Radiocarbon, Vol 56, Nr 2, 2014, p 767–778 DOI: 10.2458/56.16942
© 2014 by the Arizona Board of Regents on behalf of the University of Arizona
Proceedings of the Radiocarbon and Archaeology 7th International Symposium
Ghent, Belgium, April 2013 | Edited by Mark Van Strydonck, Philippe Crombé, and Guy De Mulder
© 2014 by the Arizona Board of Regents on behalf of the University of Arizona
RESERVOIR EFFECT OF ARCHAEOLOGICAL SAMPLES FROM STEPPE BRONZE
AGE CULTURES IN SOUTHERN RUSSIA
N Shishlina
1
• V Sevastyanov
2
• E Zazovskaya
3
• J van der Plicht
4
ABSTRACT. Human and animal bone collagen from different Steppe Bronze Age cultures, including the Yamnaya, East
Manych Catacomb, and Lola cultures, show large variations in δ
13
C and δ
15
N values. More positive values of δ
13
C and δ
15
N
are caused by change in diet and/or a more arid climate. Human bones show reservoir effects caused by aquatic diet compo-
nents. These effects can be quantied by paired dating of human bone and associated terrestrial samples. Previous research
showed reservoir corrections for several cultures of the Eurasian steppes Bronze Age, leading to a revision of the local chro-
nologies. Additional investigations showed that some paired dates do not reveal reservoir effects. Recently, our database was
extended signicantly with new
13
C,
15
N, and
14
C analyses of the Lola, Krivaya, and Babino cultures. The size of the reservoir
offset for the studied cultures is discussed.
INTRODUCTION
For Bronze Age Eurasian steppe graves, paired radiocarbon dating of samples from the contempo-
raneous archaeological context enabled the identication of variations in reservoir effects obtained
for several cultures (Shishlina et al. 2007, 2009). We studied populations that exploited the area
between the Lower Don and the Lower Volga rivers and the North Caucasus piedmont. The ex-
ploitation of this new area was based on a multicomponent mobile economy, including (1) raising of
domesticated animals (sheep/goat, cattle, and probably horse); (2) seasonal movements across vast
areas, including the coastlines of the Black and Azov seas; (3) shing and gathering of wild plants;
and (4) a system of exchange. The values of δ
13
C and δ
15
N for human bones show different diet
systems, which were reconstructed for various population groups (Shishlina et al. 2012b). Humans
are at the end of the trophic chain and, as consumers, are representatives of a mixed and multicom-
ponent diet system. In addition, previously obtained strontium data indicate that at least some of
the individuals probably moved from one place to another quite regularly since their childhood and
consumed food, which causes different isotope signals (Shishlina and Larionova, in press).
The consumption of aquatic food components, including freshwater and marine food, causes
apparent
14
C ages of dated human bones. We performed analysis of
13
C,
15
N, and
14
C isotopes from
collagen extracted from human and animal bones, including paired dating of human bones and ter-
restrial samples. In previous publications, several regional offsets were suggested based on the rst
measurements of the
14
C age of paired terrestrial and human bone samples as well as terrestrial and
aquatic samples (Shishlina et al. 2007, 2009). A reservoir offset of 400–500 yr was derived for the
Early Catacomb culture, and 300–450 yr for the East Manych Catacomb culture. However, the more
paired samples analyzed, the more complicated the situation became.
Our hypothesis is that different components of the human diet are reected in different reservoir
correction values. If the human diet contained different amounts of aquatic resources with different
14
C ages, the reservoir effect signals would be different due to geological factors as well as ecologi-
cal and economic contexts of each culture. The southern part of Russia, i.e. the Caspian and Lower
Don areas, are investigated during 5000–2000 cal BC.
1. State Historical Museum, Moscow, Russia. Corresponding author. Email: nshislina@mail.ru.
2. Institute of Geochemistry and Analytical Chemistry, Russian Academy of Sciences, Moscow, Russia.
3. Institute of Geography, Russian Academy of Sciences, Moscow, Russia.
4. Centre for Isotope Research, Groningen University, Groningen, the Netherlands; also at Faculty of Archaeology, Leiden
University, Leiden, the Netherlands.
768
N Shishlina et al.
The aim of the present article is to analyze paired
14
C dates obtained for different population groups
and discuss the following issues:
(i) The variations in reservoir effect offsets obtained for several cultures, which occupied the same
area under different ecological conditions and exploited different food resources; and
(ii) A possible correlation between human stable isotopes and reservoir effect offsets.
MATERIALS AND METHODS
This study used data from several archaeological cultures, i.e. the Eneolithic, Majkop, Yamnaya,
Steppe North Caucasus, Early Catacomb, Eastern Manych, and Lola cultures. Their time intervals
and regional distributions are shown in Table 1 and Figure 1, respectively. Archaeological human
and animal bone samples were subjected to analysis. Some data were published previously (Shish-
lina at al. 2007, 2009, 2012b). New samples of human and animal bones as well as wood and plants
coming from excavations near the village of Remontnoye in the Rostov region (southern Russia)
where
14
C dates and additional stable isotope ratios were measured.
Most stable isotope ratio measurements were made in the Institute of Geochemistry and Analytical
Chemistry, the Russian Academy of Sciences, using the isotope mass-spectrometer DELTA Plus
XP (ThermoFinnigan), linked to a Flash EA element analyzer. Each sample was measured in tripli-
cate with an analytical error for δ
13
C of ±0.2‰ and for δ
15
N of ±0.2–0.3‰. A subset of the measure-
ments was made at Groningen University. After the chemical pretreatment, the prepared collagen is
combusted and turned into CO
2
by an elemental analyzer (EA), coupled on-line with a stable isotope
mass spectrometer (MS). The EA is also used for purifying the CO
2
(Aerts-Bijma et al. 2001). In
addition, the EA/MS system enables precise measurement of the δ
13
C values.
For
14
C, the CO
2
is subsequently reduced to graphite by reacting under excess H
2
gas. This graphite
is pressed into target holders that are placed in the ion source of the accelerator mass spectrometry
(AMS). The Groningen AMS facility is based on a 2.5MV accelerator (van der Plicht et al. 2000).
For fossil bones, established quality parameters are the organic carbon and nitrogen content of the
collagen, which should be within specied ranges (DeNiro 1985). For our samples, these parameters
are good to reasonable.
The isotope ratios are reported in per mil deviation with respect to the international standards VPDB
and AIR for
13
C and
15
N, respectively. Conventional
14
C dates were measured at the Institute of Ge-
ography of the Russian Academy of Sciences using liquid scintillation (lab code IGAN). AMS dates
were obtained at Groningen University (lab code GrA). For bone samples, collagen was extracted
using an improved version of the Longin (1971) method. The
14
C dates are reported in conventional
yr BP. They are calibrated into calendar years (cal BC) using the calibration curve IntCal09 (Reimer
et al. 2009).
Table 1 Analyzed cultures and their chronologies.
Culture Time interval
Eneolithic 4300–3800 cal BC
Yamnaya 3000–2500 cal BC
Early Catacomb 2700–2400 cal BC
North Caucasus 2600–2400 cal BC
East Manych Catacomb 2500–2200 cal BC
Lola, Early Babino, Krivaya Luka 2200–1900 cal BC
769
Reservoir Effect of Archaeological Samples from Steppe Bronze Age
Figure 1 Location of the Lola, Early Babino and Krivaya Luka sites: 1 – Ostrovnoy, 2 – Mandjikiny-1, 3 – Temrta 1,
4 – Khar-Zukha, 5 – Peschany V, 6 – Yashkul 1, 7 – Khomush-Oba, 8 – Linevo.
770
N Shishlina et al.
RESULTS AND DISCUSSION
We performed
14
C dating of 47 paired samples, including pairs of both terrestrial samples (wood-
plant; wood-animal bone; animal bone-plant) and terrestrial and aquatic samples (shell-plant; sh
bone-plant). Forty such pairs result from
14
C dates of human bone and associated terrestrial samples
(animal bone, wood, and plant). The results are discussed below for the various cultures (Figure 3).
Eneolithic Culture
Only one pair of dates is presently available, i.e. cow and human bones from the Eneolithic Kh-
valynsk Cemetery II, grave 10. The offset is 220
14
C yr. We assume that other
14
C dates of human
bones available for this culture also would show this apparent age. The average value of the δ
15
N
ratio in human bone collagen from the Khvalynsk humans is +13‰ (Shishlina et al. 2012a). This
provides the size of the reservoir effect correction (Shishlina et al. 2009). People belonging to this
culture consumed quite a lot of aquatic food. Stable isotope data available for ve human bones
conrm this statement (Shishlina 2008).
The consumption of freshwater sh by the local Eneolithic population of the adjacent Dnieper Basin
in Ukraine also causes apparent ages of dated humans from the Molukhov Bugor burial ground. The
reservoir effect correction for this case is 110
14
C yr (Lillie et al. 2009).
Majkop Culture
Unfortunately, our database does not contain paired dates for the Majkop culture thus far, despite
the fact that the both human and terrestrial samples obtained from different Majkop graves were
14
C
dated. Human bones of the same cultural context are sometimes older than
14
C dates of terrestrial
samples. Terrestrial samples (pairs such as plant textile-animal bone) dating from the same grave do
not show any offset (Shishlina 2008: Table 11). Stable isotope values obtained for steppe Majkop
bones (Shishlina et al. 2012b: Table 4) show that an aquatic component in the diet of this popula-
tion is likely. Without paired
14
C dates from the same Majkop grave context, we can only speculate
that
14
C dates obtained from humans might be older than
14
C dates of terrestrial samples (animal
bones, textile plant bers). Due to the average value of the δ
15
N in Majkop human bone collagen
(human = 7), which is +14‰, we estimate that the reservoir effect can be up to several hundred
years. Future paired dating will clarify this hypothesis.
Hollund et al. (2010) analyzed four pairs obtained from Aygurskiy Majkop culture graves (human
and animal bones, including snake bones). The results yield paired dates without a signicant reser-
voir age in three cases, and with a reservoir effect in one pair (from 315 to 172 yr, grave 8) (Hollund
et al. 2010: Table 3).
Yamnaya Culture
There are eight paired dates available for this local culture including dates of humans of differ-
ent age and sex, domesticated animals, bone items made of ungulate bone, and wood samples. In
addition, one pair of human and ungulate bone of the Repino culture is available.
1
In four cases,
human bones show reservoir effects, caused by aquatic diet components.
Four out of the eight paired cases do not show any age offset. This can be explained in two alter-
native ways. One is that the human diet did not include aquatic components; rather, the diet was
based on C
3
vegetation with high δ
15
N values (13–15‰), and esh/milk of domesticated animals.
1. Some scholars consider the Repino culture an early stage of the Yamnaya culture (Trifonov 1996).
771
Reservoir Effect of Archaeological Samples from Steppe Bronze Age
An alternative explanation is that humans consumed food from freshwater resources without a res-
ervoir effect. The isotope composition of the local plants and domesticated animal bones indicates
that in some cases, due to the arid conditions of the environment, plants show elevated δ
15
N values.
Therefore, the local population as well as domesticated animals feeding on these plants also show
high δ
15
N values (Shishlina et al. 2012a,b).
Reservoir effect corrections for other dated pairs vary between 200 and 270 yr (Shishlina et al. 2009,
2012b). The new pair from Peschany V, kurgan 15, grave 6 (human bone and wood) shows a very
large reservoir effect of 860 yr for the human bone collagen. This large effect dates back to around
3000 BC when people started to occupy different pastures located across the exploited area.
Catacomb Cultures (Early Catacomb, West and East Manych Catacomb Cultures)
A lot of isotopic data are available for this period (Shishlina at al. 2012b). They clearly show the
role of aquatic food components in the diet of the local steppe population. This is conrmed by sh
bones and scales identied in the graves, as well as by impressions of nets on the bottom of some
clay pots (Shishlina 2008). The size of the reservoir effect varies from 100 to 750 yr. Isotope data
for dated human bones also show large variations. We assume that some individuals consumed
sh obtained from different reservoirs, determined by their seasonal movements. People were very
mobile at that time (Shishlina 2008) and during their seasonal movements, probably consumed
aquatic food with different isotope values and
14
C ages. This leads to apparent ages in the
14
C dates
of human bones.
There are four pairs without offsets in the Catacomb culture database, though the isotope values of
such pairs are similar to those of humans who fed on sh (Shishlina et al. 2012a). A similar situa-
tion is observed for the North Caucasus Piedmont steppes (Hollund et al. 2010). Humans from the
Ipatovo burial ground, kurgan 2, in the Stavropol region show a very small reservoir effect or no
reservoir effect at all.
Lola, Babino, and Krivaya Luka Cultures
New stable isotope and
14
C data are available for these cultures and are shown in Table 2 and Fig-
ure 3. The stable isotope values for human bones vary signicantly; the diet of the people of these
cultures does not correspond to the steppe diet of the local population, the value for which is −22.2
to –17.5‰ for δ
13
C, and +9.4 to +17.1‰ for δ
15
N (Shishlina et al. 2012a). We assume that the iso-
tope ecology of the Caspian and Lower Don steppe areas changed during a period of aridization,
which reached its peak in 2200–2000 cal BC (Borisov and Mimokhod 2011). Another possible
explanation is that some individuals from those cultures came to this part of the steppe from the
regions where food components show different isotope signals. Besides that, the
87
Sr/
86
Sr values
obtained for enamel of some individuals (Shishlina and Larionova, in press) indicate that some of
these humans were not local, so they consumed food produced in different areas. This is reected
in specic isotope values in their bone collagen, which differ from the values for the local steppe
population.
Fifteen
14
C dates are available for these three cultures; 8 human bones, 6 animal bones, and 1 wood
sample were
14
C dated. They are listed in Table 3.
