Sea of Azov through the narrow
Strait of Kerch. At a maximum depth of 15 m, the Sea of Azov is the
presently shallowest sea in the world. At present, it provides a
constant freshwater outflow to the Black Sea via the Don River
(Kosarev et al., 2008) and it was probably a large river valley during
intervals of lower sea level. However, during the late Pleistocene, it
was connected to the Caspian Sea via the Manych River valley
floodway (Major et al., 2006) from which reddish-brown “choco-
late” clays flooded into the northern and western Black Sea around
16e18 cal ka BP (Major et al., 2006). Although the Black Sea hy-
drographic configuration is largely controlled by fluvial inputs from
 S. Ferguson et al. / Quaternary International 465 (2018) 117e129
the discharge of other large rivers such as the Danube, Bug, Dniepr,
Dniestr, and the Rioni, the shallow connection to the Marmara Sea
plays a pivotal role in the hydrographic balance of the Black Sea.
This precarious situation has lead to periodic interruption and
resumption of marine flow between the two bodies of water.
3. Material and methods
DSDP Site 380 of Leg 42B is located at 42 05.940 N and
29 36.820 E in a water depth of 2107 m in the southwestern Black
Sea north of the Bosphorus Strait (Supko et al., 1978). Site 380 refers
to DSDP Hole 380 that was drilled in 1975 during DSDP Leg 42B,
using conventional drill-hole coring techniques which penetrated
to 370.5 m sub-bottom and recovered 159.5 m of sediment by
continuous coring of 40 core barrels, each of length 9.5 m. Only the
top 70 m of Site 380 (Cores 1e8) were analyzed for this project. The
units sampled (Units 1b to 1d of Ross et al., 1978) are from Unit 1
(Cores 1e5) composed of terrigenous sediments including muds
and sandy silts, and from Unit II, a diatomaceous mud, including
intervals of marine diatom species in Cores 6 and 8, and calcareous
oozes in Cores 7, 8, and at the 6/5 boundary. The lower half of Core 7
was tentatively assigned to the Emiliana huxleyi zone. The age of the
base of Unit 1d was estimated as ~100,000 yr BP (late MIS 5) and
the Unit 1d/c boundary at 42 m (Core 5, base section 4) was
assigned an age of 70,000 yr BP (Muratov et al., 1978) which is near
the start of glacial stage MIS 4.
Although Muratov et al. (1978) do not specify the basis for the
age assignments, it appears from the shipboard descriptions
(Ross et al., 1978) that it is based on the appearance of E. huxleyi
(with, at that time, a youngest first occurrence datum of 110 ka BP).
The assignment of a 70 ka BP age for the Unit 1d/c boundary is
presumably based on correlation with the termination of the fluc-
tuating warmerecooler intervals of interglacial MIS 5 and the start
of glacial conditions of MIS 4 (now dated at 71 ka BP by the Lisiecki
and Raymo (2005) marine isotope record). We support these
tentative age assignments based on the occurrence of the highest
pre-Holocene sapropelic mud (TOC > 0.6%) near the top of Core 5
and tentatively correlated with sapropel S3 (median termination
age 70.8 ka BP in North Aegean). A second sapropelic mud interval
(presumably S4 equivalent) occurs in Core 6 and a third sapropel
occurs at the bottom of Core 7 down to 76 m that would logically
correspond with Aegean sapropel S5 (121.0 ka BP). Subunit 1c
(2e42 m) are muds and sandy silts, the top of which is correlated
with Unit III of Ross et al. (1974) and which includes some sapro-
pelic mud (<0.9%) in Core 4, suggestive of the warmer MIS 3
interstadial phase. Subunit 1b (0e2 m) consists of olive-gray sap-
ropelic mud with pinstripe laminae from 0e150 cm and is corre-
lated with Unit II of Ross et al. (1974). Below 150 cm in Subunit 1b,
there is bioturbated mud, then massive dark gray muds interlay-
ered with light brown interbeds that continue to the base of Core 1
and possibly correspond with the late MIS2 chocolate clays of Black
Sea piston cores. These reddish clastic units are not present in the
Subunit 1c sediments of Core 2 (9.5e19 m) but they re-appear as
thicker beds in Core 4 (27.5e38 m) below the sapropelic mud. Core
3 at DSDP 380 was void (Ross et al., 1978). Twenty grams of sedi-
ment were sampled at 50 cm intervals wherever available between
depths of 67.47 and 0.52 m. Poor recovery from 59 to 54 m,
47.5e43 m, and no recovery from 29.5 to 19 m hindered sampling,
however 40 samples from the least disturbed sediments were
selected as suitable for analysis. Samples were processed using the
Cour (1974) method. All samples were treated with HCl and HF to
digest unwanted carbonates and silicates. ZnCl2 (density of 2.00 g/
ml) was used as a density separator to float and concentrate the
palynomorphs. The residues were sieved at 10 mm, and the
remaining fraction was mounted in glycerol (density of 1.23 g/m3).
119
An Olympus BX41 microscope and oil immersion 60 or 100
magnification objectives were used for dinoflagellate cyst analysis.
Microphotographs were taken using a QCapture camera with a
0.5 coupler, and managed with QCapture software. A minimum of
300 identifiable dinoflagellate cysts were counted on each slide.
Reference slides and bibliographic material from the Center of
Excellence in Palynology (CENEX) library were used to identify the
cysts at the species level. Identification of the dinoflagellate cysts
was based on comparison with curated dinoflagellate cysts from
the CENEX-SHELL collection and on published Black Sea studies
including those of Marret et al. (2009), Verleye et al. (2009), and
Caspian studies of Marret et al. (2004). However, for morphotypes
of Lingulodinium machaerophorum we used the classification
similar to Marret et al. (2004) for Caspian morphotypes C and D
with processes greater or less than ca.7 mm. Precisely, we divided
the L. machaerophorum into two sub-groups: “long” processes
(>5 mm but <15 mm), and “short” processes (<5 mm). No “true long”
processes as defined by (Mertens et al., 2009), with an average
process length of 15 mm or greater were observed. Therefore, the L.
machaerophorum processes present in this section are on average
smaller than L. machaerophorum processes present in the Black Sea
in other studies.
Dinoflagellate cysts recovered were grouped according to their
ecological requirements as follows (a) marine stenohaline taxa
typically restricted to marine sea-surface conditions (Ataxiodinium
choane, Impagidinium spp., Impagidinium caspienense, Impagidinium
inaequalis, Spiniferites bentorii sensu stricto, and Spiniferites mir-
abilis; (b) marine euryhaline species tolerating wider ranges of sea-
surface conditions such as Nematosphaeropsis labyrinthus, other
Spiniferites spp. (excluding Spiniferites cruciformis), Achomosphaera
cf. andalusiensis, and Lingulodinium machaerophorum (form with
long processes); c) brackish species endemic to the Paratethyan
basins, including Galeacysta etrusca, Spiniferites cruciformis and
Pyxinidopsis psilata.
Morphological changes for Galeacysta etrusca and Spiniferites
cruciformis were measured by tracking four dimensions; the
lengths and widths of the endocyst (LEN and WEN) and ectocyst (LEX
and WEX) (Fig. 2), following the methodology defined by Popescu
et al. (2009) for the Mediterranean and Paratethys Sea regions.
First, multivariate descriptive analyses (Principal Component
Analysis and Linear Discriminant Analysis) of the morphometric
variability of the two taxa were conducted based on these four
initial variables. Second, according to these first results a time series
analysis of G. etrusca based on three synthetic morphometric de-
scriptors was conducted. These descriptors are (after Popescu et al.,
2009):
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 X ¼ logð 4 Len  Wen  Lex  WexÞ, where X is the log-
transformed geometric mean of the four initial variables; it is
an estimate of the overall size of the dinoflagellate cyst;
 Y ¼ logðLen  Wen=Lex  WexÞ, where Y is the log-transformed
ratio of the endocyst size to ectocyst size;
 Z ¼ logðLen  Lex=Wen  WexÞ, where Z is the log-transformed
ratio of the lengths to widths.
All statistical analyses were conducted using the PAST v. 2.10
freeware (Hammer et al., 2001).
4. Results
4.1. Dinoflagellate cyst distribution at DSDP Site 380
More than 16,000 dinoflagellate cysts were counted during this
study, yielding a total of 23 taxa (11 in situ species and 12 reworked
species). The preservation of specimens was moderate to good, and
120 S. Ferguson et al. / Quaternary International 465 (2018) 117e129

Fig. 2. The four measured dimensions illustrated on scanning electron microscope (SEM) images of Galeacysta etrusca (left), and Spiniferites cruciformis (right).

particular effort were made to identify the “brackish endemic
species” as defined by Popescu et al. (2009) because they show a
large morphologic spectra that appear to be related to sea-surface
physical parameters including temperature, salinity and nutrient
content variability (Popescu et al., 2007, 2009). Dinoflagellate cysts
recovered include eight marine taxa (Fig. 3), with varying amounts
of stenohaline and euryhaline marine species and alternating with
larger amounts of brackish Paratethyan basin species, and some-
times with large amounts of reworked taxa. The marine species
(Fig. 3) mostly show a low relative abundance (between 1% and
12%), except in two deeper samples, where these species respec-
tively reach 16% (at 62.95 m) and 42.3% (at 67 m). The overall
assemblage at DSDP Site 380 is clearly dominated by high abun-
dance of the brackish taxa (Fig. 3; Plate 1), with Pyxidinopsis psilata
and Galeacysta etrusca making up to 90% of relative abundance, and
Spiniferites cruciformis achieving its maximum relative abundance
(less than 20%) between 0 and 10 m depths.
Additional species recovered of reworked taxa include Areoli-
geracean taxa (Cretaceous e Paleogene) (Williams et al., 1998),
Oligosphaeridium spp. (Upper Cretaceous e Paleogene) (Williams
et al., 1998), Paleogene e Eocene Enneadocysta sp. (Fensome
et al., 2007), Eocene Glaphyrocysta spp. (Williams et al., 1998),
Homotryblium sp. (Paleogene e Miocene) (Williams et al., 1998),
Late Eocene e Early Oligocene Deflandrea phosphoritica (Williams
et al., 1998), Wetzeliella symmetrica (Oligocene) (Williams et al.,
1998), Hystrichokolpoma sp. and Virgodinium sp. both with a last
appearance in the Miocene (Süto} -Szentai, 1986; Warny and Wrenn,
2002), and Miocene e Early Pliocene Spiniferites bentorii pannonicus
}-Szentai, 1986).
(Süto
Four dinoflagellate cyst zones were identified based on dino-
flagellate cyst distributions and relative abundances of paleoeco-
logical groups (Fig. 3).
Dinoflagellate cyst zone 1 (68e67 m) is defined by the highest
relative abundance of marine dinoflagellate cysts (42.3% relative

Fig. 3. Synthetic diagram showing relative abundances of the main marine, and brackish dinoflagellate species recovered from DSDP Site 380 as well as reworked dinoflagellate
cysts in samples. Lithology and lithologic subunits are from Ross et al. (1978). Core 3 was not recovered during the drilling and therefore was not sampled. Note that the top sample
is 0.52 m subsurface, not the seabed. Age estimates were assigned based on top and bottom dates provided by Hsü (1978) as well as the age model provided in Ferguson (2012).
 S. Ferguson et al. / Quaternary International 465 (2018) 117e129 121

Plate 1. Light photomicrographs of DSDP Site 380 dinoflagellate cysts with sample depth (slide number) and England Finder coordinates for each specimen. Scale bar ¼ 20 mm. 1e8
Galeacysta etrusca 1 60.02 m (2) K12/0; 2 42.45 m (3) R37/0; 3 60.02 m (2) S10/0; 4 34.46 m (2) R22/4; 5 60.02 m (3) M24/2; 6 60.02 m (3) T16/2; 7 60.02 m (2) M26/1; 8 62.50 m (1)
V15/0. 9e16 Spiniferites cruciformis 9 34.46 m (1) W46/1; 10 34.46 m (2) P38/2; 11 8.495 m (1) Q12/2; 12 0.97 m (2) O28/1; 13 2.50 m (2) V28/4; 14 33.5 m (1) J34/0; 15 41.50 m (1)
J19/3; 16 0.97 m (1) T14/4.

abundance at 67.47 m) including Spiniferites mirabilis, Spiniferites
sp., and Lingulodinium machaerophorum. The brackish species
group, with a relative abundance of 57.7%, is dominated by the most
widespread (Spain to Caspian Sea) Paratethyan relic taxon G.
etrusca with 57% (at 67.47 m). Brackish species Spiniferites cruci-
formis and P. psilata were present but in very low amounts with only
two cysts of S. cruciformis and one cyst of P. psilata observed within
this zone. Reworked dinocysts have a relative abundance of 10% at
67 m depth.
Dinoflagellate cyst zone 2 (67e38 m) is characterized by a
reduction in marine taxa which never dominate the studied section
again above dinoflagellate cyst zone 1 and by a change in the
dominant brackish species, with G. etrusca being replaced by P.
psilata by the top of this zone. G. etrusca makes up 99% of the
assemblage (at 62.5 m) in the lower part of the zone, and P. psilata
begins increasing in the middle of the zone and ends up dominating
the assemblage with 84% (at 41 m) towards the end of this zone.
Brackish S. cruciformis remains rare only reaching up to 4% within
this zone. In zone 2, marine taxa reach their highest abundance of
10% total dinoflagellate cysts at 40 m. Euryhaline taxa L. machaer-
ophorum, Spiniferites spp., and Achomosphaera cf. andalousiensis are
present and as well as the stenohaline taxon Spiniferites mirabilis.
Lastly, zone 2 is also marked by the highest and most continuous
relative abundance of reworked taxa within the studied section
reaching its highest presence of 20% at 52 m.
Dinoflagellate cyst zone 3 (38e11 m) is characterized by the
first appearance of stenohaline taxon Ataxiodinium choane, and the
continuance of assemblage dominance by brackish P. psilata. Pyx-
idinopsis psilata has a relative abundance of 95% or more in all but
one sample (82% at 16 m where G. etrusca briefly increases to 6%). S.
cruciformis is rare in this zone with a maximum relative abundance
of 1% at 16 m depth. Marine abundances remain low throughout
zone 3 with euryhaline Spiniferities spp. and L. machaerophorum
cysts appearing most consistently, Organic matter is low
122 S. Ferguson et al. / Quaternary International 465 (2018) 117e129
throughout and sediments are predominantly red and gray banded
muds with traces of “chocolate brown” laminae possibly related to
flood events from Caspian Sea overflow through Manych Strait (see
Major et al., 2006). Reworked cysts range from 0 to 2% within this
zone. Note there is a 9.5 m interval of no sediment recovery in the
middle of this zone (19e28.5 m).
Dinoflagellate cyst zone 4 (11e0.52 m) is characterized by the
highest relative abundance in S. cruciformis recovered in the stud-
ied section (up to 20% at 3.5 m), with an average of 6% throughout
the zone. This zone is also associated with minor increases in ma-
rine taxa including the first appearances of stenohaline taxa
Impagidinium sp., A. choane, and Spiniferites bentorii sensu stricto.
Euryhaline taxa Spiniferites spp., A. cf. andalusiensis, and L.
machaerophorum once again occur consistently throughout this
zone and slightly increase in low abundances here. The relative
abundance of total marine taxa reaches its peak of 6% in the up-
permost sample (0.52 m). Brackish taxon G. etrusca increases
slightly but remains relatively rare and the dinoflagellate cyst
assemblage is dominated by S. cruciformis and P. psilata. The relative
abundance of P. psilata ranges between 99% at 8 m and 62% at 3.5 m.
Reworked dinoflagellate cyst occurrence is low with a maximum
abundance of 2% (at 6.49 m, and 0.52 m) in zone 4.
4.2. Statistical morphometric analysis of Galeacysta etrusca and
Spiniferites cruciformis
Intraspecific changes in dinoflagellate cyst morphology are
known to be linked to changes in controlling environmental con-
ditions, particularly salinity as first indicated by Wall et al. (1973)
for late Pleistocene
early Holocene Black Sea assemblages of S.
cruciformis and P. psilata (as Tectatodinium psilatum). The variable
length of Lingulodinium machaerophorum processes is a well-
established case (Mertens et al., 2009, 2012), but as discussed
above, other species such as G. etrusca and S. cruciformis are also
known to have a morphology influenced by sea surface salinity, SSS
(e.g., Mudie et al., 2001, 2016; Sorrel et al., 2007; Popescu et al.,
2009). Because these brackish water indicator species are abun-
dant at our site, a detailed morphometric analysis was conducted
on the variability of their size, using only well-preserved specimens
(i.e. torn cysts and partially covered cysts were not measured). Four
dimensions (the lengths and widths of the endocyst, LEN and WEN,
and the ectocyst, LEX and WEX) were measured for both Spiniferites
cruciformis (n ¼ 65) and Galeacysta etrusca (n ¼ 1074) (Fig. 2).
Morphometric variability for both species was analyzed: (i)
through a Principal Component Analysis (PCA) and Linear
Discriminant Analysis (LDA) of Log-transformed measurements,
and (ii) by graphing the X, Y, and Z synthetic descriptors (see
Methods).
The PCA shows that the overall morphometric variability of G.
etrusca and S. cruciformis can be decomposed into three principal
components (PCs), which are independent linear combinations of
the four measured dimensions and were used in a previous study of
morphometric variability (Popescu et al., 2009). Together, these
three PC's explain 93.6% of the total variability (Table 1, Fig. 4). PC1
(57.0% of the total variability) is an overall-size descriptor (a large
PC1 value indicates a large overall size for the specimen). PC2 (21.1%
of the total variability) sets the two EN-descriptors against the two
EX-descriptors, thus being an EN:EX descriptor (a large PC2 value
indicate a large EN:EX ratio, i.e., a large endocyst size relative to the
ectocyst size). PC3 explains 15.5% of the overall variability and sets
the two L-descriptors against the two W-descriptors, thus being an
L:W descriptor (a large PC3 is related to a small L:W ratio, i.e. a more
rounded shape). The fourth PC which explains less than 7% of the
overall variability is likely a “noise-parameter” accounting for the
inter-individual variability.
Histograms of G. etrusca and S. cruciformis specimen coordinates
on PC1, PC2 and PC3 are used to evaluate how similar/dissimilar the
two species are morphometrically (Fig. 5). The large amount of
overlapping between the two species clearly gives evidence to their
overall morphometric similarities based on the linear combinations
of the four measured variables as defined by PCA. Both species
project on the same range of PC1 and PC3 values (Fig. 5A and C), but
not on the same PC2 values (Fig. 5B). This indicates that the main
source of morphometrical dissimilarity between the two species is
from the EN:EX-size ratio marking endocyst size relative to that of
the ectocyst. S. cruciformis individuals show more negative PC2-
values, and thus possess a smaller ratio of endocyst size to ectocyst
size than G. etrusca, while they both share similar “overall size”
(PC1), and L:W ratios (PC3).
Based on this first exploratory result, we further evaluated the
ability of the four measured descriptors to accurately separate the
two species through a linear Discriminant Analysis (DA) coupled
with a Hotelling's test (Hotelling's T2 ¼ 353.4, p(H0: same multivariate

65
means) ¼ 3  10 ) for equality between two multivariate means,
showing a highly significant difference between G. etrusca and S.
cruciformis. The DA results show that the two species are
morphometrically distinct, with an overall “leave-one-out” (jack-
knifed) probability of correct classification of 88.67% (Fig. 6).
Finally, mixture analyses were performed for the three syn-
thetic descriptors X, Y, and Z on the five most sampled (abundant)
levels for G. etrusca to identify possible mixtures of morphomet-
rically distinct sub-groups within this species and use it to evaluate
paleo-SSS. As illustrated by the most abundant sampled depth
(n ¼ 298) (Fig. 7), these distributions are unimodal, indicating that
G. etrusca is morphometrically homogeneous throughout our
studied section. This is in contrast to Popescu et al.'s (2009) results
for Zanclean-aged sediment at the same study site
(828.02e841.91 m in Hole 380, which indicated the presence of
four morphometrically distinct G. etrusca groups. The presence of
only one morphometric group for our study is further confirmed
by comparing the [X, Y] distribution of the sampled specimens to
those of Popescu et al. (2009) (Fig. 8). The comparison shows that,
even if including rather small-sized specimens, the G. etrusca
group present in MIS 5
1 estimated age sediment from DSDP Site
380 is similar to G. etrusca group ‘b’ defined by Popescu et al.
(2009) as a marine group, indicating that this site is character-
ized by relatively stable marine salinities at least throughout MIS
5e to 4 where G. etrusca is found in statistically significant abun-
dances. The presence of G. etrusca starts to decline at MIS 5a, depth
of 39.9 m, and continues to have a limited presence for the rest of
the studied section into MIS 1.
5. Discussion
5.1. Environmental reconstruction based on dinoflagellate
assemblages
Based on the environmental affinities described above, the high
abundances of brackish taxa recovered in all samples from DSDP
Site 380 indicate that this area of the Black Sea experienced low
sea-surface salinity ranging between about 12 and 22 psu
throughout most of the interval studied. This observation is in
accord with the persistent presence of Lingulodinium machaer-
ophorum which does not grow in salinities of less than 7 psu, and is
most common in a salinity range of ca. 13e18 psu in the modern
Ponto-Caspian region (Mertens et al., 2012). Despite the low reso-
lution sampling interval of our study, some paleoenvironmental
changes are nevertheless inferred based on observed assemblage
differences.
 S. Ferguson et al. / Quaternary International 465 (2018) 117e129 123

Fig. 4. Principal Component Analysis (PCA) of G. etrusca (crosses) and S. cruciformis (open squares) specimens based on the four measured dimensions (lengths and widths of the
endocyst and ectocyst; Fig. 3). A. PC1 vs PC2; B. PC2 vs PC3. PC1, PC2 and PC3 represent 57.0%, 21.1% and 15.5% of the overall morphometric variability, respectively. Ellipses are
bivariate 70% confidence intervals for G. etrusca (short-dashed curve) and S. cruciformis (long-dashed curve).

Table 1 interglacial interval. This connection caused SSS to increase in the

Summary of the numerical results of the correlation matrix-based PCA of G. Black Sea from a previous glacial stage (MIS 6) salinity of 5 psu up to
etrusca þ S. cruciformis. The three first PCs, together explain 93.6% of the total
22 psu for the southeastern Black Sea (S.-M. Popescu, personal
morphometric variance, and convey morphometric information analogous to the X,
Y and Z synthetic descriptors, respectively. communication). In comparison, today's seawater enters the Bos-
phorus Strait at an average bottom water salinity of 34.9 psu
PC1 PC2 PC3 PC4
(Murray et al., 2005) and mixes with lower salinity surface water as
% Expl. var. 59.55 17.34 16.29 6.82
it pours into the Black Sea to form a bottom water layer of ca. 24 psu
Correlation between original variables and PCs: (Mertens et al., 2012). However, the high freshwater river input
r PC1 PC2 PC3 PC4 relative to evapotranspiration keeps the Black Sea SSS between
LEN 0.804
0.07
0.512
0.295 about 16 psu on the shelves and 18 psu in the basins. The maximum
WEN 0.767
0.581 0.034 0.268
diversity of marine dinoflagellate cysts and the peak occurrence of
LEX 0.76 0.588
0.104 0.258
WEX 0.755 0.074 0.615
0.218 stenohaline taxa also indicates a full marine connection. However,
L. machaerophorum specimens present throughout zone 1 exhibit a
Meaning of the resulting PCs.
PC1: Overall size (large PC1-value <¼> large overall specimen size).
mixture of “long” (>5
<15 mm), and “short” (<5 mm) processes
PC2: EN vs. EX size (large PC2 <¼> small EN/EX-size ratio). (Fig. 9), but no fully developed processes >15 as found in Marmara
PC3: Lengths vs. widths (large PC3 <¼> small L/W ratio). and Aegean Seas. Absence of normal long processes indicates that
PC4: Noise. salinity conditions were relatively low in this part of the Black Sea
following the inflow of marine waters to the formerly isolated
lacustrine environment at the beginning of MIS 5e. High relative
5.1.1. Dinoflagellate cyst zone 1 (68e67 m): interglacial marine abundance of the marine morphometric group of G. etrusca along
environment with the low abundance of P. psilata characterize dinoflagellate
A relative abundance of 42.3% marine taxa at depth 67.47 m, zone 1. The presence of some reworked species in sediments at
combined with the presence of sapropel sediment and coccoliths 67 m may indicate higher erosion of Cenozoic western Black Sea
(reported by DSDP Leg 42B Ross et al., 1978) indicate a connection sediments associated with the Bosphorus inflow and subsequent
between the Aegean Sea and the Black Sea within this latest MIS 5e strengthening of the Rim Current.

Fig. 5. Frequency histograms of G. etrusca (grey) and S. cruciformis (black) specimen coordinates on PC1 (A), PC2 (B) and PC3 (C), illustrating the relatively more dissimilar values
observed between the two species for PC2 (endocyst size vs. ectocyst size) than for PC1 (overall cyst size) and PC3 (lengths vs. widths).
124 S. Ferguson et al. / Quaternary International 465 (2018) 117e129

Fig. 6. Discriminant analysis of G. etrusca (grey) vs S. cruciformis (black) specimens based on the four measured dimensions (lengths and widths of
the endocyst and ectocyst; Fig. 3). For each specimen, the Discriminant Score (DS) is calculated from the linear discriminant function
DS ¼ (22.225  log(LEN)) þ (25.513  log(WEN))
(40.197  log(LEX)) þ (
25.954  log(WEX)) þ 43.3652, clearly evidencing the main source of morphometrical difference
(the endocyst/ectocyst size ratio) as also identified by the PCA (Figs. 4 and 5). The jackknifed confusion matrix gives the number of correctly classified and misclassified
specimens for each species based on a “leave-one-out” procedure.

5.1.2. Dinoflagellate cyst zone 2 (67e38 m): late interglacial
brackish environment with brief marine fluctuations
It is unclear what drives the fluctuations observed between G.
etrusca and P. psilata throughout zone 2 which includes the time
from MIS 5d
MIS 5a. The rarity of stenohaline marine taxa in-
dicates that there was a weaker Mediterranean connection, but the
sapropelic sediments indicate presence of stratified water and not
uniformly brackish or well-mixed conditions. The presence of
common Atlantic/Mediterranean stenohaline Spiniferites spp. are
never found in the Caspian Basin suggesting that the salinity here
was higher than that of the Caspian SSS (ca. 13e15 psu). Reworked
dinoflagellate cysts reach abundances of 20% (at 52 m), which is the

Fig. 7. Gaussian (plain) and kernel (dotted) density distributions of X, Y and Z synthetic descriptors for the 298 measured specimens of G. etrusca from the most abundant level
(depth: 62.5 m) of the analyzed time series. At this depth as for all other tested depths, the three empirical distributions do not show any marked evidence of multimodality,
illustrating the morphometrical homogeneity of G. etrusca in the DSDP Site 380 studied interval.
 S. Ferguson et al. / Quaternary International 465 (2018) 117e129 125

Fig. 8. Biplot of X and Y synthetic descriptors contrasting the Galeacysta etrusca (black dots; dark-grey ellipse: 70% binomial Confidence Interval) and Spiniferites cruciformis (grey
dots and light-grey ellipse) specimens from DSDP Site 380 with the distributions of Popescu et al. (2009) morphometric groups ‘a’e‘d’ (colored ellipses, 70% binomial CI; group ‘a’ is
related to brackish conditions, group ‘b’ to marine conditions, group ‘c’ to freshwater, and group ‘d’ to high nutrient levels).

Fig. 9. Lingulodinium machaerophorum individuals present throughout studied section at DSDP Site 380. Our “long” processes are defined as being >5 mm but <15 mm in length, and
our “short” processes are defined as being <5 mm in length. No “true long” processes being 15 mm (as defined by Mertens et al. (2009)) were present. Dinoflagellate cyst
assemblage zones are shaded.
126 S. Ferguson et al. / Quaternary International 465 (2018) 117e129
maximum borehole depth indicating high river discharge causing
erosion and re-deposition of Cenozoic shelf sediments northwest of
the DSDP 380 drillhole site. More than halfway up of zone 2 (at
48 m), L. machaerophorum reaches an acme again, with a mixture of
long and short processes (Fig. 9), likely marking a brief salinity
increase and is coeval to the decrease of P. psilata and occurrence of
sapropelic mud possibly correlative with MIS 5c. These changes
suggest minor increase in SSS and SST associated with periodic
minor influxes of warmer marine waters, and a marine connection
to the Black Sea just before the Unit 1c/b boundary during MIS 1.
Alternatively, changes in evaporation relative to freshwater supply
by rainfall and rivers may have changed the salinity, as reported for
MIS 4
2 in the southeastern Black Sea (Wegwerth et al., 2016), but
that model would not account for presence of the sapropelic mud
intervals in this zone.
5.1.3. Dinoflagellate cyst zone 3 (38e11 m): cold brackish
conditions with brief salinity rises
Here, two salinity increases are indicated by two episodes of
increased abundance in L. machaerophorum with relatively long
processes (>5e15 mm (Fig. 9); at 33.5 and 16 m depths in DSDP Hole
380. These saline intervals are coeval with the presence of marine
Spiniferites spp., A. cf. andalusiensis, as well as a peak of reworked
dinoflagellate cysts immediately after both episodes and following
glacial stage MIS 4. This may indicate short-lived warmer in-
terstadials that characterize MIS 3, within a zone that appears to be
is predominantly cold with brackish conditions as also recorded by
the low percentage of marine taxa and dominance of P. psilata as
reported for MIS 4 and 2 in the SE Black Sea basin (Shumilovskikh
et al., 2014). Alternatively, these changes may again correspond to
fluctuations in evaporation relative to precipitation during MIS 3 as
reported by Wegwerth et al. (2016). No sapropelic sediment is
present, and TOC values are very low throughout the massive
banded clastic sediments. At the deepwater DSDP Site 380, there is
also a notable absence of freshwater algae throughout this zone, in
contrast to the (Shumilovskikh et al., 2014) record for the SE basin.
5.1.4. Dinoflagellate cyst zone 4 (11e0.52 m): late glacial brackish
waters with minor interglacial salinity influence
Although brackish P. psilata remains the predominant species
and reworked dinoflagellate cysts are rare (2% maximum), there are
peaks of Spiniferites cruciformis that has a wider salinity range in
surface sediments of the present-day Ponto-Caspian basins
(Zonneveld et al., 2013; Marret et al., 2015). Several minor marine
influxes in zone 4 are also indicated by the presence of marine
stenohaline taxa Impagidinium sp., Ataxiodinium choane, and S.
bentorii, along with an increase in L. machaerophorum with “long”
processes (Fig. 9) in many samples (Fig. 3). The S. cruciformis acme in
this zone (ca. 20%) is notable and corresponds to the well known late
Pleistocene intervals described for Marmara Sea and other Black Sea
cores (e.g. Mudie et al., 2004; Marret et al., 2009; Shumilovskikh
et al., 2013). The low diversity of total dinoflagellate species in this
zone and absence of peridinioid taxa which characterize both
interglacial stages MIS 5e and MIS 1 (Shumilovskikh et al., 2013),
show that the DSDP Site 380 samples do not include the mid-late
Holocene sediments where diversity increases to around 34 taxa
(Marret et al., 2009; Bradley et al., 2012).
5.2. Environmental significance of the assemblage recovered from
DSDP Site 380
Galeacysta etrusca, one of the most common species recovered at
our site, is known to occur as four biometrically determined mor-
photypes in relation to qualitative assessments of salinity or nutrient
variability (Groups ‘a’e‘d’ of Popescu et al., 2009). In their paper,
Popescu et al. (2009) indicate that the species most likely originated
from Pannonian basins (Hungary) and/or from the Eastern Paratethys
basins (Southern Romania, Black and Caspian Seas) and later
migrated to the Mediterranean Sea during late Miocene and early
Pliocene highstands. These periodic widespread migrations into the
Mediterranean possibly imply a greater sea surface temperature
(SST) tolerance and opportunistic species characteristics for this
taxon compared to the other brackish Paratethyan species.
The second endemic taxon recovered in large quantity within
this study is Spiniferites cruciformis. This species was first described
for samples of Black Sea sediments by Wall et al. (1973) and was
attributed an early Holocene age. However, since then this species
has been reported for late Miocene to early Pliocene sediments in
the Black Sea at Site 380 (Popescu, 2006), southern Romania
(Popescu, 2001), latest Miocene of Italy (Bertini, 1988), in late
Pleistocene e early Holocene sediments from Greece (Kouli et al.,
2001), and in PleistoceneeHolocene sediments in the Black, Aral,
Marmara and Caspian Seas (Mudie et al., 2001, 2002; Marret et al.,
2004; Sorrel et al., 2006; Leroy et al., 2007; Londeix et al., 2009). In
late Pleistocene sediments of Marmara Sea, this species and P. psi-
lata increase during the BøllingeAllerød warm periods (ca.
14.7e12.7 ka BP), indicating a preference for warmer surface waters
during glacialeinterglacial stage transitions (Londeix et al., 2009;
Shumilovskikh et al., 2013, 2014). The sea-surface salinity prefer-
ence of S. cruciformis is a matter of debate. Wall and Dale (1973)
described it as a “plastic” species that can present important
morphological differences. It is classically accepted as a brackish
water species (where brackish is defined as a salinity of ca. 3e11 psu
by in (Mudie et al., 2011, 2016) except by Kouli et al. (2001) that
argue for a freshwater preference of S. cruciformis based on its
occurrence in a freshwater lake in Greece. Leroy and Albay (2010)
also show that small amounts of S. cruciformis form 3 can live in
the Turkish freshwater Lake Sapanca. However, Mudie et al. (2001)
show that the salinity range of S. cruciformis form 1 in early Holo-
cene sediments of Marmara Sea is ca. 12e22 psu, as measured by
calibration with oxygen isotope values of co-occurring planktonic
foraminifera. Present-day distribution of S. cruciformis includes
sites in the fully marine Mediterranean Sea (Zonneveld et al., 2013).
The last so-called Paratethyan endemic species to dominate the
assemblage is Pyxidiniopsis psilata, which is known to appear since the
late Miocene in the Black Sea (Popescu, 2008), as well as in Black Sea
MIS 6 sediments (Shumilovskikh et al., 2013), in late Pleistocene
sediments from the Marmara Sea (Mudie et al., 2001, 2002, 2004;
Londeix et al., 2009), and in Black Sea sediments before the last con-
nections between the Marmara and Mediterranean seas which
occurred around 10 cal ka BP (Mudie et al., 2001; Popescu, 2008;
Verleye et al., 2009; Bradley et al., 2012). Generally, this species in-
creases during glacial periods possibly indicating a preference for
cooler surface waters (Popescu, 2008) but it is found in modern sed-
iments of the Caspian Sea where annual SST is 16e18 C and in sub-
tropical water off Florida (Zonneveld et al., 2013). The presence of this
species in the Black Sea where the sea surface salinity ranges between
7 and 20 psu, and its sparsity in the Caspian Sea were its associated SSS
ranges between 5 and 12 psu (Mudie, unpublished data) indicates a
probable SSS range between 7 and 12 psu as suggested by it concur-
rence with L. machaerophorum for the early Holocene of the southwest
Black Sea (Marret et al., 2009; Mertens et al., 2012).
Lingulodinium machaerophorum is one of the few cosmopolitan
marine dinoflagellate cyst species recovered in this study. As dis-
cussed previously, this species is known to show a wide range of
process lengths, which is correlated with salinity (Mertens et al.,
2009, 2012). Its presence has been recorded in coastal sediments
from subtropical to tropical regions of the Southern Hemisphere,
and from tropical to temperate regions in the Northern Hemi-
sphere. It is distributed in marine environments characterized by a
 S. Ferguson et al. / Quaternary International 465 (2018) 117e129
sea surface temperature (SST) between about 1.5  C (winter SST;
rarely found from 0 to
1.5 ) and from >10e29.1  C (summer SST),
with sea surface salinity (SSS) between 8.5 and 39.5 psu (summer e
autumn), and with nutrient ranges of 0.06e1.1 mM for phosphate
and 0.04e12.0 mM for nitrate (Marret and Zonneveld, 2003,
updated in Zonneveld et al., 2013). This species is common in low
salinity environments such as the Black Sea and Marmara Sea (Wall
and Dale, 1973; Wall et al., 1973; Mudie et al., 2001, 2004, 2011;
Marret et al., 2009), Caspian Sea (Marret et al., 2004), Baltic Sea
(Ellegaard, 2000), or in periodically hypersaline conditions of the
endorheic Aral Sea (Sorrel et al., 2006). L. machaerophorum appears
to respond to salinity stress by a reduction in process lengths from
long spines to short bulbous processes, and therefore presence of
shortened processes is commonly used as a proxy for past sea-
surface salinity (Mudie et al., 2001; Sorrel et al., 2006; Marret
et al., 2009; Mertens et al., 2009; Verleye et al., 2009). Long pro-
cesses are present during times of relatively normal marine salinity
and little to no stress, while short/bulbous processes indicate times
of low salinity and severe stress to the taxa (Mertens et al., 2009;
Verleye et al., 2009). Most of these studies (Mudie et al., 2001;
Marret et al., 2009; Mertens et al., 2012) classified L. machaer-
ophorum with processes longer than >15 mm as long processes, and
short processes ranging from <5 to 10 mm in length. However, in
our study, it is notable that true long processes were not present,
and that (as noted in the method section) the “long” processes
reported herein are >5 but <15 mm, with short processes are
defined as being <5 mm. The absence of normal long processes
(>15 mm) at DSDP Site 380 may relate to its position downstream
from the Danube Shelf where modern populations of L. machaer-
ophorum have short processes within the river discharge plume,
and indicates low salinity for the entire record (Mudie, unpublished
data). Also notable is the sparse occurrence of L. machaerophorum
cyst with bulbous processes in the DSDP Hole 380 samples; this
morphotype is rare in the Caspian Sea but it is common in more
saline waters of the Black Sea (Marret et al., 2015).
Regarding S. cruciformis and other cruciform dinocyst species,
Eaton (1996) suggested that this rare endocyst shape may be a
stress response to low salinity waters of less than 12 psu. Com-
parisons of G. etrusca and S. cruciformis distributions at DSDP Site
380 do show that cruciform taxa are correspondingly more abun-
dant at different depths, but without measurement of morpho-
metric heterogeneities within G. etrusca in the style of Popescu
et al. (2009), conclusions about endocyst shape and salinity
changes remain tenuous. Our PCA analysis of S. cruciformis mor-
phometrical variation based on endocyst: ectocyst size ratios
(Fig. 4) would also indicate a more complex relationship with SSS,
at least for MIS 5
1, as these species appear in salinity possibly
higher than 12 psu, and S. cruciformis with typical morphology can
be present in surface sediment of Marmara and the Mediterranean
Sea at salinities up to at least 29 psu (Zonneveld et al., 2013). Mudie
et al. (2001) encountered the same problem for the S. cruciformis
morphotypes found within Holocene sediments of the Black Sea
and Marmara Sea region where five morphotypes were identified
based on process lengths and development of sutural septa. Neither
multivariate regression of abundance against foraminiferal oxygen
isotope estimates of salinity nor comparisons to the sea-level curve
allowed them to propose concrete salinity reconstructions for the
different morphotypes other than an overall estimation of brackish
to low salinity marine conditions.
5.3. Marine reconnections and chronology on and timing of the last
reconnection
The main problems in trying to reconstruct a marine recon-
nection model from our DSDP Site 380 dataset are largely due to the
127
limited sample distribution, as well as the poor recovery, lack of
high resolution magnetostratigraphy, and limited microfossil age
markers for this DSDP Site. The drilling did not recover the mid-late
Holocene sediment which contains a higher diversity of dinofla-
gellate species (ca. 25e30), including many heterotrophic taxa
(Marret et al., 2009; Bradley et al., 2012; Shumilovskikh et al., 2013)
such as the Holocene endemic species Peridinim ponticum that has a
first appearance around 5.5 ka BP in the Black Sea. Furthermore no
samples were available between 28 and 17 m (most of MIS 3 ac-
cording to our age estimates), preventing precise age reconstruc-
tion and calculation of marine reconnection rates. Another issue is
the lack of an oxygen isotope curve from the Black Sea because of
the absence of planktonic foraminifera in the low salinity (<18 psu)
Black Sea although they live in the semi-marine Marmara Sea
where SSS is ca. 22e26 psu. Despite these limitations, some ob-
servations can be made by comparison with more complete late
Pleistocene sections elsewhere in the Black Sea. Verleye et al.
(2009) proposed that the Black Sea was close to fully connected
to the Aegean at 7.97 ka BP as evidenced by the presence of 95% of
marine dinoflagellate cysts (Operculodinium centrocarpum sensu
strictum and Spiniferites spp.) at a slope site southeast of Bosphorus
Strait, in the path of the marine bottom water overflow. Such
abundances of marine species were not seen at the deeper water
DSDP Site 380 north of Bosphorus Strait which is at the terminus of
the counterclockwise circulation of both surface and bottom water
in the Black Sea, but a small increase in marine palynomorphs was
observed near the top of dinoflagellate cyst zone 4. This event is
marked by increases in in situ euryhaline marine species (Spinifer-
ites spp. A. cf. andalusiensis [now extinct throughout the region but
present in latest Pleistocene] and L. machaerophorum) along with
the appearance of stenohaline marine species Ataxiodinium choane,
and the highest abundance of L. machaerophorum with “long”
processes.
Comparison can also be made with the palynological studies of
MIS 5e and MIS 4
2 in deep basins of the southeastern Black Sea
(Shumilovskikh et al., 2013, 2014; Wegwerth et al., 2016). These
studies show that in this part of the Black Sea, both the Eemian and
Holocene interglacials are marked by a succession from lacustrine
brackish taxa (P. psilata, S. cruciformis, and Caspidinium rugosum) to
more marine assemblages (L. machaerophorum, Spiniferites ramo-
sus complex and up to 20 other autotrophic and heterotrophic
dinoflagellate cyst taxa) which is attributed to inflow of saline
Mediterranean water. During MIS 5e, this marine connection per-
sisted from ca, 128e119 ka BP, with SSS reaching ca, 28e30 psu, and
between 127.5 and 121 ka BP a relatively high SST is evident due to
the high abundance of Tuberculodinium vancampoae and Bitecta-
todinium tepikiense (Shumilovskikh et al., 2013). In contrast, the
absence of these warm water indicators, lower diversity and
absence of heterotrophic taxa at DSDP Site 380 suggests that our
studied samples did not cover the full suite of interglacial
conditions.
For the interval from 64 to 25 ka BP in the southeastern Black
Sea, there is a comparable dominance of Pyxidinopsis psilata and
Spiniferites cruciformis in the dinocyst record, indicating brackish
water, with a decrease in marine indicators around 54 ka BP and a
further freshening of Black Sea from ca, 32
25 ka BP
(Shumilovskikh et al., 2014). Notably, species diversity in the SE
Black Sea records shows higher diversity of autotrophic species, but
G. etrusca is rare or absent throughout this densely sampled interval
(152 levels) and Caspidinium rugusum, a Caspian endemic dinofla-
gellate cyst is commonly present in contrast to its absence at Site
380. In the SE basin, there is also an almost ubiquitous presence of
freshwater algae which is not found at the deeper water location of
DSDP Hole 380. High resolution pollen studies on the southeastern
core 25GC1 from 418 m water depth (Shumilovskikh et al., 2014)
128 S. Ferguson et al. / Quaternary International 465 (2018) 117e129
show there were alternating arid/humid phases in the region which
were expressed as minor salinity and water level changes, with
interstadials having higher dinoflagellate cyst concentrations
probably related to increased temperature, runoff, and nutrients.
Drier conditions with increased salinity mark stages MIS 4 and MIS
2 and stadials within MIS 3. Overall, there is no indication of a full
marine reconnection in the eastern Black Sea basins between the
end of MIS 4 and MIS 2.
Minor oscillations in dinoflagellate species are observed in the
low resolution study of MIS 4
2 at DSDP Site 380, despite the
missing Core 3 interval that represents the middle part of MIS 3
(probably ~54e44 ka BP). The observed increase in marine taxa and
appearance of the typical deep water Mediterranean taxa Impagi-
dinium sp. and Nematosphaeropis clearly indicate some brief influx
of marine water to the western region which is not recorded in the
SE basin cores of Shumilovskikh et al. (2013, 2014) where N. laby-
rinthus disappears after MIS 5e. This brief salinity increase appears
to correspond to the Tarkankutian and Surozhian interstadial in-
tervals dated ca. 40e27 ka BP and is marked by a peak occurrence of
Pterocysta cruciformis on the SW Black Sea shelf (Rochon et al.,
2002). P. cruciformis is very rare at the deep water DSDP Sites, as
also found in a deep basin core from Marmara Sea where it is re-
ported for the late Pleistocene (Londeix et al., 2009).
6. Conclusions
The dinoflagellate cyst analysis of late PleistoceneeHolocene
samples from DSDP Site 380 located at the base of the western
Black Sea's continental slope is constrained by low resolution
sampling and limited age control, but it provides further evidence
for past connections between the Black and Mediterranean Seas
and the paleoenvironmental changes from late MIS 5 to early MIS
1. Brackish species Pyxidinopsis psilata, Galeacysta etrusca, and
Spiniferites cruciformis were the dominant taxa found in all sam-
ples from Site 380. P. psilata is more abundant during glacial
periods, while G. etrusca and S. cruciformis are associated with
warmer sea surface temperatures at the end of MIS 5 and end of
MIS 2, respectively. Species with a primarily marine affinity
mainly include Lingulodinium machaerophorum, Spiniferites ben-
torii, Achomosphaera cf. andalousiensis, and Nematosphaeropsis
labyrinthus; these common Mediterranean taxa are present dur-
ing the later interstadial phases of the interglacial MIS 5, and
together with presence of sapropelic muds, indicate at least
intermittent connections between the Black and Mediterranean
Seas at times of high sea levels before the glacial regression of
MIS 4 and 2. The reconnection intervals appear as small differ-
ences in sea surface salinity for the southwestern Black Sea dur-
ing the Last Glacial Climatic Cycle, towards the middle of MIS 3
when the salinity most likely fluctuated between the restricted
ranges of 12 and 22 psu as indicated by the morphometric ana-
lyses of Galeacysta etrusca. No freshwater, low salinity brackish or
high nutrient morphometric subgroups were seen in G. etrusca at
any sampled depths, indicating that the salinity in the western
Black Sea remained brackish to low-salinity marine throughout
the sampled interval at DSDP Site 380. Finally, discriminant
analysis between Galeacysta etrusca and Spiniferites cruciformis
show that the two Paratethyan relic species are morphometrically
distinct, with an overall probability of correct classification of
88.67% based on the four morphologic measurements alone. In
the deep water western Black Sea basin, G. etrusca has its last
peak abundance in MIS 5a whereas S. cruciformis is most abun-
dant in late MIS 2 and early MIS 1. Lower species diversity,
absence of heterotrophic taxa (including endemic Peridinium
ponticum) and absence of thermophiles such as Tuberculodinium
vancampoae apparently distinguish assemblages at the bathyal
western DSDP Sites from those of mid-depth cores from the
southeastern basin.
Acknowledgments
This paper is primarily a summary of S. Ferguson's M.S. thesis.
Thanks are extended to Speranta Popescu for her substantial
guidance and contribution during Ferguson's Master Thesis
(Ferguson, 2012). She thanks her committee members for
constructive remarks, and the LSU Department of Geology and
Geophysics for financial assistance. Thanks are also extended to
Martin Head (Brock University) for assistance with the identifica-
tion of some dinoflagellate cyst species and ranges of reworked
taxa, and to Crawford White for valuable figure and editing support.
We also thank two anonymous reviewers for their constructive and
helpful comments.
References
Aksu, A.E., Mudie, P.J., Rochon, A., Kaminski, M.A., Abrajano, T., Yaşar, D., 2002a.
Persistent Holocene outflow from the Black Sea to the eastern Mediterranean
contradicts Noah's flood hypothesis. GSA Today 12 (5), 4e9.
Aksu, A.E., Hiscott, R.N., Kaminski, M.A., Mudie, P.J., Gillespie, H., Abrajano, T.,
Yaşar, D., 2002b. Last glacialeHolocene paleoceanography of the Black Sea and
Marmara sea: stable isotopic, foraminiferal and coccolith evidence. Marine
Geology 190, 119e149.
Badertscher, S., Fleitmann, D., Cheng, H., Edwards, R.L., Go €ktürk, O.M., Zumbühl, A.,
Leuenberger, M., Tüysüz, O., 2011. Pleistocene water intrusions from the Med-
iterranean and Caspian Seas into the Black Sea. Nature Geoscience 4 (4),
236e239.
Bertini, A., 1988. Palinologia ed aspetti ambientali del versante adriatico dell
Appennino centro-settentrionale durante il Messiniano e lo Zancleano. PhD
thesis. University of Florence, Italy.
Bradley, L.R., Marret, F., Mudie, P.J., Aksu, A.E., Hiscott, R.N., 2012. Constraining
Holocene sea-surface conditions in the south-western Black Sea using dino-
flagellate cysts. Journal of Quaternary Science 27, 835e843.
Cour, P., 1974. Nouvelles techniques de de
tection des flux et de retombe
es polli-
niques: e
tude de la se
dimentation des pollens et des spores  a la surface du sol.
Pollen et Spores 23 (2), 247e258.
Eaton, G.L., 1996. Seriliodinium, a new late Cenozoic dinoflagellate from the Black
Sea. Review of Palaeobotany and Palynology 91, 151e169.
Ellegaard, M., 2000. Variation in dinoflagellate cyst morphology under conditions of
changing salinity during the last 200 years in the Limfjord, Denmark. Review of
Palaeobotany and Palynology 109, 65e81.
Fensome, R.A., Guerstein, G.R., Williams, G.L., 2007. The Paleogene dinoflagellate
cyst genera Enneadocysta and Licracysta gen. nov.: new insights based on ma-
terial from offshore eastern Canada and the Southern Hemisphere. Micropa-
leontology 52 (5), 385e410.
Ferguson, S., 2012. Evaluation of Pleistocene to Holocene (MIS 5 to 1) Climatic
Changes in Southwestern Black Sea: A Palynological Study of DSDP Site 380.
M.S. Thesis. Louisiana State University and Agricultural and Mechanical College,
pp. 1e59.
Hammer, Ø., Harper, D.A.T., Ryan, P.D., 2001. PAST: Paleontological statistics soft-
ware package for education and data analysis. Palaeontolia Electronica 4.
Hiscott, R.N., Aksu, A.E., Mudie, P.J., Marret, F., Abrajano, T., Kaminski, M.A., Evans, J.,
Çakirog lu, A.I., Yaşar, D., 2007. A gradual drowning of the southwestern Black
Sea shelf: evidence for a progressive rather than abrupt Holocene reconnection
with the eastern Mediterranean Sea through the Marmara Sea Gateway. Qua-
ternary International 167, 19e34.
Hsü, K.J., 1978. Stratigraphy of the lacustrine sedimentation in the Black Sea. In:
Ross, D.A., et al. (Eds.), Initial Reports of the Deep Sea Drilling Project, 42(2),
pp. 509e524.
Ivanova, E.V., Murdmaa, I.O., Karpuk, M.S., Schornikov, E.I., Marret, F., Cronin, T.M.,
Buynevich, I.V., Platonova, E.A., 2012. Paleoenvironmental changes on the
northeastern and southwestern Black Sea shelves during the Holocene. Qua-
ternary International 261, 91e104.
_
Işler, E.B., Hiscott, R.N., Aksu, A.E., 2016. Late Quaternary chronostratigraphy of the
Aegean Sea sediments: special reference to the ages of sapropels S1eS5. Turkish
Journal of Earth Sciences 25, 1e18.
Kerey, I.E., Meriç, E., Tunog lu, C., Kelling, G., Brenner, R.L., Dogan, A.U., 2004. Black
SeaeMarmara Sea quaternary connections: new data from the Bosphorus,
Istanbul, Turkey. Palaeoceanography, Palaeoclimatology, Palaeoecology 204,
277e295.
Kosarev, A.N., Kostianoy, A.G., Shiganova, T.A., 2008. The sea of Azov. In:
Kostianoy, A.G., Kosarev, A.N. (Eds.), The Handbook of Environmental Chemis-
try, 5, pp. 68e89.
Kouli, K., Brinkhuis, H., Dale, B., 2001. Spiniferites cruciformis: a fresh water dino-
flagellate cyst? Review of Palaeobotany and Palynology 113, 273e286.
 S. Ferguson et al. / Quaternary International 465 (2018) 117e129
Lericolais, G., Popescu, I., Guichard, F., Popescu, S.-M., 2007. Water-level fluctuations
in the Black Sea since the last glacial maximum. In: The Black Sea Flood
Question: Changes in Coastline, Climate, and Human Settlement, vol. 4,
pp. 437e452.
Leroy, S.A.G., Albay, M., 2010. Palynomorphs of brackish and marine species in cores
from the freshwater Lake Sapanca, NW Turkey. Review of Palaeobotany and
Palynology 160 (3), 18e188.
Leroy, S.A.G., Marret, F., Gibert, E., Chalie
, F., Reyss, J.-L., Arpe, K., 2007. River inflow
and salinity change in the Caspian Sea during the last 5500 years. Quaternary
Science Reviews 26, 3359e3383.
Lisiecki, L.E., Raymo, M.E., 2005. A Pliocene-Pleistocene stack of 57 globally
distributed benthic d18O records. Paleoceanography 20, 1e17.
Londeix, L., Herreyre, Y., Turon, J.-L., Fletcher, W., 2009. Last glacial to Holocene
hydrology of the Marmara Sea inferred from a dinoflagellate cyst record. Re-
view of Palaeobotany and Palynology 158, 52e71.
Major, C.O., Goldstein, S.L., Ryan, W.B.F., Lericolais, G., Piotrowski, A.M., Hajdas, I.,
2006. The co-evolution of Black Sea level and composition through the last
deglaciation and its paleoclimatic significance. Quaternary Science Reviews 25,
2031e2047.
Marret, F., Zonneveld, K.A.F., 2003. Atlas of modern organic-walled dinoflagellate
cyst distribution. Review of Palaeobotany and Palynology 125 (1), 121e200.
Marret, F., Leroy, S., Chalie
, F., Gasse, F., 2004. New organic-walled dinoflagellate
cysts from recent sediments of Central Asian seas. Review of Palaeobotany and
Palynology 129, 1e20.
Marret, F., Mudie, P.J., Aksu, A., Hiscott, R.N., 2009. A Holocene dinocyst record of a
two-step transformation of the Neoeuxinian brackish water lake into the Black
Sea. Quaternary International 197, 72e86.
Marret, F., Mudie, P., Mertens, K.N., Shumilovskikh, L., 2015. Atlas of dinoflagellate
cyst distributions in the Black Sea corridor. Geological Society of America Ab-
stracts and Programs 47 (7), 1e141.
Mertens, K.N., Ribeiro, S., Boumetarhan, I., Caner, H., Nebout, N.C., Dale, B., De
Vernal, A., Ellegaard, M., Filipova, M., Godhe, A., Goubert, E., Grøsfjeld, K.,
Holzwarth, U., Kotthoff, U., Leroy, S.A.G., Londeix, L., Marret, F., Matsuoka, K.,
Mudie, P.J., Naudts, L., Pen ~ a-Manjarrez, P.L., Persson, A., Popescu, S.-M.,
Pospelova, V., Sangiorgi, F., van der Meer, M.T.J., Vink, A., Zonneveld, K.A.F.,
Vercauteren, D., Vlassenbroeck, J., Louwye, S., 2009. Process length variation in
cysts of a dinoflagellate, Lingulodinium machaerophorum, in surface sediments:
investigating its potential as salinity proxy. Marine Micropalentology 70, 54e69.
Mertens, K.N., Bradley, L.R., Takano, Y., Mudie, P.J., Marret, F., Aksu, A.E., Hiscott, R.N.,
Verleye, T.J., Mousing, E.A., Smyrnova, L.L., Bagheri, S., Mansor, M., Pospelova, V.,
Matsuoka, K., 2012. Quantitative estimation of Holocene surface salinity vari-
ation in the Black Sea using dinoflagellate cyst process length. Quaternary
Science Reviews 39, 45e59.
Mudie, P.J., Aksu, A.E., Yaşar, D., 2001. Late Quaternary dinoflagellate cysts from the
Black, Marmara and Aegean seas: variations in assemblages, morphology and
paleosalinity. Marine Micropaleontology 43, 155e178.
Mudie, P.J., Rochon, A., Aksu, A.E., Gillespie, H., 2002. Dinoflagellate cysts, fresh-
water algae and fungal spores as salinity indicators in late Quaternary cores
from Marmara and Black Seas. Marine Geology 190, 203e231.
Mudie, P.J., Rochon, A., Aksu, A.E., Gillespie, H., 2004. Recent dinoflagellate cyst
assemblages in the AegeaneMarmaraeBlack Sea corridor: statistical analysis
and reinterpretation of the early Holocene Noah's Flood hypothesis. Review of
Palaeobotany and Palynology 128, 143e167.
Mudie, P.J., Marret, F., Aksu, A.E., Hiscott, R.N., Gillespie, H., 2007. Palynological evi-
dence for climatic change, anthropogenic activity and outflow of Black Sea water
during the late Pleistocene and Holocene: centennial- to decadal-scale records
from the Black and Marmara Seas. Quaternary International 167e168, 73e90.
Mudie, P.J., Leroy, S.A.G., Marret, F., Gerasimenko, N.P., Kholeif, S.E.A., Sapelko, T.,
Filipova-Marinova, M., 2011. Nonpollen palynomorphs: indicators of salinity
and environmental change in the CaspianeBlack SeaeMediterranean corridor.
The Geological Society of America Special Paper 473, 89e116.
Mudie, P., Rochon, A., Richards, K., Ferguson, S., Warny, S., 2016. Spiniferites cruci-
formis, Pterocysta cruciformis and Galeacysta etrusca: morphology and paleo-
ecology. Palynology (accepted); July 6, 2016.
Muratov, M.V., Neprochnov, Y.P., Ross, D.A., Trimonis, E.S., 1978. Basic feature of the
Black Sea late Cenozoic history based on the results of Deep Sea Drilling, Leg
42B. In: Initial Reports of the Deep Sea Drilling Project, 42 (2), pp. 1141e1148.
Murray, J.W., Stewart, K., Kassakian, S., Krynytzky, M., DiJulio, D., 2005. Oxic, suboxic
and anoxic conditions in the Black Sea. In: Gilbert, A., Yanko-Hombach, V.,
Panin, N. (Eds.), Climate Change and Coastline Migration as Factors in Human
Adaptation to the Circum-Pontic Region: from Past to Forecast, pp. 1e26.
Panin, N., Popescu, I., 2007. The Northwestern Black Sea: climatic and sea-level
changes in the late Quaternary. In: Yanko-Hombach, V., et al. (Eds.), The Black
Sea Flood Question. Springer, Dordrecht, The Netherlands, pp. 387e404.
Popescu, S.-M., 2001. Repetitive changes in Early Pliocene vegetation revealed by
high-resolution pollen analysis: revised cyclostratigraphy of southwestern
Romania. Review of Palaeobotany and Palynology 120, 181e202.
Popescu, S.-M., 2006. Late Miocene and early Pliocene environments in the
southwestern Black Sea region from high-resolution palynology of DSDP Site
380A (Leg 42B). Palaeogeography, Palaeoclimatology, Palaeoecology 238,
64e77.
129
Popescu, S.-M., 2008. Continental and Marine Environmental Changes in Europe
Induced by Global Climate Variability and Regional Paleogeographic Changes.
PhD thesis. Claude Bernard University Lyon 1, France.
Popescu, S.-M., Melinte, M.C., Suc, J.P., Clauzon, G., Quille
ve
re

, F., Süto-Szentai, M.,
2007. Earliest Zanclean age for the Colombacci and uppermost Di Tetto for-
mations of the “latest Messinian” northern Apennines: New palae-
oenvironmental data from the Maccarone section (Marche Province, Italy).
Geobios 40 (3), 359e373.
Popescu, S.-M., Dalesme, F., Jouannic, G., Escarguel, G., Head, M.J., Melinte-
Dobrinescu, M.C., Süto-szentai, M., Bakrac, K., Clauzon, G., Suc, J.-P., 2009.
Galeacysta etrusca complex: dinoflagellate cyst marker of Paratethyan influxes
to the Mediterranean Sea before and after the peak of the Messinian salinity
crisis. Palynology 33 (2), 105e134.
Rochon, A., Mudie, P.J., Aksu, A.E., Gillespie, H., 2002. Pterocysta gen. nov.: a new
dinoflagellate cyst from Pleistocene glacial-stage sediments of the Black and
Marmara seas. Palynology 26, 95e105.
Ross, D.A., Uchupi, E., Prada, K.E., Macllvaine, J.E., 1974. Bathymetry and micro-
topography of Black Sea. In: Degans, E.T., Ross, D.A. (Eds.), The Black Sea- Ge-
ology, Chemistry, and Biology, American Association of Petroleum Geologists,
vol. 20, pp. 1e10.
Ross, D.A., Neprochnov, Y.P., Erickson, A.J., Hsü, K., Hunt, J.M., Manheim, F.,
Percival, S., Senalp, M., Stoffers, P., Supko, P., Traverse, A., Trimonis, E.A., 1978.
Site 380. In: Ross, D.A., et al. (Eds.), Initial Reports of the Deep Sea Drilling
Project, 42(2), pp. 119e221.
Ryan, W.B.F., Pitman, W.C., Major, C.O., Shimkus, K., Moskalenko, V., Jones, G.A.,
Dimitrov, P., Gorür, N., Sakinç, M., Yüce, H., 1997. An abrupt drowning of the
Black Sea shelf. Marine Geology 138, 119e126.
Ryan, W.B.F., Major, C.O., Lericolais, G., Goldstein, S., 2003. Catastrophic flooding of
the Black Sea. Annual Review of Earth and Planetary Sciences 31, 525e554.
Shumilovskikh, L.S., Marret, F., Fleitmann, D., Arz, H.W., Nowaczyk, N., Behling, H.,
2013. Eemian and Holocene sea surface conditions in the southern Black Sea:
organic-walled dinoflagellate cyst record from core 22-GC3. Marine Micropa-
leontology 101, 146e160.
Shumilovskikh, L.S., Fleitmann, D., Nowaczyk, N.R., Behling, H., Marret, F.,
Wegwerth, A., Arz, H.W., 2014. Orbital- and millennial-scale environmental
changes between 64 and 20 ka BP recorded in Black Sea sediments. Climate of
the Past 10, 939e954.
Sorrel, P., Popescu, S.-M., Head, M.J., Suc, J.-P., Klotz, S., Oberha €nsli, H., 2006. Hy-
drographic development of the Aral Sea during the last 2000 years based on a
quantitative analysis of dinoflagellate cysts. Palaeogeography, Palae-
oclimatology, Palaeoecology 234, 304e327.
Sorrel, P., Popescu, S.-M., Klotz, S., Suc, J.P., Oberha €nsli, H., 2007. Climate variability
in the Aral Sea Basin (central Asia) during the late Holocene based on vegeta-
tion changes. Quaternary Research 67 (3), 357e370.
Soulet, G., Me
not, G., Lericolais, G., Bard, E., 2011. A revised calendar age for the last
reconnection of the global ocean. Quaternary Science Reviews 30, 1e8.
Supko, P., Ross, D.A., Neprochnov, Y.P., 1978. Introduction and explanatory notes, Leg
42B, Deep Sea Drilling Project. In: Ross, D.A., Neprochnov, Y.P., et al. (Eds.),
Initial Reports of the Deep Sea Drilling Project, 42 (2), pp. 3e15.
} -Szentai, M., 1986. A magyarorsza
Süto
gi Pannoniai (s.l.) re

tego €sszlet mikroplankton
vizsg
alata. Folia Comloensis 2, 25e45.
Verleye, T.J., Mertens, K.N., Louwye, S., Arz, H.W., 2009. Holocene salinity changes in
the southwestern Black Sea: a reconstruction based on dinoflagellate cysts.
Palynology 33 (1), 77e100.
Wall, D., Dale, B., 1973. Paleosalinity relationships of dinoflagellates in the late
Quaternary of the Black Sea: a summary. Geoscience and Man 7, 95e102.
Wall, D., Dale, B., Harada, K., 1973. Descriptions of new fossil dinoflagellates from
the late Quaternary of the Black Sea. Micropaleontology 19, 18e31.
Warny, S., Wrenn, J.H., 2002. Upper Neogene dinoflagellate cyst ecostratigraphy of
the Atlantic coast of Morocco. Micropaleontology 48 (3), 257e272.
Wegwerth, A., Kaiser, J., Dellwig, O., Shumilovskikh, L.S., Nowaczyk, N.R., Arz, H.W.,
2016. Northern hemisphere climate control on the environmental dynamics in
the glacial Black Sea “Lake”. Quaternary Science Reviews 135, 41e53.
Williams, G.L., Lentin, J.K., Fensome, R.A., 1998. American Association of Strati-
graphic Palynologists Contributions Series. The Lentin & Williams Index of
Fossil Dinoflagellates, 1998 ed., vol. 34, pp. 1e817
Yanko-Hombach, V., 2007. Controversy over Noah's flood in the Black Sea:geological
and foraminiferal evidence from the shelf. In: Yanko-Hombach, V., Gilbert, A.S.,
Panin, N., Dolukhanov, P.M. (Eds.), The Black Sea Flood Question:Changes in
Coastline, Climate, and Human Settlement. Springer, Dordrecht, pp. 149e203.
Yanko-Hombach, V., Mudie, P.J., Kadurin, S., Larchenkov, E., 2013. Holocene marine
transgression in the Black Sea: new evidence from the northwestern Black Sea
shelf. Quaternary International 345, 100e118.
Yanko-Hombach, V., Mudie, P.J., Kadurin, S., Larchenkov, E., 2014. Holocene marine
transgression in the Black Sea: new evidence from the northwestern Black Sea
shelf. Quaternary International 345, 110e118.
Zonneveld, K.A.F., Marret, F., Versteegh, G.J.M., Bogus, K., Bonnet, S.,
Bouimetarhan, I., Crouch, E., de Vernal, A., Elshanawany, R., Edwards, L., et al.,
2013. Atlas of modern dinoflagellate cyst distribution based on 2405 data
points. Review of Palaeobotany and Palynology 191, 1e197.