Geochemical pollution of trace metals in permafrost-affected soil in the Russian Arctic marginal environment

Abstract

The Arctic marginal environment has been considered as far from industrial areas and low population. During June–July of 2016 “Russian High Latitude” expedition, 93 samples of soil genetic horizon from 25 soil profiles dug till frozen ground were sampled from 8 islands and 2 capes of the Russian Arctic without direct anthropogenic influences. Nine trace metals (Pb, Cd, Cu, Ni, Co, Zn, Fe, Mn and Hg) were measured and quantified by energy-dispersive X-ray analysis for elemental concentrations. Through analysis of divided soil groups (Haplothels, Turbels, Historthels), the factors of organic matter and cryoturbation had a significant influence on metals’ distribution except for Fe and Mn. From summarized soil master horizons (O, A, B, C), Fe and Mn are abundant in all horizons suggesting as geochemical background values. Cu, Pb, Co and Ni are distributed specifically in different horizons with leaching and accumulation process, whereas Hg is evenly disturbed in all horizons. The correlation analysis reveals that distribution of most metals in present soils is highly depended on soil properties (pH, TOC, clay and silt). Li was selected as normalizing element for metals’ concentrations from mineral layers to establish geochemical baseline concentrations. The concentrations of trace metals have been assessed by geoaccumulation index (Igeo) and enrichment factor, showing only Co and Zn are moderately polluted and slightly polluted, and Co, Cu, Zn and Pb are enriched in topsoil. Other indices as modified degree of contamination (mCdegree) and pollution load index (PLI), mCdegree show moderate degree of pollution and PLI shows unpolluted to moderate pollution load. The ecological risk indices, e.g., ecological risk factor (Er) and potential ecological risk index, show low ecological risk potential.

Full text

ORIGINAL PAPER
Geochemical pollution of trace metals in permafrost-
affected soil in the Russian Arctic marginal environment
Xiaowen Ji .Evgeny Abakumov .Vitaly Tomashunas .Vyacheslav Polyakov .
Sergey Kouzov
Received: 8 September 2019 / Accepted: 18 April 2020
ÓSpringer Nature B.V. 2020
Abstract The Arctic marginal environment has been
considered as far from industrial areas and low
population. During June–July of 2016 ‘‘Russian High
Latitude’’ expedition, 93 samples of soil genetic
horizon from 25 soil profiles dug till frozen ground
were sampled from 8 islands and 2 capes of the
Russian Arctic without direct anthropogenic influ-
ences. Nine trace metals (Pb, Cd, Cu, Ni, Co, Zn, Fe,
Mn and Hg) were measured and quantified by energy-
dispersive X-ray analysis for elemental concentra-
tions. Through analysis of divided soil groups
(Haplothels, Turbels, Historthels), the factors of
organic matter and cryoturbation had a significant
influence on metals’ distribution except for Fe and Mn.
From summarized soil master horizons (O, A, B, C),
Fe and Mn are abundant in all horizons suggesting as
geochemical background values. Cu, Pb, Co and Ni
are distributed specifically in different horizons with
leaching and accumulation process, whereas Hg is
evenly disturbed in all horizons. The correlation
analysis reveals that distribution of most metals in
present soils is highly depended on soil properties (pH,
TOC, clay and silt). Li was selected as normalizing
element for metals’ concentrations from mineral
layers to establish geochemical baseline concentra-
tions. The concentrations of trace metals have been
assessed by geoaccumulation index (I
geo
) and enrich-
ment factor, showing only Co and Zn are moderately
polluted and slightly polluted, and Co, Cu, Zn and Pb
are enriched in topsoil. Other indices as modified
degree of contamination (mC
degree
) and pollution load
index (PLI), mC
degree
show moderate degree of
pollution and PLI shows unpolluted to moderate
pollution load. The ecological risk indices, e.g.,
ecological risk factor (E
r
) and potential ecological
risk index, show low ecological risk potential.
Keywords Trace metals Soil pollution Arctic 
Permafrost Ecological risk
Electronic supplementary material The online version of
this article (https://doi.org/10.1007/s10653-020-00587-2) con-
tains supplementary material, which is available to authorized
users.
X. Ji (&)E. Abakumov V. Tomashunas 
V. Polyakov S. Kouzov
Department of Applied Ecology, Saint Petersburg State
University, 16-Line, 29, Vasilyevskiy Island,
Saint Petersburg, Russian Federation 199178
e-mail: jixiaowen4321@qq.com
V. Polyakov
Arctic and Antarctic Research Institute, Saint Petersburg,
Russian Federation 199397
V. Polyakov
Department of Soil Science and Agrochemistry, Saint
Petersburg State Agrarian University,
Pushkin, Saint Petersburg, Russian Federation 19660
123
Environ Geochem Health
https://doi.org/10.1007/s10653-020-00587-2(0123456789().,-volV)(0123456789().,-volV)

Introduction
Climate warming in the Arctic has led the deeper
thawing of permafrost and downward movement of
the seasonally thawed active layer into preceding
frozen materials (Smith et al. 2010; Romanovsky et al.
2010; Christiansen et al. 2010; Barker et al. 2014).
This response of the active layer to increasing air
temperature can affect the geochemical composition
in the permafrost-affected soils through transport of
trace metals, enhancing migration of organic carbon as
well as biochemical processes (Muskett and Roma-
novsky 2011; Pokrovsky et al. 2011; Perreault and
Shur 2016; Patton et al. 2019). The thawing snow
water can form subsurface flow in active layer of
permafrost. As a result, the active layer will expand
downward so that more labile mineral phase can be
exposed to weathering process, which may bring the
geochemical substances as trace metals in soil pore-
water to upper layers during congeliturbation
processes.
Trace metals consist in bedrock and in soils in
nature, which are usually bound to oxides, carbonates,
sulfides and silicates (Antcibor et al. 2014). Per-
mafrost-affected soils in Arctic and subarctic store a
great amount of organic matter (Tarnocai et al. 2009),
which could form organo-mineral associations bound
with majority of trace metals (Dube et al. 2001; Hofle
et al. 2013). The climate warming may intensify the
degradation of large carbon reservoir including bound
trace metals in the permafrost-affected soils (Antcibor
et al. 2014). Additionally, the mobility of trace metals
is constrained due to their forms bound or adsorbed to
other soil substances (Dube et al. 2001). Different
trace metals have varied migration and leaching
processes (Niskavaara et al. 1997). For instance, the
leaching of Mn is enhanced in acid soil while
weakened in neutral and alkaline soils (Moskov-
chenko et al. 2017). Cu in deeper soil horizon is mostly
transported with organic substances (Niskavaara et al.
1997). However, trace metals are also an important
group of anthropogenic pollutants which belong to
long-range pollutants. They can transport from emis-
sion sources in middle latitude to the remote regions
such as Arctic and Antarctic through atmospheric
circulation patterns (Rahn and Lowenthal 1984;
Ryaboshapko et al. 1998). Although the Arctic
environment is used to consider as marginally influ-
enced by anthropogenic activities, local pollution and
long-range transport both contribute to the pollution
sources. For example, Norilsk mining industry of
nickel in the Kola Peninsula is a significant contributor
of trace metals in the Arctic, which can across several
hundred kilometers (Boyd et al. 2009; Jaffe et al. 1995;
Niskavaara et al. 1997; Zhulidov et al. 2011).
Until now, the contamination in terrestrial tundra of
the Russian Arctic ecosystem has been little studied
out of Russia with some reports only focusing on air
and soil samples from industrial emission in Kola
Peninsula (Rovinsky et al. 1995; Ryaboshapko et al.
1998; Pacyna et al. 2010; Pacyna 1995; Kashulina
2017; Lukina and Nikonov 2001; Kozlov and Barcan
2000). For example, the noticeable air pollution
adjacent to the metallurgical industries in the Kola
Peninsula called ‘‘industrial deserts’’ was reported
(Jaffe et al. 1995; Reimann et al. 1997,1999). The
exploitation of some natural resources such as oil, gas
and coal has been developed for sustaining other
industrialization in the former Soviet Union (Pryde
1991). In these soils directly influenced by atmo-
spheric deposition from emission points, Kashulina
(2017) observed extremely high concentrations of Ni
(9000 mg kg
-1
) and Cu (6000 mg kg
-1
) in the upper
organic soil horizons and also found that the degree of
technogenic degradation in soils resulted in the loss of
great amount of organic matters. The environmental
pollution in central part of the Kola Peninsula has also
been found to be associated with landscape deterio-
rations. However, most heavy industries in the Russian
Arctic have not been reported and are sporadically
distributed. The influence of these hazardous sub-
stances from atmospheric transport on soils without
direct pollution sources is little studied. Abakumov
et al. (2017) firstly reported the accumulation of trace
metals in the upper layer of Stagnic Cryosols in Belyi
Island of the Russian Arctic. Moskovchenko et al.
(2017) further found the elevated concentrations of Cd
and Pb with influence of silt, clay and organic matter
removal. From the aspect of soil vertical profile,
leaching and migration of trace metals were found in
permafrost-affected soils, especially in gleyed hori-
zon, in Lena Delta River.
The objectives of this study are to investigate (1)
accumulation of nine trace metals (Pb, Cu, Ni, Co, Zn,
Fe, Mn, Cd and Hg) in different soil types with
considered factors of cryoturbation and organic mat-
ters; (2) to investigate accumulation of trace metals in
each genetic horizons; (3) to establish the geochemical
123
Environ Geochem Health

background values for this region; (4) to measure level
of metals’ pollution and pollution load in topsoil; and
(5) to evaluate level of potential ecological risk in
topsoil of the Russian Arctic.
Materials and methods
Study area
Soil investigations were carried out from the North
European sector of Russia to East Siberia (Fig. 1).
Two islands, Vaygach and Kolguyev, are in North
European sector of Russia. Six islands, Bennett,
Gerkules, Bolshoy Begichev, Wrangel, Bolshoy
Lyakhovsky and Chelyuskin, are located in the Kara
Sea, the Laptev Sea and the East Siberian Sea.
Haranasale Cape and Valraray Cape were situated in
West and East Siberia, respectively. All investigated
areas belong to Arctic climatic zone.
Vaygach Island (69.968358 N, 59.703154 E) is
lying between the Novaya Zemlya Archipelago (split
by Kara Strait) and Russian mainland (split by the
Yugorsky Strait) and its area is approximately
3.4 km
2
. There are many rivers, thermokarst lakes
and swaps on this island. This island is surrounded by
Kara Sea (east) and Pechora Sea (west). This island is
only 157 m higher than the sea level; the geological
structure remains the same. Soils mainly contain
calcareous cleaving stones (carbonate rock). Soils
overmoisturize on the one hand and the lack of oxygen
on the other. Kolguyev Island (69.087703 N,
49.193683 E) is a circular-shaped island located in
southeastern Barents Sea to northeast of the Kanin
Peninsula. The area is about 3497 km
2
. Wetlands
consisting of morainic hills and bogs are vast on this
island. The island mostly contains limestone, and its
elevation above the sea is geologically recent. Raised
beaches are frequent. The rocks are heavily scored by
ice, but this was probably marine ice, not that of
glaciers. Bennett Island (76.685860 N, 148.956612 E)
is in the north of the East Siberian Sea, and it is the
largest island of the De Long group (150 km
2
). It
consists of Early Paleozoic, late Cretaceous, Pliocene
and Quaternary sedimentary and igneous rocks.
Gerkules Island (75.41297 N, 88.21157 E) is one of
few scattered islands in Mona Islands which are
situated in the Kara Sea and north of Taymyr
Peninsula. Bolshoy Begichev Island (74.317829 N,
112.543314 E) is lying within the Khatanga Gulf,
Laptev Sea, and its size is 1764 km
2
. Wrangel Island
(71.251415 N, 179.675427 E) is located between the
Chukchi Sea and East Siberian Sea. This island is
about 7600 km
2
in area and 125 km wide. It consists
of a central belt of low-relief mountains with the
highest elevation of 1096 m above the sea level
Fig. 1 Location of investigated sites in the Russian Arctic
123
Environ Geochem Health

(Kosko et al. 1993). The short summer is cool with
July temperatures average 2.4–3.6 °C on the south
coast, while the north coast is 1 °C on average (Quinn
and Woodward 2015). The island has a southern
coastal plain (15 km wide; a 40 km wide east–west
trending central belt of low-relief mountains, with the
highest elevations) and a roughly 25 km wide northern
coastal plain. Its soils contain metamorphosed, folded
and faulted volcanic and sedimentary rocks ranging in
age from upper Precambrian to lower Mesozoic.
Bolshoy Lyakhovsky Island (73.579338 N,
141.881764 E) is located between the Laptev Sea
and the East Siberian Sea and is the biggest of
Lyakhovsky Islands. It has 4600 km
2
in area and a
highest altitude of 270 m. It contains highly folded and
faulted Precambrian metamorphic rocks and tur-
bidites; Mesozoic turbidites and igneous rocks; and
Cenozoic sediments. Chelyuskin Island
(71.760199 N, 149.968611 E) is located at the mouth
of Taymyr Gulf, Kara Sea.
Haranasale Cape (71.25281 N, 73.30310 E) is at
northwest direction of Gydan Peninsula which is
primarily flat terraces and more than 60% of surface
underlain by permafrost (Ejarque and Abakumov
2016). Soils developed on Pleistocene sands, which
are underlain by late Quaternary alluvial sediments
and marine clays and are influenced by cryoturbation
(Walker et al. 2009). The average air temperature per
year is about -10 °C, with the lowest temperature of
-25 °C in January and the highest temperature of
8°C in August. The mean air temperature remains
above 0 °C for about 70 days per year. On average, the
annual precipitation in this area is about 325 mm yr
-1
and the annual evaporation ranges from 50 to 100 mm
(Buchkina et al. 1998). The Haranasale has a vast
diversity of soil types because of seasonally cryo-
pedogenetic processes influencing texture and struc-
ture of bedrocks, parent materials and thaw depths.
Valraray Cape (69.01480 N, 169.18026 E) is located
at northeast direction of Chukchi Peninsula which is
the farthest east peninsula of Asia. In Valraray Cape,
the relief was formed with variable lowlands and hills
with about 100 m elevation differences. Valraray
Cape is influenced by mining activities (tin, lead, zinc,
gold and coal) in hinterland of Chukchi Peninsula. The
closest mining factory (Kupol Gold mine) is about
240 km away from the sampling site. The detailed
information about soil characteristics and landscapes
is shown in Table S1, Supplementary material.
Soil sampling
Soil sampling was conducted during June–July of
2016 ‘‘Russian High Latitude’’ expedition by taking
vessel ‘‘Mikhail Somov.’’ In total, 25 soil profiles (93
samples from each genetic horizon) were dug till the
frozen ground. Soil types were classified according to
the World Reference Base for Soil Resources (WRB
2015) and were also classified to suborders of Gelisols
(Histels, Orthels and Turbels) for distinguishing effect
of cryoturbation in soils with permafrost according to
USA Soil Classification (Krasilnikov et al. 2009). The
procedures of soil sampling and site restoration were
all followed by sampling protocols for permafrost-
affected soils (Ping et al. 2013). The area of soil pits
was about 35 935 cm and depth ranged between 15
and 83 cm based on position of the frozen ground.
Each soil horizon was sampled by a small plastic
shovel after removal of litter. The samples of about
500 g were stored in polyethylene bags. Soil samples
of Haranasale and Valraray cape were taken from
coastal marshes. Most soil samples from the islands
were taken from interfluve (interstream) areas, flood
plains, vegetated tundra and bare soil. No samples
were taken from direct anthropogenic pollution
sources.
Chemical analysis
Pretreatment of soil samples
In the laboratory, plant roots and coarse rocks were
removed from the soils. The soil samples were air-
dried at room temperature for about 1 month until the
weight remained stable more than 1 week. Then,
samples were homogenized by grinding using a roller
mill and then through a 2 mm sieve. Following the soil
mixed with water at a concentration of 2 mg/mL in
PCR tubes, then dropped on glass slide and placed into
desiccator for maintaining zero moisture status.
According to thin-film method (Ba
¨chmann et al.
1992), samples were done thin in order to substantially
eliminate the absorption effects during energy-disper-
sive X-ray analysis (EDX) instrumentation and drop
casts of the solution were smoothened to obtain
appropriate calibration results. Gold plating of sam-
ples was conducted to eliminate any deflection caused
by conductive effect. Triplicates from each soil
sample were analyzed separately.
123
Environ Geochem Health

Elemental analysis through EDX
The content of lead (Pb), cadmium (Cd), copper (Cu),
nickel (Ni), cobalt (Co), zinc (Zn), iron (Fe), man-
ganese (Mn) and mercury (Hg) was analyzed by
scanning electron microscopy with energy-dispersive
X-ray spectroscopy (SEM–EDX) (JSM-6390LA, EX-
2300, JEOL, Tokyo, Japan) at soil laboratory of Dept.
Applied Ecology, Saint Petersburg State University.
SEM–EDX can measure both quantitative and qual-
itative elemental compositions by the measurement of
characteristic radiations (Ba
¨chmann et al. 1992).
SEM–EDX has the advantage of allowing to analyze
solid materials without the digestion process which is
necessary for atomic absorption spectroscopy (AAS)
or inductively coupled plasma mass spectrometry
(ICP-MS) analysis (Sitko et al. 2004). The character-
istic radiation emission of each element is entirely not
depended on any kind of chemical bonds for accu-
rately measuring elements existing in the present
samples (Ba
¨chmann et al. 1992). The characteristic
peak obtained for every element in EDX from a
sample in atomic level precision which enables to
directly convert to the chemical concentration (at level
of mg kg
-1
). Quantification of data set was performed
by the comparison between magnitudes of x-ray
counts and those of pure element standards, and
appropriately correcting for atomic number (Z),
absorption (A) and fluorescence (F), collectively
referred to ZAF correction (Wedepohl 1967; Ives
et al. 1997). Soil certified reference material (RTC,
SQC-00, Sigma-Aldrich, St. Louis, US) was further
used to check the accuracy of the method. Metal
recoveries were ranged between 87% and 99% (Cd
and Pb, respectively). Final data are taken from the
average concentrations of triplicates. In this case, 279
samples were measured.
Soil basic parameters
The pH of soil samples was potentiometrically deter-
mined with a pH meter (Ekotest-2000, Moscow,
Russia) in H
2
O extract with ratio of a 1:2.5 and 1:25
for soils with low and high organic carbon content,
respectively (DINISO10390 2005). Soil particle size
distribution was measured with a ASTM 152H soil
hydrometer (Preiser Scientific, St. Albans, WV, USA)
using method recommended by World Reference Base
for Soil Resources (WRB 2015). The homogenized
and air-dried soils were analyzed for organic carbon
(OC) and nitrogen (N) content using C/N element
analyzer (multi-N/C
Ò
2100 TOC, Analytik-jena,
Germany).
Geochemical baseline concentrations (GBCs)
and quantification of soil contamination through
indices
GBCs of trace metals in the Arctic soils
GBCs values of trace metals are crucial to evaluate the
pollution status of soils. The world average shale’s
content as background values (Rubio et al. 2000) and
average content of earth crust (Loska et al. 2004) are
currently widely used as GBC value of trace metals.
However, regional background values, especially for
permafrost-affected soils, vary with time, temperature
and landscapes. Some metals, such as Mn and Fe in
this study, vary in different soil types (Jiang et al.
1996). A normalization method depended on the
equation of a linear regression derived from the
correlation of the metal in the equation and a
conservation reference element (Tian et al. 2017). In
this study, lithium (Li) was selected as normalizing
element due to the best proxy for fine particles of soils
and sediments (Aloupi and Angelidis 2001). The
equation of geochemical baseline can be defined as:
Cm¼aCNþbð1Þ
where Cmis the concentration of metals (mg kg
-1
),
CNis the concentration of Li (mg kg
-1
) and a/b
represents the regression constant. In the x–yscatter-
plot depicted by Eq. (1), concentrations ranged in 95%
confidence interval are regarded as natural sources
(Fig. 2). Concentrations ranged out of 95% confidence
interval are regarded as having arisen from anthro-
pogenic sources. Then, concentrations derived from
the anthropogenic sources were removed so that a
linear regression was fitted by naturally sourced
concentrations with new constants as shown in
Eq. (2):
Bm¼c
€
CNþdð2Þ
where Bmand
€
CNare the GBC of metals (mg kg
-1
)in
the investigated area and normalized mean concen-
tration of Li, respectively, and cand dare the
regression constants.
123
Environ Geochem Health

123
Environ Geochem Health

Geoaccumulation index (I
geo
)
Geoaccumulation index (I
geo
) is used to understand the
present environmental status and pollution status of
trace metal regarding natural background values
(Mu
¨ller 1969) as shown below:
Igeo ¼log2
Cm
1:5Bm
 ð3Þ
The constant 1.5 is used to minimize possible
variations due to lithogenic variations (Taylor and
McLennan 1995). I
geo
was classified into seven grades
(Table S2). When I
geo
values were above 5, the
concentrations of metals may be one to two orders of
magnitudes higher than the regional GBCs (Bhuiyan
et al. 2010).
Enrichment factor (EF)
Enrichment factor (EF) was preferably used to eval-
uate metals’ contaminants than I
geo
by other authors
(Covelli and Fontolan 1997). This method normalizes
the metal concentrations compared to the normalizer
(reference) metals (Ravichandran et al. 1995), which
is commonly used to evaluate trace metals in topsoil
influenced anthropogenic pollution (Almasoud et al.
2015). Two rationing stages are used to calculate
normalized EF values that the ratio of metal and
reference metal concentration in each sample is
divided by the same ratio of the reference values
(Rubio et al. 2000) as shown below:
EF ¼
Cm
=
CLi

sample
Cm
=
CLi

reference
ð4Þ
where CLi is the concentration of Li (reference metal).
In this study, Li used as normalizer and reference
values are as same as that of I
geo
. When EF \2itis
considered to be crustal origin and EF [2 suggests
more possibility of anthropogenic sources (Liaghati
et al. 2004). If EF [10, then it will be considered as
purely anthropogenic sources (Bhuiyan et al. 2010).
Multi-order classification has been suggested by some
authors (Zinkute et al. 2017; Likuku et al. 2013)as
shown: EF \2 indicates the deficiency to minimal
enrichment; 2–5 is moderate enrichment; 5–20 is
strong enrichment; 20–40 is high–very high enrich-
ment; [40 is very strong enrichment.
Contamination factor (CF)
CF represents the pollution level directly assessed by
individual metal in the soil of each site (Mu
¨ller 1969).
It is the ratio of measured concentration of individual
metal in soil sample to GBCs value shown in the
equation:
CFi¼Cm
Bm
ð5Þ
The contamination level assessed by CF values can
be divided into four levels recommended by Likuku
et al. (2013) and Mu
¨ller (1969) as shown in Table S2.
Degree of contamination (C
degree
)
C
deg
is an aggregative indicator of soil contamination
combining CF values of all metals in each sampling
site (Hakanson 1980). This index has been used in
evaluation of contamination level in Mediterranean
sediments (Ahdy and Khaled 2009) and farming soil
(Loska et al. 2004). C
deg
assessed the degree of multi-
metal contamination with the numeric sum of each CF
values of metals expressed in the following equation:
Cdegree ¼X
n¼9
i¼1
CFið6Þ
where nis the number of selected metals in this study
and total CF values of 9 metals to calculate C
degree
in
each site of 25 sampling sites. The contamination
category was classified based on the number of metals.
Categorization is shown in Table S2.
Modified degree of contamination (mC
degree
)
A generalized method to assess the average contam-
ination degree used mC
degree
divided by the number of
metals (Machender et al. 2011; Abrahim 2019)as
followings:
mCdegree ¼Cdedree
nð7Þ
bFig. 2 Geochemical baseline concentration of Pb, Cd, Cu, Ni,
Co, Zn, Fe, Mn and Hg in soils of the Russian Arctic
123
Environ Geochem Health

The categorization of mCdegree is shown in
Table S2.
Pollution load index (PLI)
PLI was firstly proposed by Tomlinson et al. (1980)to
evaluate extent of contamination load by all measure
metals for different sites. First, Tomlinson’s approach
was successfully used in assessing the costal contam-
ination and was proved that it is a useful index to
develop a plan of action (Angulo 1996). Then, PLI has
been widely used to evaluate soil pollution with multi-
metals (Adama et al. 2016; Akoto et al. 2016;
Katsoyiannis et al. 2011; Tian et al. 2017). The
calculations are according to the following equations:
PLI ¼ðCF1CF2CFnÞ1=nð8Þ
Pollution level of PLI values is shown in Table S2.
Ecological risk factor (E
r
)
E
r
is used for evaluation of ecological risk for
individual pollutant based on toxicity response of its
own (Hakanson 1980), which can be calculated as:
ERi¼TrCFið9Þ
where Trrefers to toxicity response of each metal and
CF is the contamination factor of corresponding metal.
Nine selected metals (Pb, Cd, Cu, Ni, Co, Zn, Fe, Mn
and Hg) in this study are assigned toxic response as
follows: Zn = Fe = Mn = 1; Pb = Cu = Ni = Co =
5; Cd = 30; Hg = 80; taken from Hakanson (1980),
Xu et al. (2008) and Saleh et al. (2018). Mugos
ˇa et al.
(2016) assessed the ecological risk of polluted coastal
environment by using same responses. The catego-
rization of E
r
is given in Table S2.
Potential ecological risk index (RI)
RI is improved approach of E
r
, which can evaluate
magnitude of environment sensitivity led by toxic
metals in soil environment (Sahoo et al. 2016). RI can
be calculated as the sum of E
r
for selected metals in
soils as following Hakanson (1980) as:
RI ¼X
n¼9
i¼1
ERið10Þ
RI is calculated for nine (n) metals. The categorization
is given in Table S2.
Statistical analysis
Kolmogorov–Smirnov test, Pearson’s correlation
coefficient (p\0.05) and principal component anal-
ysis (PCA) were used for statistical association
between metal concentrations and soil physiochemical
properties by using SPSS Statistics 21 software (IBM,
Armonk, NY, USA).
Results and discussion
Soils physicochemical characterization
The results of soil characterization from different soil
groups and horizons are depicted in Table S3. Mostly
soil profiles collected from the studying sites showed
increasing pH values from topsoil (O and A) to B
horizon and decreasing pH values in the bottom C
horizon with a pH range between 4.07 and 8.87, and
generally organic layer showed a lower pH. Most soil
samples were fine-texted, with a high content of silt,
which is due to the sampling sites were lower lands
where relatively heavier textural accumulated in the
depression. A relatively higher carbon content was
found in Histosols or A/Bhorizon accumulated with
organic matters. The general organic matter contents
were high in these sites due to the low temperature,
poorly drained conditions and low mineralization,
which cause a surface layer of mold precipitated by
humic substances. The average C/N ratios (10.32) are
comparable to those in West and North Siberian soils
(Ji et al. 2019a,b; Antcibor et al. 2014; Moskovchenko
et al. 2017).
Absolute metal concentration
The maximum and minimum values of Pb, Cu, Ni, Co,
Zn, Fe, Mn and Hg varied by one to two orders of
magnitude. Generally, except major elements such as
Fe and Mn, the average concentrations showed
Zn [Ni [Hg [Co [Cu [Pb [Cd (Table 1).
The lowest concentrations were found for Hg
(0.005–0.114 mg kg
-1
) in all sites. The maximum
concentration of Pb was found on Island Wrangel,
123
Environ Geochem Health

Table 1 Concentration (numerator/denominator, mg kg
-1
), geoaccumulation (I
geo
) and enrichment factors (EF) of each metal in the Russian Arctic topsoil
Sample Pb Cd Cu Ni Co
Conc. I
geo
EF Conc. I
geo
EF Conc. I
geo
EF Conc. I
geo
EF Conc. I
geo
EF
H1 3.91–7.61
5.30
-0.5 1.06 0.065–0.093
0.08
-3.0 0.19 10.45–20.20
11.95
0.5 2.19 28.71–34.94
20.66
-0.1 1.44 35.77–48.82
41.82
2.2 6.89
VC1 5.04–7.48
6.05
-0.3 1.21 0.03–0.78
0.78
0.1 1.57 4.90–8.52
8.40
0.2 1.76 15.24–19.62
19.51
-0.6 0.98 40.00–61.10
40.56
2.1 6.59
B1 12.12–12.40
6.58
0.7 2.48 0.25–0.30
0.27
-1.6 0.51 10.15–10.42
10.29
0.5 2.13 32.22–43.54
37.88
0.5 2.18 42.49–54.39
48.44
2.6 8.96
B2 15.42–16.99
16.23
1.2 3.4 0.17–0.19
0.186
-2.1 0.36 8.50–9.64
9.07
0.4 2.02 3.92–7.14
5.53
-0.1 1.45 25.77–75.27
50.52
1.5 4.24
V1 0.97–8.03
1.34
-2.9 0.19 0.02–0.19
0.05
-3.7 0.11 0.79–1.54
1.17
-3.2 0.17 29.97–44.62
39.73
-2.9 0.20 12.23–13.33
12.78
0.5 2.2
V2 6.56–8.58
7.63
0.2 1.72 0.17–0.79
0.20
0.1 1.58 12.55–15.39
13.93
0.8 2.63 2.35–6.6
3.12
0.0 1.5 22.60–36.30
29.98
1.9 5.78
G1 3.46–8.23
5.35
-0.6 1.00 0.15–0.19
0.15
-2.3 0.31 5.11–11.43
7.99
-0.5 1.07 36.46–39.99
39.75
-3.3 0.15 70.42–85.97
80.02
3.1 13.18
BB1 10.49–11.47
10.98
0.6 2.29 0.09–0.13
0.11
-2.5 0.27 19.20–17.20
18.20
1.9 5.7 68.07–73.76
70.91
0.5 2.08 23.41–29.97
26.69
1.7 4.94
BB2 10.04–10.35
19.47
1.4 3.89 0.08–0.13
0.08
-2.5 0.27 20.33–21.24
21.16
1.5 4.26 54.11–58.42
55.02
0.4 1.99 69.32–89.45
80.09
2.6 9.35
W1 15.60–24.19
19.89
1.8 5.12 0.08–0.095
0.09
-3.0 0.19 36.80–39.26
38.03
2.5 8.23 68.07–73.76
70.91
1.3 3.69 28.69–56.78
33.13
2.0 5.82
W2 17.61–18.47
19.07
1.3 3.82 0.03–0.04
0.07
-3.3 0.15 28.15–31.7
29.82
2.1 6.25 54.11–58.42
55.02
1.0 2.92 24.542–38.7
28.02
2.1 6.38
W3 18.30–21.07
19.92
1.5 4.22 0.04–0.06
0.05
-3.5 0.14 26.58–29.37
28.28
2.0 5.93 54.89–61.08
60.37
0.9 2.74 24.93–30.25
26.79
1.7 4.98
W4 10.30–16.47
11.55
0.5 2.06 0.02–0.10
0.03
-4.9 0.05 26.01–38.44
28.79
1.9 5.45 28.08–38.04
29.32
-0.1 1.4 13.33–37.04
33.68
1.9 5.55
W5 14.41–16.28
16.28
1.1 3.26 0.03–0.75
0.38
0.0 1.51 24.69–42.81
34.06
2.6 8.98 52.19–116.25
72.73
2.0 5.81 31.21–220.23
49.99
4.6 36.27
W6 38.68 2.4 7.74 0.39 -0.9 0.78 8.84 0.3 1.85 7.45 -2.0 0.37 40.007 2.1 6.59
W7 22.67 1.6 4.53 0.31 -1.3 0.62 28.55 2.0 5.99 11.51 -1.4 0.58 40.396 2.1 6.65
123
Environ Geochem Health

Table 1 continued
Sample Pb Cd Cu Ni Co
Conc. I
geo
EF Conc. I
geo
EF Conc. I
geo
EF Conc. I
geo
EF Conc. I
geo
EF
BL1 10.50–11.92
10.97
0.7 2.39 0.06–0.09
0.07
-3.5 0.14 10.17–11.545
10.43
0.5 2.13 28.67–29.77
29.24
-0.1 1.43 22.42–28.8
26.69
1.6 4.39
BL2 8.91–9.42
8.91
-0.1 1.37 0.06–0.21
0.18
-1.8 0.44 10.12–13.83
10.40
1.0 2.9 21.79–25.06
23.30
-0.5 1.09 30.03–53.22
31.24
2.5 8.77
BL3 9.51–11.04
10.82
0.3 1.9 0.057–0.06
0.06
-3.5 0.14 12.22–14.83
13.34
0.8 2.56 25.09–26.01
25.61
-0.3 1.25 20.61–26.97
25.56
1.6 4.44
BL4 3.48–16.47
16.29
-1.1 0.7 0.202–0.30
0.21
-1.3 0.61 11.21–14.21
13.35
1.0 2.98 9.63–22.45
22.33
-0.4 1.12 29.00–40.24
38.83
2.1 6.59
C1 8.14–9.19
6.98
0.3 1.84 0.053–0.06
0.05
-3.5 0.14 13.73–27.54
17.50
1.9 5.77 50.90–70.92
51.21
1.2 3.55 40.90–52.27
49.38
2.4 8.13
C2 11.37 0.6 2.27 0.14 -2.4 0.28 33.57 2.2 7.04 35.79 0.3 1.79 40.02 2.1 6.59
C3 3.76–7.04
5.97
-0.3 1.19 0.10–0.14
0.11
-2.7 0.23 26.55–33.20
27.79
1.9 5.57 114.05–157.57
155.86
2.4 7.79 44.47–61.08
59.65
2.3 7.32
K1 4.29–7.34
6.53
0.0 1.47 0.15–0.36
0.20
-1.0 0.73 13.78–16.68
15.16
0.9 2.89 17.26–30.57
19.13
-0.8 0.86 27.90–40.78
40.26
2.1 6.59
K2 3.03–11.27
4.84
0.6 2.25 0.02–0.29
0.02
-1.3 0.59 5.68–14.01
9.88
1.0 2.94 16.35–31.08
25.33
0.1 1.55 18.68–34.16
19.93
1.9 5.63
Max 38.68 2.4 7.74 0.79 0.1 1.58 42.82 2.6 8.98 157.57 2.4 7.79 220.24 4.6 36.27
Min 0.97 -2.9 0.19 0.02 -4.9 0.05 0.79 -3.2 0.17 3.00 -3.3 0.15 12.33 0.5 2.2
Mean/median 11.52 0.44 2.25 0.14 -2.2 0.28 18.39 1.06 2.94 40.48 -0.08 1.45 39.03 2.13 6.59
Zn Mn Hg Fe
Conc. I
geo
EF Conc. I
geo
EF Conc.
a
I
geo
EF Conc.
a
I
geo
EF
H1 46.80–51.03
50.15
1.2 3.46 432–984
713
-1.7 0.48 8.6–103
37.47
-1.7 0.46 12,614–29,903
23,361
-0.6 0.97
VC1 67.84–86.75
36.86
2.8 10.62 454–768
642
-1.8 0.43 9.9–58.5
24.34
-2.3 0.30 12,025–18,753
22,689
-0.7 0.94
B1 93.76–135.81
114.79
2.7 10.05 978–987
983
–1.2 0.66 16–26
21
-2.5 0.26 11,930–14,875
13,403
-1.4 0.56
B2 73.77–79.96
76.87
2.0 5.91 816–967
891
-1.3 0.59 64–90
77
-0.6 0.96 23,947–34,335
29,141
-0.3 1.21
123
Environ Geochem Health

Table 1 continued
Zn Mn Hg Fe
Conc. I
geo
EF Conc. I
geo
EF Conc.
a
I
geo
EF Conc.
a
I
geo
EF
V1 6.08–50.65
28.36
-1.7 0.45 421–672
587
-1.9 0.39 5.9–52
24.6
-2.3 0.3 12,445–33,150
12,445
-1.5 0.52
V2 49.89–75.00
57.05
1.9 5.55 683–922
820
-1.5 0.55 16–114
44
-1.4 0.55 8326–34,029
24,224
-0.6 1.00
G1 8.80–10.40
9.60
-1.2 0.65 455–887
705
-1.7 0.47 52–114
76
-0.7 0.95 19,475–33,593
24,398
-0.6 1.01
BB1 64.36–69.78
67.07
1.8 5.16 459–668
563
-2.0 0.38 10–70
40
-1.6 0.50 13,401–32.876
23,138
-0.6 0.96
BB2 67.83–95.24
70.51
2.2 7.04 537–860
670
-1.7 0.45 9–50
19
-2.7 0.24 9631–30,209
18,532
-1.0 0.77
W1 95.71–119.53
107.62
2.6 8.84 631–931
781
-1.5 0.52 15–20
17
-2.8 0.21 11,907–15,482
13,694
-1.4 0.57
W2 89.41–95.95
90.77
2.2 7.1 672–814
730
-1.6 0.49 5–35
12
-3.3 0.15 6315–29,036
18,392
-1.0 0.76
W3 83.99–95.65
91.62
2.1 6.21 653–897
782
-1.5 0.52 13–49
30
-2.0 0.38 13,105–34,203
23,459
-0.6 0.97
W4 82.25–85.96
83.96
2.0 6.08 536–967
688
-1.7 0.46 13–47
32
-1.9 0.40 18,102–20,169
19,461
-0.9 0.81
W5 86.27–123.51
109.54
2.4 8.1 426–872
604
-1.9 0.4 11–34
26
-2.2 0.33 9246–34,574
20,952
-0.8 0.87
W6 57.25 1.5 4.23 906 -1.3 0.6 33.97 -1.8 0.43 12,505 -1.5 0.52
W7 85.02 2.1 6.29 516 -2.1 0.34 21.88 -2.5 0.26 23,622 -0.6 0.98
BL1 50.76–59.68
56.67
1.5 4.19 742–947
740
-1.6 0.49 14–40
28
-2.1 0.35 8143–32,083
22,243
-0.7 0.92
BL2 57.34–60.48
58.00
1.5 4.29 588–723
641
-1.8 0.43 10–40
24
-2.3 0.30 10,620–33,515
20,478
-0.8 0.85
BL3 54.46–56.84
56.55
1.5 4.18 488–981
635
-1.8 0.42 14–41
29
-2.0 0.36 6901–32,866
17,962
-1.0 0.74
BL4 87.19–139.77
105.24
2.8 10.34 497–848
618
-1.9 0.41 17–47
30
-2.0 0.38 17,454–30,142
24,394
-0.6 1.01
C1 75.00–82.29
76.49
2.0 6.09 403–970
776
-1.5 0.52 15–44
28
-2.1 0.35 13,461–19,486
18,335
-1.0 0.76
123
Environ Geochem Health

while relatively high Cd values were found in Valraray
Cape and high variability of Cd concentrations was
found in Island Wrangel and Island Bennet. The
highest median concentration of Cu and Ni was found
in Island Chelyuskin and Island Wrangel. The content
of Co did not vary strongly with highest values in
Island Wrangel. The highest variability of concentra-
tions was found for Zn from 6.08 to 143.56 mg kg
-1
.
Only Island Kolguyev and Island Gerkules showed a
relatively low content of Zn. Island Wrangel was used
as Soviet military outpost, which might be the reason
for the highest concentration of Cd, Cu, Ni, Co. The
differences of metal distribution in different sites may
be both due to different geographical environments
and different anthropogenic sources.
The range of Fe and Mn concentrations was within
previous ranges from Siberian Arctic (Antcibor et al.
2014;Jietal.2019a,b). The Cd concentrations in the
soils studied here were similar to those reported in
coastal lowland of Spitsbergen, Svalbard (Marques
et al. 2017; Halbach et al. 2017). However, the
concentration of Cd in this study was observed to be
much higher in organic layer (O/histic horizon) in soil
profiles. This is in line with the observation in Belyi
Island (Moskovchenko et al. 2017) and in the Lena
Delta (Antcibor et al. 2014). The highest concentration
of Cd (0.79 mg kg
-1
) was similar to mean content of
Cd in Histosols (0.78 mg kg
-1
) on the world scale
(Meharg 2011). The average Pb concentrations
(16.3 mg kg
-1
) in surficial layers (above C horizon)
were slightly higher than those found in Lena Delta
(Antcibor et al. 2014) and Yamal-Gydan Peninsula (Ji
et al. 2019a,b;Moskovchenko2013) and were one
order of magnitude higher than those reported in Belyi
Island (\0.5 mg kg
-1
) (Moskovchenko et al. 2017)
and were comparable to the average of Arctic circle
(AMAP 2005). Comparison with Pb levels of surface
soils (0–5 cm) influenced by anthropogenic activities in
Svalbard (mean = 16.3 mg kg
-1
)(Marquesetal.
2017), average Pb concentrations in our study were
still obviously lower. Pb in natural soils is inherited
from parent materials. However, widespread Pb anthro-
pogenic emission causes the majority of soils enriched
this metal, especially in the top horizon (Meharg 2011).
Interestingly, only soil profile BB2 enriched apparently
higher Pb concentrations in O horizon with declining
trend from A to C horizon. In Island Wrangel, Pb
concentrations in all sites were all higher than other
investigated areas, and there was no vertically
Table 1 continued
Zn Mn Hg Fe
Conc. I
geo
EF Conc. I
geo
EF Conc.
a
I
geo
EF Conc.
a
I
geo
EF
C2 46.5 1.2 3.44 640 -1.8 0.43 10.48 -3.5 0.13 19,713 -0.9 0.82
C3 65.42–93.96
65.80
1.7 4.84 557–750
633
-1.8 0.42 17–25
21
-2.5 0.26 15,442–30,809
23,206
-0.6 0.96
K1 17.95–33.31
27.47
0.4 2.03 424–963
783
-1.5 0.52 16–36
31
-2.0 0.39 8031–27,621
18,506
-1.0 0.77
K2 22.29–49.18
33.96
1.3 3.64 474–807
658
-1.8 0.44 14–36
25
-2.3 0.31 7388–34,004
19,771
-0.9 0.82
Max 143.56 2.8 10.62 346,574.24 -1.2 0.66 0.11 -0.6 2.4 29,141.000 -0.3 1.21
Min 6.08 -1.7 0.45 6315.28 -2.1 0.34 0.005 -3.5 -2.9 12,445.000 -1.5 0.52
Mean/Median 67.86 1.62 5.55 12,032.66 -1.6 0.46 0.03 -2.1 0.35 20,320.960 -0.9 0.84
The underlined number is average value
a
The values 910
-3
123
Environ Geochem Health

decreasing trend of Pb in all soil profile. However, we
cannot determine this level belongs to geology back-
ground due to uncertainty of the current active layer
soils age and fallout of aerosol particles from the
atmosphere. The average Zn, Ni, Co and Cu concen-
trations were generally higher than other natural Arctic
soils (Ji et al. 2019a;Antciboretal.2014;Moskov-
chenko et al. 2017). Hg concentrations on average
(0.035 mg kg
-1
) were considered to be higher due to
when it compares with the highest concentration of
worldwide basis that Histosols in Canada
(0.04 mg kg
-1
)(Franketal.1976) and paddy soils in
Japan (0.035 mg kg
-1
) (Kitagishi and Yamane 1981)
and Vietnam (300 mg kg
-1
)(Survey1970). Although
accumulation of Hg in soils is controlled by precipita-
tion and organic complex formation, we cannot exclude
the contribution of Hg from anthropogenic sources.
Metal distribution within different soil groups
In order to distinguish the different distribution of
trace metals in soil with/without cryoturbation and
decomposed organic material, the great soil groups for
each determined soil suborders were based on US soil
classification (Krasilnikov et al. 2009). The soils were
divided into great groups of the Orthels suborder
(Aquorthels, Historthels, Haplorthels), Turbels subor-
der (Histoturbels, Aquiturbels, Psammoturbels) and
Histels suborder (Fibristels). Figure 3reveals log-
transformed distribution of nine trace metals for each
determined soil group. Fe and Mn have the largest
median values and Hg has the lowest median values in
all soil groups. The first combined group consisting of
Fibristels and Historthels (Fig. 3a) with the high
content of organic carbon (up to 44%) is observed
by higher median and lower range of Zn and Ni values,
followed by higher median and range of Co values.
The second combined group which contains all
varieties of Turbels with a lower content of organic
carbon (median = 1.2%) was observed by higher
median and lower range of Co and lower range and
median value of Pb (Fig. 3b). The third group
(Fig. 3c) represents the soil groups of Orthels with
median organic content of 6.1%, where Zn is observed
Fig. 3 Log-boxplot comparison of variation and gravimetric
concentrations (mg kg
-1
) of nine measured metals in soil
groups determined by three suborders of Gelisols (soil with
permafrost within 1 m from the soil surface), i.e., Orthels,
Histels and Turbels, according to USA Soil Classification. The
transformation of these soil groups to WRB classification is
listed as: Fibristels &Folic Histosols, Historthels &Histic
Cryosols, Psammoturbels &Turbic Cryosols (Arenic),
Aquiturbels &Turbic Cryosols (Reductaquic), Histo-
turbels &Histic Turbic Cryosols, Aquorthels &Cryosols
(Reductaquic), Haplorthels &Haplic Cryosols
123
Environ Geochem Health

to be higher median values while only Hg has the
largest range values which is the largest in all soil
groups. However, the distribution of Cd and Pb has the
largest range in organic soil, and the distribution of Ni
and Zn has the largest rang in soils with cryoturbation.
A significantly higher range of Fe values was charac-
terized in Orthels suborder.
Metal distribution in different genetic horizons
In the permafrost-affected soils, the soil pedon varied
greatly in the active layers due to both effects of the
change of thawing depth susceptible to the atmo-
spheric temperature changing and cryoturbation pro-
cess during winter. Additionally, parent materials for
the permafrost-affected soils are not clear due to
different processes of relief formation during history.
Therefore, it is difficult to differentiate the distribution
of trace metals from upper layers and substrates. We
only divided all genetic horizons from 25 soil profiles,
for which we identified O, A, B and C master horizons.
Where the horizon of maximum leaching E was
observed, it was included in horizon A. The variability
and levels of Mn appeared to be similar in all horizons,
and Fe has the highest concentration in all horizons
with bigger range observed in O and A horizon that
can be due to more intensive biogeochemical process
and organic enrichment of Fe. Fe and Mn oxides occur
in soils as coating on soil particles, such as concretions
or nodules and fillings in cracks and veins (Meharg
2011). In thawing period, more active redox process
occurs in permafrost-affected soils especially in upper
layers lead the formation of concretion with dominat-
ing Fe and Mn accumulation. This result is consistent
with the previous study geoaccumulation of Fe and
Mn in the north of Yamal Peninsula, Russian Arctic (Ji
et al. 2019a). In comparison with background values
of Fe and Mn in other Russian Siberian regions
(Antcibor et al. 2014; Moskovchenko et al. 2017;
Abakumov et al. 2017), it can be speculated that Fe
and Mn levels in this study represent geochemical
background values of weathering bedrocks.
Other metals showed a pattern in each horizon as
Co [Ni [Hg (O)/Ni [Co [Hg (A)/Co [Hg [
Ni (B)/Ni [Co &Hg(C) [Cu [Pb [Cd (Fig. 4).
Cu and Pb have the second lowest levels (Cu [Pb) in
all horizons, while Cu accumulated more in C horizon
followed by O and A horizon; and Pb showed
Fig. 4 Log-boxplot comparison of variation and gravimetric
concentrations of nine measured metals in divided four great soil
genetic horizons (O: the layer of organic matter on surface of a
mineral soil; A: topsoil of mineral soil horizon with a little
organic matter and low clay; B: subsoil below A horizon where
maximum accumulation of clay and silt occur; C: weathered
rock/parent above the permafrost/frozen soil table which has not
been acted upon by the soil-forming process)
123
Environ Geochem Health

relatively comparable levels in all horizons with the
highest variability in A horizon. The mobility of Cu in
soil profile is highly associated with pH (low pH =
weak mobility) (Bradl 2005) and thence the bigger
retention capacity of O horizon rich in organic/humic
materials (pH = 4.4–4.6) may prevent the adsorption
on clay fraction and leaching process to A horizon for
Cu. However, the higher Cu levels in the bottom of C
horizon may reflect Cu derived from parent materials
in the active layer and deeper permafrost. The
homogenous vertical distribution of Pb in soil profiles
may indicate that Pb may originate from soil itself,
whereas the observed higher ranges of Pb in A horizon
could be due to not all investigated soil pedons have an
O horizon; and secondly, Pb is considered to have
affinity to insoluble humic substances (Angehrnbetti-
nazzi et al. 1989). Co, Ni and Hg have the highest
distribution in all horizons. Co was the highest in O
and B horizons, which is explained by two aspects as
(1) Co mostly has two oxidation stages (Co
2?
and
Co
3?
). Co is more mobile during weathering in
oxidizing acid environments, whereas the high sorp-
tion by Mn and Fe oxides, and clay minerals;
therefore, Co does not migrate in a soluble phase
(Meharg 2011). For A horizon, eluviation occurs with
transport of elements in dissolved soil solution and
less clay is contained. (2) Adsorption, weathering and
transforming mechanisms of Co forms are not only
depended on variable oxidation of Co, but also on
microbial activity (Ainsworth et al. 1994). Numerous
species of bacteria and archaea are concentrated in A
horizon (Johnson et al. 2005; Wilkinson and Hum-
phreys 2005). These bioturbations would either trans-
port Co to upper O horizon or downward B horizon. Ni
was found to be relatively higher in A and C horizon.
Organic matter is believed to mobilize Ni from
carbonates and oxides as well as to decline Ni sorption
on clays (Rieuwerts et al. 1998), thus Ni bound to
organic ligands could not be particularly strong that
can be leached and accumulated in A horizon.
Although surface soil Ni concentrations can reflect
soil-forming processes and contamination, Ni concen-
trations in soil profile are highly dependent on the Ni
content in parent rock. Considering previously
reported Ni concentrations from soils in Lena Delta
River, which is very similar to our soil type and has the
same forming age—the early Holocene (Antcibor
et al. 2014), the distribution of Ni still can reflect the
geochemical background. Interestingly, Hg showed
high values in all horizons. However, Hg abundance in
the Earth’s crust is still not clear (Meier et al. 2016).
Hg is not mobile by weathering in several ionic species
as previous reports pointed out the accumulation of Hg
in soil is determined by formation of organic complex
and dissolution processes after precipitation (Fis-
chwasser 1990; Rundgren et al. 1992; McNeal and
Rose 1974). Besides, the significant higher content of
Hg was observed compared with that in Lena Delta
River, which may indicate the anthropogenic sources
of Hg.
The correlation between trace metals and soil
properties
All data of nine metals’ concentrations and soil
properties from each genetic horizon (Table S3) were
firstly tested by Kolmogorov–Smirnov test with a
result of non-normal distribution. Then, Spearman
correlation between the concentrations of trace metals,
total organic carbon (TOC), total nitrogen (TN) and
particle size fraction was used. The result (Table S4)
showed a significant positive correlation between Zn
and Ni concentrations and TOC, which may be
relevant to the uptake of Zn by vegetation (mostly
moss and lichens) and the binding Zn to organic
matters (Meharg 2011). Although only Oxytropis has
undergone intense speciation in the Arctic (Pre
´vost
et al. 1987), effects of Ni on mineralization in the
Arctic tundra soils are still unknown. A positive
relationship with clay fraction was found for Mn and
Hg. Other metals in this study did not show clear
dependence on any considered soil properties. Some
positive correlation between metals themselves has
been observed as Co correlated with Ni and Cd, Cu
correlated with Ni and Hg correlated with Fe. This
cannot represent the same sources of these metals due
to high variability of metals’ distribution in the soil
profile and mixture sources of anthropogenic and
geology background. Therefore, a principal compo-
nent analysis (PCA) was conducted to reveal the
relationship between the different trace metals, indi-
cating a common source.
Four rotated factors for trace metals and soil
properties are shown in Table S5. Factor 1 reveals
22% of the total variance and showed the significantly
positive loadings on TOC, pH, clay and silt fraction,
Pb, Cu, Ni, Co, Zn, Fe, Mn, Hg. This observation
indicates that all metals’ concentration except Cd in
123
Environ Geochem Health

investigated soils is highly depending on organic
matter content and soil particle size distribution. Most
metals can be more easily absorbed by organic and
mineral components in most soil types (Meharg 2011).
Factor 2 (Fe, Mn and Zn) explained 21% of the total
variance showing soil geobackground values of these
three metals in this region. Factor 3 (Zn, Pb and Cd) is
predominated by Zn and indicates that Cd is strongly
associated with Zn in its geochemistry (Meharg 2011).
It might also indicate that most Pb and Cd originate
from the weathering of mineral soil, but instead is
probably originating from atmospheric deposition due
to the similar concentrations with the study in
Svalbard (Halbach et al. 2017). Factor 4 only
contributed 12% to the total variance. High loadings
on TOC, clay fraction, Pb, Hg, Cu and Co can be
interpreted as the association of Co and Pb with clay
minerals and organic matters, i.e., Co can be absorbed
by clay minerals in oxidizing acidic environments
during weathering; Pb and Cu can be incorporated in
clay minerals by oxidization of their sulfides during
weathering and also suggested an anthropogenic
source. From three main factors between trace metals
and soil properties from each genetic horizon
(Tables S6–S9), some information can be drawn that
most metals except Hg and Mn were associated with
organic matter in each horizon. Co occurred together
with clay minerals in O horizon. Some metals as Pb,
Cu and Ni were accumulated in silt texture soils in A/
Bhorizon, while Cu, Zn and Fe were prone to clayed
soil in deeper C horizon. Hg may originate from a
different source from all other metals.
GBCs of trace metals in topsoil of Russian Arctic
soils
The concentrations of trace metals used for calculation
of GBC values were taken from mineral layers without
organic matter accumulation and leaching from
organic layers (B and C horizon). In this case,
Histosols were not taken for GBC values’ calculation.
The results of GBC of metals (mg kg
-1
) in the Russian
Arctic soils are 11.50 for Pb, 0.13 for Cd, 10.76 for Cu,
51.84 for Ni, 42.94 for Co, 66.88 for Zn, 24,131.85 for
Fe, 733.10 for Mn and 0.03 for Hg. Compared with our
previous investigation in hinterland of Yamal Penin-
sula (Ji et al. 2019b), Russian subarctic regions, the
GBC values in this study were all significantly higher
except for Cd (0.14 mg kg
-1
). Three aspects may
explain this difference: first, different GBCs
approaches were used that relative cumulative fre-
quency versus concentrations was used in soils in
Yamal Peninsula (Ji et al. 2019b). This approach
allows to eliminate anomaly values while no normal-
izing element was used. Second, the concentrations of
metals from material materials (C horizon) were not
considered. Thirdly, the weathering of soil is more
intensive and lower temperature in the Russian Arctic
marginal environment, causing the variability of soil
types and transport of metals. Besides, the GBC values
in this study were also relatively higher than those in
background topsoil (0–15 cm) reported in the Russian
Siberia, i.e., Belyi Island (Moskovchenko et al. 2017;
Abakumov et al. 2017) and Lena Delta River (Antci-
bor et al. 2014). For permafrost-affected soils in tundra
ecosystem, GBCs values are impossible to define the
range of values since the significant fluctuation of
thawing layer each summer. Most metals are prone to
be accumulated in organic matter and some of metals
are bound with Fe/Mn oxides (Meharg 2011). Addi-
tionally, the direct pollution sources of metals in the
northern boreal areas are rare so that air transport is the
main source of anthropogenic metals which are
deposited on the surface soils. Therefore, using metal
concentrations from the mineral layers (clay accumu-
lation) with Li normalization can better reflect
geochemical substrates.
Geoaccumulation and enrichment of trace metals
in topsoil
I
geo
values of Cu and Ni had the greatest range of
-3.2 to 0.1 and -3.3 to 2.4, respectively. The range
of other metals followed the degressive orders of I
geo
values as Cd (-4.9 to 0.1), Zn (-1.7 to 2.8), Co
(0.5–4.6), Hg (-3.5 to -0.6), Fe (-1.5 to -0.3)
and Mn (-2.1 to -1.2). The highest contamination
site was W5 with the highest I
geo
values of 4.6 for Co
(Fig. S1a), belonging to the highly to extremely
polluted. Co was also only found to be highly polluted
in site G1 (I
geo
= 3.1). Mean I
geo
values showed Co
was moderately to highly polluted (I
geo
= 2–2.5) in
most investigated areas except it showed to be
moderately polluted (I
geo
= 1.2) in Island Gerkules
and moderately polluted (I
geo
= 3.1) in Island Vay-
gach (Fig. S1b). Zn showed to be moderately to highly
polluted in Valraray Cape, Island Bennett and Island
Wrangel with the mean I
geo
range of 2.1–2.8, and
123
Environ Geochem Health

moderately polluted and unpolluted–slightly polluted
Zn was observed in Island Bolshoy Lyakhovsky (mean
I
geo
= 1.8), Island Chelyuskin (mean I
geo
= 1.6) and
Island Kolguyev (mean I
geo
= 0.9). The relatively
higher mean I
geo
values of Cu were found in Island
Wrangel (1.9), Island Chelyuskin (2.0) and Island
Kolguyev (1.0). Mean I
geo
values of Pb were only
found to be moderately polluted in Island Bennett
(1.0) and Island Wrangel (1.4). Other metals in all
investigated areas appeared to be unpolluted status.
Taking Li as normalizing element, EF values of all
metals were analyzed. The result of mean EF values
(in case any single value being too high or too low
compared to the rest of the samples, median has been
taken as mean) was ordered as Co (6.59) [Zn
(5.55) [Cu (2.94) [Pb (2.25) [Ni (1.45) [Fe
(0.85) [Mn (0.46) [Hg (0.35) [Cd (0.28)
(Table 1). According to the categorization of mean
EF values (EF values = 2–5, moderately enriched; EF
values = 5–20, strongly enriched), Pb and Cu were
moderately enriched, and Co and Zn were strongly
enriched in the topsoil (EF values = 5–20), also
indicating anthropogenic origin (EF values [2). As
for Ni, Mn, Hg and Cd, they had a minimal enrichment
(EF values \2), suggesting a crustal origin. Sample
site W5 showed a high to very high enrichment of Co,
and all sites are moderate to strong enrichment of Co
(Fig. S2). This may be possibly due to mining and
Table 2 Contamination
factor (CF), degree of
pollution (C
degree
), modified
degree of pollution
(mC
degree
) and pollution
load index (PLI) of nine
measured metals in the
topsoil
Sample CF C
degree
mC
degree
PLI
Pb Cd Cu Ni Co Zn Mn Hg Fe
H1 1.1 0.2 2.2 1.4 6.9 3.5 0.5 0.5 1.0 17.13 1.90 1.14
VC1 1.2 1.6 1.8 1.0 6.6 10.6 0.4 0.3 0.9 24.39 2.71 1.45
B1 2.5 0.5 2.1 2.2 9.0 10.0 0.7 0.3 0.6 27.77 3.09 1.55
B2 3.4 0.4 2.0 1.4 4.2 5.9 0.6 1.0 1.2 20.15 2.24 1.58
V1 0.2 0.1 0.2 0.2 2.2 0.4 0.4 0.3 0.5 4.52 0.50 0.33
V2 1.7 1.6 2.6 1.5 5.8 5.5 0.5 0.6 1.0 20.85 2.32 1.67
G1 1.0 0.3 1.1 0.2 13.2 0.7 0.5 1.0 1.0 18.79 2.09 0.83
BB1 2.3 0.3 5.7 2.1 4.9 5.2 0.4 0.5 1.0 22.28 2.48 1.48
BB2 3.9 0.3 4.3 2.0 9.4 7.0 0.4 0.2 0.8 28.26 3.14 1.53
W1 5.1 0.2 8.2 3.7 5.8 8.8 0.5 0.2 0.6 33.19 3.69 1.66
W2 3.8 0.2 6.3 2.9 6.4 7.1 0.5 0.2 0.8 28.02 3.11 1.44
W3 4.2 0.1 5.9 2.7 5.0 6.2 0.5 0.4 1.0 26.09 2.90 1.56
W4 2.1 0.1 5.5 1.4 5.5 6.1 0.5 0.4 0.8 22.27 2.47 1.17
W5 3.3 1.5 9.0 5.8 36.3 8.1 0.4 0.3 0.9 65.53 7.28 2.74
W6 7.7 0.8 1.9 0.4 6.6 4.2 0.6 0.4 0.5 23.12 2.57 1.36
W7 4.5 0.6 6.0 0.6 6.7 6.3 0.3 0.3 1.0 26.25 2.92 1.49
BL1 2.4 0.1 2.1 1.4 4.4 4.2 0.5 0.4 0.9 16.44 1.83 1.13
BL2 1.4 0.4 2.9 1.1 8.8 4.3 0.4 0.3 0.8 20.43 2.27 1.26
BL3 1.9 0.1 2.6 1.3 4.4 4.2 0.4 0.4 0.7 16.01 1.78 1.07
BL4 0.7 0.6 3.0 1.1 6.6 10.3 0.4 0.4 1.0 24.14 2.68 1.35
C1 1.8 0.1 5.8 3.5 8.1 6.1 0.5 0.4 0.8 27.14 3.02 1.49
C2 2.3 0.3 7.0 1.8 6.6 3.4 0.4 0.1 0.8 22.79 2.53 1.27
C3 1.2 0.2 5.6 7.8 7.3 4.8 0.4 0.3 1.0 28.59 3.18 1.53
K1 1.5 0.7 2.9 0.9 6.6 2.0 0.5 0.4 0.8 16.25 1.81 1.21
K2 2.3 0.6 2.9 1.6 5.6 3.6 0.4 0.3 0.8 18.18 2.02 1.34
Max 7.7 1.6 9.0 7.8 36.3 10.6 0.7 1.0 1.2 65.5 7.3 2.7
Min 0.2 0.1 0.2 0.2 2.2 0.4 0.3 0.1 0.5 4.5 0.5 0.3
Mean 2.5 0.5 4.0 2.0 7.7 5.6 0.5 0.4 0.8 23.94 2.66 1.38
123
Environ Geochem Health

metallurgical factories are abundant in the Russian
Siberia as important sources of Co, for example,
company Norilsk that processes copper and nickel,
and it is one of the biggest metallurgical factories in
the world (Shevchenko et al. 2003). Co is primarily
extracted as a by-product of nickel and copper ores,
which is consistent with moderately strong enrichment
of Cu.
Soil pollution level
The analysis of contamination factor reveals the
pollution level above background baseline values.
The average pollution levels (CF values) showed in
order as Co (7.7) [Zn (5.6) [Cu (4.0) [Pb
(2.5) [Ni (2.0) [Fe (0.8) [Cd = Mn (0.5) [Hg
(0.4) (Table 2). According to the categorization of CF
values (Table S2), Zn and Cu belonged to high
pollution (3 BCF \6) and Co has extremely high
pollution (CF C6), whereas Pb and Ni have moderate
pollution, and Cd, Mn, Fe and Hg have low pollution.
All investigated areas in the Russian Arctic have
extremely high pollution of Co except for Island
Vaygach (Fig. S3). Extremely high pollution of Zn
was observed in Island Wrangel, Bennett and Valraray
Cape. Island Bolshoy Lyakhovsky has the most
amount of metals’ pollution as Zn and Ni with high
pollution and Cu and Co with extremely high
pollution.
Table 3 Ecological risk
(E
r
) and potential risk factor
(RI) of nine measured
metals in the topsoil
Sample E
r
RI
Pb Cd Cu Ni Co Zn Mn Hg Fe
H1 5.30 5.58 10.96 7.18 34.44 3.46 0.48 37.00 0.97 105.36
VC1 6.06 46.96 8.81 4.91 32.94 10.62 0.43 24.00 0.94 135.66
B1 12.41 15.24 10.65 10.89 44.79 10.05 0.66 21.00 0.56 126.22
B2 17.00 10.68 10.10 7.23 21.22 5.91 0.59 77.00 1.21 150.96
V1 0.97 3.41 0.83 0.98 10.98 0.45 0.39 24.00 0.52 42.52
V2 8.58 47.49 13.16 7.49 28.88 5.55 0.55 44.00 1.00 156.70
G1 5.00 9.24 5.36 0.75 65.88 0.65 0.47 76.00 1.01 164.35
BB1 11.47 8.15 28.52 10.39 24.68 5.16 0.38 40.00 0.96 129.72
BB2 19.47 7.97 21.32 9.94 46.76 7.04 0.45 19.00 0.77 132.72
W1 25.60 5.68 41.16 18.44 29.09 8.84 0.52 17.00 0.57 146.90
W2 19.08 4.65 31.27 14.61 31.89 7.10 0.49 12.00 0.76 121.83
W3 21.08 4.05 29.65 13.72 24.92 6.21 0.52 30.00 0.97 131.12
W4 10.30 1.54 27.27 7.02 27.74 6.08 0.46 32.00 0.81 113.23
W5 16.28 45.36 44.88 29.06 181.36 8.10 0.40 26.00 0.87 352.31
W6 38.68 23.52 9.27 1.86 32.94 4.23 0.60 34.00 0.52 145.63
W7 22.67 18.72 29.93 2.88 33.26 6.29 0.34 21.00 0.98 136.07
BL1 11.93 4.08 10.66 7.17 21.97 4.19 0.49 28.00 0.92 89.43
BL2 6.86 13.13 14.51 5.45 43.83 4.29 0.43 24.00 0.85 113.33
BL3 9.52 4.10 12.81 6.27 22.21 4.18 0.42 29.00 0.74 89.26
BL4 3.48 18.45 14.90 5.61 32.94 10.34 0.41 30.00 1.01 117.15
C1 9.19 4.12 28.87 17.73 40.66 6.09 0.52 28.00 0.76 135.94
C2 11.37 8.51 35.19 8.95 32.94 3.44 0.43 10.40 0.82 112.04
C3 5.97 6.93 27.83 38.97 36.62 4.84 0.42 21.00 0.96 143.54
K1 7.35 21.96 14.45 4.32 32.94 2.03 0.52 31.00 0.77 115.33
K2 11.27 17.80 14.70 7.77 28.14 3.64 0.44 25.00 0.82 109.57
Max 38.68 47.49 44.88 38.97 181.36 10.62 0.66 77.00 1.21 352.31
Min 0.97 1.54 0.83 0.75 10.98 0.45 0.34 10.40 0.52 42.52
Mean 12.68 14.29 19.88 9.98 38.56 5.55 0.47 30.42 0.84 132.68
123
Environ Geochem Health

Overall pollution level in the Russian Arctic
topsoil
Degree of contamination (C
deg
) is based on total CF
values of each metal for the site contamination. Most
of the site showed C
deg
values were less than 1 time of
the sample number (Table 2), which is classified as
low degree of pollution (Table S2). Few sites as B1,
BB2, W1, W2, W3, W7, BL4, C1, K1, especially for
Island Wrange, showed C
deg
values were 1–2 times of
the sample number, indicating moderate degree of
pollution, whereas only site W5 in Island Wrange was
2–4 times of the sample number, belonging to high
degree of pollution. C
deg
can be modified to mC
deg
defining a comprehensive evaluation (G. M. S.
Abrahim and Parker 2008). In this way, most of the
sites were classified into moderate degree of pollution
(2 BmC
degree
\4). Only three sites (H1, BL1 and
BL3) resulted in nil to very low degree of pollution
(mC
degree
\1.5) and site V1 resulted in low degree of
pollution (2 BmC
degree
\4). Site W5 evidenced high
degree of pollution (4 BmC
degree
\8). Considering
the whole investigated sites in the Russian Arctic, the
mean values (2.66) showed moderate degree of
pollution (Table 2).
PLI index is the best way to show the pollution load
on each sampling site through calculation of geometric
tendency (Manna and Maiti 2018). This study showed
there was an unpolluted to moderate pollution load in
the entire investigated areas with the mean value of
1.38 (Table 2). Only V1 and G1 showed no pollution
load (1 \PLIB1), while W5 revealed moderate
pollution load. Therefore, the most seriously polluted
site is located in W5 (Island Island Wrangel) which is
covered by moss and lichen tundra without particular
pollution sources. However, Island Wrange was used
as Soviet military outpost, which may explain the
metal pollution sporadically distributed in this island.
Ecological risk of trace metals pollution
The result of ecological risk (E
r
) for nine measured
metals varied significantly as Pb ranged from 0.97 to
38.68, Cd from 1.54 to 47.49, Cu from 0.83 to 44.88,
Ni from 0.75 to 38.97, Co from 10.98 to 181.36, Zn
from 0.45 to 10.62, Mn from 0.34 to 0.66, Hg from
10.40 to 77.00 and Fe from 0.52 to 1.21 (Table 3). Zn,
Fe and Mn with less toxic response showed very low
mean ecological risk (E
r
= 0.47–0.84), while Pb, Cu,
Ni and Co with moderate toxic response also showed a
low ecological risk with relatively higher E
r
values
ranged from 9.98 to 38.56. For Cd and Hg having high
toxic response, the ecological risk appeared to be low
as well (E
r
= 14.29 for Cd and 30.42 for Hg). Sensitive
response of biologic tundra environment to the toxic
compounds can be assessed by potential ecological
risk (RI) for individual sampling site. The result of
average RI value for entire investigated Russia Arctic
was 132.68 (Table 3), which is low ecological risk
potential (RI \150). Specifically, it is found that site
G1, V2, B2 had moderately ecological risk potential,
whereas only W5 had high ecological risk potential
(Fig. S4).
Conclusions
In this study, we investigated 9 trace metals from 25
soil profiles with 93 genetic organic and mineral
horizons from active layer of permafrost-affected soil
in the Russian Arctic during summer. Different
degrees of organic matter decomposition and cryotur-
bation are two important factors for permafrost-
affected soils, causing great variability of metals’
distribution. From the analysis of soil genetic hori-
zons, levels of Fe and Mn are considered to be
geochemical background while other metals showed
different distribution in different horizons where upper
layers (O and A horizon) have significantly higher
variability for most metals. Correlation analysis and
PCA revealed that trace metals are highly depended on
soil properties and Co, Cu, Pb, Hg may derive from
anthropogenic sources. According to I
geo
, EF, CF
indexes evaluated through normalized geochemical
baseline concentrations (GBCs), Co, Zn, Cu and Pb
have polluted status in the topsoil in this area. Zn, Co
and Cu have higher polluted status. According to
ecological indexes for all nine metals (E
r
,mC
deg
,PLI
and RI), Island Wrangel has the highest values for all
these indices. Taking these sites as whole, this Russian
Arctic marginal soils moderate degree of pollution and
low degree of ecological risk.
Although low ecological risk potential is presented
so far, with the climate changing, the status of metals
highly associated with soil properties (especially for
carbon content) can be greatly changed. We suggest
that long-term monitoring of trace metals with more
sampling sites in soils of the Arctic marginal
123
Environ Geochem Health

environment is needed for more accurate ecological
assessment.
Acknowledgements This work was supported by Grant of
Russian Foundation for Basic research (18-44-890003, 19-416-
890002, and 19-05-50107) and by a Grant of Saint Petersburg
State University ‘‘Urbanized ecosystems of the Russian Arctic:
dynamics; state and sustainable development (Grant No.
39377455).’’ The authors are grateful to Dr. Julia Antcibor
from Institute of Soil Science, University of Hamburg, for her
assistance in laboratory. We would like to thank Miss Yu Su
from the School of Visual Arts at BFA Computer Art for helping
with data visualization and Miss Kuznetsova Ekaterina from
School of Journalism and Communication, Tsinghua
University, for helping with the Russian translation. We are
very grateful to two anonymous reviewers for carefully pointing
out the mistakes and giving good suggestions for revision of our
manuscript.
References
Abakumov, E., Shamilishviliy, G., & Yurtaev, A. (2017). Soil
polychemical contamination on Beliy Island as key back-
ground and reference plot for Yamal region. Polish Polar
Research, 38(3), 313–332. https://doi.org/10.1515/popore-
2017-0020.
Abrahim, G. (2019). Holocene sediments of Tamaki Estuary:
Characterisation and impact of recent human activity on
an urban estuary in Auckland, New Zealand.
Abrahim, G. M. S., & Parker, R. J. (2008). Assessment of heavy
metal enrichment factors and the degree of contamination
in marine sediments from Tamaki Estuary, Auckland, New
Zealand. Environmental Monitoring and Assessment,
136(1), 227–238. https://doi.org/10.1007/s10661-007-
9678-2.
Adama, M., Esena, R., Fosu-Mensah, B., & Yirenya-Tawiah, D.
(2016). Heavy metal contamination of soils around a hos-
pital waste incinerator bottom ash dumps site. Journal of
Environmental and Public Health.https://doi.org/10.1155/
2016/8926453.
Ahdy, H. H. H., & Khaled, A. (2009). Heavy metals contami-
nation in sediments of the western part of Egyptian
Mediterranean Sea. Australian Journal of Basic and
Applied Sciences, 3(4), 3330–3336.
Ainsworth, C. C., Gassman, P. L., Pilon, J. L., & Van Der Sluys,
W. G. (1994). Cobalt, cadmium, and lead sorption to
hydrous iron oxide: Residence time effect. Soil Science
Society of America Journal, 58, 1615–1623. https://doi.
org/10.2136/sssaj1994.03615995005800060005x.
Akoto, O., Nimako, C., Asante, J., & Bailey, D. (2016). Heavy
metals enrichment in surface soil from abandoned waste
disposal sites in a hot and wet tropical area. Environmental
Processes, 3(4), 747–761. https://doi.org/10.1007/s40710-
016-0183-x.
Almasoud, F. I., Usman, A. R., & Al-Farraj, A. S. (2015). Heavy
metals in the soils of the Arabian Gulf coast affected by
industrial activities: Analysis and assessment using
enrichment factor and multivariate analysis. Arabian
Journal of Geosciences, 8(3), 1691–1703. https://doi.org/
10.1007/s12517-014-1298-x.
Aloupi, M., & Angelidis, M. O. (2001). Normalization to
lithium for the assessment of metal contamination in
coastal sediment cores from the Aegean Sea, Greece.
Marine Environmental Research, 52(1), 1–12. https://doi.
org/10.1016/s0141-1136(00)00255-5.
AMAP. (2005). Heavy metals in the Arctic. Oslo: Rctic Moni-
toring and Assessment Programme.
Angehrnbettinazzi, C., Thoni, L., & Hertz, J. (1989). An attempt
to evaluate some factors affecting the heavy metals–metal
accumulation in a forest sand. International Journal of
Environmental Analytical Chemistry, 35(2), 69–79. https://
doi.org/10.1080/03067318908028380.
Angulo, E. (1996). The Tomlinson Pollution Load Index applied
to heavy metal, ‘Mussel-Watch’ data: A useful index to
assess coastal pollution. Science of the Total Environment,
187(1), 19–56. https://doi.org/10.1016/0048-
9697(96)05128-5.
Antcibor, I., Eschenbach, A., Zubrzycki, S., Kutzbach, L.,
Bolshiyanov, D., & Pfeiffer, E. M. (2014). Trace metal
distribution in pristine permafrost-affected soils of the
Lena River delta and its hinterland, northern Siberia,
Russia. Biogeosciences, 11(1), 1–15. https://doi.org/10.
5194/bg-11-1-2014.
Ba
¨chmann, K., Steigerwald, K., & Klenk, H. (1992). Develop-
ment of a method for direct determination of solids as thin
films by energy dispersive X-ray fluorescence analysis.
Microchimica Acta, 108(3), 299–302. https://doi.org/10.
1007/BF01242440.
Barker, A. J., Douglas, T. A., Jacobson, A. D., McClelland, J.
W., Ilgen, A. G., Khosh, M. S., et al. (2014). Late season
mobilization of trace metals in two small Alaskan arctic
watersheds as a proxy for landscape scale permafrost active
layer dynamics. Chemical Geology, 381, 180–193. https://
doi.org/10.1016/j.chemgeo.2014.05.012.
Bhuiyan, M. A. H., Parvez, L., Islam, M. A., Dampare, S. B., &
Suzuki, S. (2010). Heavy metal pollution of coal mine-
affected agricultural soils in the northern part of Bangla-
desh. Journal of Hazardous Materials, 173(1–3), 384–392.
https://doi.org/10.1016/j.jhazmat.2009.08.085.
Boyd, R., Barnes, S. J., De Caritat, P., Chekushin, V. A.,
Melezhik, V. A., Reimann, C., et al. (2009). Emissions
from the copper–nickel industry on the Kola Peninsula and
at Noril’sk, Russia. Atmospheric Environment, 43(7),
1474–1480. https://doi.org/10.1016/j.atmosenv.2008.12.
003.
Bradl, H. (2005). Heavy metals in the environment: Origin,
interaction and remediation (Vol. 6, p. 282). Academic
Press.
Buchkina, N. P., Zuyev, V. S., & Balashov, E. V. (1998). Effects
of tracked vehicles on the morphological and physical
properties of tundra soils. Soil and Tillage Research, 48(4),
317–324. https://doi.org/10.1016/s0167-1987(98)00158-5.
Christiansen, H. H., Etzelmu
¨ller, B., Isaksen, K., Juliussen, H.,
Farbrot, H., Humlum, O., et al. (2010). The thermal state of
permafrost in the nordic area during the international polar
year 2007–2009. Permafrost and Periglacial Processes,
21(2), 156–181. https://doi.org/10.1002/ppp.687.
Covelli, S., & Fontolan, G. (1997). Application of a normal-
ization procedure in determining regional geochemical
123
Environ Geochem Health

baselines. Environmental Geology, 30(1), 34–45. https://
doi.org/10.1007/s002540050130.
DINISO10390. (2005). Soil quality: Determination of pH (DIN
ISO 10390:2005).
Dube, A., Zbytniewski, R., Kowalkowski, T., Cukrowska, E., &
Buszewski, B. (2001). Adsorption and migration of heavy
metals in soil. Polish Journal of Environmental Studies,
10(1), 1–10.
Ejarque, E., & Abakumov, E. (2016). Stability and biodegrad-
ability of organic matter from Arctic soils of Western
Siberia: Insights from C-13-NMR spectroscopy and ele-
mental analysis. Solid Earth, 7(1), 153–165. https://doi.
org/10.5194/se-7-153-2016.
Fischwasser, K. (1990). Environmental inorganic chemistry—
properties, processes, and estimation methods. Edited by I.
Bodek; W. Lyman; W. F. Reehl and D. H. Rosenblatt.—
NEw York/Oxford/Beijing/Frankfurt/Sao Paulo/Sydney/
Tokyo/Toronto: Pergamon Press 1988. ISBN 0-08-
036833-6, $ 150.00. Internationale Revue der gesamten
Hydrobiologie und Hydrographie, 75(5), 681-681, https://
doi.org/10.1002/iroh.19900750509.
Frank, R., Ishida, K., & Suda, P. (1976). Metals in agricultural
soils of Ontario. Canadian Journal of Soil Science, 56(3),
181–196. https://doi.org/10.4141/cjss76-027.
Hakanson, L. (1980). An ecological risk index for aquatic pol-
lution-control—A sedimentological approach. Water
Research, 14(8), 975–1001. https://doi.org/10.1016/0043-
1354(80)90143-8.
Halbach, K., Mikkelsen, O., Berg, T., & Steinnes, E. (2017). The
presence of mercury and other trace metals in surface soils
in the Norwegian Arctic. Chemosphere, 188, 567–574.
https://doi.org/10.1016/j.chemosphere.2017.09.012.
Hofle, S., Rethemeyer, J., Mueller, C. W., & John, S. (2013).
Organic matter composition and stabilization in a polyg-
onal tundra soil of the Lena Delta. Biogeosciences, 10(5),
3145–3158. https://doi.org/10.5194/bg-10-3145-2013.
Ives, M., Lewis, D. B., & Lehmberg, C. (1997). Depth profile
analysis of multilayer Ni–Fe alloy coatings by glow dis-
charge optical emission spectroscopy (GDOES) and
energy dispersive X-ray (EDX) linescan—A comparative
study. Surface and Interface Analysis, 25(3), 191–201.
https://doi.org/10.1002/(sici)1096-9918(199703)25:3%
3c191:Aid-sia218%3e3.3.Co;2-2.
Jaffe, D., Cerundolo, B., Rickers, J., Stolzberg, R., & Baklanov,
A. (1995). Deposition of sulfate and heavy-metals on the
Kola–Peninsula. Science of the Total Environment,
160–61, 127–134. https://doi.org/10.1016/0048-
9697(95)04350-a.
Ji, X., Abakumov, E., Antcibor, I., Tomashunas, V., Knoblauch,
C., Zubzycki, S., et al. (2019a). Influence of anthropogenic
activities on metals in arctic permafrost: A characterization
of benchmark soils on the Yamal and Gydan Peninsulas in
Russia. Archives of Environmental Contamination and
Toxicology, 76(4), 540–553. https://doi.org/10.1007/
s00244-019-00607-y.
Ji, X., Abakumov, E., & Polyakov, V. (2019b). Assessments of
pollution status and human health risk of heavy metals in
permafrost-affected soils and lichens: A case-study in
Yamal Peninsula, Russia Arctic AU—Ji, Xiaowen. Human
and Ecological Risk Assessment: An International Journal.
https://doi.org/10.1080/10807039.2018.1490887.
Jiang, D. Z., Teng, E. J., & Liu, Y. L. (1996). The contribution of
difference on the element background values in soils and
the analysis of variance of single factor on soil groups.
Environmental Monitoring China, 2, 21–24.
Johnson, D. L., Domier, J. E. J., & Johnson, D. N. (2005).
Reflections on the nature of soil and its biomantle. Annals
of the Association of American Geographers, 95(1), 11–31.
https://doi.org/10.1111/j.1467-8306.2005.00448.x.
Kashulina, G. M. (2017). Extreme pollution of soils by emis-
sions of the copper–nickel industrial complex in the Kola
Peninsula. Eurasian Soil Science, 50(7), 837–849. https://
doi.org/10.1134/S1064229317070031.
Katsoyiannis, A., Sweetman, A. J., & Jones, K. C. (2011). PAH
molecular diagnostic ratios applied to atmospheric sources:
A critical evaluation using two decades of source inventory
and air concentration data from the UK. Environmental
Science and Technology, 45(20), 8897–8906. https://doi.
org/10.1021/es202277u.
Kitagishi, K., & Yamane, I. (1981). Heavy metal pollution in
soils of Japan. Tokyo: Japan Scientific Societies Press.
Kosko, M. K., Cecile, M. P., Harrison, J. C., Ganelin, V. G.,
Khandoshko, N. V., & Lopatin, B. G. (1993). Geology of
Wrangel Island, between Chukchi and east Siberian seas,
northeastern Russia. Bulletin 461, Geological Survey of
Canada, Ottawa Ontario, 101.
Kozlov, M. V., & Barcan, V. (2000). Environmental contami-
nation in the central part of the Kola Peninsula: History,
documentation, and perception. AMBIO A Journal of the
Human Environment, 29(8), 512–517.
Krasilnikov, P., Marti, J.-J. I., Arnold, R., & Shoba, S. (2009). A
handbook of soil terminology, correlation and classifica-
tion. London: Earthscan.
Liaghati, T., Preda, M., & Cox, M. (2004). Heavy metal distri-
bution and controlling factors within coastal plain sedi-
ments, Bells Creek catchment, southeast Queensland.
Australia. Environment International, 29(7), 935–948.
https://doi.org/10.1016/s0160-4120(03)00060-6.
Likuku, A. S., Mmolawa, K. B., & Gaboutloeloe, G. K. (2013).
Assessment of heavy metal enrichment and degree of
contamination around the copper–nickel mine in the Selebi
Phikwe Region, Eastern Botswana. Environment and
Ecology Research, 1(2), 32–40. https://doi.org/10.13189/
eer.2013.010202.
Loska, K., Wiechuła, D., & Korus, I. (2004). Metal contami-
nation of farming soils affected by industry. Environment
International, 30(2), 159–165. https://doi.org/10.1016/
S0160-4120(03)00157-0.
Lukina, N., & Nikonov, V. (2001). Assessment of environ-
mental impact zones in the Kola Peninsula forest ecosys-
tems. Chemosphere, 42(1), 19–34.
Machender, G., Dhakate, R., Prasanna, L., & Govil, P. K.
(2011). Assessment of heavy metal contamination in soils
around Balanagar industrial area, Hyderabad, India. Envi-
ronmental Earth Sciences, 63(5), 945–953. https://doi.org/
10.1007/s12665-010-0763-4.
Manna, A., & Maiti, R. (2018). Geochemical contamination in
the mine affected soil of Raniganj Coalfield—A river basin
scale assessment. Geoscience Frontiers, 9(5), 1577–1590.
https://doi.org/10.1016/j.gsf.2017.10.011.
Marques, M., Sierra, J., Drotikova, T., Mari, M., Nadal, M., &
Domingo, J. L. (2017). Concentrations of polycyclic
123
Environ Geochem Health

aromatic hydrocarbons and trace elements in Arctic soils:
A case-study in Svalbard. Environmental Research, 159,
202–211. https://doi.org/10.1016/j.envres.2017.08.003.
McNeal, J. M., & Rose, A. W. (1974). The geochemistry of
mercury in sedimentary rocks and soils in Pennsylvania.
Geochimica et Cosmochimica Acta, 38(12), 1759–1784.
https://doi.org/10.1016/0016-7037(74)90160-4.
Meharg, A. A. (2011). Trace elements in soils and plants. 4th
edition. By Kabata-Pendias A. Boca Raton, FL, USA: CRC
Press/Taylor & Francis Group (2010), pp. 548. Experi-
mental Agriculture, 47(4), 739–739, https://doi.org/10.
1017/s0014479711000743.
Meier, M. M. M., Cloquet, C., & Marty, B. (2016). Mercury
(Hg) in meteorites: Variations in abundance, thermal
release profile, mass-dependent and mass-independent
isotopic fractionation. Geochimica et Cosmochimica Acta,
182, 55–72. https://doi.org/10.1016/j.gca.2016.03.007.
Moskovchenko, D. V. (2013). Ecogeochemistry of oil and gas
development areas of West Siberia. Novosibirsk: Geo. (in
Russian).
Moskovchenko, D. V., Kurchatova, A. N., Fefilov, N. N., &
Yurtaev, A. A. (2017). Concentrations of trace elements
and iron in the Arctic soils of Belyi Island (the Kara Sea,
Russia): Patterns of variation across landscapes. Environ-
mental Monitoring and Assessment. https://doi.org/10.
1007/s10661-017-5928-0.
Mugos
ˇa, B., Ðurovic
´, D., Nedovic
´-Vukovic
´, M., Barjaktarovic
´-
Labovic
´, S., & Vrvic
´, M. (2016). Assessment of Ecological
Risk of Heavy Metal Contamination in Coastal Munici-
palities of Montenegro. International Journal of Environ-
mental Research and Public Health, 13(4), 393.
Mu
¨ller, G. (1969). Index of geoaccumulation in sediments of the
rhine river. Geojournal, 2, 108–118.
Muskett, R. R., & Romanovsky, V. E. (2011). Alaskan per-
mafrost groundwater storage changes derived from
GRACE and ground measurements. Remote Sensing, 3(2),
378–397. https://doi.org/10.3390/rs3020378.
Niskavaara, H., Reimann, C., Chekushin, V., & Kashulina, G.
(1997). Seasonal variability of total and easily leachable
element contents in topsoils (0–5 cm) from eight catch-
ments in the European Arctic (Finland, Norway and Rus-
sia). Environmental Pollution, 96(2), 261–274. https://doi.
org/10.1016/s0269-7491(97)00031-6.
Pacyna, J. M. (1995). The origin of Arctic air-pollutants—
Lessons learned and future-research. Science of the Total
Environment, 160–61, 39–53. https://doi.org/10.1016/
0048-9697(95)04343-y.
Pacyna, E. G., Pacyna, J. M., Sundseth, K., Munthe, J., Kind-
bom, K., Wilson, S., et al. (2010). Global emission of
mercury to the atmosphere from anthropogenic sources in
2005 and projections to 2020. Atmospheric Environment,
44(20), 2487–2499. https://doi.org/10.1016/j.atmosenv.
2009.06.009.
Patton, A. I., Rathburn, S. L., & Capps, D. M. (2019). Landslide
response to climate change in permafrost regions. Geo-
morphology, 340, 116–128. https://doi.org/10.1016/j.
geomorph.2019.04.029.
Perreault, P., & Shur, Y. (2016). Seasonal thermal insulation to
mitigate climate change impacts on foundations in per-
mafrost regions. Cold Regions Science and Technology,
132, 7–18. https://doi.org/10.1016/j.coldregions.2016.09.
008.
Ping, C.-L., Clark, M. H., Kimble, J. M., Michaelson, G. J.,
Shur, Y., & Stiles, C. A. (2013). Sampling protocols for
permafrost-affected soils. Soil Horizons. https://doi.org/10.
2136/sh12-09-0027.
Pokrovsky, O. S., Shirokova, L. S., Kirpotin, S. N., Audry, S.,
Viers, J., & Dupre, B. (2011). Effect of permafrost thawing
on organic carbon and trace element colloidal speciation in
the thermokarst lakes of western Siberia. Biogeosciences,
8(3), 565–583. https://doi.org/10.5194/bg-8-565-2011.
Pre
´vost, D., Bordeleau, L. M., Caudry-Reznick, S., Schulman,
H. M., & Antoun, H. (1987). Characteristics of rhizobia
isolated from three legumes indigenous to the Canadian
high arctic: Astragalus alpinus,Oxytropis maydelliana,
and Oxytropis arctobia.Plant and Soil, 98(3), 313–324.
https://doi.org/10.1007/bf02378352.
Pryde, P. R. (1991). Environmental management in the Soviet
Union (Cambridge Soviet Paperbacks; 4, Vol. Accessed
from https://nla.gov.au/nla.cat-vn1080305). Cambridge:
Cambridge University Press.
Quinn, J. A., & Woodward, S. L. (2015). Earth’s landscape: An
encyclopedia of the world’s geographic features (Vol. 2,
p. 887). ABC-CLIO.
Rahn, K. A., & Lowenthal, D. H. (1984). Elemental traces of
distant regional pollution aerosols. Science, 223(4632),
132–139. https://doi.org/10.1126/science.223.4632.132.
Ravichandran, M., Baskaran, M., Santschi, P. H., & Bianchi, T.
S. (1995). History of trace metal pollution in Sabine-
Neches Estuary, Beaumont, Texas. Environmental Science
and Technology, 29(6), 1495–1503. https://doi.org/10.
1021/es00006a010.
Reimann, C., Boyd, R., deCaritat, P., Halleraker, J. H., Kashu-
lina, G., Niskavaara, H., et al. (1997). Topsoil (0–5 cm)
composition in eight arctic catchments in northern Europe
(Finland, Norway and Russia). Environmental Pollution,
95(1), 45–56. https://doi.org/10.1016/s0269-
7491(96)00102-9.
Reimann, C., Halleraker, J. H., Kashulina, G., & Bogatyrev, I.
(1999). Comparison of plant and precipitation chemistry in
catchments with different levels of pollution on the Kola
Peninsula, Russia. Science of the Total Environment, 243,
169–191. https://doi.org/10.1016/s0048-9697(99)00390-3.
Rieuwerts, J. S., Thornton, I., Farago, M. E., & Ashmore, M. R.
(1998). Factors influencing metal bioavailability in soils:
Preliminary investigations for the development of a critical
loads approach for metals. Chemical Speciation and
Bioavailability, 10(2), 61–75. https://doi.org/10.3184/
095422998782775835.
Romanovsky, V. E., Drozdov, D. S., Oberman, N. G., Malkova,
G. V., Kholodov, A. L., Marchenko, S. S., et al. (2010).
Thermal state of permafrost in Russia. Permafrost and
Periglacial Processes, 21(2), 136–155. https://doi.org/10.
1002/ppp.683.
Rovinsky, F., Pastukhov, B., Bouyvolov, Y., & Burtseva, L.
(1995). Present day state of background pollution of the
natural environment in the Russian Arctic in the region of
the Ust-Lena Reserve. Science of the Total Environment,
160–161, 193–199. https://doi.org/10.1016/0048-
9697(95)04356-6.
123
Environ Geochem Health

Rubio, B., Nombela, M. A., & Vilas, F. (2000). Geochemistry of
Major and Trace Elements in Sediments of the Ria de Vigo
(NW Spain): An Assessment of Metal Pollution. Marine
Pollution Bulletin, 40(11), 968–980. https://doi.org/10.
1016/S0025-326X(00)00039-4.
Rundgren, S., Ruhling, A., Schluter, K., & Tyler, G. (1992).
Mercury in soil: Distribution, speciation and biological
effects. A review of the literature and comments on critical
concentrations. Copenhagen: Nordic Council of Ministers.
Ryaboshapko, A., Gallardo, L., Kjellstro
¨m, E., Gromov, S.,
Paramonov, S., Afinogenova, O., et al. (1998). Balances of
oxidized sulfur and nitrogen over the former Soviet Union
territory. Atmospheric Environment, 32(4), 647–658.
https://doi.org/10.1016/S1352-2310(97)00298-7.
Sahoo, P. K., Equeenuddin, S. M., & Powell, M. A. (2016).
Trace elements in soils around coal mines: Current sce-
nario, impact and available techniques for management.
Current Pollution Reports, 2(1), 1–14. https://doi.org/10.
1007/s40726-016-0025-5.
Saleh, S. M. K., Thabit Amer, A., & Al-Alawi, A. (2018).
Potential ecological risk of heavy metals in surface sedi-
ments from the Aden coast, Southern Yemen. IOSR Jour-
nal of Environmental Science, Toxicology and Food
Technology (IOSR-JESTFT), 12(10), 42–55. https://doi.
org/10.9790/2402-1210024255.
Shevchenko, V., Lisitzin, A., Vinogradova, A., & Stein, R.
(2003). Heavy metals in aerosols over the seas of the
Russian Arctic. Science of the Total Environment, 306(1),
11–25. https://doi.org/10.1016/S0048-9697(02)00481-3.
Sitko, R., Zawisza, B., Jurczyk, J., Buhl, F., & Zielonka, U.
(2004). Determination of high Zn and Pb concentrations in
polluted soils using energy-dispersive X-ray fluorescence
spectrometry. Polish Journal of Environmental Studies,
13(1), 91–96.
Smith, S. L., Romanovsky, V. E., Lewkowicz, A. G., Burn, C.
R., Allard, M., Clow, G. D., et al. (2010). Thermal state of
permafrost in North America: A contribution to the inter-
national polar year. Permafrost and Periglacial Processes,
21(2), 117–135. https://doi.org/10.1002/ppp.690.
Survey, G. (1970). Mercury in the environment: A compilation
of papers on the abundance, distribution, and testing of
mercury in rocks, soils, waters, plants, and the atmosphere.
New York: US Government Printing Office.
Tarnocai, C., Canadell, J. G., Schuur, E. A. G., Kuhry, P.,
Mazhitova, G., & Zimov, S. (2009). Soil organic carbon
pools in the northern circumpolar permafrost region. Glo-
bal Biogeochemical Cycles.https://doi.org/10.1029/
2008gb003327.
Taylor, S. R., & McLennan, S. M. (1995). The geochemical
evolution of the continental crust. Reviews of Geophysics,
33(2), 241–265. https://doi.org/10.1029/95RG00262.
Tian, K., Huang, B., Xing, Z., & Hu, W. Y. (2017). Geochemical
baseline establishment and ecological risk evaluation of
heavy metals in greenhouse soils from Dongtai, China.
Ecological Indicators, 72, 510–520. https://doi.org/10.
1016/j.ecolind.2016.08.037.
Tomlinson, D. L., Wilson, J. G., Harris, C. R., & Jeffrey, D. W.
(1980). Problems in the assessment of heavy-metal levels
in estuaries and the formation of a pollution index. Hel-
golander Meeresuntersuchungen, 33(1–4), 566–575.
https://doi.org/10.1007/bf02414780.
Walker, D. A., Leibman, M. O., Epstein, H. E., Forbes, B. C.,
Bhatt, U. S., Raynolds, M. K., et al. (2009). Spatial and
temporal patterns of greenness on the Yamal Peninsula,
Russia: Interactions of ecological and social factors
affecting the Arctic normalized difference vegetation
index. Environmental Research Letters. https://doi.org/10.
1088/1748-9326/4/4/045004.
Wedepohl, K. H. (1967). Thisen, R—Quantitative electron
microprobe analysis. Naturwissenschaften, 54(2), 52.
Wilkinson, M. T., & Humphreys, G. S. (2005). Exploring
pedogenesis via nuclide-based soil production rates and
OSL-based bioturbation rates. Australian Journal of Soil
Research, 43(6), 767–779. https://doi.org/10.1071/
sr04158.
WRB. (2015). World reference base (WRB) for soil resources,
international soil classification system for naming soils and
creating legends for soil maps. Rome; Food and Agricul-
ture Organization of the United Nation (FAO).
Xu, Z., Ni, S., Tuo, X., & Zhang, C. (2008). Calculation of heavy
metals’ toxicity coefficient in the evaluation of potential
ecological risk index. Environmental Science and Tech-
nology, 31(2), 112–115.
Zhulidov, A. V., Robarts, R. D., Pavlov, D. F., Kamari, J.,
Gurtovaya, T. Y., Merilainen, J. J., et al. (2011). Long-term
changes of heavy metal and sulphur concentrations in
ecosystems of the Taymyr Peninsula (Russian Federation)
North of the Norilsk Industrial Complex. Environmental
Monitoring and Assessment, 181(1–4), 539–553. https://
doi.org/10.1007/s10661-010-1848-y.
Zinkute, R., Taraskevicius, R., Jankauskaite, M., & Stankevi-
cius, Z. (2017). Methodological alternatives for calculation
of enrichment factors used for assessment of topsoil con-
tamination. Journal of Soils and Sediments, 17(2),
440–452. https://doi.org/10.1007/s11368-016-1549-4.
Publisher’s Note Springer Nature remains neutral with
regard to jurisdictional claims in published maps and
institutional affiliations.
123
Environ Geochem Health