Content uploaded by Irina P. Panyushkina
Author content
All content in this area was uploaded by Irina P. Panyushkina on Sep 01, 2016
Content may be subject to copyright.
TRENDS IN ELEMENTAL CONCENTRATIONS OF TREE RINGS
FROM THE SIBERIAN ARCTIC
IRINA P. PANYUSHKINA
1
*, VLADIMIR V. SHISHOV
2
, ALEXI M. GRACHEV
3
, ANASTASIA A. KNORRE
4
,
ALEXANDER V. KIRDYANOV
2,4
, STEVEN W. LEAVITT
1
, EUGENE A. VAGANOV
2,4
, EUGENE P. CHEBYKIN
5
,
NATALIA A. ZHUCHENKO
5
,and MALCOLM K. HUGHES
1
1
Laboratory of Tree-Ring Research, University of Arizona, 1215 E. Lowell St., Tucson, AZ 85721, USA
2
Siberian Federal University, 79 Svobodniy Ave., Krasnoyarsk, 660041, Russia
3
Institute of Geography, RAS, 29 Staromonetniy Pereulok, Moscow, 119017, Russia
4
Sukachev Institute of Forest SB RAS, Akademgorodok, Krasnoyarsk, 660036, Russia
5
Limnological Institute SB RAS, 3 Ulan-Batorskaya St., Irkutsk, 664033, Russia
ABSTRACT
The biogeochemistry and ecology of the Arctic environment have been heavily impacted by
anthropogenic pollution and climate change. We used ICP-MS to measure concentrations of 26
elements in the AD 1300–2000 tree rings of larch from the Taymyr Peninsula in northern Siberia for
studying the interaction between environmental change and wood chemistry. We applied a two-stage
data reduction technique to identify trends in the noisy measurement data. Statistical assessment of
variance of normalized time series reveals pronounced depletion of xylem Ca, Mg, Cl, Bi and Si
concentrations and enrichment of P, K, Mn, Rb, Sr and Ba concentrations after ca. AD 1900. The
trends are unprecedented in the 700-year records, but multiple mechanisms may be at work and
difficult to attribute with certainty. The declining xylem Ca and Mg may be a response to soil
acidification from air pollution, whereas increasing P, K, and Mn concentrations may signal changes
in root efficiency and excess water-soluble minerals liberated by the permafrost thaw. The changes
seem consistent with mounting stress on Arctic vegetation. This study supports the potential of tree
rings for monitoring past and ongoing changes in biogeochemistry of Arctic ecosystems related to
pollution and permafrost thaw.
Keywords: wood chemistry, dendrochemistry, Larix decidua, biogeochemical cycling, soil acidifi-
cation, permafrost thawing, Arctic pollution, Norilsk Nickel smelting complex, ICP-MS.
INTRODUCTION
The landscape of the Siberian Arctic has been
roiling in a cauldron of environmental change over
the course of the 20
th
Century. Tree cover has
increased as northern tree limit has expanded
both northward and upward (Kharuk et al. 2006;
Kirdyanov et al. 2012; Golubeva et al. 2013), but
there have been concomitant adverse impacts on
forest health and productivity from pollution load-
ing (Nilsson et al. 1998; Kirdyanov et al. 2014) and
climate change (Lloyd et al. 2011). Furthermore,
permafrost thawing has influenced biogeochemical
cycling with various environmental feedbacks
(Shur and Jorgenson 2007; Yarie and Van Cleve
2010; Keuper et al. 2012). Plant moisture stress
from the deepening of the active melting layer of
permafrost soils has contributed to unusual decline
in forest growth in the high north (Barber et al.
2000; Girardin et al. 2014), with additional stress
from acidification of the environment and soil
(Shortle et al. 1997; Nilsson et al. 1998; Shur and
Jorgenson 2007; Rice and Herman 2011). The
global acceleration of nitrification, anthropogenic
acidification, and deposition of contaminants has
also greatly altered biogeochemistry of Arctic eco-
systems (Galloway et al. 2003; Smith et al. 2011).
Acid compounds induced by air pollution over
the Arctic observed since the 1970s have been
*Corresponding author: ipanyush@email.arizona.edu
TREE-RING RESEARCH, Vol. 72(2), 2016, pp. 67–77
DOI: http://dx.doi.org/10.3959/1536-1098-72.02.67
Copyright © 2016 by The Tree-Ring Society 67
attributed to industrial pollution originating in
mid-latitudes (AMAP Assessment 2006), but pollu-
tion over the Arctic has been present even earlier
(Law and Stohl 2007).
The effect of these environmental changes cas-
cades through plant-soil interactions. Soil leaching
related to nitrification is a well-documented phenom-
enon in Arctic and boreal ecosystems (Van Miegroet
and Cole 1984; Sverdrup et al. 1994). In cold environ-
ments, deposition of acid anions, e.g. SO
4
5
and
NO
3
−
, affects tree uptake of important macronutri-
ents, e.g. Ca, P, Mn andK (Agren et al. 2012). Aci dity
effects extend to leaching and transport of base
cations Ca
2+
,K
+
and Mg
2+
within the soil profile
and their depletion from the exchangeable pool
available for tree uptake (Hogberg et al.2006).Con-
sequently, long-term (decadal) environmental altera-
tions in Siberian boreal forests can influence
exchange of elements between soil and trees.
Given these likely changes in the Siberian Arc-
tic chemical environment, one approach to long ret-
rospective monitoring of biogeochemistry changes
is through elemental analysis of tree rings (e.g.
Hall et al. 1975; Hughes et al. 1980; Baes and
McLaughlin 1984; Guyette et al. 1992; Padilla and
Anderson 2002; Kuang et al. 2008). Concentrations
of elements in tree rings have been applied to
reconstructing soil pH and atmospheric pollution,
and in monitoring sulfur deposition and metal
contamination (Cutter and Guyette 1993; Chen
et al. 2010; Doucet et al. 2012), and sometimes
molar ratios of elements (e.g. Ca/Mg, Mg/Mn or
Ca/Al) have been found to be more diagnostic for
environmental effects than xylem elemental con‐
centrations alone (DeWalle et al. 1999; Kuang
et al. 2008).
Our project endeavored to produce long, highly-
resolved and well-replicated records of concentra-
tions of a broad suite of chemical elements in tree
rings from the Taymyr Peninsula of Russia. These
records not only provide early baseline information
on chemical composition hundreds of years ago,
but also serve as a bellwether for increasing impacts
of the recent human-induced changes of the last
century. This paper statistically analyzes the vari-
ance of the chemical record derived from larch
tree rings extending back 700 years, far exceeding
the length described in previous studies (Padilla
and Anderson 2002), and the results of Principal
Component analysis form the basis for considering
the contributions of various possible environmental
influences.
MATERIALS AND METHODS
Site Settings and Tree Collection
Tree rings of larch (Larix gmelinii Rupr.)
collected on the Kotuy River catchment of the Tay-
myr Peninsula were used for this study (Figure 1).
Larch is an optimal choice for dendrochemistry in
the region, consistent with the common ideals of a
long-lived conifer tree species, wide range of geo-
graphical distribution, distinct heartwood, a small
number of rings in sapwood, and low heartwood
moisture content (Cutter and Guyette 1993). The
site is a 36-kmtransect along the Kotuy River terraces
between ca.70u539N102u559E (350 m a.s.l.) and
70u379N103u239E(160–350 m a.s.l.). Wood cross-
sections from vigorous-looking trees with full foliage
and no evidence of anthropogenic injury or fire scars
were originally collected for a dendroclimatic study
(Naurzbaev and Vaganov 2000). We sub-sampled
cross-sections of 16 trees from the collection of
several hundred specimens gathered at the site.
The study area is located at the northernmost
limit of tree growth in Eurasia and the world. The
landscape encompasses sparse vegetation (less than
30%tree coverage) growing on polygonal tundra
soils with permafrost and thermokarst topography.
The constraints of regional topography and the pre-
vailing surface winds (westerlies) result in open-air
transport into the area from the west, north and
south. The seasonal intensification of the Siberian
High and the Siberian Low as well as the Arctic
front play a key role in the atmospheric circulation
of this region as for the entire Eurasian continent.
The site is ca. 400 km northeast of the edge of an
85-km zone of direct contamination impact from
the Norilsk Nickel smelting complex (Figure 1),
the largest source of heavy metals and sulfate-S pol-
lutants in the world since the 1940s (EPR 2010).
Tree-Ring Sampling
Our sampling selected individual wood speci-
mens with large growth rings to provide sufficient
68 PANYUSHKINA, SHISHOV, GRACHEV, KNORRE, KIRDYANOV, LEAVITT, VAGANOV, CHEBYKIN,
ZHUCHENKO, and HUGHES
material per ring group for analysis. The age of sub-
sampled tree rings was defined with crossdating
against a well-replicated Taymyr tree-ring width
chronology (Naurzbaev and Vaganov 2000). Small
radial blocks ca.1-cmwideby1-cmthickwerecut
from 16 tree cross-sections, which provided replica-
tion of 5–6 trees over the outer 450 years of the
700-yr record and at least 3 trees for the remaining
early years (Supplementary Figure S1A). The rings
were separated into 5-year or 10-year groups with a
mass of dry wood at least 100 mg (Figure S1B).
Because of small mass of ring growth, we mostly iso-
lated 10-year groups, which determined the decadal
resolution of the resulting tree-ring records.
Analytical Analysis
Concentrations of 26 chemical elements were
measured for each sub-sampled ring-group (Figure
S1B). Chemical pre-treatment of wood and analyt-
ical measurements were done at the Limnological
Institute SBRAS (Irkutsk, Russia). The sub-
sampled wood was digested in nitric acid (HNO
3
)
(Sheppard et al. 2008). The concentrations of Li,
B, Na, Mg, Al, Si, P, Cl, K, Ca, Cr, Mn, Fe, Ni,
Cu, Zn, Rb, Sr, Zr, Ag, Cd, Sn, I, Ba, Pb and Bi
were measured on an Agilent 7500 quadrupole
Inductively Coupled Plasma Mass Spectrometer
(ICP-MS). Concentrations of nitrogen and sulfur
were not measured because pretreatment of the
wood samples utilizes nitrogen (HNO
3
), and the
carrier gas (argon) contains small amount of sulfur.
Rigorous measures for quality control and reliable
element detection included: use of stainless-steel
scalpels for ring separation, work in a cleanroom
environment with sterile chemical dishware, regu-
lar calibration checks of ICP-MS operational para-
meters and replication of measurements (Dahlquist
and Knoll 1978; Sheppard et al. 2008). To enhance
the measuring accuracy of trace element detection,
we measured each wood sub-sample 3 to 6 times.
ICP-MS measurements were calibrated with a
multi-element standard solution “2A Standard”
([Ag], [Al], [Ba], [Ca], [Cr], [Cu], [Fe], [K], [Mg],
[Mn], [Na], [Ni], [Rb], [Sr] and [Zn]510.08 ppb).
A standard of Lake Baikal water was used for
Figure 1. Location of the tree-ring site (triangle) in Siberian Arctic and the Norilsk Nickel mining and smelting complex (star). World
map at lower right insert denotes the study region.
Trends in Larch Dendrochemistry from the Taymyr Peninsula, Siberia 69
other elements (Na, Mg, Si, S, Cl, K, Ca) (Suturin
et al. 2008). We present here concentration of ele-
ments in ppb (mg/kg) with respect to the mass of
dry wood sample prior to adding acid. Analytic
uncertainty of metal concentration measurements
(Na, Mg, Al, K, Ca, Mn, Fe, Ni, Cu, Zn, Rb, Sr
and Bi) in excess of 0.1 ppb (or 0.05 ppm referenced
to the mass of dry wood sample) were less 630%,
while analytic uncertainty of other elements (B,
Si, P, Cl, Zr, Ag, Cd, Sn and I) was slightly higher
than 630%. See more details on analytical meth-
ods applied in this study in Grachev et al. (2013).
Statistical Analysis of Time Series
To constrain high variance and differences of
concentration measurements associated with chem-
ical properties and biochemical cycling of studied
elements, we normalized tree-ring time series to
their mean and standard deviation (Z-scores) with
the formula:
Zc¼
c
c
rc
;
where Z
c
5standardized value whose mean 50and
standard deviation 51; c5the measured concen-
tration;
c5mean; σ
c
5standard deviation. Z-scores
were calculated for each series of element mea‐
surements per sampled tree then averaged into a
tree-ring chronology of an element concentration
(Figure S1C). Each tree-ring chronology (henceforth
designated “element time series”) includes 3-5 trees
for any given year from 1300 to 2000 (Figure S1A).
To detect common variance among time series
of measured element concentrations, we applied a
two-stage data reduction process of cluster analysis
followed by principal components analysis (PCA).
These data reduction processes are routinely applied
to data mining of multivariate tree-ring data and cli-
mate series (e.g. LaMarche and Fritts 1971; Shishov
and Vaganov 2010). We believe this is the first appli-
cation of these techniques to dendrochemical series.
First, we ran cluster analysis (Späth 1980) that
agglomerated an optimum number of most-interre-
lated groups for 26 element time series (Figure
S1D). The optimal distance between clusters was
measured with Pearson correlation using Ward’s
method (Ward 1963) that minimizes the sum of
squares of neighboring clusters. Second, we examined
the clustered set of element time series with Principal
Component analysis (Figure S1E). Factor loadings
of principal components (PCs) were calculated
with the Varimax (orthogonal) rotation method to
maximize variance of loadings across a correlated
assemblage of variables (Jolliffe 2002).
RESULTS
Trends in the Variance of Element Time Series
The raw decadal/pentadal concentration mea-
surement series are quite variable across the range
of elements analyzed (Figure S2; Vaganov et al.
2013). For example, some elements (e.g. Ca and
Mg) show a long-term decline, others are generally
low in concentration punctuated by abrupt high
peaks (e.g. Fe and Cr), and yet others increase
towards the end of their records (e.g. K and Rb).
The mean of most measurement time series have
a small standard error indicating a strong coher-
ence among the individual-tree element records.
There was no consistent changein concentration
coincident with the timing of the splices to form the
700-yr series. Some elements (e.g. Ba) showed no
change coincident with the splices, while other ele-
ments did so at AD 1700 (Bi, Cd) and AD 1900
(Rb, Mn). Elements with abrupt changes coincident
with the AD 1700 and 1900 splices did not show a
similar change coincident with the AD 1600 splices.
The diagnostic statistical approach we used
effectively classifies linear and non-linear interac-
tions in the variance of the normalized element
time series. The cluster analysis identified four clas-
ses within which element concentrations are most
interrelated through last 700 years (Table 1). The
number of element time series included in a single
cluster varies from four to ten. The PCA implemen-
ted for the element time series of a single cluster
quantifies a common domain of signals recorded
across the elemental concentrations previously
selected by clusters. The first two clusters (cluster
#1 and #2) captured the variance patterns in con-
centrations of various metals and several biologi-
cally essential trace elements: the Fe-Zr-Cd-Sn-Pb
assemblage and the Li-B-Na-Al-Cr-Ni-Cu-Zn-Ag-I
assemblage, respectively. These assemblages show
no significant changes of element concentrations
70 PANYUSHKINA, SHISHOV, GRACHEV, KNORRE, KIRDYANOV, LEAVITT, VAGANOV, CHEBYKIN,
ZHUCHENKO, and HUGHES
with time. The two other clusters (cluster #3and
#4) captured time-series of macro- and micronutri-
ent elements whose concentrations have been con-
siderably changing after ca.1900.
The two principal components derived from
the element time series assembled in the third
cluster (#3, Si-P-Cl-Bi) share 56%and 26%of
common variance (Table 1). The first principle
component with significant loadings (PC1) pools
Bi, Cl and Si series together and shows an acceler-
ated decline after ca. 1900 (Table 2, Figure 2a).
The second PC (PC2) presents a positive trend after
ca. 1940 expressing the signal mainly associated
with the loading of the phosphorous concentration
variable (Table 2, Figure 2b).
The three PCs of the fourth cluster (#4, Mg-K-
Ca-Mn-Rb-Sr-Ba) explain 45%,27%and 12%of
common variance in the contributing series (Table
1). The PC1 of this cluster links dominant variance
between the Mg and Ca series and shows a negative
trend over the last 200 years (Table 2, Figure 2c). In
contrast, PC2 and the PC3 present positive trends
after ca. 1800 and 1900, respectively, and their factor
loadings are highest for Mn and Rb concentrations
and for K, Sr, and Ba concentrations, respectively
(Table 2, Figure 2d–e).
Significance of Detected Trends
ANOVA modeling applied to the PC vari-
ables for two separate intervals, 1305–1900 and
1900–1995, confirms statistically high significance
of changes in the variance of element concentra-
tions detected by the cluster analysis and PCA. A
few element concentrations even follow the trends
in the raw measurements.
The timing of the concentration rise and
enrichment trends correlates between some of the
elements. A sharp peak in the concentrations of
several PCs emerges during the second half of the
20
th
Century (1950–2000), the last 50 years of the
element time series. Evidence suggests these end-
of-record shifts cannot be explained by higher cat-
ion mobility in sapwood (living outer xylem tissue
where upward water conduction in the tree trunk
takes place), as has been observed in many studies
with shorter element time series (Helmisaari and
Siltala 1989; Smith et al. 2009). For example,
although some elements, e.g. P and K, have a pro-
pensity to accumulate in sapwood through the
function of protoplasts (Smith and Shortle 1994),
Table 2. Factor loadings for PCs of cluster #3 and #4 with detected trend after ca. 1900. Each PC denotes a sign of the fitted trends
shown in Figure 2. Bold font denotes significant correlation coefficients.
Cluster #3
Elements
PC1 PC 2 Cluster #4
Elements
PC 1 PC 2 PC3
Negative Trend Positive Trend Negative Trend Positive Trend Positive Trend
Si 0.32 0.27 Mg 0.41 0.05 −0.10
P−0.16 0.80 K−0.20 0.08 0.62
Cl 0.40 0.07 Ca 0.51 0.07 −0.29
Bi 0.51 −0.39 Mn 0.19 0.55 −0.33
Rb −0.14 0.45 0.15
Sr 0.09 −0.14 0.42
Ba 0.07 −0.22 0.35
Table 1. Results of cluster and principal component (PC)
analysis of 26 element time series for the period 1300–2000.
PC#Eigenvalue
Total
Variance
%
Cumulative
Eigenvalue
Cumulative
Variance, %
Cluster #1: Fe, Zr, Cd, Sn, Pb
11.43 28.54 1.43 28.54
21.14 22.85 2.57 51.39
30.96 19.19 3.53 70.59
40.88 17.57 4.41 88.16
Cluster #2: Li, B, Na, Al, Cr, Ni, Cu, Zn, Ag, I
13.87 38.74 3.87 38.74
21.82 18.22 5.69 56.96
31.39 13.96 7.09 70.92
40.84 8.35 7.93 79.26
Cluster #3: Si, P, Cl, Bi
12.23 55.69 2.23 55.69
21.03 25.65 3.25 81.34
Cluster #4: Mg, K, Ca, Mn, Rb, Sr, Ba
13.17 45.25 3.17 45.25
21.89 27.0 5.06 72.25
30.83 11.89 5.89 84.14
Trends in Larch Dendrochemistry from the Taymyr Peninsula, Siberia 71
sapwood of larch in our study accounts for only
20–25 outer rings so sapwood accumulation cannot
explain post-1900 trends. Furthermore, in our
study the sub-sampled tree-ring specimens contain
two segments of sapwood, i.e. ca. 1875–1900 and
1975–2000 (Figure S1A). For the most part, trends
in raw tree-ring element concentrations in sapwood
of the wood from the end of the 19
th
Century are
Figure 2. Results of principal component analysis of normalized element concentrations from 1300 to 2000 for cluster #3: PC1 (a)
and PC2 (b), and cluster #4: PC1 (c), PC2 (d) and PC3 (e). Table 2 shows the factor loadings of PCs from the cluster #3 and #4.
Elements of major contribution to the PC variance are denoted on the plots. Thick line shows trend estimated with distance-weighted
least squares fitting, and thin dashed lines are 95%-confidence intervals calculated with linear regression.
72 PANYUSHKINA, SHISHOV, GRACHEV, KNORRE, KIRDYANOV, LEAVITT, VAGANOV, CHEBYKIN,
ZHUCHENKO, and HUGHES
different from those in the 20
th
Century sapwood
(Figure S2), so we must reject the prospect that
the elemental trends at the end of the 20
th
Century
are simply an artifact of sapwood processes.
Pearson correlation analysis between monthly
precipitation and temperature from the nearby
Khatanga weather station and the normalized 26–
element tree-ring series for the interval 1936–2000
shows no significant correlations, indicating the
variance of tree-ring element series is not strongly
influenced by weather. The chemical signature of
declining Ca-Mg and Bi-Cl-Si, and rising P-Mn-
Rb-K-Sr-Br in the PCs of the studied tree rings
over 700 years seems therefore most likely a
response to alteration of soil chemistry affecting
the availability of these exchangeable minerals in
soil. Competing environmental factors may be at
play to produce these long-term changes in soil
and larch chemistry.
DISCUSSION
Our project has produced novel and provoca-
tive records of tree-ring chemistry that may be related
to recent and ancient changes in chemical environ-
ment experienced in the Taymyr area, perhaps in
part amplified by climatic warming. Although very
remote, it turns out that multiple conditions and
anthropogenic processes may contribute to the obser-
vations, particularly those of the last century. Even
though our field data set is not sufficient to precisely
identify primary cause (s), this section considers pos-
sible mechanisms influencing the tree-ring elemental
data, which could promote exploration with data
from disparate existing studies and help focus fur-
ther investigations (Vaganov et al. 2013).
Changes in Soil Nutrient Bioavailability and
Tree Uptake
The trends in elemental composition in tree
rings can be influenced by various factors that
affect their concentration and availability in soils.
The overall bulk quantity of elements in soils, con-
tributed from various sources and processes can
influence the element concentrations in tree rings.
Baseline concentrations of many elements avail-
able to plants in soils derive from weathering of
the local bedrock. However, element concentra-
tions can sometimes be greatly enhanced by air-
borne transport of natural and anthropogenic
atmospheric inputs (Nriagu 1989; Law and Stohl
2007). There are more than a few varied forest
locations where evidence of changing soil chemical
composition and soil pH correlated with tree-ring
element concentrations and pollution in the late
20
th
Century (Johnson et al.2008;Chenet
al.2010).
Furthermore, processes affecting availability
and uptake of nutrients by trees can also determine
the trends in elemental composition of tree rings.
The availability of nutrients to trees is a function
of total concentration in soil solution, which will
be influenced by both soil pH and soil mineral
exchangeable pool (Berthrong et al. 2009). The
input of H+during acid deposition can alter the
chemistry of soils by altering weathering, cation
exchange, and mobility and availability of ions in
the soil (Lawrence et al. 1995; DeHayes et al.
1999). Excess acidity may adversely affect plant
nutrition if it contributes to uptake of excess toxic
metals or to deficiency of essential nutrient cations
such as calcium and potassium (Hirschuk 2004;
Lautner and Fromm 2009). Soil acidification (low
soil pH) can decrease exchangeable Ca as acid-
liberated Al cations occupy cation-exchange sites
on clays and displace the atmospheric-origin base
cations (e.g. Ca, K, Mg, Mn) from the soil
exchange complex, contributing to their leaching
out of the soil column (Warby et al. 2009).
Connection to Soil Acidification and Pollution
In the Taymyr region, polygonal tundra soil
generally has low organic matter content, gleying
and acidity (Goryachkin 2010). Soils in wet set-
tings of the Kotuy are near neutral (pH 7.3) on
average but soils from dry microenvironments
are slightly acidic (pH 5.3) (Schmidt 1999). The
study site had dry soil conditions with maximum
thaw depth up to 60 cm (field observation of
authors). The estimated negative trend in the
xylem Ca and Mg concentrations at the site
(cluster #3 PC1) may signal soil acidification
effects on base-cation loss. Similarly, the detected
negative trend in xylem chlorine concentrations
Trends in Larch Dendrochemistry from the Taymyr Peninsula, Siberia 73
(cluster #4 PC1) may suggest low Cl availability
in the soil and possible decrease of soil salinity
caused by excess of water or fluctuations of water
table. Tree uptake and transport of Cl is highly
sensitive to soil salinity and more saline condition
of soil boosts Cl-uptake by roots and vice versa
(Broadley and White 2001). The observed
increase of xylem Mn concentration can indicate
ongoing soil acidification. Studies on reconstruc-
tions of soil pH with tree rings suggest that
xylem concentration of Mn correlates negatively
with soil pH (Kogelmann and Sharpe 2006).
The acidity may be mobilizing Mn that might
otherwise be locked up in mineral oxides. Like
Mn, Ba concentration in trees also has a negative
relationship with soil pH (Broadley and White
2001) and has been used to monitor sulfur
transport and deposition into soils (Cutter and
Guyette 1993).
The largest local source of sulfur deposition,
the Norilsk smelting complex, is about 500 km
west of the tree-ring site. The area of larch dieback
impacted by the Norilsk pollution has been
increasing for the last 60 years and presently
reaches ca. 85 km from Norilsk (Ivshin and Shiya-
tov 1996; Voronin and Ziganshin 1999; Kirdyanov
et al. 2014). There is no direct evidence of detri-
mental effects of the Norilsk pollution on the Tay-
myr forest (Golubeva et al. 2013) although
modeled trajectories indicate active transport of
airborne sulfur to the site at times (Figure S3). His-
torically, the smelting complex has been operating
since the 1940s, and the metal production tripled
between the 1960s and 1980s. In 1983 the emis-
sions of sulfur dioxide and mineral dust from the
Norilsk metal production reached a maximum of
2483 thousand tons and 73.7 thousand tons,
respectively (EPR 2010). More recently, air pollu-
tion in 1990–1999 remained high, as the Norilsk
Nickel production released an average of 2066.5
thousand tons of sulfur dioxide and 25.5 thousand
tons of particles into the air each year (EPR 2010).
Our tree-ring observations of post-1950 chemical
concentrations at the study site may signal soil pH
changes, perhaps linked to consequences of both
local and global sources of Arctic acidification after
the 1940s.
Possible Connection to Permafrost Thaw
The trend of increased xylem K, P, Rb and Sr
concentrations in the tree-ring records over the last
century is puzzling. The K
+
activity in soils with
low pH should decline (as occurs for Ca
2+
and
Mg
2+
) when presence of H
+
and Al
3+
cations
increases and the base ions are leached away. Nor
should high P be expected in most Arctic soils
with low fertility because of low mobility and avail-
ability of this element to plants (Burton et al. 2001).
It therefore seems likely that the interrelated con-
centrations of these elements are driven by a mech-
anism distinct from soil acidification. Acquisition
of phosphorous by trees mainly occurs through dif-
fusion (Barber 1984). We hypothesize that changes
in larch root morphology and absorption efficiency
would be the primary mechanisms for increased
uptake. Root-induced increase of soluble P, K
and Rb in alkaline or slightly acid soils has been
reported as a function of both decreasing soil pH
(acidification) and warming air temperature (Bur-
ton et al. 2001; Hinsinger 2001; Ruess et al. 2006).
Considering the fact that climate of the Taymyr
Peninsula and soil temperature have been warming
steadily since 1880 and 1920, respectively (Figure
S4; Naurzbaev and Vaganov 2000; Park et al.
2014), several factors may contribute to the root-
induced mineral modification, including ongoing
thaw of permafrost soils and deepening of its active
layer (Fedotov et al. 2012).
Permafrost thawing contributes to release of
old carbon accumulated over thousands of years
and changes the net carbon exchange in the Arctic
and boreal forest (Hobbie et al. 2002; Schuur et
al. 2009). Mineralization of soluble elements dur-
ing the decomposition of organic matter may
alter their concentration in the soil solution and
on adsorption sites. Even so, the rate of release
of base cations from the permafrost thaw is con-
troversial (Keuper et al. 2012), and the thawing
and associated nutrient release are coupled with
tree nutrient uptake (Hobbie et al. 2002;
Herzschuh et al. 2013). Furthermore, many stud-
ies report significant impact of both drought and
moisture excess on the soil exchangeable cation
pool throughout the soil profile in the Arctic
caused by climate warming (Kreuzwieser and
Gessler 2010).
74 PANYUSHKINA, SHISHOV, GRACHEV, KNORRE, KIRDYANOV, LEAVITT, VAGANOV, CHEBYKIN,
ZHUCHENKO, and HUGHES
CONCLUSIONS
The 700-year tree-ring records of elemental
concentrations from the Taymyr Peninsula indicate
dramatic changes in tree-ring concentrations of P,
Ca, Mg, Bi, Cl, Si, Mn, K, Rb, Sr and Ba in larch
after ca. 1900. However, many possible influences
on soil and plant chemistry (e.g. local pollution
and regional pollution inputs, acidity, permafrost
thawing) operate in the Taymyr area and the
record of elemental concentrations in the tree rings
is not sufficient by itself to identify the cause(s) with
certainty.
The negative trend in tree uptake of soluble
Ca
2+
-Mg
2+
and Bi
3+
-Cl
−
-Si
4+
, and the positive
trends in soluble P, Mn
2+
-Rb
+
and K
+
-Sr
2+
-Ba
2+
suggest ecologically important changes in biogeo-
chemical cycling of major nutrients available to
the trees, which could be identified through the his-
torical patterns of elemental composition of tree
rings. The signal of low Ca-Mg-Cl and high P-
Mn-K exposes a possible diagnostic relationship
between concentrations of these elements, and sup-
ports the potential of tree rings in addressing how
the concentrations of nutrient elements in soils
have changed with time.
We propose that our tree-ring records of
wood chemistry relate to changes in soil chemis-
try driven by both climate changes and Arctic
pollution. Although no soil chemistry measure-
ments were available at the site, the long-term
trends in our data may be a consequence of pH
fluctuations of permafrost soil with increasing
soil acidity at decadal resolution in this region
and impact of permafrost thawing water on soil.
The geographical scale of the estimated long-
term changes in wood chemistry of larch needs
further investigation.
ACKNOWLEDGMENTS
This study was supported by the U.S. Civilian
Research & Development Foundation (CRDF
grant RUG1-2950-KR-09) and the Russian Science
Foundation (RSF #14-14-00219, development of
statistical approach). We appreciate the helpful
comments of numerous reviewers and we especially
thank K. Smith for his diligent and insightful edito-
rial contributions.
REFERENCES CITED
AMAP Assessment, 2006. Acidifying Pollutants, Arctic Haze,
and Acidification in the Arctic. Arctic Monitoring and Assess-
ment Program Report, Oslo, Norway.
Agren, G. I., J. A. M. Wetterstedt, and M. F. K. Billberger,
2012. Nutrient limitation on terrestrial plant growth –Model-
ing the interaction between nitrogen and phosphorus. New
Phytology 194(4):953–960.
Baes III, C. F., and S. B. McLaughlin, 1984. Trace elements in
tree rings: Evidence of recent and historical air pollution.
Science 224:494–497.
Barber, S. A., 1984. Soil Nutrient Bioavailability: A Mechanistic
Approach. John Wiley & Sons, New York.
Barber, V. A., G. P. Juday, and B. P. Finney, 2000. Reduced
growth of Alaskan white spruce in the twentieth century
from temperature-induced drought stress. Nature 406:
668–673.
Berthrong, S. T., E. G. Jobbágy, and R. B. Jackson, 2009. A
global meta-analysis of soil exchangeable cations, pH, carbon,
and nitrogen with afforestation. Ecological Applications 19
(8):2228–2241.
Broadley, M. R., and P. J. White, 2001. Chloride in soils and its
uptake and movement within the plant: A review. Annals of
Botany 88:967–988.
Burton, A. J., K. S. Pregitzer, R. W. Ruess, R. L. Hendrick, and
M. F. Allen, 2001. Fine root respiration rates in North Amer-
ican forests: Effects of nitrogen concentration and temperature
across biomes. Oecologia 131:559–568.
Chen, L., F. H. Wu, T.-W. Liu, J. Chen, Z.-J. Li, Z.-M. Pei, and
H.-L. Zheng, 2010. Soil acidity reconstruction based on tree
ring information of a dominant species Abies fabri in the sub-
alpine forest ecosystems in southwest China. Environmental
Pollution 158:3219–3224.
Cutter, B. E., and R. P. Guyette, 1993. Anatomical, chemical,
and ecological factors affecting tree species choice in dendro-
chemistry studies. Environmental Quality 22:611–619.
Dahlquist, R. L., and J. W. Knoll, 1978. Inductively coupled
plasma-atomic emission spectrometry: Analyses of biological
materials and soils for trace and ultra-trace elements. Applied
Spectroscopy 32:1–30.
DeHayes, D. H., P. G. Schaberg, G. J. Hawley, and G. R. Strim-
beck, 1999. Acid rain impacts calcium nutrition and forest
health. BioScience 49:789–800.
DeWalle, D. R., J. S. Tepp, B. R. Swistock, W. E. Sharpe, and
P. J. Edwards, 1999. Tree-ring cation response to experimental
watershed acidification in West Virginia and Maine. Journal
of Environmental Quality 28:229–309.
Doucet, A., M. M. Savard, C. Bégin, J. Marion, A. Smirnoff,
and T. B. M. J. Ouarda, 2012. Combining tree-ring metal con-
centrations and lead, carbon and oxygen isotopes to recon-
struct peri-urban atmospheric pollution. Tellus-B 64:19005,
http://dx.doi.org/10.3402/tellusb.v64i0.19005
EPR, 2010. Environmental Protection Report for the Region of
Krasnoyarsk Krai, ca. 2009. Krasnoyarsk, Russia.
Fedotov, A. P., M. A. Phedorin, I. V. Enushchenko, K. E.
Vershinin, M. S. Melgunov, and T. V. Khodzher, 2012. A
reconstruction of the thawing of the permafrost during the
Trends in Larch Dendrochemistry from the Taymyr Peninsula, Siberia 75
last 170 years on the Taimyr Peninsula (East Siberia, Russia).
Global and Planetary Change 98–99:139–152.
Galloway, J. N., J. D. Aber, J. W. Erisman, S. P. Seitzinger,
R. W. Howarth, E. B. Cowling, and B. J. Cosby, 2003. The
nitrogen cascade. BioScience 53(4):341–356.
Girardin, M. P., X. J. Guo, R. De Jong, C. Kinnard, P. Bernier,
and F. Raulier, 2014. Unusual forest growth decline in boreal
North America covaries with the retreat of Arctic sea ice.
Global Change Biology 20:851–866.
Golubeva, E., A. Hofgaard, and K. Silenchuk, 2013. The mor-
phometric structure of the Larix gmellinii recruitment at the
northern limit of its range in the forest-tundra ecotone.
Geography, Environment, Sustainability 3(6):86–93.
Goryachkin, S. V., 2010. Soil Cover of the North: Patterns, Genesis,
Ecology and Evolution. GEOS Publ., Moscow, Russia.
Grachev, A. M., E. A. Vaganov, S. W. Leavitt, I. P. Panyush-
kina, E. P. Chebykin, V. V. Shishov, N. A. Zhuchenko, A.
A. Knorre, M. K. Hughes, and M. M. Naurzbaev, 2013.
Methodology for development of a 600-year tree-ring multi-
element record for larch from the Taymyr Peninsula, Russia.
Journal of Siberian Federal University-Biology 5(1):61–72.
Guyette, R. P., G. S. Henderson, and B. E. Cutter, 1992. Recon-
structing soil pH from manganese concentrations in tree-rings.
Forest Science 38:727–737.
Hall, C., M. K. Hughes, N. W. Lepp, and G. J. Dollard, 1975.
Cycling of heavy metals in woodland ecosystems. In Proceed-
ings of International Conference on Heavy Metals in the Envi-
ronment, pp. 227–246. Institute for Environmental Studies,
University of Toronto, Canada.
Helmisaari, H.-S., and T. Siltala, 1989. Variation in nutrient
concentrations of Pinus sylvsetris stems. Scandinavian Journal
of Forest Research 4:443–451.
Herzschuh, U., L. A. Pestryakova, L. A. Savelieva, L. Heinecke,
T. Böhmer, B. K. Biskaborn, A. Andreev, A. Ramisch, A. L.
C. Shinneman, and H. J. B. Birks, 2013. Siberian larch forests
and the ion content of thaw lakes form a geochemically func-
tional entity. Nature 4(2408), http://dx.doi.org/10.1038/
ncomms3408.
Hinsinger, P., 2001. Bioavailability of soil inorganic P in the rhi-
zosphere as affected by root-induced chemical changes: A
review. Plant and Soil 237:173–195.
Hirschuk, K. D., 2004. The calcium conundrum. Both versatile
nutrient and specific signal. Plant Physiology 136:2438–2442.
Hobbie, S. E., K. J. Nadelhoffer, and P. Högberg, 2002. A syn-
thesis: The role of nutrients as constraints on carbon balances
in boreal and arctic regions. Plant and Soil 242(1):163–170.
Hogberg, P., H. Fan, M. Quist, D. Binkley, and C. O. Tamm,
2006. Tree growth and soil acidification in response to
30 years of experimental nitrogen loading on boreal forest.
Global Change Biology 12(3):489–499.
Hughes, M. K., N. W. Lepp, and D. A. Phipps, 1980. Aerial
heavy metal pollution and terrestrial ecosystems. Advances in
Ecological Research 11:218–327.
Iijima, Y., T. Ohta, A. Kotani, A. N. Fedorov, Y. Kodama, and
T. C. Maximov, 2013. Sap flow changes in relation to perma-
frost degradation under increasing precipitation in an eastern
Siberian larch forest. Ecohydrology, http://dx.doi.org/10.
1002/eco.1366.
Ivshin, A. P., and V. S. Shiyatov, 1996. The assessment of sub-
tundra forests degradation by dendrochronological methods
in the Norilsk industrial area. Dendrochronologia 13:113–126.
Johnson, D. W., D. E. Todd, C. F. Trettin, and P. J. Mulhol-
land, 2008. Decadal changes in potassium, calcium, and mag-
nesium in a deciduous forest soil. Soil Science Society of
America Journal 72:1795–1805.
Jolliffe, I. T., 2002. Principal Component Analysis, Springer
Series in Statistics 2nd ed. Springer, New York.
Keuper, F., P. M. van Bodegom, and E. Dorrepaal, 2012. A frozen
feast: Thawing permafrost increases plant-available nitrogen in
subarctic peatlands. Global Change Biology 18(6):1998–2007.
Kharuk, V. I., K. J. Ranson, S. N. Im, and M. M. Naurzbaev,
2006. Forest–tundra larch forests and climatic trends. Russian
Journal of Ecology 37:291–298.
Kirdyanov, A. V., F. Hagerdorn, A. Knorre, E. V. Fedotova, E.
A. Vaganov, M. M. Naurzbaev, P. A. Moiseev, and A.
Rigling, 2012. 20
th
century tree-line advance and vegetation
changes along an altitudinal transect in the Putorana Moun-
tains, northern Siberia. Boreas 41:56–67.
Kirdyanov, A. V., V. S. Myglan, A. V. Pimenov, A. A. Knorre,
A. K. Ekart, and E. A. Vaganov, 2014. The die-off dynamics
of Siberian larch under the impact of pollutants emitted by the
Norilsk enterprises. Contemporary Problems of Ecology 7(6):
679–684.
Kogelmann, W. J., and W. E. Sharpe, 2006. Soil acidity and man-
ganese in declining and nondeclining sugar maple stands in
Pennsylvania. Journal of Environmental Quality 35:433–441.
Kreuzwieser, J., and A. Gessler, 2010. Global climate change
and tree nutrition: Influence of water availability. Tree Physi-
ology 30(9):1221–1234.
Kuang, Y. W., D. Z. Wen, G. Y. Zhou, G. W. Chu, F. F. Sun,
and J. Li, 2008. Reconstruction of soil pH by dendrochemistry
of Masson pine at two forested sites in the Pearl River Delta,
South China. Annals of Forestry Science 65(804):685–696.
LaMarche Jr., V. C., and H. C. Fritts, 1971. Anomaly patterns
of climate over the western United States, 1700–1930, derived
from principal components analysis of tree-ring data. Monthly
Weather Review 99(2):138–142.
Lautner, S., and J. Fromm, 2009. Calcium-dependent physiolog-
ical processes in trees. Plant Biology 12:268–274.
Law, K. S., and A. Stohl, 2007. Arctic air pollution: Origins and
impacts. Science 315(5818):1537–1540.
Lawrence, G. B., M. B. David, and W. C. Shortle, 1995. A new
mechanism for calcium loss in forest-floor soils. Nature
378:162–164.
Lloyd, A. H., Bunn, A. G., and Berner, L., 2011. A latitudinal
gradient in tree growth response to climate warming in the
Siberian taiga. Global Change Biology 17:1935–1945.
Naurzbaev, M. M., and E. A. Vaganov, 2000. Variation of early
summer and annual temperature in east Taymir and Putoran
(Siberia) over the last two millennia inferred from tree rings.
Journal of Geophysical Research, 105(D6):7317–7326.
Nilsson, S., K. Blauberg, E. A. Samarskaya, and V. I. Kharuk,
1998. Pollution stress of Siberian forests. In Air Pollution in
the Ural Mountains, edited by I. Linkov, and R. Wilson,
pp. 31–54. Kluwer Academic & Plenum Publishers, Dor-
drecht, The Netherlands.
76 PANYUSHKINA, SHISHOV, GRACHEV, KNORRE, KIRDYANOV, LEAVITT, VAGANOV, CHEBYKIN,
ZHUCHENKO, and HUGHES
Nriagu, J. O., 1989. A global assessment of natural sources of
atmospheric trace metals. Nature 338:47–49.
Padilla, K. L., and K. A. Anderson, 2002. Trace element concen-
tration in tree-rings biomonitoring centuries of environmental
change. Chemosphere 49:575–585.
Park, H., A. B. Sherstiukov, A. N. Fedorov, I. V. Polyakov, and J.
E. Walsh, 2014. An observation-based assessment of the influ-
ences of air temperature and snow depth on soil temperature
in Russia. Environmental Research Letters 9(6):064026 (7 pp.).
Rice, K. C., and J. S. Herman, 2011. Acidification of Earth: An
assessment across mechanisms and scales. Applied Geochemis-
try 27:1–14.
Ruess, R. W., R. Hendrick, J. G. Vogel, and B. Sveinbjornsson,
2006. The role of fine roots in the functioning of Alaskan bore-
al Forests. In Alaska’s Changing Boreal Forest, edited by F. S.
Chapin III, M. Oswood, K. van Cleve, L. A. Viereck, and D.
L. Verbyla, pp. 189–201. Oxford University Press, New York.
Schmidt, N., 1999. Microbial Properties and Habitats of Perma-
frost Soils on Taimyr Peninsula, Central Siberia. Ph.D. disser-
tation, Christian-Albrechts University of Kiel, Germany.
Schuur, E. A. G., J. G. Vogel, K. G. Crummer, H. Lee, J. O.
Sickman, and T. I. Osterkamp, 2009. The effect of permafrost
thaw on old carbon release and net carbon exchange from tun-
dra. Nature 459:556–559.
Sheppard, P. R., M. H. Ort, K. C. Anderson, M. D. Elson, L.
Vazquez-Selem, A. W. Clemens, N. C. Little, and R. J. Speak-
man, 2008. Multiple dendrochronological signals indicate the
eruption of Paricutin volcano, Michoacan, Mexico. Tree-
Ring Research 64(2):97–108.
Shishov, V. V., and E. A. Vaganov, 2010. Dendroclimatological
evidence of climate changes across Siberia. In Environmental
Change in Siberia: Earth Observation, Field Studies and Model-
ing, edited by H. Balzter, pp. 101–115. Springer, London, UK.
Shortle, W. C., K. T. Smith, R. Minocha, G. B. Lawrence, and
M. B. David, 1997. Acidic deposition, cation mobilization,
and biochemical indicators of stress in healthy red spruce.
Journal of Environmental Quality 26:871–876.
Shur, Y. L., and M. T. Jorgenson, 2007. Patterns of permafrost
formation and degradation in relation to climate and ecosys-
tems. Permafrost and Periglacial Processes 18:7–19.
Smith, K. T., and W. C. Shortle, 1994. Tree biology and dendro-
chemistry. In Tree Rings, Environment and Humanity, edited
by J. S. Dean, D. M. Meko, and T. W. Swetnam, pp. 629–
635. University of Arizona Press, Tucson, USA.
Smith, K. T., W. C. Shortle, J. H. Conolly, R. Minocha, and J.
Jellyson, 2009. Calcium fertilization increases the concentration
of calcium in sapwood and calcium oxalate in foliage of red
spruce. Environmental and Experimental Botany 67:277–283.
Smith, S. J., J. Van Aardenne, Z. Klimont, R. J. Andres, A.
Volke, and S. Delgado Arias, 2011. Anthropogenic sulfur
dioxide emissions: 1850–2005. Atmospheric Chemistry and
Physics 11:1101–1116.
Späth, H. , 1980. ClusterAnalysis Algorithms for DataReduction and
Classification of Objects. Horwood Press, Chichester, England.
Suturin, A. N., L. F. Paradina, V. N. Epov, A. R. Semenov, V. I.
Lozhkin, and L. L. Petrov, 2003. Preparation and assessment
of a candidate reference sample of Lake Baikal deep water.
Spectrochimica Acta 58:277–288.
Sverdrup, H., P. Warfvinge, and B. Nihlgård, 1994. Assessment
of soil acidification effects on forest growth in Sweden. Water,
Air, and Soil Pollution 78(1–2):1–36.
Vaganov,E. A., A. M. Grachev,V. V. Shishov,I. P. Panyushkina, S.
W. Leavitt, A. A. Knorre, E. P. Chebykin, and O. V. Menyailo,
2013. Elemental composition of tree rings: A new perspective in
Biogeochemistry. Doklady Biologic al Sciences 453:375–379.
Van Miegroet, H., and D. W. Cole, 1984. The impact of nitrifi-
cation on soil acidification and cation leaching in a red alder
ecosystem. Journal of Environmental Quality 13:586–590.
Voronin, V. I., and R. A. Ziganshin, 1999. Dendrochronlogical
and geochemical studies in the drying forests affected by aerial
emissions from the Norilsk metallurgical factory in order to
interpret large-scale examination of toxic damage to stand of
trees. Letopis’Prirody Taymyrskogo Zapovednika 15:1–56.
Warby, R. A. F., C. E. Johnson, and C. T. Driscoll, 2009. Con-
tinuing acidification of organic soils across the Northeastern
USA: 1984–2001. Soil Science Society of America Journal
73:274–284.
Ward, J. H., 1963. Hierarchical grouping to optimize an objec-
tive function. Journal of the American Statistical Association
58:236–244.
Yarie, J., and K. Van Cleve, 2010. Long-term monitoring
of climatic and nutritional effects on tree growth in interi-
or Alaska. Canadian Journal of Forest Research 40:
1325–1335.
Zhulidov, A. V., R. D. Robarts, D. F. Pavlov, J. Kämäri, T. Y.
Gurtovaya, J. J. Meriläinen, and I. N. Pospelov, 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 Mon-
itoring and Assessment 181(1–4):539–553.
Received 4 March 2014; accepted 28 May 2016.
Supplementary Material is available at http://
www.treeringsociety.org/TRBTRR/TRBTRR.htm
Trends in Larch Dendrochemistry from the Taymyr Peninsula, Siberia 77
