Palaeogeography, Palaeoclimatology, Palaeoecology 419 (2015) 23–36
Contents lists available at ScienceDirect
Palaeogeography, Palaeoclimatology, Palaeoecology
journal homepage: www.elsevier.com/locate/palaeo
Palaeoecology of Late Glacial and Holocene profundal Ostracoda of
pre-Alpine lake Mondsee (Austria) — A base for further (palaeo-)
biological research
Tadeusz Namiotko a,⁎, Dan L. Danielopol b, Ulrich von Grafenstein c, Stefan Lauterbach d, Achim Brauer d,
Nils Andersen e, Matthias Hüls e, Krystyna Milecka f, Angel Baltanás g, Walter Geiger h, DecLakes Participants i
a
University of Gdańsk, Faculty of Biology, Department of Genetics, Laboratory of Limnozoology, Wita Stwosza 59, 80-308 Gdańsk, Poland
University of Graz, Institute of Earth Sciences (Geology & Palaeontology), Heinrichstrasse 26, A-8010 Graz, Austria
c
Laboratoire des Sciences du Climat et de l’Environnement, UMR, Centre National de la Recherche Scientifique – Commissariat à l'Énergie Atomique (CNRS-CEA), Orme des Merissiers,
91191 Gif-sur-Yvette, France
d
GFZ German Research Centre for Geosciences, Section 5.2 – Climate Dynamics and Landscape Evolution, Telegrafenberg, 14473 Potsdam, Germany
e
Christian Albrechts University, Leibniz Laboratory for Radiometric Dating and Stable Isotope Research, Max-Eyth-Strasse 11-13, 24118 Kiel, Germany
f
Adam Mickiewicz University, Faculty of Geographical and Geological Sciences, Department of Biogeography and Palaeoecology, Dzięgielowa 27, 61-680 Poznań, Poland
g
Autonomous University of Madrid, Faculty of Sciences, Department of Ecology, 28049 Madrid, Spain
h
Budinskygasse 16/18, 1190 Vienna, Austria
i
Soumaya Belmecheri (LSCE, Gif-sur-Yvette, now at Pennsylvania State Univ.), Marc Desmet (ISTO, Univ. F. Rabelais, Tours), Helmut Erlenkeuser (Leibniz Lab., Kiel),
Jérôme Nomade (Lab. Géodynamique Chaines Alpines, Grenoble)
b
a r t i c l e
i n f o
Article history:
Received 14 February 2014
Received in revised form 18 August 2014
Accepted 10 September 2014
Available online 19 September 2014
Keywords:
Late Glacial
Holocene
Profundal ostracod succession
Geometric morphometrics
Valve preservation
Mondsee
a b s t r a c t
Succession of profundal ostracod palaeoassemblages in response to environmental changes during the Late
Glacial and Holocene was studied in a ~ 15-m-long sediment sequence from pre-Alpine lake Mondsee
(Austria). First local ostracod assemblage zone LOAZ-1 (prior to 15,700 a BP), with low abundances of
Leucocythere mirabilis and Limnocytherina sanctipatricii followed by Cytherissa lacustris, corresponds to the
Pleniglacial phase of clastic-detrital sedimentation at relatively high rates. Most of the key species of LOAZ-2
(15,590 to 13,940 a BP, including the Pleniglacial–Late Glacial transition), i.e. limnocytherids, Fabaeformiscandona
cf. harmsworthi, F. tricicatricosa, C. lacustris and Candona candida, reveal the significant association with high Al
contents and low sedimentation rates and are classified as preferring low-productivity conditions. In contrast,
Candona neglecta, dominant in LOAZ-3 (13,820 to 9960 a BP, palynologically defined as Bölling–Early Holocene)
and in the last LOAZ-4 (9780 a BP to present) as well as Cypria ophtalmica (second key species in LOAZ-4) show
relationship with higher productivity, elevated sedimentation rates and decreasing Al contents.
Furthermore, valve biodegradation and shape disparity at different time periods and in relation to sedimentation
rate changes were investigated in A-3 juveniles of Candona neglecta. The highest percentage of valves
biodegraded by Actinobacteria was recorded during the Allerød, Younger Dryas and Early Holocene, when the
lowest sedimentation rates occurred, whereas significantly lower frequencies of biodegraded valves were
recorded during the mid-Holocene and Late Holocene, when sediment accumulation was much higher. It is
also hypothesised that the degree of the valve shape variation was related to the lake productivity. During the
Allerød, Younger Dryas and Early Holocene a valve shape variation significantly lower than that recorded during
Late Holocene intervals, characterised by intensified lake productivity, was observed. Additionally, an agenda of
potential questions and approaches that should be considered and form the core of further (palaeo-)biological
research projects is offered.
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
⁎ Corresponding author. Tel.: +48 58 5236101.
E-mail addresses: [email protected] (T. Namiotko),
[email protected] (D.L. Danielopol), [email protected]
(U. von Grafenstein), [email protected] (S. Lauterbach),
[email protected] (A. Brauer), [email protected] (N. Andersen),
[email protected] (M. Hüls), [email protected] (K. Milecka),
[email protected] (A. Baltanás), [email protected] (W. Geiger).
http://dx.doi.org/10.1016/j.palaeo.2014.09.009
0031-0182/© 2014 Elsevier B.V. All rights reserved.
Being common in non-marine waters, small bivalved ostracod
crustaceans are nowadays used almost routinely as versatile palaeoproxies in environmental reconstructions alongside other indicators.
Their low-magnesium calcitic valves, often abundant and wellpreserved in lacustrine Quaternary sediments, are valuable objects for
palaeoenvironmental studies on the habitat type and succession as
24
T. Namiotko et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 419 (2015) 23–36
well as on chemical composition of past ambient water and provide a
source for chemical analysis of trace elements and stable isotopes,
which are known to reflect water and air temperature, chemistry and
productivity (Holmes, 2001; De Deckker, 2002; von Grafenstein, 2002;
Boomer et al., 2003; Curry, 2003; Ito et al., 2003; Horne et al., 2012).
The multi-disciplinary project DecLakes (Decadal Holocene and
Lateglacial variability of the oxygen isotopic composition in precipitation over Europe reconstructed from deep-lake sediments) was
primarily aimed at the reconstruction of European decadal climate
variability as expressed in the stable oxygen isotope composition of
past precipitation derived from ostracod valves recovered from the
profundal sediments of deep pre-Alpine and lowland lakes. As a part
of the DecLakes project, the present study focused on two aspects of
the ostracod material obtained from a deep pre-Alpine lake Mondsee
in Austria. The first one addressed the ostracod assemblage succession
in relation to environmental changes in the deep lacustrine habitat
during the Late Glacial to Holocene and the link between the ostracod
sequential distribution and other limnological proxies. The second
aspect of this study had arisen from the need of searching and testing
ostracod-inferred palaeoenvironmental methods (Horne et al., 2012).
It dealt with the palaeoecological information that can be extracted
from the morphological variation and the preservation state of ostracod
valves. In this context, we studied traces of biodegradation of the valves
as well as disparity in valve shape using geometric morphometrics at
different time periods and in relation to sedimentation rate changes.
2. Study site
Lake Mondsee (47°49′ N, 13°24′ E; 481 m above sea level) is located
ca. 25 km east of Salzburg in the Salzkammergut lake district of the
Upper Austrian Alpine Foreland (Fig. 1). It is a reasonably large (surface
area 13.8 km2, volume 0.51 km3) and deep lake (maximum water depth
68 m, mean water depth 37 m) with a moderately developed shore line
(28.3 km in length) and a water retention time of about 1.7 years
(Müller, 1979; Jagsch and Megay, 1982; Dokulil and Teubner, 2012).
The bathymetry is characterised by a shallower north-western basin
(up to 48 m deep) and a deeper southern basin (maximum depth
68 m, Jagsch and Megay, 1982). The lake is holomictic and stratifies
regularly during summer and sporadically in winter, thus switching between dimictic and monomictic mixing regimes (Dokulil and Teubner,
2012). As many other lakes, lake Mondsee underwent anthropogenic
eutrophication in the late 1960s and early 1970s but considerably improved since then and can presently be classified as oligo-mesotrophic
(Dokulil and Teubner, 2012). Three main streams (the Fuschler Ache,
Wangauer Ache and Zeller Ache) discharge into the north-western
basin, whereas only some smaller creeks (e.g. Kienbach) flow into the
southern lake basin. There is one outflow (Seeache), which discharges
(average outflow 9.2 m2 s−1) into lake Attersee, the last in a chain of
lakes within the catchment area (Jagsch and Megay, 1982).
About half of the catchment area (~247 km2) is covered by forests;
the remaining part is mainly used for agriculture (meadows, pastures
and arable land) and touristic recreation (Beiwl, 2008; Klug and
Jenewein, 2010). The southwestern shoreline of lake Mondsee mainly
follows a thrust zone dividing the catchment into two geological
units: the steeply sloped Northern Calcareous Alps (Mesozoic limestones and dolomites) in the south and the gently sloped hills of the
Flysch Zone (siliciclastic Cretaceous sediments, partly covered by
Quaternary deposits) in the north (Fig. 1; van Husen, 1989). At present,
the local climate is temperate with warm summers (mean annual,
January and July air temperatures of 8.7 °C, 0.5 °C and 17.8 °C,
respectively) and relatively high precipitation (annual average of
1550 mm with a maximum in spring/summer) (climate data for the
period 1971–2000, Central Institute for Meteorology and Geodynamics
(ZAMG), Vienna, Austria).
Previous work on lake Mondsee ostracods dates back to the 1980s
when intensive studies on recent and fossil distribution and autecology
of ostracod species in the littoral and profundal have been undertaken
(Danielopol et al., 1985, 1988, 1990a, 1993; Geiger, 1993). These studies
focused on Cytherissa lacustris, an endobenthic indicator species of
environmental changes related to eutrophication processes at the
sediment–water interface in recent and historical times. Investigations
of the fossil record were mainly based on short sediment cores (mostly
Fig. 1. Bathymetric map of lake Mondsee (coring location indicated by a white point) as well as location of Mondsee in Europe and simplified catchment map with the two major
geological units.
T. Namiotko et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 419 (2015) 23–36
30–40 cm in length), but also included a low-resolution study of two
cores of 7 and 13 m in length, recovered from 48 m and 67 m water
depth, respectively (Handl, 1990; Danielopol et al., 1993).
3. Material and methods
3.1. Fieldwork, chronology and geochemical analyses
A continuous, about 15-m-long sediment sequence (MO-05),
comprising the Late Glacial and Holocene, was recovered in June 2005
from the southern basin of lake Mondsee (47° 48′ 41″ N, 13° 24′ 09″ E,
62 m water depth; Fig. 1) by using a 90-mm-diameter UWITEC piston
corer (Lauterbach et al., 2011a). The chronology of the Holocene
part of this sediment record (0–1129 cm, Lithozone V according to
Lauterbach et al., 2011a) is based on microscopic varve counting,
supported by accelerator mass spectrometry 14C dating of terrestrial
plant macrofossils and 137Cs dating, while for the Late Glacial part
(1129–1273 cm, Lithozones IV–II) the age model is based on wigglematching between the oxygen isotope record derived from benthic
ostracods and the Greenland ice core isotope record (for details see
Lauterbach et al., 2011a). The age model for the previously undated
lowermost part of the sediment core (Lithozone I according to
Lauterbach et al., 2011a) was established by extrapolating the average
sedimentation rate of 1 mm a−1 at the base of Lithozone IIa down to
the base of the sediment record, yielding a basal age of 16,793 a BP at
1493 cm sediment depth (all ages are reported as “a BP”, referring to
calendar years before AD 1950). However, due to the absence of any
independent chronological proof, the age model within Lithozone I
must be considered uncertain.
In order to characterise relative changes in the content of major
elements for the complete Late Glacial and large parts of the Holocene,
impregnated sediment blocks were measured by micro-X-ray fluorescence (μXRF) scanning (200 μm resolution) by using an EAGLE III XL
μXRF spectrometer (Rh X-ray tube, 40 kV, 300 μA, 250 μm spot size,
60 s counting time). Element intensities for Ca, Mg and Al are expressed
semi-quantitatively as counts per second (cps) (Lauterbach et al.,
2011a; Swierczyński et al., 2013).
3.2. Ostracod analyses
For studying the ostracod succession throughout the sediment
profile, 75 sediment slices of 0.5 cm thickness (ca. 16 cm3) were
prepared following standard techniques (Griffiths and Holmes, 2000;
Holmes, 2001). After drying at 60 °C for 24 h, samples were weighted,
disaggregated in dilute H2O2 (10%), wet-sieved through a 125-μm mesh
and rinsed in pure ethanol before drying. Ostracod remains
(disarticulated valves, carapaces and their larger fragments) were
identified under a microscope at up to 500 × magnification using
identification keys by Absolon (1978), Griffiths and Holmes (2000)
and Meisch (2000), with taxonomy and nomenclature following the
latter. In each sample, total ostracod abundances were standardised
as number of valves (one carapace = two valves) per gram of dry
sediment. Abundances of particular species/taxa were presented as
relative percentages on a stratigraphical diagram using the software
C2 v. 1.5 (Juggins, 2007), plotting abundances vs. depth against the
background of major lithostratigraphical units (Lithozones I–V) and
local pollen assemblage zones (LPAZ) according to Lauterbach et al.
(2011a). Unidentified valves and valves of the littoral species were
combined in the stratigraphical diagram into one category (“other
species”) and were not included in the quantitative analysis. The
ostracod samples from particular sediment layers were analysed and
local ostracod assemblage zones (LOAZ) were defined by Unweighted
Pair Group Mean Average (UPGMA) constrained hierarchical clustering
using the software PAST ver. 1.94b (Hammer et al., 2001) as well as
unconstrained ordination by Principal Coordinates Analysis (PCO) and
Similarity Percentages Analysis (SIMPER) using the software PRIMER
25
ver. 6.1.10 (Clarke and Gorley, 2006); all these procedures based on
species/taxa standardised counts (percentages) and the Bray–Curtis
similarity coefficient. To determine the relationship between ostracod
species distribution and environmental predictor variables (sedimentation rate and relative contents of Al, Ca and Mg as well as the ratios Al/Ca
and Mg/Ca) the distance-based linear model (DISTLM) was used based
on 44 samples of ostracod data and fourth-root transformed environmental variables with the Best Selection Procedure (examining all
possible combinations of predictor variables and providing the best 1variable model, 2-variable model, and so on) and An Information
Criterion (AIC) as a selection criterion. A significance of the relationship
between the ostracod-derived multivariate dataset and individual
variables, when considered alone and ignoring all other variables, was
also checked by the marginal tests (Anderson et al., 2008). The best
fitted model was visualised in a two-dimensional ordination triplot using the distance-based redundancy analysis (dbRDA). Both
analyses (DISTLM and dbRDA) were performed using the software
PERMANOVA + for PRIMER (Anderson et al., 2008). The relative
contents of Al, Ca and Mg (as well as the ratios Al/Ca and Mg/Ca)
were chosen as environmental predictor variables of ostracod distribution because changes of concentrations of these elements and/or in
their proportions in the studied sediment could denote not only the
source of the deposited material (allochthonous Al-rich siliciclastic
detrital input from the Flysch Zone or Mg-rich dolomitic material
from the Northern Calcareous Alps) but may also indicate: a) intensified
biological productivity of the lake when marked by rising content of
endogenic Ca (and the corresponding decrease of the allochthonous
matter flux of Al- and Mg-rich material) and b) a cause or source of an
increased sedimentation, autochthonous (accumulation of biogenic
particles as varved calcite mud) versus allochthonous (high Al and Mg
contents) (see Lauterbach et al., 2011a for details). Sedimentation rate
was also included in the DISTLM and dbRDA analyses as this variable
(coupled with the type of the sediment, i.e. organogenic or mineral) is
a well known driving factor of ostracod distribution (see e.g. Löffler,
1986 for data on Cytherissa lacustris).
To study the morphological variation and the preservation state of
ostracod valves at different time periods in relation to sedimentation
rate changes, the juvenile A-3 stage of the most abundant species
Candona neglecta was chosen. Details on the general postembryonic
developmental trajectories of the valve shape of Candona neglecta
(as compared to Candona candida) are presented in Danielopol et al.
(2008). The juvenile A-3 stage valves are of special interest for morphological and palaeoecological studies because we have noticed their
following characteristics: 1) they do not display the sexual dimorphism
which is present in valves of the A-1 and A-2 juvenile stages (besides
shape differences, the “Anlagen” of the seminiferous tubules are visible
on living juvenile males); 2) the general aspect of the outlines for the
opposite valves is similar and no statistically significant difference
exist in our material (see also below); 3) they are preserved in a higher
number than valves of the last juvenile stages (A-1 and A-2) or those of
the adult specimens, but being already well calcified can be picked out
from sediment without much difficulty and finally give precise isotopic
measurements (Andersen unpubl.); and 4) their deposition after animal
moulting or death is near the sediment surface, namely they accumulate
in the first centimetre of the sediment surface layer, a peculiarity
already noticed by Decrouy et al. (2012), hence chemical and biological
information from the sediment–water interface may be easily stored in
these valves. The juvenile A-3 stage C. neglecta valves were routinely
recognised and further analysed from several sections representing
selected time windows within the lake sediment succession: 1) the
Allerød (All) at ~ 13,620–13,660 years before present (a BP), 2) the
Younger Dryas (YD) at ~ 12,080–12,100 a BP, 3) the Early Holocene
(EH) at ~ 10,920–10,940 a BP, 4) the 8.2 ka BP cold event (8.2 ka)
at ~ 8030–8170 a BP, 5) the Late Holocene (LH) at ~ 1060–1360 a
BP, and (6) sub-recent times (SR) at ~ 20–70 a BP. Considering the
chronology of this sediment record and the correlation with the global
26
T. Namiotko et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 419 (2015) 23–36
climatic fluctuations see Lauterbach et al. (2011a). Between 34 and 64
right valves were inspected in selected samples to estimate the degree
of valve biodegradation in each of the five time windows 1–5 listed
above. The percentage of valves with more than the half of their surface
covered by traces of micro-borings sensu Danielopol et al. (1986) (i.e. a
dendritic network of thin shallow channels on the valve surface)
was correlated (Pearson product moment correlation) with the average
sedimentation rate within the respective interval according to
Lauterbach et al. (2011a). We looked to this possible positive statistical
correlation between the degree of valve biodegradation and the
sediment accumulation rate because in a previous study dealing with
Mondsee sediments and ostracods (Danielopol et al., 1986) we noticed
differences between habitats with low sediment accumulation or
during different time periods when the trophic state of the lake changed
from the oligo-mesotrophic to more eutrophic one.
Prior to main analysis of the degree of the valve shape variation
in relation to the lake productivity, we tested the shape dissimilarity
of the right and left valves of the A-3 juvenile valves of C. neglecta
using the 1-way layout of the non-parametric test ANOSIM implemented in the PRIMER v. 6 (Clarke and Gorley, 2006). ANOSIM is a
randomisation/permutation test working on rank similarities arising
from all pairs of replicates within and between samples or sites
(in this case the opposite valves). A Global test statistic R is calculated
(Clarke and Warwick, 2001, chapter 6) which is approximately zero
when similarities between and within the samples are the same. The
opposite extreme R value is 1, when the dissimilarity between samples
is 100%. In our case the dissimilarity of the opposite valve shapes was
low with the Global R values between 0.1 and 0.3, and with probabilities
p = 0.01 or higher, meaning that there was a slight partial overlapping
due to the fact that the left valve covers the right one at the periphery.
Because no significant shape difference was statistically visible, we
used in our further analyses both the left and right valves. Consequently,
a total of 542 left and right intact valves were selected from all studied
samples, representing a short span of time (6–20 years) in each of
the six defined time windows. The valves were photographed with a
DS-5M Nikon digital camera fitted to a transmitted light microscope
Nikon E-200. The valve shapes, expressed as lateral outline projections,
were investigated for their variability using common procedures of
geometric morphometrics and multivariate statistic analysis (Baltanás
and Danielopol, 2011). Geometric morphometrics shape analysis was
accomplished with the software Morphomatica (Linhart et al., 2007)
and the valve shape disparity in a given sample was expressed as a
mean of Euclidean distance to centroid (MDC) ± standard error (SE).
The MDC values were correlated (Spearman rank order correlation)
with the sedimentation rate as well as with the relative contents of Ca
and Al. For the statistical treatment of the data, the Principal Coordinates
Analysis (PCO) and the test of homogeneity of dispersions (PERMDISP)
were used as implemented in the software PERMANOVA + for PRIMER
v.6 and explained by Anderson et al. (2008). The PCO was used to
document that within an empirical morphospace, on the one hand,
the shape of the A-3 juvenile valves of C. neglecta from the Younger
Dryas is not significantly dissimilar as compared with the shape of
those valves from other time slices, like the Late Holocene, but on the
other hand, the shape variability (expressed as dispersion within the
2D-morphospace) differs markedly. A more precise quantitative and
statistical analysis was performed using PERMDISP, based on the
dissimilarities of Euclidean distances. PERMDISP in our case gives a
measure of the degree of morphological variability of the valve shapes
within and between the studied samples.
4. Results
4.1. Ostracod succession
The studied sediment core, comprising the entire Holocene and Late
Glacial and thus covering more than the last ~ 15,000 years (Lauterbach
et al., 2011a), yielded in total 26,557 valves of 16 ostracod taxa:
Darwinula stevensoni (Brady & Robertson), Candona candida (Müller),
C. neglecta Sars, Fabaeformiscandona cf. harmsworthi (Scott), F. protzi
(Hartwig), F. tricicatricosa (Diebel & Pietrzeniuk), Pseudocandona sp.
(juveniles of undetermined species), Cyclocypris laevis (Müller),
C. ovum (Jurine), Cypria ophtalmica (Jurine), Ilyocypris sp. (juveniles of
undetermined species), Potamocypris sp. (early stage juveniles of one
or more species), Limnocythere inopinata (Baird), Limnocytherina
sanctipatricii (Brady & Robertson), Leucocythere mirabilis Kaufmann
and Cytherissa lacustris (Sars). Valves of two other species, Herpetocypris
reptans (Baird) and Cypridopsis vidua (Müller), were found in the
sediment layers studied only for the isotope analysis. Among the most
dominant species, nine are here considered to form the autochthonous
profundal component of the fossil assemblages and were included in
the quantitative analyses: C. neglecta (45.7% of the total number of the
recovered valves) followed by C. ophtalmica (14.5%), F. tricicatricosa
(12.8%), L. sanctipatricii and L.mirabilis (9.3%, considered together as
one taxon since these two species were represented predominantly by
juvenile valves which are hardly to be assign to a given species) and
C. lacustris, F. cf. harmsworthi, C. candida and F. protzi (1.9–3.4% each).
In addition, Potamocypris sp., of which few valves were found mostly
in the basal part of the studied sediment sequence (48 valves in total),
was also added to the group as it may constitute an autochthonous
species in the early successional stage of lake Mondsee. Owing to the
fragmentary and/or immature nature of a number of valves belonging
to the subfamily Candoninae (6.7% of the total number of the recovered
valves), specific determinations proved difficult in several samples.
Hence these valves (most probably representing the five profundal
species listed above) were lumped together under one taxon
Candoninae juv. 128 valves (b 0.5% of the total number of the recovered
valves) of the remaining six species typical for shallow limnic habitats
(Darwinula stevensoni, Pseudocandona sp., Cyclocypris laevis, C. ovum,
Ilyocypris sp. and Limnocythere inopinata) are here believed to be
allochthonous in the deep bottom zone and thus not included in the
further analyses.
The sediment samples differed greatly in the number of the ostracod
valves recovered, with the total abundances varying between N 250
valves per gram of dry sediment (e.g. around 7200 and 8600 a BP as
well as 12,100 a BP) and less than 10 (prior to ca. 14,890 a BP), with a
mean ± SD of 82 ± 75.2 valves per gram dry sediment (Fig. 2).
Overall, the valve preservation was good to very good, which,
coupled with the presence of both adults and a full range of large and
small juveniles, strongly suggests that the studied palaeoassemblages
were autochthonous. Four major intervals of ostracod stratigraphy
(referred here as local ostracod assemblage zones LOAZ) were
recognised based on species composition and abundances using the
UPGMA clustering and PCO ordination (Figs. 2 and 3) as well as the
SIMPER analysis (Table 1). The UPGMA cluster analysis shows a
dichotomy between bottom samples (before ca. 13,940 a BP) and
those from the upper part of the core. Within the former group, the 12
bottom-most samples, referred here as LOAZ-1, split out quite clearly
from more heterogeneous samples constituting LOAZ-2 (Fig. 3A). The
two other LOAZs (LOAZ-3 and LOAZ-4) of the upper part of the core
are more similar to each other. Although it does not enforce any
hierarchy upon the dataset, the PCO scatter plot (Fig. 3B) is in good
accordance with the dendrogram (Fig. 3A). Relative distances between
LOAZ-4, LOAZ-3 and LOAZ-2 are quite evident along the first PCO axis
(explaining 66.5% of the total variation), whereas the PCO2 (21.5% of
the total variation) distinguishes between LOAZ-1 and LOAZ-4 as one
group and LOAZ-2 and LOAZ-3 as the other.
LOAZ-1 (1490.5–1379.0 cm, ca. 16,780 to 15,690 a BP) was
characterised by very low abundances (on average less than 5 valves
per gram dry sediment, Fig. 2), a very high within-group average
similarity between individual samples (80.3%, Table 1) and was clearly
dominated by Leucocythere mirabilis and Limnocytherina sanctipatricii
(the former more abundant than the latter) showing very high relative
T. Namiotko et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 419 (2015) 23–36
27
Fig. 2. Stratigraphical diagram of the lake Mondsee ostracod dataset plotting percentage abundances of individual species/taxa and total ostracod abundances (per gram dry sediment) as
well as six environmental variables (sedimentation rates in mm a−1, Ca, Mg and Al contents in counts per second (cps) and the ratios Mg/Ca and Al/Ca) versus depth (cm) and age (a BP)
of the lake Mondsee sediment record along with the four established local ostracod assemblage zones (LOAZ), lithozones (I–V) and local pollen assemblage zones (LPAZ, AL = Allerød,
Bø = Bølling, EH = Early Holocene, OlderD = Older Dryas, OldestD = Oldest Dryas, YD = Younger Dryas,), two latter following Lauterbach et al. (2011a).
percentages (together 80.4% on average, Fig. 2, Table 1). These species
were accompanied by Potamocypris sp. (5.7% on average), Cytherissa
lacustris (3.9%) and, appearing gradually towards the top of this
zone, by four Candoninae species, of which Fabaeformiscandona cf.
harmsworthi (7.9%) was the most abundant (Figs. 2 and 3C).
LOAZ-2 (1369.5–1211.0 cm, ca. 15,590 to 13,940 a BP) was richer in
ostracod remains than LOAZ-1, with total abundances gradually
increasing towards the top of this zone (from 3 in the bottom-most
layer to 72 valves per gram of dry sediment at the top, with an average
of 16 valves per gram of dry sediment, Fig. 2). This interval was
characterised by a significant gradual decrease of the proportional
representation of Limnocytherinae (down to 28.5% on average) and
nearly disappearing of Potamocypris sp. as well as an increasing
abundance of Candoninae species (Fabaeformiscandona cf. harmsworthi,
F. tricicatricosa, Candona neglecta and C. candida) and Cytherissa lacustris
(Figs. 2, 3C and Table 1). It has to be stressed that abundances of the
particular candonid species were the most underestimated in this
LOAZ as several small juvenile and/or broken valves which could not
be identified to the species level constituted the highest fraction of the
recovered valves compared to the other three zones (Fig. 2). However,
LOAZ-2 was the most heterogeneous of the four assemblage zones,
having the lowest within-group similarity (65.4%, Table 1). A closer
inspection of the ostracod succession within LOAZ-2 (Figs. 2 and 3A)
reveals that this heterogeneity may be related not only to the gradual
decline of Limnocytherinae and the rise of F. cf. harmsworthi and
F. tricicatricosa but mostly to the steep, twofold decline of C. neglecta
and appearance of Cypria ophtalmica, allowing the definition of two
subzones LOAZ-2a and LOAZ-2b. The percentage abundance of
C. neglecta was considerably high (17.3% on average) prior to ca.
14,690 a BP (subzone LOAZ-2a) but reduced (8.9% on average) parallel
with the appearing of C. ophtalmica after ca. 14,600 a BP (subzone
LOAZ-2b).
LOAZ-3 (1201.0–1049.5 cm, ca. 13,820 to 9960 a BP) contained the
highest number of ostracod valves (mean overall abundance of 151
valves per gram of dry sediment, Fig. 2) and shows a high withingroup average similarity (80.2%, Table 1). In all LOAZ-3 sediment samples, the most abundant species were C. neglecta and F. tricicatricosa,
both having high counts (on average 246 and 151 valves per sample,
respectively) and high percentages (mean ± SD = 46.0 ± 7.1% and
26.6 ± 7.6%, respectively, Fig. 2 and Table 1). The two dominant species
were accompanied by Cytherissa lacustris and Candona candida with low
but nearly constant relative abundances (on average 6.0% and 4.2%,
respectively), which appeared similar to those in the previous LOAZ-2
(Table 1, Fig. 2). Remarkably, F. cf. harmsworthi became completely
extinct, whereas two species of Limnocytherinae continued decreasing
gradually from the base (22–31%) towards the top (5–8%) of this
zone. In contrast, C. ophtalmica was sparse at the bottom (0.2%) but
substantially increasing its frequency towards the top (15%).
The last LOAZ-4 (1040.0–10.0 cm, ca. 9780 a BP to present) was
clearly dominated by C. neglecta (mean percentage ± SD = 63.7 ±
15.8) associated with C. ophtalmica (28.0 ± 12.0) (Table 1). Other
candonid species (F. tricicatricosa and C. candida) as well as two species
of Limnocytherinae and C. lacustris, still significant in the previous
LOAZ-3, became sparse and eventually completely extinct in the
LOAZ-4 (Fig. 2). Worth mentioning is also a transient peak of
Fabaeformiscandona protzi between ca. 8860 and 7210 a BP (Fig. 2).
The total ostracod abundances in the LOAZ-4 were generally high
but varied greatly and declined in the uppermost sediment layers
28
T. Namiotko et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 419 (2015) 23–36
T. Namiotko et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 419 (2015) 23–36
Table 1
Results of Similarity Percentage Analysis (SIMPER) showing the average Bray–Curtis similarities (%) between sediment samples belonging to each of the four distinguished local
ostracod assemblage zones LOAZ-1 to LOAZ-4 (row Within), the average Bray–Curtis similarities (%) between sediment samples of each pair of different LOAZs (below diagonal in
the upper part matrix), and the list of indicator species for each LOAZ with their average
percentage abundances (first value) and percentage contributions to the within-group
similarity (second value). Species codes: Cc = Candona candida, Cl = Cytherissa lacustris,
Cn = Candona neglecta, Co = Cypria ophtalmica, Fh = Fabaeformiscandona cf.
harmsworthi, Ft = Fabaeformiscandona tricicatricosa, Lsp = Limnocytherina sanctipatricii
and Leucocythere mirabilis.
LOAZ-1
LOAZ-1
LOAZ-2
LOAZ-3
LOAZ-4
Within
Key species
39.80
17.22
3.78
80.26
Lsp: 80.41,
92.72
LOAZ-2
LOAZ-3
49.57
19.37
65.37
Lsp: 28.46, 28.32
Fh: 23.03, 23.45
Ft: 17.69; 18.68
Cn: 13.82, 14.61
Cl: 10.50, 9.45
Cc: 4.93, 4.91
53.30
80.19
Cn: 45.98, 52.22
Ft: 26.62, 27.70
Lsp: 12.98, 10.05
Cl: 6.01, 5.53
Cc: 4.15, 3.09
LOAZ-4
77.85
Cn: 63.65, 70.11
Co: 28.00, 27.30
(min. = 7, max. = 272, mean ± SD = 113 ± 61.6 valves per gram of dry
sediment).
To identify a potential parsimonious model for the MO-05 core
ostracod data in response to environmental variables, a distancebased linear model (DISTLM) was used. The predictor data consisted
of six environmental variables (sedimentation rate and relative
contents of Al, Ca and Mg within the sediments as deduced from μXRF
element scanning as well as the ratios Al/Ca and Mg/Ca, each variable
fourth-root transformed), whereas the response data included percentage abundances of nine autochthonous profundal species/taxa from 44
sediment samples with total ostracod abundances usually N 10 valves
per gram of dry sediment for which environmental data were available
(these samples represented three ostracod zones LOAZ-2, 3 and 4). In
the marginal tests every individual environmental variable, except the
relative Ca content, when considered alone and ignoring all other
variables, showed a significant relationship with the ostracod-derived
multivariate dataset (p b 0.05, Table 2). It can be also revealed that
the Al content alone explained 47% of the variability in the response
dataset, and other variables (contents of Mg, Al/Ca ratio and sedimentation rate) also individually explained substantial portions (close to 17%
or more) of the variation in the ostracod data (Table 2). The best model
on the basis of the AIC criterion (with the lowest AIC value of 283.45 and
correlation of 0.591) had two variables: sedimentation rate and the Al
content. Five other models (but with three variables), however,
achieved AIC values within 2 units of the best model, suggesting a
reasonable amount of redundancy among the environmental variables
due to substantial inter-correlations between these (Table 2). Fig. 4
visualises the best DISTLM model with the two variables as a dbRDA
ordination plot where the first axis captured 95% of the fitted model
variation and 56% of the total variation. However, there is still residual
variation in the original data matrix (the two axes together explain
only 59% of the total variation) and the two environmental variables,
sedimentation rate and Al content (and other variables correlated
with them), have multiple partial correlation with the dbRDA1 of
0.455 and − 0.890, respectively. Six of the ostracod species/taxa
(Fabaeformiscandona tricicatricosa, Candona candida, Cytherissa lacustris,
Limnocytherinae, Fabaeformiscandona cf. harmsworthi and Potamocypris
sp.) are located on the left side of the triplot (Fig. 4) and are negatively
correlated (≥ 0.48) with the dbRDA1. These species were most
29
Table 2
Results of the distance-based linear model (DISTLM) for the ostracod data from the
lake Mondsee studied core using the Best Selection Procedure and the AIC selection
criterion of the fourth-root transformed environmental sediment variables (SedRate =
sedimentation rate, % = percent of variation in the ostracod dataset explained by environmental variable, R2 = proportion of the variation in the ostracod data set that is explained by environmental data matrix in a given model, RSS = residual sum of squares).
Marginal tests
Variable
Sum of squares (trace)
SedRate
Mg
Al
Ca
Mg/Ca
Al/Ca
10,165
14,295
27,941
2612
5153
13,521
Best models
AIC
R2
283.45
284.37
284.85
284.85
285.03
285.08
0.591
0.601
0.597
0.597
0.595
0.595
Pseudo-F
p
%
0.003
0.001
0.001
0.152
0.024
0.002
17.24
24.24
47.38
0.43
0.74
22.93
RSS
No variables
Selections
24,096
23,511
23,771
23,773
23,867
23,895
2
3
3
3
3
3
SedRate, Al
SedRate, Ca, Al/Ca
SedRate, Al, Al/Ca
SedRate, Al, Ca
SedRate, Mg, Al
SedRate, Al, Mg/Ca
8.746
13.436
37.809
1.946
4.021
12.493
characteristic for LOAZ-2 and, to a lesser extent, LOAZ-3, whose
sediments were characterised by relatively high Al contents and low
sedimentation rates (19.1 cps and 0.71 mm a−1, respectively). On the
other hand, Candona neglecta and Cypria ophtalmica, clearly positively
correlated with the dbRDA1 (0.785 and 0.676, respectively), are situated on the right side of the plot and dominated in the samples of LOAZ-4
(C. neglecta also in LOAZ-3) with comparatively low Al contents and
high sedimentation rates (11.3 cps and 1.12 mm a −1, respectively).
4.2. Preservation state and morphological variation of Candona neglecta
juvenile valves
Table 3 presents percentage of valves with micro-borings (i.e.
filamentous traces and holes left by Actinobacteria on the calcite wall,
Fig. 5) in relation to the sedimentation rate for the five selected time
intervals Allerød, Younger Dryas, Early Holocene, the 8.2 ka BP cold
event and Late Holocene with each sample set from these intervals
comprising a short and comparable span of time (ca. 13–40 years).
The frequency of valve biodegradation was the highest in the Allerød
(87% studied valves with micro-borings) followed by the Early Holocene and Younger Dryas (85% valves with micro-borings), when
sedimentation rate was the lowest (from 0.24 mm a−1 in the Allerød
to 0.50 mm a−1 in the Early Holocene). In contrast, significantly lower
frequencies of micro-bored valves were recorded during the midHolocene (8.2 ka BP cold event) and Late Holocene (64% and 56%,
respectively), when sedimentation rates amounted to 0.75 and
1.25 mm a−1, respectively. Hence, the percentage of valves with traces
of micro-borings appeared to be negatively correlated with the sedimentation rate (statistically significant Pearson correlation = −0.945,
p = 0.015).
Results of the shape variability of 542 valves of the Candona neglecta
juveniles A-3 stage studied with geometric morphometrics and multivariate statistics are shown in Table 4 and Fig. 6. Valve shape disparity
expressed as the mean of Euclidean distances to centroid (MDC ±
standard error) in samples from the Allerød (ca. 13,650 a BP) to the
8.2 ka BP cold event varied between 13.4 ± 0.57 in the Younger Dryas
(Fig. 6C) to 15.7 ± 0.99 in the Early Holocene (Table 4), with a mean
of 14.36, thus being clearly lower than during the Late Holocene (ca.
Fig. 3. Lake Mondsee ostracod data analysis: average linkage UPGMA cluster analysis along with five lithozones (A), Principal Coordinates Analysis PCO (B), the same PCO but
with superimposed circles representing percentage abundance of Candona neglecta (C1), Limnocytherina sanctipatricii and Leucocythere mirabilis (C2) and Fabaeformiscandona
tricicatricosa (C3).
30
T. Namiotko et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 419 (2015) 23–36
Fig. 4. Distance-based redundancy analysis (dbRDA) ordination plot for the fitted best
distance-based linear model (DISTLM) of the lake Mondsee ostracod dataset versus two
selected predictor environmental variables (sedimentation rate and Al content) visualised
as vectors corresponding to multiple partial correlations with the first two dbRDA axes
as well as with overlying vectors of Pearson simple linear correlations of individual
species/taxa with the dbRDA1 and dbRDA2.
1350 a BP) and sub-recent times, where it ranged from 16.7 ± 1.12 to
22.6 ± 2.21 (Table 4), with he mean value of 19.25. The difference of
the shape disparity between the Younger Dryas sample YD-1 with one
of the lowest MDC values and the Late Holocene samples LH-1 and
LH-2 with the highest MDC (Fig. 6) appeared statistically significant
(test of homogeneity of dispersions PERMDISP t = 4.663, p = 0.001).
Statistically significant differences were also recorded between the
following samples: 8.2 ka-1 + 8.2 ka-2 versus LH-1 + LH-2 (t =
4.229, p = 0.001), LH-1 versus 8.2 ka-2 (t = 1.994, p = 0.003), LH-2
versus 8.2 ka-1 (t = 3.551, p = 0.003), LH-2 versus 8.2 ka-2 (t =
4.791, p = 0.001), LH-3 versus 8.2 ka-1 (t = 3.395, p = 0.003) and
LH-3 versus 8.2 ka-2 (t = 4.715, p = 0.001). Additionally, the MDC
values of the studied samples were positively correlated with the
sedimentation rate (statistically significant Spearman rank order
correlation = 0.62, p = 0.022) as well as negatively correlated with
the relative Ca and Al contents (statistically significant Spearman
rank order correlation = − 0.67, p = 0.011 and − 0.61, p = 0.025,
respectively).
5. Discussion
5.1. Ostracod species richness and autochthonous versus allochthonous
palaeoassemblages
In total, valves of (at least) 18 ostracod species were recovered from
the MO-05 core sediments, thus revealing a relatively high species
richness compared to Late Glacial to Holocene records from other
deep European postglacial lakes (Griffiths, 1995). All identified ostracod
species (except Fabaeformiscandona cf. harmsworthi) are known from
Central and Western European lacustrine habitats in both recent times
(Meisch, 2000) and the Quaternary (Griffiths, 1995). The name of
Fabaeformiscandona cf. harmsworthi is given here in open nomenclature
to indicate the provisional identification since the valves from lake
Mondsee differed slightly from those originating from the high Arctic
sites within the present range of the species (Namiotko et al., 2007,
2009).
Environmental requirements of the eight most abundant species
(Candona candida, C. neglecta, Cypria ophtalmica, Cytherissa lacustris,
Fabaeformiscandona protzi, F. tricicatricosa, Leucocythere mirabilis and
Limnocytherina sanctipatricii) clearly reflect ecological conditions of
the profundal to sublittoral bottom zones of oligotrophic to moderately
eutrophic European postglacial lakes (Meisch, 2000; Namiotko et al.,
2012). Most of these species are also well-known to be distinctive for
the early successional phases in such lakes, especially in Alpine and
pre-Alpine lakes (e.g. Löffler, 1975, 1983, 1997).
The presence of both adults and numerous juveniles of almost all
developmental stages of the eight most abundant species (and F. cf.
harmsworthi), coupled with the relatively good preservation of their
valves, implies that the studied ostracod palaeoassemblages were
predominantly autochthonous, i.e. preserved in situ thanatocoenoses
sensu Boomer et al. (2003), and as such may well reflect the environmental conditions in which the ostracods lived. The presence of valves
of the remaining species/taxa in the MO-05 core sediments, which
are known to occur mostly in the lake littoral, may be due to a postmortem transport from shallow water deposits. However, one cannot
reject the in situ deposition of the valves of at least some of those
species, e.g. Cyclocypris laevis or Cypridopsis vidua, which have been
sporadically reported also from greater depths (e.g. down to N 70 m in
lake Neuchâtel, Switzerland: Monard, 1920). However, most probably
the valves of rare species in the MO-05 sediments (with the exception
of Potamocypris sp. from the bottom section of the studied core) are
not of significant importance, and because they have never been
found in the deep zone of Mondsee (Danielopol et al., 1993), are
believed to be allochthonous in the studied sediment record.
All collected species but two (Fabaeformiscandona cf. harmsworthi
and Cyclocypris laevis) have previously been recorded in Mondsee
either as living specimens or empty valves (Danielopol et al., 1985,
1988, 1993; Fabaeformiscandona tricicatricosa listed therein as
Fabaeformiscandona caudata Kaufmann). However, the occurrence of
C. laevis in lake Mondsee is not unexpected, as this nearly ubiquitous
species often lives in lacustrine habitats (Meisch, 2000) and has been
formerly reported from other deep (pre-)Alpine lakes (e.g. in Lake
Constance: Löffler, 1969). In contrast, F. harmsworthi is considered
nowadays an Arctic species of unknown ecology. Hence, the discovery
of valves resembling this species in the first zones of the ostracod stratigraphy of the lake Mondsee sediments, which comprise the interval
directly after the Last Glacial Maximum, constitutes an interesting
record (for further discussion see Namiotko et al., 2007, 2009).
As Cypria ophtalmica is a well-known nektobenthic species, often
living in the phytolittoral of lakes (e.g. Szlauer-Łukaszewska, 2012),
one could ask if the occurrence of this species in the profundal habitats
of Mondsee is not allochtonous rather than autochthonous. However,
presence of this species as living and fossil in other deep parts of
Mondsee has been documented by Handl (1990) and Danielopol et al.
(1993). Cypria ophtalmica (forma lacustris) has also been recorded as
completing its life cycle in Lake Geneva at a depth of 70 m by Decrouy
et al. (2012), and has been reported from greater depths (e.g. at
330 m in Loch Ness, Scotland by Griffiths et al., 1993). Finally, Casale
and Danielopol (1990) showed that C. ophtalmica in the benthic habitat,
of lake Mondsee (at a depth of 40 m) is represented by individuals with
poorly pigmented valves resembling those populations which commonly lives in groundwater habitats. Thus, it could be that the records
from Mondsee actually refer to other species — Cypria lacustris Sars.
However, as in our opinion C. lacustris is hardly distinguishable from
C. ophtalmica based solely on the valve shape, we prefer a more widely
used name C. ophtalmica (for further discussion on differences and
possible synonymy of C. ophtalmica and C. lacustris see Meisch, 2000).
5.2. Ostracod palaeoassemblages succession in response to major postglacial
environmental changes in lake evolution
The succession of the ostracod thanatocoenoses in the studied
core sediments is discussed here in relation to the five main lithostratigraphical units (Lithozones I–V) established by Lauterbach et al.
(2011a) based on geochemical and microfacies analysis. Backed up by
pollen analysis and stable oxygen and carbon isotope measurements
T. Namiotko et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 419 (2015) 23–36
Table 3
Percentage of the A-3 stage juvenile valves of Candona neglecta with micro-borings (% MB)
related to the sedimentation rate (SedRate) in the selected sediment samples of the studied lake Mondsee core representing a short span of time in five different time windows:
AL = Allerød, YD = Younger Dryas, EH = Early Holocene, 8.2 ka = the 8.2 ka BP cold
event, LH = Late Holocene.
Time
window
Age
(a BP)
SedRate
(mm a−1)
Time span (a)
No samples
Total no
valves
% MB
LH
8.2 ka
EH
YD
AL
1300
8200
11,100
12,100
13,600
1.25
0.75
0.50
0.36
0.24
20
13
20
28
40
5
2
2
2
2
41
34
62
52
55
56.10
64.03
85.48
84.62
87.27
on ostracods, the lithozones reflect the major palaeoenvironmental
phases of lake evolution (Lauterbach et al., 2011a).
The first ostracod biozone LOAZ-1 (prior to ca. 15,700 a BP) corresponds to the basal part of the Pleniglacial Lithozone I (Figs. 2 and 3A),
which is characterised by clastic-detrital sedimentation at relatively
high rates (approx. 1 mm a− 1), indicating extensive erosion of the
catchment in a cold terrestrial environment dominated by sparse herb
and shrub vegetation (Lauterbach et al., 2011a). The earliest ostracod
assemblages which colonised lake Mondsee in the Pleniglacial (already
around 16,800 a BP) contained juveniles of limnocytherid species
(Leucocythere mirabilis and/or Limnocytherina sanctipatricii) accompanied by single valves of juvenile Potamocypris sp. (Fig. 2). Towards the
top of LOAZ-1, prior to ca. 15,600–15,700 a BP, assemblages became
progressively more diverse as Cytherissa lacustris and first candonids
(Fabaeformiscandona cf. harmsworthi, F. tricicatricosa and Candona
candida) appeared. Although the interpretation of the precise sequence
of species appearance in the bottom-most samples of LOAZ-1 is hampered by the low valve abundance (Fig. 2), it seems that L. mirabilis
and C. lacustris were the first to arrive. This pioneer (pre-Allerød)
species composition (excepting F. cf. harmsworthi) is quite typical for
most profundal sediment records of (pre-)Alpine lakes (Löffler, 1975,
1997; Schwalb et al., 1998; Belis et al., 2008) or maar lakes of the Eifel
mountain range in western Germany (e.g. Scharf, 1993) and indicates
oligotrophic and well-oxygenated conditions (Löffler, 1975, 1997). It is
also in concordance with a sediment record previously obtained from
a coring site at 67 m water depth where valves of C. lacustris and
L. mirabilis were also found in pre-Allerød sediments, i.e. prior to
14,000 a BP (Danielopol et al., 1993).
Ostracod subzone LOAZ-2a, which shows an increase of both the
total valve abundance and the proportional representation of candonid
species (simultaneous with the decline of limnocytherid species, Fig. 2),
corresponds with the top of Lithozone I. Since sediment composition
remains unchanged, it is difficult to explain why under still low primary
productivity conditions and predominantly allochthonous sedimentation with Mg-rich dolomitic material from the Northern Calcareous
Alps and Al-rich siliciclastic detrital input from the Flysch Zone (i.e.
with low endogenic calcite contents) ostracods became more abundant
in general and candonids (especially F. cf. harmsworthi and
31
F. tricicatricosa) got to co-dominating with limnocytherids and
Cytherissa lacustris. Perhaps, once established in the previous LOAZ-1,
populations of candonid species might just start to grow at that time
without a direct link to environmental conditions. Also Candona
neglecta, although appearing already at the end of LOAZ-1, established
a stable and abundant population in LOAZ-2a (Fig. 2).
A shift from subzone LOAZ-2a to LOAZ-2b coincides with the transition from Lithozone I to II between 14,565 and 14,625 a BP (Figs. 2 and
3A), reflecting the Pleniglacial to Late Glacial transition (Lauterbach
et al., 2011a). The onset of the Late Glacial is defined by the rapid
warming reflected by the δ18O increase and, as a response, a rather
sluggish lake productivity increase evidenced by the first appearance
and the consequent slight rise of endogenic calcite. However, allochthonous matter flux (high Al and Mg contents), still prevailed until ca.
14,100 a BP, when dwarf shrub (Juniperus, Salix and Betula) and
herbaceous vegetation was reduced in favour of expanding Pinus
forests, and the catchment stabilised and biological productivity intensified as marked by a rise in calcite contents (Lauterbach et al., 2011a).
Ostracod assemblages responded to this environmental change by a further decline of limnocytherids (on average to 24% from 31% in LOAZ-2a
and 80% in LOAZ-1) in favour of Fabaeformiscandona cf. harmsworthi
(up to 28% on average) and F. tricicatricosa (up to 21% on average)
(Fig. 3C) as well as by the appearance of Cypria ophtalmica (Fig. 2).
Though a single valve of the latter species was found already at the
top of LOAZ-1, its regular occurrence, though with a low abundance, is
only observed in LOAZ-2b. All species present in LOAZ-2b except
for Candona neglecta and Cypria ophtalmica displayed in the dbRDA
ordination a significant association with high Al contents and low
sedimentation rates (Fig. 4), and may be classified as favouring rather
low-productivity lacustrine conditions. On the contrary, C. neglecta
and C. ophtalmica show an association with higher productivity, indicated by higher sedimentation rates and decreasing Al contents. These two
environmental variables which were chosen in the DISTLM model
(Table 2) should be interpreted cautiously as being necessarily the
only causative predictor variables. Nevertheless, these may be acting
as proxies for some other important variables that were either omitted
from the model for the reasons of parsimony (e.g. Ca content or Mg/Ca
ratio) or were not measured.
LOAZ-2b turned into the LOAZ-3 between 13,940 and 13,820 a BP
(200–300 years prior to the end of Lithozone II), which coincides with
the first short-term Late Glacial cooling episode (Mo-LG1) and the
middle of palynologically defined Bölling, with the further expansion
of Pinus coupled with the frequent occurrence of heliophytic plants
(Poaceae, Chenopodiaceae, Artemisia and Juniperus) (Lauterbach et al.,
2011a). The beginning of LOAZ-3 is marked by a rapid and substantial
(almost two-fold) increase of the total ostracod abundance (up to
115–183 valves per gram dry sediment), the rise of relative abundance
of C. neglecta (up to 32%) as well as the decline and subsequent
complete extinction of F. cf. harmsworthi at the beginning of the Allerød
(Fig. 2). This ostracod biozone continued until 9960–9780 a BP throughout the entire Lithozones III and IV (equivalent to the Allerød and
Younger Dryas, respectively) as well as the base of Lithozone V (Early
Fig. 5. Right valves (outer view) of the A-3 juvenile stage of Candona neglecta from the Younger Dryas section (1149.5–1150.0 cm, ca. 12,080 a BP) of the lake Mondsee sediment core with
microborings (filamentous traces and holes left by Actinobacteria which bored into the chitin layer and dissolved the calcite wall). Scale bars = 0.1 mm.
32
T. Namiotko et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 419 (2015) 23–36
Table 4
Valve shape disparity expressed as the mean of Euclidean distances to centroid (MDC) ± standard error (SE) of the A-3 stage juveniles of Candona neglecta related to the sedimentation
rate (SedRate) in the selected sediment samples of the studied lake Mondsee core representing short spans of time in six different time windows: AL = Allerød, YD = Younger Dryas,
EH = Early Holocene, 8.2 ka = 8.2 ka BP cold event, LH = Late Holocene and SR = sub-recent along with the palaeobotanical information according to Lauterbach et al. (2011a).
Time window
SR-2
SR-1
LH-3
LH-2
LH-1
8.2 ka-2
8.2 ka-1
EH-2
EH-1
YD-2
YD-1
AL-2
AL-1
Core depth (cm)
Age (cal yr BP)
top
bottom
top
bottom
15.5
21.5
226.5
227.5
228.5
919.5
929.5
1099.5
1100.0
1149.5
1150.0
1195.0
1195.5
20.0
26.0
227.5
228.5
229.5
920.0
930.0
1100.0
1100.5
1150.0
1150.5
1195.5
1196.0
19
42
1337
1344
1350
8025
8158
10,917
10,927
12,076
12,090
13,615
13,635
36
65
1344
1350
1356
8031
8165
10,927
10,937
12,090
12,104
13,635
13,655
Time
span
SedRate
(mm/yr)
No
samples
No
valves
MDC ± SE
Palaeobotanical record
15
18
7
6
6
6.6
6.6
10
10
14
14
20
20
2.27
1.96
1.43
1.67
1.67
0.75
0.75
0.50
0.50
0.36
0.36
0.25
0.25
7
7
2
2
2
1
1
1
1
1
1
1
1
29
37
34
21
35
37
30
54
53
48
60
58
46
16.678
17.661
21.073
22.597
18.216
13.486
14.813
13.830
15.736
13.399
13.794
14.844
14.970
Poaceae dominate, herbs and open plant communities
changed by man
Fagus dominates, presence of Picea, Abies, Corylus and
Carpinus; no permanent human settlement
Holocene) (Figs. 2 and 3A). During LOAZ-3 limnocytherinae continued
their progressive decrease (down to 5–8% at the end of the zone) in
contrast to C. ophtalmica significantly increasing its frequency (15%),
whereas F. tricicatricosa had high abundances in the entire sequence of
this interval (27% on average) (Fig. 2, Table 1). The rise in the total
ostracod abundance parallel to the massive calcite precipitation in
Lithozone III and at the base of Lithozone V (70–85% CaCO3) suggests
favourable conditions for most of the profundal species during warming
Fig. 6. Morphological variability analysis of the juvenile A-3 valves of Candona neglecta
from the lake Mondsee sediment core: superposition of left valve outlines of adults and
A-1 to A-3 juveniles (A), two left A-3 juvenile valves with the maximal contrasting shapes
from the Late Holocene and superposition of their outlines standardised by equal surface
(B), example of an ordination by the Principal Coordinates Analysis (PCO) of juvenile A-3
valves originating from the Younger Dryas (green triangles) and the Late Holocene
(reverse blue triangles) and displaying significant shape disparity differences (C).
±1.1183
±1.2468
±1.4739
±2.2050
±1.2738
±0.6415
±0.9374
±0.6320
±0.9903
±0.5695
±0.5764
±0.6888
±0.8908
Picea dominates with high % of Ulmus and presence
of Tilia and Fraxinus
Pinus dominates, appearance of Picea, Abies, Tilia, Alnus
and Quercus
Pinus dominates, increase of Juniperus, Artemisia and
Chenopodiaceae
Pinus dominates, Dryopteris filix-mas present
in the Allerød and at the onset of the Holocene, indicated by a slight
productivity increase (low minerogenic input and high amounts of
endogenic calcite) but still low organic matter contents (only a slight
increase in total organic carbon) and the lowest sedimentation rates
(ca. 0.45 mm a− 1 on average) (Lauterbach et al., 2011a). Ostracod
response to the Younger Dryas cooling was almost unnoticeable. Only
in one sample (1159.5–1160.0 cm, ca 12,360 a BP), which corresponds
to the interval of the lowest Ca and total inorganic carbon contents
(Lauterbach et al., 2011a), the total ostracod abundance drastically
dropped (to 40 valves compared to 210 and 250 valves per gram dry
sediment in the samples below and above) as well as Cytherissa lacustris
and limnocytherids temporarily increased their abundances (from 7 to
14% and from 8 to 10%, respectively) (Fig. 2). However, analysis at
higher (decadal) resolution is definitely needed to draw any reliable
conclusions in that case. The same applies to the Younger Dryas–
Holocene transition in order to study in detail the ostracod response
to the warming at the onset of the Holocene.
A significant shift in the ostracod record, marked by the gradual
disappearance of Cytherissa lacustris as well as progressive substantial
decline of Fabaeformiscandona tricicatricosa, limnocytherids and
Candona candida, synchronous with the clear domination of Candona
neglecta associated with Cypria ophtalmica (Fig. 2) characterises the
transition between LOAZ-3 and LOAZ-4. This change in the ostracod
palaeoassemblages was associated with the increasing abundance of
organic matter in the sediment and increasing sedimentation rates
(from ca. 0.6 to N 2.0 mm a−1 throughout Lithozone I) as well as with
the decrease of allochthonous matter flux and intensified biochemical
calcite precipitation (Lauterbach et al., 2011a). High abundances of
Cypria ophtalmica and Candona neglecta according to the dbRDA analysis
are apparently related to such conditions (Fig. 4). These two species are
also the most widely distributed ostracods in lake Mondsee today, occurring in both profundal and littoral habitats (Danielopol et al.,
1993). Cypria ophtalmica is considered to be a generalist, widely distributed nektobenthic species inhabiting almost every type of inland
waterbodies, often doing well in organically enriched sediments or
even stressed environments, including organically polluted and
hypoxic sites (Meisch, 2000) but preferring waters with high calcium
content N 72 mg Ca2 + dm−3 (Hiller, 1972). The high abundance of
C. ophtalmica valves in LOAZ-4 (Fig. 2) is considered to be related to
the productivity of the lake. This assertion is based on the observation
of Handl (1990) who noted that in the Late Holocene of the shallow
pre-Alpine lake Halleswiessee in Austria this species increases its
densities during the time where the lake displays a mesotrophic
state, with high accumulation of organic matter, slight allochthonous
mineralogenic influxes and high values of total phosphorus in the
sediments as documented by Behbehany (1987). Candona neglecta
also appears to be tolerant to reduced oxygen contents in water
T. Namiotko et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 419 (2015) 23–36
(Hagerman, 1967; Danielopol et al., 1985, 1993) and has a preference
for rather high calcium concentrations (Vesper, 1975). On the contrary,
Cytherissa lacustris due to its long 2-year life cycle, inbenthic mode of
life, low mobility and fecundity as well as relatively heavy specific
weight is not able to avoid oxygen depletion (despite a high tolerance
towards hypoxia) during summer stagnation conditions in organically
enriched, fluffy deposits with high sedimentation rates (Danielopol
et al., 1988; Geiger, 1993). Therefore, most probably the expansion of
C. ophtalmica and C. neglecta and the synchronous retreat of C. lacustris
(and perhaps also limnocytherids and F. tricicatricosa) in LOAZ-4 were
rather caused by a lowering of oxygen contents coupled with increasing
organic matter loads and resulting alteration of sediment structure (i.e.
change in the trophic status of the lake) than by changes of water
temperature, which most probably remained fairly constant (4–6 °C
as today: Geiger, 1990a) in the deepest part of lake Mondsee throughout the studied period.
The ostracod succession in the last interval of the studied sediment
record (most of LOAZ-4) appeared to be reasonably coherent with
that from a 730-cm-long sediment core recovered in 1987 from 48 m
water depth in the north-western basin of lake Mondsee, covering at
least the last 6000 years (Danielopol et al., 1993). In that sedimentary
record also Candona neglecta clearly dominates (reaching up to 30–
80%) and together with Fabaeformiscandona tricicatricosa (indicated
herein as F. caudata) and Cypria ophtalmica accounts 70–90% the assemblages throughout the profile, except for the topmost layers (40–0 cm)
where a decrease in species richness and abundances is observed, likely
due to modern anthropogenic eutrophication (Danielopol et al., 1993).
However, in contrast to the 1987 sediment core, F. tricicatricosa in the
presently studied sediment core occurred after ca. 6000 a BP only at
very low abundances (on average less than 0.5%), whereas Cytherissa
lacustris, which is present almost throughout the core recovered in
1987 (with relative abundances N 5%), became almost completely
extinct at ca. 6000 a BP at 62 m water depth (the MO-05 core).
The environmental reconstruction based on the ostracod record
preserved in the studied sediments from 62 m water depth is also
quite consistent with studies on ostracod sequences from other (pre-)
Alpine lakes (Löffler, 1975; Schwalb et al., 1998). In many of these
lakes a remarkable change in species composition and abundances
occurred during the early Holocene due to the onset of more organic
sedimentation in addition to improved nutrient conditions (Löffler,
1975; Schwalb et al., 1998). However, lake-specific differences in ostracod successions (e.g. early major changes prior to the onset of the
Holocene, lack of some species typical for most lakes or influences of
entering rivers introducing fluvial species to the profundal zone) are
commonly observed due to complex local factors (Löffler, 1975;
Schwalb et al., 1998). It is nevertheless also true that so far only little information exists on complete ostracod successions from the Late Glacial
to recent times in the profundal zone of deep postglacial European lakes
(see the review by Griffiths, 1995) as well as on ecological requirements
of most of profundal-dwelling species (e.g. F. tricicatricosa, L. mirabilis,
L. sanctipatricii, and even of possible differences in autecology between
C. neglecta and C. candida, compare e.g. Meisch, 2000) that consequently
hampers thorough comparisons and interpretations.
To conclude, the data on the ostracod succession presented
here show the potential benefits of a more detailed (higher resolution)
examination of ostracods from lake Mondsee sediments, aiming at more
refined palaeoenvironmental reconstruction.
5.3. Ostracod valve biodegradation as a proxy of the sedimentation rate
It is considered that in many cases the filamentous traces and holes
present on dead ostracod valves are made by Actinobacteria (review in
Danielopol et al., 1986), which are one type of the microorganisms
degrading chitin. The most common of these bacteria in freshwater
sediments are species which are aerobic, and, although they could be
found in several cm within the sediment, the sediment–water interface
33
is the most active site of microbial chitin mineralisation in the lake
environment (Rux, 1983). Oxygen and chitin availability within the
sediment are the main factors resulting in decreasing abundance
gradient of Actinobacteria throughout the sediment depth (Danielopol
et al., 1986 and references therein). Therefore, one could expect that
the activity of chitinolytic bacteria in decomposing ostracod valves
will be lower if the valves are deposited under hypoxic conditions
and/or are rapidly buried in habitats of high sedimentation, compared
to those valves remaining longer at the sediment surface in oxygenrich habitats of a low sedimentation rate. Indeed, a similar correlation
supporting this hypothesis was evidenced by Danielopol et al. (1986),
who found in short sediment cores from 47 m water depth in lake
Mondsee (b30-cm-long, covering ~ 50 years) that the percentage of
corroded valves of Candona neglecta deposited during a period of low
sedimentation rates (basal part of the profile) was much higher (up to
70–80%) than during a period of higher sedimentation rates and periodical anoxia (b 40%).
The present results revealed a similar relationship between the
degree of valve biodegradation and sediment accumulation rate but
proved this relation for a much longer sediment record. The data point
out that in deep lacustrine conditions during periods of low sedimentation rates (0.24–0.50 mm a−1 in the Allerød, Younger Dryas and Early
Holocene) exposure of ostracod valves was longer, and consequently,
abundance of micro-boring organisms leaving traces on valves
was higher than during periods with high sedimentation rates
(0.75 to 1.25 mm a− 1 during the 8.2 ka BP cold event and the Late
Holocene) (Table 3). This taphonomic aspect merits further detailed
(palaeo-)ecological studies in order to understand the way bacteria
attack the valves deposited on the bottom of the lake upon the death
of the ostracods.
5.4. Ostracod valve shape variability as a possible proxy of lake productivity
A study of the intra-specific shape variability of ostracod valves
using morphometric techniques represents an important aspect in
evolutionary (palaeo-)ecology. For two widely distributed non-marine
ostracod species, Eucypris virens (Jurine) and Limnocythere inopinata,
Baltanás and Geiger (1998) show that valve shapes under fieldconditions and/or in laboratory controlled cultures display significant
intra- and inter-population variability. Baltanás et al. (2000b), considering the shape variability of E. virens valves note that it increases significantly under fluctuating environmental conditions but the exact causes
which determine this variability remain poorly understood. Further, it
was mentioned that it remains very elusive for the knowledge on the
shape variability to estimate the relative genetic and environmental
contributions to the total phenotypic valve disparities for a given
species (Baltanás et al., 2002). It is also interesting to mention the
convergent reaction of carapace shape transformation under given
salinity concentrations. Martens (1985) for Mytilocypris henricae (Chapman) and Yin et al. (1999) for Limnocythere inopinata report that
specimens exposed to high salinity concentration display more elongated valve shapes. We showed above that in the Late Holocene samples
with a higher degree of valve disparity the shape changed from more
or less dorsally rounded outline to an elongated one (Fig. 6).
Considering the specific case of C. neglecta, Baltanás et al. (2000a),
mention that the disparity of the adult valves of this species differs
(i.e. it is smaller) from that of the related species C. candida. This points
out to inter-specific differences in the capacity to realise the valve shape.
Finally, it is important to remark that Baltanás and co-workers stress
repeatedly in the above mentioned publications (Baltanás and Geiger,
1998; Baltanás et al., 2000a,b) that understanding the relationship
between the shape variability of ostracod carapaces and the external
ecological drivers can be useful for further palaeoenvironmental
reconstructions.
Using an example from marine ostracods, Allmon and Ross (2001)
discuss the potential importance of nutrients for evolutionary changes
34
T. Namiotko et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 419 (2015) 23–36
of ostracods. Their discourse was based on the carapace diversity
of fossil Loxoconcha populations recovered from different palaeoenvironments of a series of atolls in the Indo-Pacific. For the present
data of Candona neglecta from lake Mondsee we also favour as a starting
hypothesis for a further more thorough research the idea that there is a
relationship between the variability of the lake productivity and the
degree of disparity of ostracod valve shapes. Below we explain our
hypothetical scenario and offer arguments for its plausibility.
First, it is well known that climate and especially air temperature has
an impact on the aquatic life in both marine and limnic systems. As an
example, Cronin and Raymo (1997) point out for deep-sea benthic
species diversity that air temperature changes play a role on the dynamics of the food production on which the deep-sea benthic fauna relies.
Such changes can be recurrent on different time-scales. Yasuhara et al.
(2008) show that the abrupt cooling during the Younger Dryas
produced within a short period of time a collapse of deep-sea ecosystems on the western coast of the North Atlantic and this was clearly
visible for ostracod assemblages. In our case, the lowest shape variability was in the sample from the Younger Dryas (Table 4), when air
temperature was much lower than during the next periods, which
may be inferred from the palaeobotanical records (Table 4 here and
Lauterbach et al., 2011a) and especially from the δ18O changes
(Lauterbach et al., 2011a; von Grafenstein own data). The opposite
appears for the samples of the Late Holocene and also for the subrecent ones. Temperature in the hypolimnion of lake Mondsee at depths
between 60 and 68 m varies on both the monthly and yearly temporal
scales within a low range of 1–2 degrees, i.e. between 4 and 6 °C
(cf. Geiger, 1990a for intra-annual values and Dokulil et al., 2006 for
inter-annual values), therefore we suspect that the water temperature
in the profundal of lake Mondsee, from where the MO-05 core was
recovered, does not play a major role in the different degrees of variability of the valve shapes. Therefore we favour as an explanation the
biogenic production of the lake, namely the plankton production
(both phyto-and zooplankton), which developed differently during
the lake history here explored. The organic material produced by the
plankton accumulates on the bottom of the lake at profundal sites and
determine additional local ecological changes like the change in the
oxygen concentration at the sediment–water interface and in the
uppermost sediment layers where ostracods generally live (Decrouy
et al., 2012 and see below for C. neglecta) and where the water chemistry also can change, like the change of pH caused by an increase of the
carbon dioxide concentration (cf. information specifically for lake
Mondsee in Danielopol et al., 1990b). Considering this general situation
we tried to better identify the potential ecological constraints which
could determine an increase in the degree of disparity of the A-3 valves
for C. neglecta at the studied core site in lake Mondsee.
Liepolt (1935) mentions C. neglecta about 1 km south from the MO05 core site in lake Mondsee. He mentions that the biogenic production
of the lake at the time was of mesotrophic type, which means that
during the summer successions of well-developed phyto- and zooplankton populations existed. The organic matter which was produced
accumulated as the organogenic sediment on the bottom of the lake,
especially in the profundal areas. At present, such situation can be
found in the same area north from the site MO-05 at a water depth of
50 m in front of the Mooswinkel bay. Yin and Geiger (1995) show for
this site and depth the presence of a developed C. neglecta population
(cf. their Fig. 3). Dokulil and Skolaut (1986) noticed during an intense
study of one year in 1982 the change of the trophic state of lake
Mondsee from eutrophic to mesotrophic with the seasonal succession
of algal blooms. Similar observations but in a more palaeolimnologic
context and restricted only to diatoms were noticed by Schmidt
(1991). During the same period, Danielopol et al. (1985, 1988) studied
intensively the ostracod fauna in lake Mondsee. Living specimens of
C. neglecta were sampled in the upper profundal zone of lake Mondsee.
Under laboratory conditions Danielopol et al. (1988) exposed specimens of C. neglecta on organogenic sediments with a high content of
decaying algae. It was noticed that after one week the living ostracods
were covered with sticky particles on the inner side of the valves
and on the limbs, making the movement on the sediment very difficult
(see Fig. 5 and further explanation in Danielopol et al., 1988). This could
apply also to the A-3 juveniles of C. neglecta which live in the profundal
of lakes preferentially in the top one centimetre of the sediment
(Decrouy et al., 2012).
In a study on the succession of cladoceran Eubosmina-species
(typical zooplankton component in lakes with increased biogenic
production), Nauwerk (1991) mentions that during the Late Holocene,
in a period more or less equivalent to our 226–230 cm sample of the
MO-05 core, an increased productivity in lake Mondsee occurred.
Studying the same sediment section, Handl (1990) shows that the
most abundant species present at this interval was C. neglecta (cf. his
Fig. 14). Considering the sediment quality of the Late Holocene section
of the MO-05 core mentioned above, one should note a reduced fraction
of allochthonous minerogenic fraction with a high Al content as a
marker of the Flysch influx. Therefore the high sedimentation rate in
this case (Table 4) is apparently the result of the accumulation of higher
biogenic particles (varved calcite mud).
It is well-known that higher summer temperatures increase the
production of phytoplankton. In our case the Late Holocene section,
where the highest disparity for A-3 valves of C. neglecta was noticed,
corresponds also with the pollen zone dominated by Fagus (Table 4).
As compared with previous pollen zones, where pollen of Pinus and
Picea dominated, it is a clear sign for a rather mild climate which favours
an increase of the lake productivity. Interesting enough, for the Late
Holocene samples with the high range of valve shape disparity, no
signs of permanent human settlement appears. It is well known that
Odilo, the duke of Bavaria, founded the Mondsee Abbey during
748 AD, about 130 years later than our evaluated age for the three
Late Holocene samples here discussed.
In conclusion, the information we offer here is certainly conjectural
and our hypothesis that productivity of lake Mondsee induced a higher
morphological disparity of the ostracod valve shape has now to be
verified with additional observations from both field and laboratory.
6. Conclusion
The information on the Mondsee ostracods, obtained within
this multi-disciplinary approach is certainly useful for palaeoecological
reconstructions and proves that ostracods are multi-proxy microfossils.
Therefore, the research approaches used here are planned to be applied
to the other two lakes of the DecLakes programme, lake Hańcza in
Poland and lake Iseo in northern Italy, for which sedimentological,
geochemical and paleoecological data already exist (Lauterbach et al.,
2011b, 2012). Such approaches are to be encouraged and emulated in
future for other sedimentary ostracod records. However, there is still
much to document and understand about ostracod ecology and life
history in deep lacustrine conditions to test the proxy methods and
inspire confidence in the efficacy of such quantitative estimates of
palaeoenvironmental changes. While some progress has been made in
our understanding of the impact of the environment which play a role
on the realisation of such morphological traits like the valve shape and
preservation, on local ecological distributions or lifecycles, several
fundamental aspects of such associations remain largely vague or the
data are known for only a very few species (just to quote a good example of studies on Cytherissa lacustris by e.g. Danielopol et al., 1990b;
Geiger, 1990a,b). Therefore, improving knowledge of profundal ostracod ecology is one of the most needed future (palaeo-)biological
research area.
One of the most promising candidate for further studies appears
Candona neglecta, species commonly found in deep lacustrine habitats
and considered by Danielopol et al. (2008) as an evolutionary unit
with a mosaic of ecological traits slightly differing on the interpopulation level (see Sterelny, 1999 for this conceptual view). A
T. Namiotko et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 419 (2015) 23–36
research programme conceived on the study of juveniles of C. neglecta,
especially the A-3 stage, is motivated by several arguments already
listed above in the Section 3.2. Other advantages are the facts that the
development time of the A-3 to moult into the next stage should be
much shorter than that of the A-1 stage (Danielopol unpubl.) which
moves into the adult stage, and that the valves of that stage also offer
excellent results for the oxygen (and carbon) isotope studies (von
Grafenstein et al., 1994, 1999; Lauterbach et al., 2011a and further
unpubl. data of the DecLakes project). Hence, it is encouraged here to
combine field and laboratory research approached at various time
and space scales in order to better understand how ostracods perceive
their sedimentary environment as physical and chemical substrate
within they live, because successful application of both wellestablished and new ostracod-inferred proxy methods needs a biological
basis for their assumptions and a critical awareness of their limitations.
Acknowledgements
This study, carried out within the frame of the European Science
Foundation EUROCORES Programme EuroCLIMATE (contract No.
ERAS-CT-2003-980409 of the European Commission, DG Research,
FP6, ESF project DecLakes no. 04-ECLIM-FP29), has been made possible
thanks to the funding from the national agencies FWF (Austria, project
no. I35-B06), DFG (Germany, project no. BR2208/2-2, AN554/1-2) and
CNRS (France). We thank A. Danielopol, J. Knoblechner, K. Minati, M.
Pichler and G. Roidmayr (at that time Inst. Limnol., Austrian Acad. Sci.,
Mondsee) and L. Namiotko (Univ. Gdańsk) for picking and preparing
ostracods for analysis. R. Niederreiter (UWITEC, Mondsee) and H.
Höllerer (Inst. Limnol., Mondsee) are thanked for long-year assistance
with the sampling equipment. We are also indebted to the Inst. Water
Ecol., Fish. & Lake Res. in Scharfling, particularly to A. Jagsch, for providing both logistical support in organising the field lab and information on
lake Mondsee. W. Piller (Univ. Graz and Austrian Acad. Sci. through its
Comm. Stratigraph. Palaeontol. Res. Austria) supported logistically the
completion of the final part of the project. R. Schmidt (at the time Inst.
Limnol., Mondsee) and T. Swierczyński (German Res. Center Geosci.,
Potsdam) are thanked for helpful information. Attendance of T.
Namiotko at the 17th Int. Symp. on Ostracoda in Rome was possible
thanks to the PalSIRP Sepkoski Grant 2013 and a grant no. ZB 530L155-D032-12 from the University of Gdańsk. Finally, we thank I.
Mazzini (CNR-IGAR, Rome) and two anonymous reviewers for their
careful reading of the manuscript and the resulting constructive
comments.
References
Absolon, A., 1978. Die Gattung Candona (Ostracoda) im Quartär von Europa. Rozpravy
Česk. Akad. Věd 88, 1–75.
Allmon, W.D., Ross, R.M., 2001. Nutrients and evolution in the marine realm. In: Allmon,
W.D., Bottjer, D.J. (Eds.), Evolutionary Paleoecology. The Ecological Context of
Macroevolutionary Change. Columbia Univ. Press, New York, pp. 105–148.
Anderson, M., Gorley, R.N., Clarke, K.R., 2008. PERMANOVA + for PRIMER: Guide to
Software and Statistical Methods. PRIMER-E Ltd., Plymouth.
Baltanás, A., Danielopol, D.L., 2011. Geometric morphometrics and its use in ostracod
research: a short guide. Joannea Geol. Paläontol. 11, 235–272.
Baltanás, A., Geiger, W., 1998. Intraspecific morphological variability: morphometry
of valve outlines. In: Martens, K. (Ed.), Sex and ParthenogenesisEvolutionary
Ecology of Reproductive Modes in Non-marine Ostracods. Backhuis Publ., Leiden,
pp. 127–142.
Baltanás, A., Namiotko, T., Danielopol, D.L., 2000a. Biogeography and disparity within the
genus Cryptocandona (Crustacea, Ostracoda). Vie Milieu 50, 297–310.
Baltanás, A., Otero, M., Arqueros, G., Rossetti, G., Rossi, V., 2000b. Ontogenetic changes
in the carapace shape of the non-marine ostracod Eucypris virens (Jurine).
Hydrobiologia 419, 65–72.
Baltanás, A., Alcorlo, P., Danielopol, D.L., 2002. Morphological disparity in populations
with and without sexual reproduction: a case study in Eucypris virens (Crustacea,
Ostracoda). Biol. J. Linn. Soc. 75, 9–19.
Behbehany, A.R., 1987. Sedimentations- und Klimageschichte des Spät- und Postglazials
im Bereich der Nördlichen Kalkalpen (Salzkammergutseen, Österreich). Göttinger
Arb. Geol. Paläontol. 34, 1–120.
35
Beiwl, C., 2008. Atlas der natürlichen Seen Österreichs mit einer Fläche ≥ 50 ha.
Morphometrie – Typisierung – Trophie. Bundesamt für Wasserwirtschaft, Vienna.
Belis, C.A., Finsinger, W., Ammann, B., 2008. The late glacial-Holocene transition as
inferred from ostracod and pollen records in the Lago Piccolo di Aviglina (Northern
Italy. Palaeogeogr. Palaeoclimatol. Palaeoecol. 264, 306–317.
Boomer, I., Horne, D.J., Slipper, I.J., 2003. The use of ostracodes in paleoenvironmental
studies, or what can we do with an ostracode shell? In: Park, L.E., Smith, A.J. (Eds.),
Bridging the Gap. Trends in the Ostracode Biological and Geological Sciences.
Paleontological Society Papers 9, pp. 153–179.
Casale, L.M., Danielopol, D.L., 1990. To see and not to be seen: the evolutionary problems
of the Ostracoda. 47–48. Bulletin de l'Institut de Géologie du Bassin d'Aquitaine,
Bordeaux, pp. 227–238.
Clarke, K.R., Gorley, R.N., 2006. PRIMER v6: User Manual/Tutorial. PRIMER-E, Plymouth.
Clarke, K.R., Warwick, R.M., 2001. Changes in Marine Communities, an Approach to
Statistical Analysis and Interpretation. PRIMER-E, Plymouth.
Cronin, T.M., Raymo, M.E., 1997. Orbital forcing of deep-sea benthic species diversity.
Nature 385, 624–626.
Curry, B.B., 2003. Linking ostracodes to climate and landscape. In: Park, L.E., Smith, A.J.
(Eds.), Bridging the Gap. Trends in the Ostracode Biological and Geological Sciences.
Paleontological Society Papers 9, pp. 223–246.
Danielopol, D.L., Geiger, W., Tölderer-Farmer, M., Orellana, C.P., Terrat, M.N., 1985. The
Ostracoda of Mondsee: spatial and temporal changes during the last fifty years. In:
Danielopol, D.L., Schmidt, R., Schultze, E. (Eds.), Contribution to the Paleolimnology
of the Trumer Lakes (Salzkammergut) and the Lakes Mondsee, Attersee and Traunsee
(Upper Austria). Limnologisches Institut, Mondsee, pp. 99–121.
Danielopol, D.L., Casale, L.M., Olteanu, R., 1986. On the preservation of carapaces
of some limnic ostracods: An exercise in actuopalaeontology. Hydrobiologia 143,
143–157.
Danielopol, D.L., Geiger, W., Tölderer-Farmer, M., Orellana, C.P., Terrat, M.N., 1988. In
search of Cypris and Cythere. A report of the evolutionary ecological project on limnic
Ostracoda from the Mondsee (Austria). In: Hanai, T., Ikeya, N., Ishizaki, K. (Eds.),
Evolutionary Biology of Ostracoda. Kodansha/Elsevier, Tokyo, pp. 485–500.
Danielopol, D.L., Carbonel, P., Colin, J.P. (Eds.), 1990a. Cytherissa (Ostracoda) — The
Drosophila of Paleolimnology 47–48. Bulletin de l'Institut de Géologie du Bassin
d'Aquitaine, Bordeaux, pp. 1–310.
Danielopol, D.L., Casale, L., Rogulj, B., Strobl, J., Maier, K. (Eds.), 1990b. Spatial Distribution
of Cytherissa lacustris Living in Mondsee 47–48. Bulletin de l'Institut de Géologie du
Bassin d'Aquitaine, Bordeaux, pp. 139–166.
Danielopol, D.L., Handl, M., Yin, Y., 1993. Benthic ostracods in the pre-alpine
deep lake Mondsee. Notes on their origin and distribution. In: McKenzie, K.G.,
Jones, P.J. (Eds.), Ostracoda in the Earth and Life Sciences. Balkema, Rotterdam,
pp. 465–480.
Danielopol, D.L., Baltanás, A., Namiotko, T., Geiger, W., Pichler, M., Reina, M., Roidmayr, G.,
2008. Developmental trajectories in geographically separated populations of nonmarine ostracods: morphometric applications for palaeoecological studies. Senckenb.
Lethaea 88, 183–193.
De Deckker, P., 2002. Ostracod palaeoecology. In: Holmes, J.A., Chivas, A.R. (Eds.),
The Ostracoda. Applications in Quaternary Research. Geophysical Monograph 131,
pp. 121–134.
Decrouy, L., Vennemann, T.W., Ariztegui, D., 2012. Sediment penetration depths of epiand infaunal ostracods from Lake Geneva (Switzerland). Hydrobiologia 688, 5–23.
Dokulil, M.T., Skolaut, C., 1986. Succession of phytoplankton in a deep stratifying lake:
Mondsee, Austria. Hydrobiologia 138, 9–24.
Dokulil, M.T., Teubner, K., 2012. Deep living Planktothrix rubescens modulated by environmental constraints and climate forcing. Hydrobiologia 698, 29–46.
Dokulil, M.T., Teubner, K., Jagsch, A., 2006. Climate change affecting hypolimnetic water
temperaturees in deep alpine lakes. Verh. Int. Ver. Theor. Angew. Limnol. 29,
1285–1288.
Geiger, W., 1990a. The role of oxygen in the disturbance and recovery of the Cytherissa
lacustris population in Mondsee. 47–48. Bulletin de l'Institut de Géologie du Bassin
d'Aquitaine, Bordeaux, pp. 167–190.
Geiger, W., 1990b. Field and laboratory studies on the life cycle of Cytherissa lacustris
(Sars) (Crustacea, Ostracoda), with special emphasis on the role of temperature. 47–
48. Bulletin de l'Institut de Géologie du Bassin d'Aquitaine, Bordeaux, pp. 191–208.
Geiger, W., 1993. Cytherissa lacustris (Ostracoda, Crustacea): Its use in detecting and
reconstructing environmental changes at the sediment–water interface. Verh. Int.
Ver. Theor. Angew. Limnol. 25, 1102–1107.
Griffiths, H.I., 1995. European Quaternary freshwater Ostracoda: a biostratigraphic and
palaeobio-geographic primer. Scopolia 34, 1–168.
Griffiths, H.I., Holmes, J.A., 2000. Non-marine ostracods and Quaternary palaeoenvironments.
Quaternary Research Association Technical Guide 8, pp. 1–188.
Griffiths, H.I., Martin, D.S., Shine, A.J., Evans, J.G., 1993. The ostracod fauna
(Crustacea, Ostracoda) of the profundal benthos of Loch Ness. Hydrobiologia
254, 111–117.
Hagerman, L., 1967. Ostracods of the Tvärminne area, Gulf of Finland. Comment. Biol. Soc.
Sci. Fenn. 30, 1–12.
Hammer, Ø., Harper, D.A.T., Ryan, P.D., 2001. PAST: Paleontological Statistics software
package for education and data analysis. Palaeontol. Electron. 4, 1–9.
Handl, M., 1990. Paläolimnologische Untersuchungen an Spät-und Postglazialen
Sedimenten des Halleswies- und Mondsees (Oberösterreich) (Palynologie und
Ostracoda)(Dissertation) Univ. Salzburg, Salzburg.
Hiller, D., 1972. Untersuchungen zur Biologie und Ökologie limnischer Ostracoden aus der
umgebung von Hamburg. Arch. Hydrobiol. Suppl. 40, 400–497.
Holmes, J.A., 2001. Ostracoda. In: Smol, J.P., Birks, H.J.B., Last, W.M. (Eds.), Tracking
Environmental Change Using Lake SedimentsZoological indicators 4. Kluwer Academic Publishers, Dordrecht, pp. 125–151.
36
T. Namiotko et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 419 (2015) 23–36
Horne, D.J., Holmes, J.A., Rodriguez-Lazaro, J., Viehberg, F.A., 2012. Overview and future
prospects for ostracod applications in Quaternary palaeoclimatology. In: Horne, D.J.,
Holmes, J.A., Rodriguez-Lazaro, J., Viehberg, F.A. (Eds.), Ostracoda as Proxies for
Quaternary Climate Change. Elsevier B.V., Amsterdam, pp. 305–315.
Ito, E., De Deckker, P., Eggins, S.M., 2003. Ostracodes and their shell chemistry: Implications for paleohydrologic and paleoclimatologic applications. In: Park, L.E., Smith, A.
J. (Eds.), Bridging the Gap. Trends in the Ostracode Biological and Geological Sciences.
Paleontological Society Papers 9, pp. 119–151.
Jagsch, A., Megay, K., 1982. Mondsee. In: Wurzer, E. (Ed.), Seenreinhaltung in Österreich:
Limnologe, Hygiene, Mafinahmen Erfolge. Bundesministerium für Land- und
Forstwirtschaft, Wasserwirtschaft 6, pp. 155–163 Wien.
Juggins, S., 2007. C2 Version 1.5 User guide. Software for Ecological and Palaeoecological
Data Analysis and Visualisation. Newcastle University, Newcastle upon Tyne.
Klug, H., Jenewein, P., 2010. Spatial modelling of agrarian subsidy payments as an input
for evaluating changes of ecosystem services. Ecol. Complex. 7, 368–377.
Lauterbach, S., Brauer, A., Andersen, N., Danielopol, D.L., Dulski, P., Hüls, M., Milecka, K.,
Namiotko, T., Obremska, M., von Grafenstein, U., DecLakes Participants, 2011a. Environmental responses to Lateglacial climatic fluctuations recorded in the sediments of
pre-Alpine Lake Mondsee (northeastern Alps). J. Quat. Sci. 26, 253–267.
Lauterbach, S., Brauer, A., Plessen, B., Dulski, P., Zamelczyk, K., Milecka, K., Danielopol, D.L.,
Namiotko, T., Hüls, M., Andersen, N., Von Grafenstein, U., DecLakes Participants,
2011b. Multiproxy evidence for Lateglacial to mid-Holocene environmental and
climatic changes in Northeastern Poland. Boreas 40, 57–72.
Lauterbach, S., Chapron, E., Brauer, A., Hüls, M., Gilli, A., Arnaud, F., Piccin, A., Nomade, J.,
Desmet, M., von Grafenstein, U., Participants, DecLakes, 2012. A sedimentary record
of Holocene surface runoff events and earthquake activity from Lake Iseo (Southern
Alps, Italy). The Holocene 22, 749–760.
Liepolt, R., 1935. Limnologische Untersuchungen der Ufer- und Tiefenfauna des Mondsees
und dessen Stellung zur Seetypenfrage. Int. Rev. Gesamten Hydrobiol. 32, 164–235.
Linhart, J., Brauneis, W., Neubauer, W., 2007. Morphomatica, version 1.6.0.1. http://
palstrat.uni-graz.at/methodsinostracodology.htm (University of Graz, Graz).
Löffler, H., 1969. Recent and subfossil distribution of Cytherissa lacustris (Ostracoda) in
lake Constance. Mitt. Int. Ver. Theor. Angew. Limnol. 17, 240–251.
Löffler, H., 1975. The evolution of ostracode faunas in Alpine and Prealpine lakes and their
value as indicators. Bull. Am. Paleontol. 65, 433–443.
Löffler, H., 1983. Aspects of the history and evolution of alpine lakes in Austria.
Hydrobiologia 100, 143–152.
Löffler, H., 1986. Ostracod analysis. In: Berglund, B.E. (Ed.), Handbook of Holocene
Palaeoecology and Palaeohydrology. Jon Wiley and Sons, Chichester, pp. 693–702.
Löffler, H., 1997. The role of ostracods for reconstructing climatic change in Holocene and
Late Pleistocene lake environment in Central Europe. J. Paleolimnol. 18, 29–32.
Martens, K., 1985. Effects of temperature and salinity on postembryonic growth in
Mytilocypris henricae (Chapman) (Crustacea, Ostracoda). J. Crustac. Biol. 5, 258–272.
Meisch, C., 2000. Freshwater Ostracoda of Western and Central Europe. Spektrum
Akademischer Verlag, Heidelberg.
Monard, A., 1920. La faune profonde du lac de Neuchâtel. Bull. Soc. Neuchâtel. Sci. Nat. 44,
65–236.
Müller, G., 1979. Grundlagendaten für Fuschlsee, Mondsee and Attersee, sowie gesamte
Einzugsgebiet. Arbeiten aus dem Labor Weyregg 3, pp. 10–14.
Namiotko, T., Pichler, M., Danielopol, D.L., Roidmayr, G., DecLakes Team, 2007. An Arctic
ostracod species (Crustacea: Ostracoda) in Late Glacial and Early Holocene sediments
of lake Mondsee (Austria). Geophys. Res. Abstr. 9, 01372.
Namiotko, T., Danielopol, D.L., Pichler, M., von Grafenstein, U., 2009. Occurrence of an
Arctic ostracod species Fabaeformiscandona harmsworthi (Scott, 1899) (Ostracoda,
Candonidae) in Late Glacial sediments of lake Mondsee. Crustaceana 82, 1209–1212.
Namiotko, T., Namiotko, L., Wysocka, A., 2012. Distribution of subfossil ostracod
assemblages in lacustrine profundal sediments of north-eastern Poland. Rev.
Micropaleontol. 55, 17–27.
Nauwerk, A., 1991. The history of the Eubosmina in lake Mondsee (Upper Austria).
Hydrobiologia 225, 87–103.
Rux, T., 1983. Microbial mineralization of chitin in a freshwater habitat, and purification of
the chitinase from Pseudomonas sp(MSc Thesis) Ball State University, Muncie,
Indiana.
Scharf, B., 1993. Ostracoda (Crustacea) from eutrophic and oligotrophic maar lakes of the
Eifel (Germany) in the Late and Post Glacial. In: McKenzie, K.G., Jones, P.J. (Eds.),
Ostracoda in the Earth and Life Sciences. Balkema, Rotterdam, pp. 453–464.
Schmidt, R., 1991. Recent re-oligotrophication in Mondsee (Austria) as indicated by
sediment, diatom and chemical stratigraphy. Verh. Int. Ver. Theor. Angew. Limnol.
24, 963–967.
Schwalb, A., Hadorn, P., Thew, N., Straub, F., 1998. Evidence for Late-Glacial and Holocene
environmental changes from subfossil assemblages and sediments of Lake Neuchâtel,
Switzerland. Palaeogeogr. Palaeoclimatol. Palaeoecol. 140, 307–323.
Sterelny, K., 1999. Species as ecological mosaics. In: Wilson, R.A. (Ed.), Species: New
Interdisciplinary Essays. The MIT Press, Cambridge, Mass., pp. 119–138.
Swierczyński, T., Lauterbach, S., Dulski, P., Delgado, J., Merz, B., Brauer, A., 2013. Mid- to
late Holocene flood frequency changes in the northeastern Alps as recorded in varved
sediments of Lake Mondsee (Upper Austria). Quat. Sci. Rev. 80, 78–90.
Szlauer-Łukaszewska, A., 2012. Ostracod assemblages in relation to littoral plant
communities of a shallow lake (lake Świdwie, Poland). Int. Rev. Hydrobiol. 97,
262–275.
van Husen, D., 1989. Blatt 65 – Mondsee. Geologische Karte der Republik Österreich 1:50
000. Geologische Bundesanstalt, Wien.
Vesper, B., 1975. Ein Beitrag zur Ostracodenfauna Schleswig-Holsteins. Mitteilungen aus
dem hamburgischen zoologischen Museum und Institut 72, pp. 97–108.
von Grafenstein, U., 2002. Oxygen-isotope studies of ostracods from deep lakes. In:
Holmes, J.A., Chivas, A.R. (Eds.), The Ostracoda. Applications in Quaternary Research.
Geophysical Monograph 131, pp. 249–266.
von Grafenstein, U., Erlenkeuser, H., Kleinmann, A., Müller, J., Trimborn, P., 1994.
J. Paleolimnol. 11, 349–357.
von Grafenstein, U., Erlenkeuser, H., Brauer, A., Jouzel, J., Johnsen, S.J., 1999. A midEuropean decadal isotope-climate record from 15,500 to 5,000 years B.P. Science
284, 1654–1657.
Yasuhara, M., Cronin, T.M., deMenocal, P.B., Okahashi, H., Linsley, B.K., 2008. Abrupt
climate change and collapse of deep-sea ecosystems. Proc. Natl. Acad. Sci. U. S. A.
105, 1556–1560.
Yin, Y., Geiger, W., 1995. Spatial distribution of ostracods related to abiotic factors in
Mooswinkl Bay, Mondsee, Austria. In: Řiha, J. (Ed.), Ostracoda and Biostratigraphy.
A.A. Balkema, Rotterdam, pp. 381–388.
Yin, Y., Geiger, W., Martens, K., 1999. Effects of genotype and environment on phenotypic
variability in Limnocythere inopinata (Crustacea: Ostracoda). Hydrobiologia 400,
85–114.