Reconstruction of palaeoenvironment and ancient human activities at Obrovac-type settlements (Serbia) using a geochemical approach

Abstract

This study aims to determine the palaeoenvironmental characteristics and activity patterns of Obrovac-type archaeological sites in Western Serbia, dated to the Late Neolithic/Early Eneolithic period, ∼5th millennium BC. These mound-like sites, enclosed by a wide ditch, that are not known in other parts of the central Balkan area, have long intrigued archaeologists investigating their origin and function over the last few decades.
In this study, for the first time, organic-geochemical analysis of paleosol samples from the Obrovac-type sites was applied with the aim of palaeoenvironmental reconstruction. Additionally, organic carbon content and anion analysis of 58 subsoil samples from these settlements were performed to determine the use of space and activity zones.
The analysis of biomarkers from selected sites suggests significant plant biodiversity in the Mačva region during the Late Neolithic/Early Eneolithic. Distribution of n-alkanes with the maximum at n-C25 and predominance of C30 hop-22(29)-ene among hopanoids in samples from Obrovac type-sites indisputably indicates that macrophytes are a dominant source of organic matter, implying a marshy and floodplain depositional environment. On the other side, a strong signal of long-chain n-alkanes indicates the input of terrestrial plants into the precursor biomass, confirming that this environment was habitable for the first settlers in this region. Anion-based analysis delineates certain activity zones, demonstrating that Obrovac type-sites manifest rather complex spatial behavior despite their relatively small size and available space.

Full text

Quaternary International xxx (xxxx) xxx
Please cite this article as: Gorica Veselinović, Quaternary International, https://doi.org/10.1016/j.quaint.2021.09.001
Available online 6 September 2021
1040-6182/© 2021 Elsevier Ltd and INQUA. All rights reserved.
Reconstruction of palaeoenvironment and ancient human activities at
Obrovac-type settlements (Serbia) using a geochemical approach
Gorica Veselinovi´
c
a
,
*
, Boban Tripkovi´
c
b
, Nevena Anti´
c
a
, Aleksandra ˇ
Sajnovi´
c
a
,
Milica Kaˇ
sanin-Grubin
a
, Tomislav Tosti
c
, Kristina Penezi´
c
d
a
University of Belgrade, Department of Chemistry, ICTM, Njegoˇ
seva 12, 11000, Belgrade, Serbia
b
University of Belgrade, Faculty of Philosophy-Department of Archaeology, ˇ
Cika Ljubina 18-20, 11000, Belgrade, Serbia
c
University of Belgrade, Faculty of Chemistry, Studentski trg 12-16, 11000, Belgrade, Serbia
d
University of Novi Sad, Biosense Institute, Zorana Đinđi´
ca 1, 21000, Novi Sad, Serbia
ARTICLE INFO
Keywords:
Late Neolithic/Early Eneolithic
Wetlands
Activity zones
Biomarkers
Anions
ABSTRACT
This study aims to determine the palaeoenvironmental characteristics and activity patterns of Obrovac-type
archaeological sites in Western Serbia, dated to the Late Neolithic/Early Eneolithic period, ~5
th
millennium
BC. These mound-like sites, enclosed by a wide ditch, that are not known in other parts of the central Balkan
area, have long intrigued archaeologists investigating their origin and function over the last few decades.
In this study, for the rst time, organic-geochemical analysis of paleosol samples from the Obrovac-type sites
was applied with the aim of palaeoenvironmental reconstruction. Additionally, organic carbon content and anion
analysis of 58 subsoil samples from these settlements were performed to determine the use of space and activity
zones.
The analysis of biomarkers from selected sites suggests signicant plant biodiversity in the Maˇ
cva region
during the Late Neolithic/Early Eneolithic. Distribution of n-alkanes with the maximum at n-C
25
and predomi-
nance of C
30
hop-22(29)-ene among hopanoids in samples from Obrovac type-sites indisputably indicates that
macrophytes are a dominant source of organic matter, implying a marshy and oodplain depositional envi-
ronment. On the other side, a strong signal of long-chain n-alkanes indicates the input of terrestrial plants into
the precursor biomass, conrming that this environment was habitable for the rst settlers in this region. Anion-
based analysis delineates certain activity zones, demonstrating that Obrovac type-sites manifest rather complex
spatial behavior despite their relatively small size and available space.
1. Introduction
Usage of organic geochemistry and other chemical analytical
methods has particularly shifted the traditional approach of archaeo-
logical research, from methods of absolute dating to the palae-
oenvironment reconstruction based on determining the quantity,
diversity, and origin of organic matter (Evershed, 2008; Lejay et al.,
2016; Mileto et al., 2017; Veselinovi´
c et al., 2021), or revealing a
different variety of human activity based on determining levels of
available phosphates and other chemical elements (Fernandez et al.,
2000; Holliday and Gartner, 2007; Salisbury et al., 2013). n-Alkanes are
widely used in archaeological research (Evershed, 2008; Bai et al., 2009;
Eckmeir and Wiesenberg, 2009; Duan and Xu, 2012; Li et al., 2016).
Their abundance and distribution could provide useful information
about organic matter sources, palaeoclimate conditions, palae-
oenvironmental characteristics, thermal maturity, and biodegradation
(Pancost et al., 2002; Peters et al., 2005). The precursors of short-chain
n-C
15
, n-C
19
, and particularly n-C
17
n-alkanes are algae and cyanobac-
teria, mid-chain n-C
23
and n-C
25
n-alkanes originate from macrophytes,
whereas odd long-chain homologs n-C
27
, n-C
29
, and n-C
31
are derived
from terrestrial vascular plants (Eglinton and Hamilton, 1967; Ficken
et al., 2000; Peters et al., 2005; Eckmeir and Wiesenberg, 2009; Bush
and McInerney, 2013). Some investigations have shown that the domi-
nance of n-C
31
is mostly related to grass vegetation and/or higher plants,
such as ash and maple (Duan and He, 2011; Feurdean et al., 2021;
Nelson et al., 2017). However, there are many processes that can affect
distributions of n-alkanes, and some of them are anthropogenic pollu-
tion and microbiological activity (Grimalt et al., 1988; Bouloubassi and
* Corresponding author.
E-mail address: gorica.veselinovic@ihtm.bg.ac.rs (G. Veselinovi´
c).
Contents lists available at ScienceDirect
Quaternary International
journal homepage: www.elsevier.com/locate/quaint
https://doi.org/10.1016/j.quaint.2021.09.001
Received 21 April 2021; Received in revised form 1 September 2021; Accepted 2 September 2021

Quaternary International xxx (xxxx) xxx
2
Saliot, 1993; Eckmeir and Wiesenberg, 2009). Microbiological reworked
organic matter in different environmental media can be recognized by
the presence of Unresolved Complex Mixtures (UCM), which typically
have a chromatographic output between n-C
16
and n-C
22
and/or
n-alkane maximum among even short-chain homologs n-C
18
or n-C
20
(Grimalt et al., 1988; Farrington and Quinn, 2015).
An advantage of using n-alkanes in palaeoenvironmental studies is
their abundance and relatively ease of analysis using gas chromatog-
raphy, coupled for example to a ame ionization detector (GC-FID),
mass spectrometer (GC–MS), or combustion isotope ratio mass spec-
trometer (GC-C-IRMS) systems.
In this study, a geochemical approach was used to investigate
Obrovac-type sites, archaeological settlements unique to Western Serbia
that formed during the Late Neolithic and Early Eneolithic. The main
objectives of this work are a) to derive more information about the
palaeoenvironmental conditions and b) to provide new data about past
activities and give insight on space use and activity zones at the
Obrovac-type sites. An organic-geochemical analysis of selected paleo-
sol samples taken from three different settlements was done to provide
more information about ancient biodiversity and environmental condi-
tions. The research is based on the isolation and identication of specic
biomarkers in the saturated fraction of the extracted organic matter. In
addition, organic carbon content and anion analysis of 58 subsoil sam-
ples from two sites was performed to better understand past activities
and the use of space, helping to provide an environmental perspective on
the formation and occupation of these atypical sites in the region.
2. Archaeological and geographical settings
2.1. Archaeological background
During the 5
th
millennium BC, signicant demographic, cultural and
economic changes were observed in the central Balkans (modern-day
Serbia). The rst half of the period is marked by the Late Neolithic Vinˇ
ca
culture (5300–4500 BC) (Bori´
c, 2009; Orton, 2012; Tasi´
c et al., 2016)
with its style of dark-burnished pottery; long-lasting fortied settle-
ments with a population of few hundred to several thousand people;
domestic buildings that consist of several rooms with elaborate interior
space; an economy based on agriculture and animal husbandry; an
intensive exchange of exotic products made of Spondylus and Glycymeris
shell, and Carpathian obsidian (Tripkovi´
c, 2013). In the middle of the 5
th
millennium BC, the once large cultural area was fragmented into several
different cultural styles (Bubanj, Tiszapolgar, Bodrogkeresztur, Lengyel)
that sometimes occurred in the same area or the same location. Large
settlements of the Vinˇ
ca period were disappearing, and smaller settle-
ments of several households were founded predominantly in the high-
lands, in difcult-to-access areas. The houses were small, without
symbolic elaboration of their interior space. Agriculture was still
commonly practiced, but animal husbandry was the dominant compo-
nent in the economy. The former interregional exchange network was
interrupted, and exotic products were rarely found in settled locations
(Bulatovi´
c et al., 2020; Milanovi´
c, 2019; Tripkovi´
c, 2013). The Early
Copper Age (Eneolithic) in the Balkan area is generally dated to between
4400 and 3900 BC (Bulatovi´
c, 2018).
In the area of Western Serbia (district of Maˇ
cva), 245 sites from the
late Neolithic and only 30 from the Early Copper Age have been recor-
ded (Stoji´
c and Cerovi´
c, 2011), indicating signicant demographic and
social changes. These changes were apparently not sudden, since to-
wards the end of the Neolithic distinctive pastoral communities were
settled in the highland areas of the central Balkans, as well as in Western
Serbia (Trbuhovi´
c and Vasiljevi´
c, 1983; Bulatovi´
c et al., 2017).
One of the characteristics of Maˇ
cva and Western Serbia during the
Late Neolithic and Early Eneolithic is Obrovac type sites, which were not
reported in other areas in the central Balkans. These are small mound-
like sites, in most cases less than 50 m in diameter, enclosed by a wide
ditch. Interest in Obrovac-type sites began during the early 1970s when
they were extensively surveyed, some of them excavated, and singled
out as a distinctive category of regional archaeological site (Trbuhovi´
c
and Vasiljevi´
c, 1975, 1983, 1983). So far, 48 of them have been docu-
mented. Based on the work of the first investigators, two mutually
related interpretations regarding these settlements were offered: i) that
they were seasonally occupied by farmers/livestock keepers from
nearby larger settlements; or ii) that the sites were permanent settle-
ments of small groups of people consisting of one or two households who
adapted to life in the wetlands (Trbuhovi´
c and Vasiljevi´
c, 1975, 1983,
1983; Chapman, 1981). Their explanation was primarily based on 1) the
restricted area available for living, 2) the remains of one to two isolated
buildings at the excavated sites, and 3) modern perception of the Maˇ
cva
plain as an area of agricultural instability due to seasonal ooding.
Unfortunately, the original research was poorly documented, and
further research programs were not implemented until relatively
recently (Tripkovi´
c et al., 2013).
Most recently, small-scale excavations provided new insights into the
cultural stratigraphy of these sites. Geophysical prospection of three
sites conrmed the earlier assumptions about 1–2 buildings enclosed by
a ditch. Based on the material culture from new excavations, a Late
Neolithic or Early Eneolithic age was conrmed (Tripkovi´
c et al., 2017;
ˇ
Soˇ
si´
c Klindˇ
zi´
c and Tripkovi ´
c, 2018). The remains of burnt buildings
made of wattle and daub were detected and, in some cases, pits lled up
with cultural layers were revealed. Pottery, chipped stone artifacts and
clay weights were associated with buildings (Tripkovi´
c et al., 2017;
Tripkovi´
c, 2020). The simple cultural stratigraphy of the sites, and a
relatively restricted number of artifacts, indicate short occupations, but
the presence of a stable building with multiple oors and oven, as is the
case at the site of Obrovˇ
cina in Ratkovaˇ
ca near Dublje (Tripkovi´
c et al.:
in preparation), implies that the duration of the occupation could also
vary between localities.
2.2. Geographical settings
The Maˇ
cva region is in west/northwestern Serbia, bordered on the
north, west, and east by the Sava and Drina rivers and to the south by the
Cer Mountain (Fig. 1). It is predominantly an agricultural area that is
characterized by the versatility of the water regime caused by the often-
seasonal ooding of the Sava and Drina rivers (Markovi´
c, 1970). High
groundwater levels in the Maˇ
cva and pseudogley or mineral bog type of
soils, with stagnating water, signicantly contribute to the hydrographic
instability of the region. Also, extensive geoarchaeological coring of six
sites has shown that ditches were probably lled with water as indicated
by water-saturated gley sediments that represent the inll of the ditches,
suggesting microenvironmental mosaics in which relatively dry areas
with developed soil were scattered across wider zones characterized by
high moisture (Tripkovi´
c and Penezi´
c, 2017). Hence, wetland habitat is
here referred to as an environment saturated with water periodically,
but the duration of water retention could vary. At the Early Eneolithic
site ˇ
Sanac-Izba, near Lipolist, the available plant and animal remains
point to an economy based on T. monococcum, T. dicoccum, T. aesti-
vum/durum, Sus scrofa domesticus, Bos taurus and Ovis/Capra, although
the low quantity of the preserved remains does not allow statistical
processing (Tripkovi´
c et al., 2017). The three Obrovac-type sites that
were not destroyed by recent agricultural activities were chosen for this
study (Fig. 1).
2.3. ˇ
Sanac at Obrovˇ
cine, Dublje (SD)
The ˇ
Sanac at Obrovˇ
cine site (44◦47′6.87"N, 19◦31′49.37"E, Fig. 1) is
a small mound around 35 m in diameter and up to 2 m high, enclosed by
a ditch. The site was explored for the rst time during 1971 through a
prole section of three superimposed layers. The deepest horizon is
dark-brown soil, considered to be an articial foundation platform. It is
overlain by a reddish zone of burnt daub, containing some pottery,
above which is a “yellowish loess-like” sediment layer up to 1 m thick.
G. Veselinovi´
c et al.

Quaternary International xxx (xxxx) xxx
3
On the top of the prole is a thin humus layer. Besides the variety of
pottery types, two stone axes were also revealed (Trbuhovi´
c and
Vasiljevi´
c, 1973). Based on the material culture, the site was generally
dated to the Early Eneolithic Bubanj culture (mid- 5
th
millennium BC
(Stoji´
c and Cerovi´
c, 2011)). Excavation of a small trench (2 ×2 m) in
2017, including extensive site coring, has conrmed previous observa-
tion of site stratigraphy, chronology, and material culture while its
formation and use still needs to be re-evaluated (Tripkovi´
c, 2017).
2.4. ˇ
Sanac at Obrva (SNO)
The ˇ
Sanac at Obrva site (44◦48′48.25′′ N, 19◦35′10.92′′ E) is around
40 m in diameter and is surrounded by a wide ditch that is still visible.
The site elevation was reportedly up to 3 m, but today not more than 1.5
m. The site is dated to the Vinˇ
ca culture (Trbuhovi´
c and Vasiljevi´
c 1975,
1983, 1983). Site drilling and excavation of a small stratigraphic trench
(2 ×2 m) were conducted in 2017, revealing a stratigraphy of up to 110
cm. As in the case of the ˇ
Sanac at Obrovˇ
cine site, dark soil is found at the
site base, covered with a cultural layer followed by a thin humus section
on top. There were few areas of burnt daub, some possibly deposited in a
pit, while a small quantity of non-diagnostic pottery (that could date to
either the Late Neolithic or Early Eneolithic) and animal bones scattered
across the trench. Samples for analysis were taken out of the excavated
area as well as from the cross-section of the site.
2.5. Liki´
ca forest (LS)
The Liki´
ca forest site is situated in the forested area on the bank of a
small stream (44◦44′36.44"N, 19◦35′28.69"E). It is a mound 40 m in
diameter with a height of up to 4 m. The site is encircled by a wide ditch
that is seasonally water-lled, creating an island connected to land by an
approximately 15 m long and 1.5 m wide causeway. The site was
excavated in 1971, but no further data relating to stratigraphy and
building structures was recorded. It is dated to the Late Neolithic/Early
Copper age (Trbuhovi´
c, Vasiljevi´
c, 1975), with some pottery possibly
dating to the Bronze age (Stoji´
c and Cerovi´
c, 2011). In 2017 a geo-
archaeological coring campaign was conducted to inspect the site
stratigraphy as well as to collect sediment samples for the analysis
presented here.
3. Materials and methods
3.1. Samples
3.1.1. Sampling for determining the palaeoenvironment
For the organic-geochemical study, a total of ve samples from three
different archaeological sites were analyzed. These samples originate
from the layers of paleosol underlying the archaeological deposits at
each of the three investigated sites. According to their stratigraphic
position, they represent soil formed during the early Holocene, before a
more permanent human occupation of these locations. These layers
were sampled with the primary goal of determining the environmental
conditions prior to settlement foundation. At the ˇ
Sanac at Obrovˇ
cine
site, three samples (SD 1–3) were taken from the vertical section of the
archaeological trench, starting at 15 cm below the cultural horizon with
a 15 cm distance between them. One sample originates from the site
ˇ
Sanac at Obrva (SNO) and one from the site Liki´
ca forest (LS). Prior to
sampling, the surface of the outcrop was scraped in order to expose fresh
layers to avoid possible contamination. Bulk samples were collected
using a sampling shovel and stored in labeled glass jars, and were
immediately transferred to the laboratory where they were air dried and
prepared for further analyses.
3.1.2. Sampling for determining activity zones
Fifty-eight bulk samples for anion analysis were collected at two sites
- ˇ
Sanac at Obrovˇ
cine and ˇ
Sanac at Obrva. The samples were collected
from the archaeological layers from two cross-sections at each site using
a Dutch auger at a depth of approximately 30–40 cm (Fig. 2). In this
way, the optimal number of samples collected from different zones of the
site enabled the determination of possible anion distribution patterns
suggesting activity zones.
All geochemical analyses were done at Department of Chemistry –
Institute of Chemistry, Technology and Metallurgy; Faculty of Chemis-
try; and Department of Archaeology – Faculty of Philosophy at the
Fig. 1. a) Regional map of Serbia; b) Maˇ
cva region with labeled Obrovac-type sites, in red color: 1-ˇ
Sanac at Obrovˇ
cine, Dublje; 2-ˇ
Sanac at Obrva; 3-Liki´
ca ˇ
suma; c)
photograph of ˇ
Sanac at Obrovˇ
cine, Dublje. (For interpretation of the references to color in this gure legend, the reader is referred to the Web version of this article.)
G. Veselinovi´
c et al.

Quaternary International xxx (xxxx) xxx
4
University of Belgrade.
3.2. Organic geochemical analysis
Soluble organic matter from ~30 g of sediment samples was
extracted using the Soxhlet apparatus with an azeotropic mixture of
methanol and dichloromethane for 32 h. The total lipid extract was
separated into saturated and aromatic fractions using column chroma-
tography over SiO
2
and Al
2
O
3
and n-hexane and benzene as eluents,
respectively. For 10 mg of organic extract, 2.55 g SiO
2
and 1.65 g Al
2
O
3
,
25 ml n-hexane and 38.5 ml benzene were used. Subsequently, the
saturated and aromatic hydrocarbons were analyzed by gas
chromatography-mass spectrometry (GC-MS). GC-MS was performed
using an Agilent 7890A gas chromatograph (HP-5MS capillary column,
30m ×0.25
μ
m, 0.25
μ
m lm thickness, Helium carrier gas 1.5 cm
3
min
−1
, FID) coupled to an Agilent 5975C mass selective detector (70
eV). The column was heated from 80 to 310 ◦C at a rate of 2 ◦C min
−1
.
The individual peaks were determined by comparison with literature
data and based on a mass spectra (library: NIST5a). For the calculation
of biomarker parameters, a relative abundance of the compounds ob-
tained by integration of peak areas (software GCMS Data Analysis) in the
appropriate mass chromatograms was used. The mass fragmentograms
of the saturated fraction used for the interpretation of biomarkers are m/
z 71 for n-alkanes and isoprenoids and m/z 123 and m/z 191 for
terpenoids.
3.3. Specic biomarker parameters
Distributions and relative amounts of n-alkanes have been used for
the calculation of specic parameters which could indicate the origin
and depositional environment of organic matter. Carbon preference
index (CPI) for the full amplitude of n-alkanes is used to examine the
odd-over-even carbon number predominance and was calculated ac-
cording to the equation: CPI =½ x [(∑
odd
C
17
-C
33
)/(∑
even
C
16
-C
32
) +
(∑
odd
C
17
-C
33
)/(∑
even
C
18
-C
34
)] (Tissot and Welte, 1984). CPI values
higher than 1 mean a predominance of odd over even chain lengths and
indicate terrestrial plant leaf wax origin, and/or limited diagenesis,
and/or no petroleum contamination (Bray and Evans, 1961; Eglinton
and Hamilton, 1967; Carr et al., 2014). Odd-over-even predominance
(OEP) values were calculated by equation OEP =(n-C
27
+n-C
29
+n-C
31
+n-C
33
)/(n-C
26
+n-C
28
+n-C
30
+n-C
32
) according to Peters et al.
(2005). To qualitatively explore the relative input of terrestrial versus
aquatic organic matter, the ratio of higher plant n-alkanes to aquatic
n-alkanes was measured using the terrestrial to aquatic ratio (TAR)
following Bourbonniere and Meyers (1996): TAR =(n-C
27
+n-C
29
+
n-C
31
)/(n-C
15
+n-C
17
+n-C
19
). The ACL
25-33
(average chain length)
parameter =(25 x n-C
25
+27 x n-C
27
+29 x n-C
29
+31 x n-C
31
+33 x
n-C
33
)/(n-C
25
+n-C
27
+n-C
29
+n-C
31
+n-C
33
) is the most common
parameter used for reconstructing vegetation and may be linked to the
predominance of higher taxonomic plants over lower taxonomic plants.
ACL
25-33
values higher than 29.5 indicate the greater contribution of
grass vegetation or/and higher plants in precursor biomass (Duan and
He, 2011). A proxy ratio P
aq
has been calculated according to Ficken
et al. (2000): P
aq
=(C
23
+C
25
)/(C
23
+C
25
+C
29
+C
31
) to reect the
submerged/oating aquatic macrophyte input to organic matter relative
to that from terrestrial plants.
3.4. Determination of moisture, organic carbon and anion content
The samples were pulverized in agate mortar with pestle, weighed on
an analytical balance and dried in the oven at 105 ◦C to constant mass in
order to determine moisture content.
The content of organic carbon was determined from dried samples
heated at 600 ◦C for at least 6 h or until a constant mass was reached.
Analysis of the most abundant anions such as uoride, chloride, ni-
trite, nitrate, bromide, oxalate, formate, and phosphate was performed
using an Ion chromatograph DIONEX ICS 3000. The Dionex ICS-3000
chromatographic set-up consisted of a single pump, a conductivity de-
tector (ASRS ULTRAII (4 mm), recycle mode), and an eluent generator
(potassium hydroxide) with a Chromeleon® Chromatography Work-
station and Chromeleon6.7 Chromatography Management Software. All
the separations were performed using an IonPac AS15 Analytical, 4 mm
×250 mm and IonPac AG15 Guard, 4 mm ×50 mm columns. The ow
rate of the mobile phase was 1.00 ml min
−1
, and the elution was a
gradient in the following order: 0–3.5 min, 0.75 mM KOH; 3.5–7 min,
7.5 mM KOH; 7–20 min, 60 mM KOH; 20–25 min, 60 mM KOH; 25–30
min, 60–0.75 mM KOH; 30–35 min, 0.75 mM KOH. In addition, the
following conditions were applied: a column temperature of 30 ◦C,
conductivity cell temperature of 35 ◦C, and a suppressor current of 149
mA. The concentrations of anions were determined from the calibration
curve using the standard square method. The stock solution concentra-
tion was 20 ppm for uoride, whereas all the other anions had a
Fig. 2. Cross-sections of ˇ
Sanac at Obrovˇ
cani, Dublje (SD) and ˇ
Sanac at Obrva (SNO) sites and sampling positions. Numbers indicate distance in metres and the labels
of individual samples.
G. Veselinovi´
c et al.

Quaternary International xxx (xxxx) xxx
5
concentration of 1000 ppm. The dilution was made according to the
estimated concentration in solute.
4. Results and discussion
4.1. Bulk geochemical parameters of the paleosol samples
The content of extractable organic matter (total lipid extract) is
relatively small in all analyzed samples, ranging from 107 to 391
μ
g/g
(Table 1). The lowest amount was extracted from the sample LS, while
the highest amount of organic extract was determined in the sample
SNO. All samples are characterized by a relatively high amount of the
aromatic fraction relative to the saturated hydrocarbon fraction in the
total lipid extract (Table 1).
4.2. The molecular composition of the organic matter of the paleosol
samples
The aliphatic hydrocarbon fraction of investigated samples consists
mainly of n-alkanes, isoprenoid alkanes, diterpanes and hopanes.
However, the most abundant compounds are n-alkanes.
4.2.1. n-Alkanes
The distribution of n-alkanes is similar in all tested samples (Fig. 3
and Fig. 4), with some distinctions.
All three samples taken vertically from ˇ
Sanac at Obrovˇ
cine in Dublje
show a bimodal n-alkane distribution (Fig. 3) with a maximum at short-
or mid-chain and long-chain n-alkanes, respectively. Sample SD 1 has a
maximum at C
22
, while samples SD 2 and SD 3 have a maximum at C
31
in
the overall distribution of n-alkanes. These samples are also character-
ized by the signicant abundance of mid-chain n-alkanes C
23
and C
25
,
which is especially emphasized in sample SD 3 (Fig. 3). In these samples,
the presence of UCMs has been noticed, which indicates a small
contribution of microbiologically reworked organic matter (Grimalt
et al., 1988).
Sample SNO showed unimodal n-alkane distribution with odd n-
alkane domination and a maximum at C
25
(Fig. 4). The distribution
pattern of n-alkanes in sample LS is bimodal with a maximum at C
23
among mid-chain n-alkanes and C
31
among long-chain n-alkanes
(Fig. 4).
Specic parameters were calculated based on the distributions and
relative amounts of n-alkanes and shown in Table 2. CPI values vary
from 1.47 to 1.86, and OEP values are in the range of 2.04–3.01 and
decrease with sample depth (Table 2). The highest TAR value was 5.69
found in sample SNO and the lowest in sample SD 1 (2.39; Table 2), but
it becomes higher along the SD prole. Values of ACL
25-33
are in the
range of 28.18–29.01 (Table 2). Values of (n-C
27
+n-C
29
)/(n-C
31
+n-C
33
)
ratio are in the range of 0.96 (SD 3) to 1.35 (SNO) (Table 2). P
aq
proxy
ratio ranges from 0.45 to 0.54 (Table 2).
4.2.2. Terpanes
The typical distribution and abundance of terpanes for all investi-
gated samples are given in Fig. 5. The presence of hopanoid C
30
hop-22
(29)-ene, which is also known as a diploptene, characterizes samples
from all three sites, ˇ
Sanac at Obrovˇ
cani, Dublje (SD), ˇ
Sanac at Obrva
(SNO) and Liki´
ca forest (LS).
4.2.3. Diterpanes
Sample SNO is characterized by the presence of pimarane, sandar-
acopimaradiene (m/z 272) and dehydroabietane (m/z 270), which are
known plant metabolites (Fig. 6). This sample also shows the prominent
presence of 16
α
(H)-phyllocladane (Fig. 6).
4.3. Concentrations of anions and organic carbon in samples from SD and
SNO proles
Analysis of the spatial distribution of different anions was done in
order to determine space use and activity zones for the Obrovac-type
settlements. Concentrations and spatial distribution of characteristic
anions, as well as organic carbon content, are given in Table 3 and
Figs. 7 and 8.
Table 1
Bulk organic-geochemical parameters for ˇ
Sanac at Obrovˇ
cani, Dublje (SD),
ˇ
Sanac at Obrva (SNO) and Liki´
ca forest (LS).
Sample SD 1 SD 2 SD 3 SNO LS
Bitumen (
μ
g/g) 324 157 195 391 107
Saturated fraction (%) 1.02 4.76 1.46 5.13 11.47
Aromatic fraction (%) 4.08 11.11 4.38 8.97 22.95
Fig. 3. Distribution of n-alkanes in samples at ˇ
Sanac at Obrovˇ
cani, Dublje
(SD 1–3).
G. Veselinovi´
c et al.

Quaternary International xxx (xxxx) xxx
6
5. Discussion
5.1. Palaeoenvironment reconstruction using biomarkers
Based on the analysis of biomarkers from the three enclosed sites,
similarities and differences between their particular locales can be
established. Bimodal n-alkane distribution in SD 1–3 samples (Fig. 3)
indicates a mixed origin of organic matter and contribution of vascular
and nonvascular plants as well as microorganisms in precursor biomass.
The predominance of short-chain n-alkanes in the sample SD 1 indicates
the contribution of lower organisms, algae and bacteria (Tissot and
Welte, 1984; Meyers and Ishiwatari, 1993; Peters et al., 2005; Rao et al.,
2011), but also could indicate biodegradation (Peters et al., 2005;
Eckmeir and Wiesenberg, 2009). In all the other samples, SD 2, SD 3,
SNO and LS, the C
max
(homolog with maximum abundance) is at C
23
or
C
25
(Figs. 3 and 4). These middle molecular weight n-alkanes were often
observed in Sphagnum species (Baas et al., 2000; Nott et al., 2000;
Pancost et al., 2002; Nichols et al., 2006; Huang et al., 2010) and
non-emergent aquatic macrophytes (Neto et al., 1998; Ficken et al.,
2000; Andersson et al., 2011) that represent the typical vegetation in the
wetlands. Therefore, a maximum at the mid-chain n-alkanes likely in-
dicates marshy conditions during the formation of organic matter. This
assumption is supported by values of the P
aq
parameter that are in the
range from 0.45 to 0.54, which refer to macrophyte impact indicating a
marshy environment (El Nemr et al., 2016). The results of this research
conrm the previous wetland hypothesis (Trbuhovi´
c and Vasiljevi´
c
Fig. 4. n-Alkane distribution for ˇ
Sanac at Obrva (SNO) and Liki´
ca forest (LS) samples.
Table 2
Specic organic geochemical parameters of n-alkane fraction in ˇ
Sanac at Obrovˇ
cani, Dublje (SD 1–3), ˇ
Sanac at Obrva (SNO) and Liki´
ca forest (LS) samples.
Sample Locality n-alkane
max
CPI
(C
16
–C
33
)
OEP (n-C
27
+n-
C
29
)
ACL
25-
33
TAR P
aq
Environmental context
/(n-C
31
+n-
C
33
)
SD 1 Sanac at Obrovcina, Dublje,
~15 cm depth
n-C
22
; n-
C
31
1.64 2.36 1.11 28.70 2.39 0.48 marshy conditions, with higher terrestrial plant input
SD 2 Sanac at Obrovcina, Dublje,
~30 cm depth
n-C
23
; n-
C
31
1.64 2.52 0.99 29.01 2.76 0.45 marshy conditions, with pronounced higher
terrestrial plants input
SD 3 Sanac at Obrovcina, Dublje,
~45 cm depth
n-C
25
; n-
C
31
1.86 3.01 0.96 28.68 4.29 0.50 marshy conditions, with strong higher terrestrial
plant input, more grass vegetation
SNO Sanac at Obrva, ~45 cm
depth
n-C
25
1.82 2.04 1.35 28.18 5.69 0.54 marshy conditions, woody plants input
LS Likica Forest, ~45 cm depth n-C
23
; n-
C
31
1.47 2.66 1.31 28.53 3.17 0.50 marshy conditions, with higher terrestrial plant input
Fig. 5. m/z 191 fragmentogram found in sample ˇ
Sanac at Obrovˇ
cani, Dublje (SD 2).
G. Veselinovi´
c et al.

Quaternary International xxx (xxxx) xxx
7
1975, 1983, 1983; Tripkovi´
c, 2020a) but also provide more precise in-
formation regarding the environment in which particular sites encircled
by a ditch formed and functioned.
Additionally, the dominance of C
30
hop-22(29)-ene, also referred to
as a diploptene, among hopanoids, in all investigated samples represents
one of the specic indicators of a wetland environment. Diploptene is
the simplest hopanoid, and it is considered an essential component for
most prokaryotes (Venkatesan, 1988). First, it was found in ferns
(Rohmer et al., 1984) and then in a few cyanobacteria (Ourisson et al.,
1987). Huang et al. (2010) conrmed the presence of diploptene in four
moss species (Sphagnum palustre, Aulacomnium palustre, Polytrichum
commune and Hypnum revolutum) and postulated that mosses could be an
important contributor of diploptene.
Among long-chain n-alkanes, the maximum is at C
31
in all samples,
indicating clear land plant inputs into these samples, where woody
plants display n-alkane distributions dominated by C
27
or C
29
while
grasses produce distributions dominated by C
31
(Cranwell, 1973; Eck-
meier and Wiesenberg, 2009). Feurdean et al. (2021) conrmed that the
n-alkane maximum at n-C
31
originates from grasses, but it is also known
to potentially be associated with other higher plants such as ash and
maple (Nelson et al., 2017), as well as conifers (Schwark et al., 2002).
Many other parameters indicate higher terrestrial plant input in organic
matter formation, such as CPI, (n-C
27
+n-C
29
)/(n-C
31
+n-C
33
), ACL
25-33
,
and OEP (Table 2). Even though the observed bimodal distributions
show the mixed origin of organic matter, the TAR parameter also shows
a higher input of terrestrial compared to aquatic precursors, and that
this dominance is increasing with sample depth (Table 2) along with the
SD prole, being the highest in the SNO samples.
Diterpanes 16
α
(H)-phyllocladane and pimarane have been identied
at the ˇ
Sanac at Obrva site (Fig. 6). Generally, the presence of these
compounds indicates that some of the conifer families, such as Pinus,
Picea, Abies, Taxus, and Juniperus, possibly contributed to organic
matter formation (Romero-Sarmiento et al., 2011; Stojanovi´
c et al.,
2012). These conifers probably grew at the periphery of wetlands and
indicate the ecological complexity of the investigated area. Dehy-
droabietanes are well-studied terpenes produced by conifers, together
with a variety of mono- and sesquiterpenes, as signicant components of
resin (Phillips and Croteau, 1999). They have been used extensively in
palaeobotany, palaeoethnobotany, and organic geochemistry as markers
for conifers (Otto and Wilde, 2001; Buckley and Evershed, 2001), but
they are also widespread in cyanobacteria (Costa et al., 2016).
The results suggest signicant biodiversity in the Maˇ
cva region
during the Early Holocene, when the rst Late Neolithic/Early Eneo-
lithic settlers established the Obrovac type-sites. Analyzed samples from
all three sites (SD, SNO, LS) point to water-logged soils, but there is also
a strong signal of odd long-chain n-alkanes, especially n-C
31
. Some
investigations have shown that the dominance of n-C
31
is mostly related
to grass vegetation and/or higher plants, such as ash and maple (Duan
and He, 2011; Feurdean et al., 2021; Nelson et al., 2017). This would
suggest that, despite the fact that this was a wetland environment, the
presence of higher plants made these sites favorable for short or long
periods of occupation. The presence of higher plants suggests the
availability of resources (e.g., construction material), that would enable
settlement formation.
Most of the Obrovac-type sites were encircled by a ditch. Inll of
these ditches surrounding the sites consists of water-saturated sandy
gley sediment (Tripkovi´
c and Penezi´
c, 2017), suggesting long-term
water presence, either in the form of standing or owing water. Occa-
sionally, the Obrovac-type sites were located within an old meander,
sometimes surrounded by an oxbow lake. During ancient times, seasonal
water table movements could have been drastic. Most probably, even in
the occasions where the ditch was a natural feature in the landscape that
was used by the prehistoric inhabitants, it seems that it was also main-
tained by them.
5.2. Activity patterns at Obrovac-type sites based on anions distribution
and organic carbon content
Prehistoric deposits mainly consist of animal bones, ceramics, lithics
and other material culture remains, but also ash, organic matter and
organic waste, which alter the chemical and physical properties of the
occupied soil (Salisbury et al., 2013). In this study, anion signatures are
identied, and certain activities are suggested. Concentrations of
investigated anion and organic carbon contents along proles SNO and
SD are given in Table 3, and the spatial distribution of selected anions
are depicted in Figs. 7 and 8.
The high concentrations of sulfates, as well as high concentrations of
phosphates, associated with lower concentrations of carbonates, reveal a
denser cluster of deposited organic matter located in the northern part of
the SNO prole (Table 3, Fig. 7). From the elevated concentrations of
phosphates surrounding the north part of the prole, it could be
assumed that there was a possible area used for food preparation in the
vicinity (Parnell and Terry, 2002) that could include activities such as
cooking and storage. In the central part of SNO, the concentration of
nitrate, nitrite, acetate, chloride, as well as phosphates is increased, and
this may indicate plant storage or the presence of a livestock pen. This
could possibly be an area where animals and their fodder were kept. In
the eastern part there is an increased concentration of nitrates and this
may represent a garden since, in the open space, nitrites are oxidized to
nitrates in the presence of oxygen. Nitrates also accumulate in plant
biomass (Brye et al., 2003; Strahm and Harrison, 2006). At the same
place, there is also an elevated concentration of organic carbon, which
could also be explained by a tended area of plants. Finally, the southern
part, where the carbonate concentration is elevated, may represent a
sleeping area or some other area without intensive activities.
The central-southwest area of the SD site contains high concentration
of phosphates, sulfates, nitrite and, to an extent, nitrate. If we consider
low concentrations of carbonates and organic carbon in the same sam-
ples, as well as high concentrations of phosphates, it probably designates
a cooking area and activities such as food processing or storage (Fig. 8).
A supposed cooking area was located close to a building with a hearth
detected during excavation (Trbuhovi´
c and Vasiljevi´
c, 1973). The
central-southeast prole contains high reservoirs of phosphates, sul-
fates, carbonates, and acetates, as well as organic carbon content that
indicate activities such as food storage. By contrast, the zone in the
northeast part of the site shows a high concentration of nitrate and ni-
trite. That area is free of indicative chemical signals, which could
designate activities that could not be dened at this level of
examination.
The Obrovac-type sites are presumed to be places for living, which
has been proven for previously excavated sites ˇ
Sanac-Izba at Lipolist
(Tripkovi´
c et al., 2017) or Obrovac in Ratkovaˇ
ca (Lug)-Dublje
Fig. 6. Characteristic diterpanes found in sample ˇ
Sanac at Obrva (SNO).
G. Veselinovi´
c et al.

Quaternary International xxx (xxxx) xxx
8
(Tripkovi´
c et al.: in preparation). In these locations, burnt wattle and
daub remains and other archaeological materials suggest intensive use
of settled places. The assumption is that the other Obrovac-type sites
might have also been used for living, but further analyses are needed to
explain the extent of occupation or differences in cultural remains. Small
quantities of cultural remains at the SNO or SD sites indicate their
relatively short periods of usage, which also brings to light the possi-
bility that such places were distinct locations of human occupation.
Interestingly, anion-based analysis delineates certain activity zones
at both SNO and SD sites, which means that the Obrovac type-sites
manifest rather complex spatial behavior that contradicts their size,
available space, and poorly preserved cultural remains. The spatial ac-
tivities tentatively proposed by this study need to be tested and validated
through an extensive, multidisciplinary excavation processes in the
future.
6. Conclusions
The subject of this research was small archaeological mound sites
encircled by a ditch located in Western Serbia. This research not only
conrms previous wetland hypotheses for the ecological contexts of
these sites but also provides more precise information on the environ-
ment in which particular sites formed and used. The analysis of bio-
markers from the three enclosed sites suggests signicant biodiversity
within the Maˇ
cva region at the time when the Late Neolithic/Early
Eneolithic inhabitants established the Obrovac type-sites. Based on the
distribution of n-alkanes with a maximum at n-C
25
as well as a pro-
nounced presence of hopanoid C
30
hop-22(29)-ene (diploptene), it could
be concluded that macrophytes were the dominant source of organic
matter, which implies that these sites were established in an environ-
ment predominantly formed under wetland and oodplain inuence.
The signicant abundance of long-chain n-alkanes indicates the pres-
ence of terrestrial plants, pointing to a habitable environment for the
Table 3
Concentrations of anions (mg/g) and organic carbon, C org. (%) in samples from SD and SNO proles.
Fluoride Chloride Carbonate Nitrite Sulfate Bromide Nitrate Phosphate Acetate Formate Oxalate C org.
SD1 0 m 1.43 33.48 94.89 38.62 74.82 9.64 56.23 8.01 177.74 45.54 6.62 1.43
SD1 4 m 0.90 41.91 2.80 7.88 38.59 0.23 44.74 0.03 42.85 19.84 4.20 0.90
SD1 6 m 0.90 64.06 24.53 1.48 54.46 0.53 100.24 0.37 86.31 4.63 6.90 0.90
SD1 8 m 1.86 29.74 7.02 24.25 77.33 0.67 55.90 0.70 103.16 39.56 12.24 1.86
SD1 10 m 1.67 16.66 19.78 28.98 35.17 1.49 76.81 0.48 123.09 28.51 12.59 1.67
SD1 12 m 2.48 46.95 23.41 44.45 64.87 1.49 121.73 0.36 142.57 11.46 1.93 2.48
SD 1 14 m 1.58 50.80 132.93 73.89 53.85 70.63 236.20 1.02 20.77 10.61 16.11 1.58
SD 1 16 m 1.20 26.61 76.53 30.37 55.95 0.13 72.63 1.52 18.00 10.16 7.31 1.20
SD1 18 m 3.04 32.51 3.29 39.05 53.45 0.79 62.05 1.21 71.38 4.22 23.18 3.04
SD 1 20 m 1.00 48.64 33.18 34.23 39.61 0.17 75.44 0.81 20.89 13.24 6.36 1.00
SD 1 22 m 2.48 24.07 62.99 34.24 56.23 0.45 91.55 1.68 25.29 13.46 1.99 2.48
SD 1 23 m 1.31 19.40 21.18 25.63 39.37 1.12 44.01 0.51 10.03 12.05 1.71 1.31
SD2 0 m 3.73 35.16 56.08 50.66 51.76 1.00 65.75 0.95 180.54 12.76 4.34 3.73
SD2 2 m 1.79 33.86 2.75 13.97 44.73 0.61 28.92 0.42 186.97 35.87 1.21 1.79
SD2 6 m 1.45 29.09 17.76 43.02 61.07 0.46 206.05 0.38 169.43 1.95 2.05 1.45
SD2 8 m 4.54 35.97 23.34 9.88 61.74 0.36 141.52 0.89 44.88 0.88 20.68 4.54
SD2 10 m 1.48 13.47 5.72 15.39 29.96 0.19 45.95 1.73 59.30 31.39 8.49 1.48
SD2 12 m 2.72 48.61 18.01 36.04 40.49 0.52 63.24 0.36 191.64 0.89 0.02 2.72
SD2 14 m 3.94 31.11 12.25 43.90 39.04 0.56 125.07 1.44 39.06 12.37 0.05 3.94
SD2 16 m 3.70 24.90 61.70 31.97 62.07 0.08 188.15 1.15 24.40 20.09 18.10 3.70
SD2 18 m 3.72 45.52 47.53 31.00 51.23 0.60 92.87 0.60 23.73 0.61 8.96 3.72
SD2 20 m 2.51 23.35 16.22 6.33 37.42 1.70 45.87 0.52 11.52 5.26 14.31 2.51
SD2 22 m 2.58 21.83 2.27 6.43 35.20 0.57 42.73 0.51 37.26 0.86 4.73 2.58
SD2 24 m 4.39 51.13 177.65 31.93 51.30 0.25 99.92 0.75 55.09 0.17 15.88 4.39
SD2 25 m 4.15 83.97 75.49 49.96 49.03 0.47 78.76 0.48 45.87 0.72 4.16 4.15
SNO1 0 m 22.87 71.18 6.65 9.39 51.29 0.33 133.54 0.75 100.37 2.16 33.62 22.87
SNO1 2 m 8.58 40.23 14.40 21.26 35.25 0.52 61.02 0.15 50.01 40.86 9.15 8.58
SNO1 4 m 8.27 89.33 23.33 102.32 91.26 0.31 89.33 1.86 91.33 18.32 12.37 8.27
SNO1 6 m 9.18 92.74 14.16 143.30 80.02 0.25 100.04 1.96 85.29 23.05 20.75 9.18
SNO1 8 m 8.71 39.64 34.16 22.76 33.45 0.47 160.53 0.46 52.87 2.69 8.58 8.71
SNO1 10 m 10.59 47.60 16.11 44.23 54.94 0.07 95.54 0.53 77.73 1.82 10.90 10.59
SNO1 12 m 6.49 90.44 54.13 39.67 58.59 0.68 90.98 0.46 34.17 17.86 24.40 6.49
SNO1 14 m 7.36 58.41 170.49 96.69 63.31 2.27 147.23 0.76 44.73 18.95 4.54 7.36
SNO1 16 m 3.95 55.90 95.93 7.52 52.02 0.24 54.45 0.10 32.53 17.89 4.69 3.95
SNO1 20 m 11.07 68.33 27.46 124.37 77.33 0.35 151.35 0.35 29.77 25.26 22.88 11.07
SNO1 21 m 10.33 113.46 110.77 79.69 78.81 0.08 120.66 0.15 91.10 15.36 8.98 10.33
SNO1 22 m 7.83 61.21 97.55 73.85 57.36 0.50 141.58 0.68 12.23 0.33 15.66 7.83
SNO1 24 m 8.29 61.37 30.01 143.40 62.60 0.03 136.44 0.17 32.14 25.65 18.32 8.29
SNO1 26 m 7.89 52.79 301.63 74.24 39.48 0.10 137.88 0.92 50.08 22.23 14.39 7.89
SNO1 28 m 10.72 26.37 16.18 105.32 34.40 0.97 90.16 0.38 72.54 10.25 6.80 10.72
SNO1 30 m 7.63 34.24 68.85 116.65 40.51 0.44 118.31 0.40 24.22 16.68 10.33 7.63
NO2 0 m 9.32 44.80 11.81 165.29 76.32 0.57 126.57 0.37 45.52 26.13 12.91 9.32
SNO2 2 m 4.23 74.33 32.35 84.33 112.65 0.41 96.32 1.26 54.33 11.87 21.26 4.23
SNO2 4 m 8.26 38.66 40.36 84.17 43.43 0.67 87.85 1.20 70.37 1.77 9.89 8.26
SNO2 6 m 6.86 39.55 9.28 124.21 42.73 0.07 146.84 0.35 34.39 26.22 13.06 6.86
SNO2 8 m 8.18 28.88 46.08 108.13 41.77 0.71 48.61 0.50 38.60 12.72 14.62 8.18
SNO2 10 m 9.47 156.83 21.33 30.38 55.21 0.86 126.98 36.17 66.38 2.02 23.54 9.47
SNO2 12 m 5.57 49.45 190.98 86.20 52.54 0.56 155.15 0.66 13.63 4.29 11.92 5.57
SNO2 14 m 11.58 58.56 88.52 46.66 49.16 0.18 85.47 0.11 10.16 11.22 5.79 11.58
SNO2 16 m 5.16 26.86 196.88 6.12 35.23 0.22 99.80 0.79 45.02 0.87 3.69 5.16
SNO2 18 m 6.67 44.75 50.15 14.81 80.17 0.52 254.32 0.44 44.66 10.43 1.99 6.67
SNO2 20 m 12.24 72.24 124.64 81.26 91.17 0.71 86.06 0.87 10.54 3.23 15.26 12.24
SNO2 22 m 9.45 44.57 148.71 104.74 31.28 0.82 64.80 36.91 26.84 12.10 21.04 9.45
G. Veselinovi´
c et al.

Quaternary International xxx (xxxx) xxx
9
Fig. 7. Spatial distribution of anion concentrations and suggested activity zones along with prole ˇ
Sanac at Obrva (SNO).
Fig. 8. Spatial distribution of anion concentrations and C org. content along ˇ
Sanac at Obrovˇ
cani, Dublje (SD) prole and proposed activity zones.
G. Veselinovi´
c et al.

Quaternary International xxx (xxxx) xxx
10
rst settlers in this region. Due to the lack of extensive archaeological
excavations, the spatial distribution of different on-site activities is of
limited scope. Based on the soil anion composition, spatial signals have
been recognized, and certain activities suggested, such as cooking areas,
places for food storage and preparation, sleeping areas etc. If proven,
this would mean that the Obrovac type-sites manifest rather complex
spatial behavior despite their size and available space. Their spatial
delineation indicates the complexity of cultural and spatial practices at
these sites, where more detailed investigations are yet to be conducted.
These initial studies set an important baseline for future, more detailed
analyses, and the approach used here could be applied as a screening
technique for sediment and soil analyses in archaeology in the region.
Data availability
The authors conrm that the data supporting the ndings of this
study are available within the article [and/or] its supplementary
materials.
CRediT authorship contribution statement
Gorica Veselinovi´
c: Conceptualization, Methodology, Visualiza-
tion, Writing – original draft, preparation, Writing – review & editing.
Boban Tripkovi´
c: Supervision, Writing – review & editing. Nevena
Anti´
c: Investigation, Data curation. Aleksandra ˇ
Sajnovi´
c: Methodol-
ogy, Data curation, Writing – review & editing. Milica Kaˇ
sanin-Grubin:
Data curation, Visualization, Writing – review & editing. Tomislav
Tosti: Methodology, Investigation, Data curation. Kristina Penezi´
c:
Conceptualization, Visualization, Writing – review & editing.
Declaration of competing interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Acknowledgments
The study was supported by the Ministry of Education, Science and
Technological Development, Republic of Serbia (Grant No. 451-03-9/
2021-14/200026, 451-03-9/2021-14/200168 and 451-03-9/2021-14/
200358). The authors would like to acknowledge Dr. Patalano, and one
further anonymous reviewer, for their useful comments, which greatly
improved this paper.
References
Andersson, R., Kuhry, P., Meyers, P., Zegühr, Y., Crill, P., M¨
orth, M., 2011. Impacts of
paleohydrological changes on n-alkane biomarker compositions of a Holocene peat
sequence in eastern European Russian Arctic. Org. Geochem. 42, 1065–1075.
https://doi.org/10.1016/j.orggeochem.2011.06.020.
Baas, M., Pancost, R., van Geel, B., Sinninghe Damst´
e, J.S., 2000. A comparative study of
lipids in Sphagnum species. Org. Geochem. 31, 535–541. https://doi.org/10.1016/
S0146-6380(00)00037-1.
Bai, Y., Fang, X., Nie, J., Wang, Y., Wu, F., 2009. A 697 preliminary reconstruction of the
paleoecological and paleoclimatic history of the Chinese Loess Plateau from the
application of biomarkers. Palaeogeogr. Palaeoclimatol. Palaeoecol. 271, 161–169.
https://doi.org/10.1016/j.palaeo.2008.10.006.
Bori´
c, D., 2009. Absolute dating of metallurgical innovations in the Vinˇ
ca culture of the
Balkans, in metals and societies: studies in honour of barbara S. Ottaway. In:
Kienlin, T.L., Roberts, B.W. (Eds.), Universit¨
atsforschungen zur pr¨
ahistorischen
Arch¨
aologie 169, 191–245.
Bouloubassi, I., Saliot, A., 1993. Investigation of anthropogenic and natural organic
inputs in estuarine sediments using hydrocarbon markers (NAH, LAB, PAH).
Oceanol. Acta 16, 145–161.
Bourbonnier, R.A., Mayers, P.A., 1996. Anthropogenic inuences on hydrocarbon
contents of sediments deposited in eastern Lake Ontario since 1800. Environ. Geol.
28, 22–28. https://doi.org/10.1007/s002540050074.
Brye, K.R., Slaton, N.A., Savin, M.C., Norman, R.J., Miller, D.M., 2003. Short-term effects
of land leveling on soil physical properties and microbial biomass. Soil Sci. Soc. Am.
J. 67, 1405–1417. https://doi.org/10.2136/sssaj2003.1405.
Buckley, S.A., Evershed, R.P., 2001. Organic chemistry of embalming agents in Pharaonic
and Graeco-Roman mummies. Nature 413, 837–841. https://doi.org/10.1038/
35101588.
Bulatovi´
c, A., Bankoff, A.H., Powell, W., Filipovi´
c, V., 2020. Some remarks on the genesis
of the early eneolithic in the central Balkans. Starinar 70, 9–40.
Bulatovi´
c, A., Filipovi´
c, V., Grigori´
c, R., 2017. Loznica, Kulturna Stratigraja
Praistorijskih Lokaliteta U Jadru, Rađevini I Azbukovici. Arheoloˇ
ski institut,
Beograd.
Bulatovi´
c, A., Vander Linden, M., Gori, M., 2018. New contributions to the absolute
chronology at the early eneolithic cultures in the central Balkans. Starinar 68, 19–32.
Bush, R.T., McInerney, F.A., 2013. Leaf wax n-alkane distributions in and across modern
plants: implications for paleoecology and chemotaxonomy. Geochem. Cosmochim.
Acta 117, 161–179. https://doi.org/10.1016/j.gca.2013.04.016.
Carr, A.S., Boom, A., Grimes, H.L., Chase, B.M., Meadows, M.E., Harris, A., 2014. Leaf
wax n-alkane distributions in arid zone South African ora: environmental controls,
chemotaxonomy and palaeoecological implications. Org. Geochem. 67, 72–84.
Chapman, J., 1981. The Vinˇ
ca Culture of South-East Europe. Studies in Chronology,
Economy and Society. International Series I - 117. British Archaeological Reports,
Oxford.
Costa, M.S., Rego, A., Ramos, V., Afonso, T.B., Freitas, S., Preto, M., Lopes, V.,
Vasconcelos, V., Magalh˜
aes, C., Le˜
ao, P.N., 2016. The conifer biomarkers
dehydroabietic and abietic acids are widespread in Cyanobacteria. Science Report 6,
23436. https://doi.org/10.1038/srep23436.
Cranwell, P.A., 1973. Chain-length distribution of n-alkanes from lake sediments in
relation to post-glacial environmental change. Freshw. Biol. 3, 259–265. https://doi.
org/10.1111/j.1365-2427.1973.tb00921.x.
Duan, Y., He, J., 2011. Distribution and isotopic composition of n-alkanes from grass,
reed and tree leaves along a latitudinal gradient in China. Geochem. J. 45, 199–207.
https://doi.org/10.2343/geochemj.1.0115.
Duan, Y., Xu, L., 2012. Distributions of n-alkanes and their hydrogen isotopic
composition in plants from Lake Qinghai (China) and the surrounding area. Appl.
Geochem. 27, 806–814.
Eckmeier, E., Wiesenberg, G.L.B., 2009. Short-chain n-alkanes (C
16-20
) in ancient soil are
useful molecular markers for prehistoric biomass burning. J. Archaeol. Sci. 36,
1590–1596. https://doi.org/10.1016/j.jas.2009.03.021.
Eglinton, G., Hamilton, R.J., 1967. Leaf epicuticular waxes. Science 156, 1322–1335.
https://doi.org/10.1126/science.156.3780.1322.
El Nemr, A., Moneer, A.A., Ragab, S., El Sikaily, A., 2016. Distribution and sources of n-
alkanes and polycyclic aromatic hydrocarbons in shellsh of the Egyptian Red Sea
coast. Egypt. J. Aquat. Res. 42, 121–131. https://doi.org/10.1016/j.
ejar.2016.05.003.
Evershed, R.P., 2008. Organic residue analysis in archaeology: the archaeological
biomarker revolution. Archaeometry 50, 6. https://doi.org/10.1111/j.1475-
4754.2008.00446.x.
Farrington, J.W., Quinn, J.G., 2015. Unresolved Complex Mixture" (UCM): a brief history
of the term and moving beyond it. Mar. Pollut. Bull. 96, 29–31. https://doi.org/
10.1016/j.marpolbul.2015.04.039.
Feurdean, A., Grindean, R., Florescu, G., Tant¸˘
au, I., Niedermeyer, E.M., Diaconu, A.-C.,
Hutchinson, S.M., Nielsen, A.B., Sava, T., Panait, A., Braun, M., Hickler, T., 2021.
The transformation of the forest steppe in the lower Danube Plain of southeastern
Europe: 6000 years of vegetation and land use dynamics. Biogeosciences 18,
1081–1103. https://doi.org/10.5194/bg-18-1081-2021.
Ficken, K.J., Li, B., Swain, D.L., Eglinton, G., 2000. An n-alkane proxy for the
sedimentary input of submerged/oating freshwater aquatic macrophytes. Org.
Geochem. 31, 745–749. https://doi.org/10.1016/S0146-6380(00)00081-4.
Grimalt, J.O., Torras, E., Albaig´
es, J., 1988. Bacterial reworking of sedimentary lipids
during sample storage. Adv. Org. Geochem. 13, 741–746. https://doi.org/10.1016/
0146-6380(88)90096-4.
Holliday, T.V., Gartner, G.W., 2007. Methods of soil P analysis in archaeology.
J. Archaeol. Sci. 34, 301–333. https://doi.org/10.1016/j.jas.2006.05.004.
Huang, X., Wang, C., Xue, J., Mayers, P.A., Zhang, Z., Tan, K., Zhang, Z., Xie, S., 2010.
Occurrence of diploptene in moss species from the Dajiuhu Peatland in southern
China. Org. Geochem. 41, 321–324. https://doi.org/10.1016/j.
orggeochem.2009.09.008.
Lejay, M., Alexis, M., Qu´
en´
ea, K., Sellami, F., Bon, F., 2016. Organic signatures of
fireplaces: experimental references for archaeological interpretations, Org.
Geochem. (Tokyo. 1967) 99, 67–77. https://doi.org/10.1016/j.
orggeochem.2016.06.002.
Li, Y., Yang, S., Wang, X., Hu, J., Cui, L., Huang, X., Jiang, W., 2016. Leaf wax n-alkane
distributions in Chinese loess since the Last Glacial Maximum and implications for
paleoclimate. Quat. Int. 399, 190–197. https://doi.org/10.1016/j.
quaint.2015.04.029.
Meyers, P.A., Ishiwatari, R., 1993. Lacustrine organic geochemistry-an overview of
indicators of OM sources and diagenesis in lake sediments. Org. Geochem. 20,
867–900. https://doi.org/10.1016/0146-6380(93)90100-P.
Milanovi´
c, D., 2019. Insight into the regional distribution and geographic setting of the
Vinˇ
ca and BSK settlements in the Central Balkans and its implications. Starinar 69,
61–84.
Mileto, S., Kaiser, E., Rassamakin, Y., Evershed, R., 2017. New insights into the
subsistence economy of the Eneolithic Dereivka culture of the Ukrainian North-
Pontic region through lipid residues analysis of pottery vessels. J. Archaeol. Sci. Rep.
13, 67–74. https://doi.org/10.1016/j.jasrep.2017.03.028.
Nelson, D., Knohl, A., Sachse, D., Schefuß, E., Kahmen, A., 2017. Sources and abundances
of leaf waxes in aerosols in central Europe. Geochem. Cosmochim. Acta 198,
299–314. https://doi.org/10.1016/j.gca.2016.11.018.
G. Veselinovi´
c et al.

Quaternary International xxx (xxxx) xxx
11
Neto, E.V.S., Hayes, J.M., Takaki, T., 1998. Isotopic biogeochemistry of the neocomian
lacustrine and upper aptian marine-evaporitic sediments of the potiguar basin,
northeastern Brazil. Org. Geochem. 28, 361–381. https://doi.org/10.1016/S0146-
6380(98)00007-2.
Nichols, J.E., Booth, R.K., Jackson, S.T., Pendall, E.G., Huang, Y., 2006. Paleohydrologic
reconstruction based on n-alkane distributions in ombrotrophic peat. Org. Geochem.
37, 1505–1513. https://doi.org/10.1016/j.orggeochem.2006.06.020.
Nott, C.J., Xie, S., Avsejs, L.A., Maddy, D., Chambers, F.M., Evershed, R.P., 2000. N-
Alkane distributions in ombrotrophic mires as indicators of vegetation change
related to climatic variation. Org. Geochem. 31, 231–235. https://doi.org/10.1016/
S0146-6380(99)00153-9.
Orton, D., 2012. Herding, settlement, and chronology in the balkan neolithic. Eur. J.
Archaeol. 15, 5–40. https://doi.org/10.1179/1461957112Y.0000000003.
Otto, A., Wilde, V., 2001. Sesqui-, Di-, and triterpenoids as chemosystematic markers in
extant conifers - a review. Bot. Rev. 67, 141–238. https://doi.org/10.1007/
BF02858076.
Ourisson, G., Rohmer, M., Poralla, K., 1987. Prokaryotic hopanoids and other
polyterpenoid sterol surrogates. Annu. Rev. Microbiol. 41, 301–333. https://10.114
6/annurev.mi.41.100187.001505.
Pancost, R.D., Baas, M., van Geel, B., Sinninghe Damst´
e, J.S., 2002. Biomarkers as
proxies for plant inputs to peats: an example from a sub-boreal ombrotrophic bog.
Org. Geochem. 33, 675–690. https://doi.org/10.1016/S0146-6380(02)00048-7.
Parnell, J.J., Terry, R.E., 2002. Soil chemical analysis applied as an interpretive tool for
ancient human activities in piedras negras, Guatemala. J. Archaeol. Sci. 29,
379–404. https://doi.org/10.1006/jasc.2002.0735.
Peters, K.E., Walters, C., Moldowan, J.M., 2005. The Biomarker Guide, Biomarkers and
Isotopes in the Environment and Human History. University Press, Cambridge.
Phillips, M.A., Croteau, R.B., 1999. Resin-based defenses in conifers. Trends Plant Sci. 4,
184–190. https://doi.org/10.1016/S1360-1385(99)01401-6.
Rao, Z.G., Wu, Y., Zhu, Z.Y., Jia, G.D., Henderson, A., 2011. Is the maximum carbon
number of long-chain n-alkanes an indicator of grassland or forest? Evidence from
surface soils and modern plants. Chin. Sci. Bull. 56, 1714–1720. https://doi.org/
10.1007/s11434-011-4418-y.
Rohmer, M., Bouvier-Nave, P., Ourisson, G., 1984. Distribution of hopanoid triterpenes
in prokaryotes. J. Gen. Microbiol. 130, 1137–1150. https://doi.org/10.1099/
00221287-130-5-1137.
Romero-Sarmiento, M., Riboulleau, A., Vecoli, M., Versteegh, G., Laggoun-D´
efarge, F.,
2011. Aliphatic and aromatic biomarkers from carboniferous coal deposits at dunbar
(east lothian, scotland): palaeobotanical and palaeoenvironmental signicance.
Palaeogeogr. Palaeoclimatol. Palaeoecol. 309, 309–326. https://doi.org/10.1016/j.
palaeo.2011.06.015.
Salisbury, B.R., Bert´
ok, G., B´
acsmegi, G., 2013. Integrated prospection methods to dene
small-site settlement structure: a case study from neolithic Hungary. Archaeol.
Prospect. 20, 1–10. https://doi.org/10.1002/arp.1442.
Schwark, L., Zink, K., Lechtenbeck, J., 2002. Reconstruction of postglacial to early
Holocene vegetation history in terrestrial Central Europe via cuticular lipid
biomarkers and pollen records from lake sediments. Geology 30, 463–466. https://
doi.org/10.1130/00917613(2002)030<0463:ROPTEH>2.0.CO;2.
ˇ
Soˇ
si´
c-Klindˇ
zi´
c, R., Tripkovi´
c, B., 2018. Okresani kameni artefakti sa ranoeneolitskog
lokaliteta ˇ
Sanac-Izba (Lipolist, zapadna Srbija). Arhaika 6, 1–26.
Stojanovi´
c, K., ˇ
Zivoti´
c, D., ˇ
Sajnovi´
c, A., Cvetkovi´
c, O., Nytoft, H.P., Scheeder, G., 2012.
Drmno lignite eld (Kostolac basin, Serbia): origin and palaeoenvironmental
implications from petrological and organic geochemical studies. J. Serb. Chem. Soc.
77, 1109–1127. https://doi.org/10.2298/JSC111126017S.
Stoji´
c, M., Cerovi´
c, M., 2011. ˇ
Sabac: Cultural Stratigraphy of Prehistoric Sites in the
Drina Valley. Archaeological Institute. ˇ
Sabac: National Museum, Belgrade.
Strahm, B.D., Harrison, R.B., 2006. Nitrate sorption in a variable-charge forest soil of the
Pacic Northwest. Soil Sci. 171, 313–321. https://doi.org/10.1097/01.
ss.0000209355.76407.16.
Tasi´
c, N., Mari´
c, M., Bronk Ramsey, Ch, Kromer, B., Barclay, A., Bayliss, A., Beavan, N.,
Gaydarska, B., Whittle, A., 2016. Vinˇ
ca-belo brdo, Serbia: the times of a tell.
Germania 93, 1–75.
Tissot, B.P., Welte, D.H., 1984. Petroleum Formation and Occurrence, second ed.
Springer, Heidelberg.
Trbuhovi´
c, V., Vasiljevi´
c, M., 1973. Rekognosciranja u Podrinju. ˇ
Sanac u Dublju, SO
Bogati´
c – neolitsko naselje. Arheoloˇ
ski pregled 15, 133.
Trbuhovi´
c, V., Vasiljevi´
c, M., 1975. “Obrovci” poseban tip neolitskih naselja u zapadnoj
Srbiji. Starinar 24–25, 157–161.
Trbuhovi´
c, V., Vasiljevi´
c, M., 1983. Najstarije Zemljoradniˇ
cke Kulture U Podrinju.
Narodni muzej, ˇ
Sabac.
Tripkovi´
c, B., Penezi´
c, K., 2017. On-site and off-site in western Serbia: a
geoarchaeological perspective of Obrovac-type settlements. Quat. Int. 429, 35–44.
https://doi.org/10.1016/j.quaint.2015.12.004, 2017.
Tripkovi´
c, B., 2013. Household and Community. House and Settlement Histories in the
Late Neolithic of the Central Balkans. Faculty of Philosophy, Belgrade.
Tripkovi´
c, B., 2017. Nastanak lokaliteta tipa „Obrovac“: stratigrafsko vrednovanje stare
hipoteze. Arhaika 5, 65–78.
Tripkovi´
c, B., 2020. The neolithic and postneolithic settlement mounds of western
Serbia. In: Tasi´
c, N.N., Urem-Kotsou, D., Buri´
c, M. (Eds.), Making Spaces into Places.
The North Aegean, the Balkans and Western Anatolia in the Neolithic. British
Archaeological reports, Oxford, pp. 141–148.
Tripkovi´
c, B., 2020a. Veliki Svet I mali Svetovi. Struktura I Znaˇ
cenje Arheoloˇ
skih Praksi
U Srbiji. Filozofski fakultet, Beograd.
Tripkovi´
c, B., Cerovi´
c, M., Buli´
c, D., 2013. Kulturno Nasleđe Severozapadne Srbije:
Lokaliteti Tipa „Obrovac”, ˇ
Cetrdeset Godina Kasnije. Rezultati Novih Arheoloˇ
skih
Istraˇ
zivanja U Severozapadnoj Srbiji I Susednim Teritorijama. Srpsko arheoloˇ
sko
druˇ
stvo i Zavod za zaˇ
stitu spomenika kulture, Beograd, Valjevo, pp. 45–56.
Tripkovi´
c, B., Cerovi´
c, M., Filipovi´
c, D., Tripkovi´
c, A., ˇ
Zivaljevi´
c, I., 2017. ˇ
Sanac-Izba kod
Lipolista, lokalitet tipa “obrovac”: stratigraja i relativna hronologija. Glasnik
Srpskog arheoloˇ
skog druˇ
stva 33, 47–72.
Venkatesan, M.I., 1988. Diploptene in antarctic sediments. Geochem. Cosmochim. Acta
52, 217–222. https://doi.org/10.1016/0016-7037(88)90070-1.
Veselinovi´
c, G., ˇ
Zivoti´
c, D., Penezi´
c, K., Kaˇ
sanin-Grubin, M., Mijatovi´
c, N., Malbaˇ
si´
c, J.,
ˇ
Sajnovi´
c, A., 2021. Geochemical characterization of sediments from the
archaeological site Vinˇ
ca – belo brdo, Serbia. Catena 196, 104914. https://doi.org/
10.1016/j.catena.2020.104914.
G. Veselinovi´
c et al.