Assessing the geochemical fingerprint of the 2010 Eyjafjallajökull tephra through instrumental neutron activation analysis: a trace element approach

Abstract

The first characterization by INAA of the tephra erupted during the most explosive phase of the 2010 Eyjafjallajökull (Iceland) eruption is here presented. To evaluate the homogeneity of the tephra and fragmentation processes not only bulk samples were considered, but also other grain size fractions and previously published data. Concentrations of 42 elements were determined. Specific attention was given to incompatible elements, which appeared to be the most significative in order to define a geochemical fingerprint of the event.

Full text

Assessing the geochemical fingerprint of the 2010 Eyjafjallajo
¨kull
tephra through instrumental neutron activation analysis: a trace
element approach
Giovanni Baccolo
1,2,3
•Massimiliano Clemenza
3,4
•Barbara Delmonte
2
•
Niccolo
`Maffezzoli
3,4,5
•Massimiliano Nastasi
3,4
•Ezio Previtali
3,4
•
Valter Maggi
2,3
Received: 18 December 2014
ÓAkade
´miai Kiado
´, Budapest, Hungary 2015
Abstract The first characterization by INAA of the
tephra erupted during the most explosive phase of the 2010
Eyjafjallajo
¨kull (Iceland) eruption is here presented. To
evaluate the homogeneity of the tephra and fragmentation
processes not only bulk samples were considered, but also
other grain size fractions and previously published data.
Concentrations of 42 elements were determined. Specific
attention was given to incompatible elements, which ap-
peared to be the most significative in order to define a
geochemical fingerprint of the event.
Keywords Eyjafjallajo
¨kull Instrumental neutron
activation analysis Tephra Incompatible elements
Introduction
The 39 day long volcanic eruption of Eyjafjallajo
¨kull oc-
curred in 2010 (April 14th–May 22nd, Southern Iceland)
had a significative impact on European air traffic [1,2].
The ash plumes generated during the explosive phases of
the event caused a main disruption of air traffic and a
consequent economical damage [3]. The production of fine
tephra, enhanced by the interaction between a high vis-
cosity magma and the melted water released by the over-
lying glacier, was remarkable [4]. The wide dispersion of
the material across Europe and the North Atlantic was
triggered by the peculiar atmospheric patterns occurring in
that period [5,6]. The development, diffusion and depo-
sition of the tephra were observed and monitored at dif-
ferent sites, using ground-based, airborne and remote
sensing techniques [7,8]. The eruption attracted an ex-
tensive scientific interest, which resulted in volcanological
[9–11], atmospheric [2,6,8,12], aerodynamical [13–15],
environmental [16,17] and geochemical studies [18–20].
Here we present an accurate geochemical characteriza-
tion of the tephra produced during the 2010 Eyjafjallajo
¨kull
eruption based on its chemical composition and especially
on trace elements. The attention given to this event en-
couraged several studies to investigate the elemental
composition of the eruption products using different ana-
lytical techniques [13,18,19,21,22]. A comparison be-
tween these measurements and our data is important not
only to assess the homogeneity of the samples with respect
to collection sites and timings, but also to understand if
there are differences between the considered analytical
techniques.
Tephra composition will also be compared to some well
established geochemical references: middle ocean ridge
basalt (MORB, [23–25]), oceanic island basalt (OIB, [25])
and upper continental crust (UCC, [26]). MORB and OIB
are indicative of the average composition of the oceanic
volcanism, occurring along the oceanic ridges (MORB) or
associated to hot mantle plumes (OIB) [27]. Specific at-
tention will be paid to some elemental groups which are
classically used in order to develop geochemical
&Giovanni Baccolo
giovanni.baccolo@mib.infn.it
1
Graduate School in Polar Sciences, University of Siena,
Via Laterina 8, 53100 Siena, Italy
2
Department of Environmental Sciences, University of
Milano-Bicocca, P.zza della Scienza 1, 20126 Milan, Italy
3
INFN, Section of Milano-Bicocca, P.zza della Scienza 3,
20126 Milan, Italy
4
Department of Physics, University of Milano-Bicocca,
P.zza della Scienza 3, 20126 Milan, Italy
5
Present Address: Centre for Ice and Climate, Juliane Maries
Vej 30, 2100 Copenhagen, Denmark
123
J Radioanal Nucl Chem
DOI 10.1007/s10967-015-4092-7

fingerprints and to understand the genetic processes in-
volved in the formation of such samples. They are: large
ion lithophile elements (LILE), high field strength elements
(HFSE) and rare earth elements (REE). These elements,
also known as incompatible elements, share high ionic radii
and charges. Resulting from these ionic properties the term
incompatible refers to the difficulty to enter the crystalline
reticules which are typically found in the earth mantle,
dominated by the presence of mafic iron minerals [28].
Thus when mantle rock melting occurs incompatible ele-
ments are easily fractionated, since they are removed from
the solid phase and are concentrated in the melt phase, i.e.
the magma [27].
Assessing an accurate geochemical characterization of
the Eyjafjallajo
¨kull tephra is not only important from a
merely geochemical and volcanological perspective, but
also for chronostratigraphic purposes. The wide dispersion
of the tephra over Europe and the accurate knowledge of
the ash properties and distribution make this event a can-
didate to define a new reference event for ice core se-
quences in Alpine and Greenlandic glaciers, where the ash
plume arrived, although its concentration in the snow-ice is
expected to be very low [7,29,30]. The identification and
source attribution of volcanic ash layers provide important
chronological constraints for refining ice dating and
establishing cross-correlations among climatic and pa-
leoclimatic sequences as ice cores [31]. Within this context
the 2010 Eyjafjallajo
¨kull eruption could offer an opportu-
nity to define a new temporal marker valid for the twenty
first century. Usually the most common tracer for volcanic
events used in ice core studies is represented by peaks in
sulphate concentration or visual recognition. The sulphate
approach is problematic in areas like the Alpine one, sur-
rounded by heavy industrialized regions which emit large
quantities of sulphates in the atmosphere [32]. Also visual
identification of tephra layers in ice cores is difficult when
particle concentration is reduced. The definition of a geo-
chemical fingerprint of the event will be useful to identify
the tephra layer using chemical methods and discriminating
it from common mineral dust deposition.
Materials and method
Samples
A sample of fine tephra produced during the first explosive
phase of the Eyjafjallajo
¨kull eruption (April 15th) was
collected a few kilometers away from the eruptive site. It
consists of incoherent material; no grains exceeding 1 mm
were present. After having been dried the sample was
stored in a closed vessel until it was analyzed. The original
material was divided into three parts using a sonic dry sifter
system (from Advantech Ó) which allows to separate dif-
ferent grain size fractions. Two grain size fractions were
extracted from the original ash sample (bulk). The first one
is represented by particles smaller than 63 lm, corre-
sponding to the size class which defines fine ash. The
second one is represented by particles smaller than 20 lm;
this size was selected because generally this is the fraction
transported at long range distances. Three aliquots (about
100 mg, dry mass) of each grain size class were prepared in
1 mL polyethylene vials for the successive analyses.
Neutron activation analysis
The elemental composition of the tephra was determined
by instrumental neutron activation analysis (INAA). Its
application to geological samples and specifically to tephra
bulk samples is well established [33–36]. Samples were
irradiated at Laboratory of Applied Nuclear Energy
(LENA, Pavia, Italy), where a 250 kW Triga Mark II nu-
clear reactor is installed [37,38]. To maximize the number
of observable radionuclides three irradiations and four data
acquisitions were carried out (see Table 1). Gamma-ray
energy spectra were acquired using hyper-pure germanium
detectors, HPGe (at the Radioactivity Laboratory of Mi-
lano-Bicocca University): a coaxial p-type ORTEC HPGe
and a well-type ORTEC HPGe (GWL series). Six reference
materials (RM, NIST 1645, 2704, 2709 and 2710; USGS
BCR2 and AGV2) were irradiated with the samples for
calibration purposes. For a detailed description of the
procedure and of the detectors see [39]. Concentration
uncertainties were calculated taking into account the
uncertainties associated to the RM certified concentrations,
to balance weighing operations and to the statistical ana-
lysis of c-spectra, including dead time correction. For the
short irradiation also irradiation and cooling times were
considered; in the case of Mg and Al final uncertainty kept
into account also the occurrence of interference reactions
from fast neutrons. The uncertainty related to the temporal
fluctuations of the neutron flux during the short irradiation
was evaluated comparing the activation rates (counts
s
-1
lg
-1
) associated to the different RM, each one mea-
sured at a different time (Fig. 1).
Results and discussion
Concentrations and comparison with published data
The concentrations of 42 elements were determined, in-
cluding trace elements and all the major elements except
for Si (see Table 2). Concentrations span from 8.25 % m/m
for Fe, to 5 ppb for Au. For 23 elements the relative
uncertainty is lower than 10 %, for 16 elements (Mg, Al,
J Radioanal Nucl Chem
123

K, Ca, Ti, As, Mo, Sb, Ba, Eu, Dy, Ho Tm, Lu, Hf, Th) it
lies in the 10–20 % range and for three elements it exceeds
20 % (Br 30 %, W 37 %, Au 27 %). The uncertainties
fluctuations are explained by the different signal to back-
ground ratios. The signal is not only proportional to the
absolute abundance of the considered element within the
sample, it depends also on the isotopic abundance of the
activated nuclide, on its neutron capture cross section and
decay branching ratios and on detector efficiency for the
measured c-emission. For these reasons some trace ele-
ments (for example Sc, Co, Cs, see Table 2) show lower
relative uncertainties with respect to other major elements
(for example Ca, Ti, K, see Table 2).
Thanks to the attention paid to this event by the scien-
tific community it was possible to compare the composition
of the tephra obtained in our study by INAA to other ones
[13,18,19,21,22]. Different techniques were used in these
works: ICP-OES, ICP-MS, EDS and wet analysis. All the
data considered for the comparison refer to bulk samples,
with the exception of trace element concentrations from
[18] which are relative to single grain measurements.
Measurements are also comparable with respect to time,
since all refer to the tephra produced during the first ex-
plosive phase of Eyjafjallajo
¨kull eruption (15–19 April),
which was the most significative for fine tephra production
and dispersion in the atmosphere. The concentration of
some selected elements determined in the different studies
are reported in Fig. 2. For most of the elements (34) the
observed differences considering all the studies are below
20 %, relative to our values taken as 100 %. Elements
showing the highest differences are V (31 %), Cr (25 %),
Co (30 %), Ni (26 %) and Sb (110 %). In the case of Cr
and Co, the high deviations could be related to fragmen-
tation process involving these elements, also related to
Table 1 Main information about the irradiation of the samples and the acquisition of the c-spectra; neutron fluxes were taken from [38]
Irradiation Average
sample
mass
(mg)
Neutron flux
(n s
-1
cm
-2
)
Irradiation
time
Number of
acquisitions
Cooling
time
Acquisition
time (s)
Observed elements
Short
irradiation
30 (7.40 ±0.95) 910
12
60 s 1 200–400 s 300 Na, Mg, Al, Ca, Ti, V, Mn
Medium
irradiation
30 (2.40 ±0.24) 910
12
6 h 1 5 days 2500 Na, K, Ca, Sc, Cr, Fe, Co, As, Br, Rb,
Mo, Sb, Cs, Ba, La, Ce, Nd, Sm, Tb,
Ho, Yb, Lu, Hf, W, Au, Th, U
Long
irradiation
80 (2.40 ±0.24) 910
12
20 h 2 40–50 days 15,000 Sc, Cr, Fe, Co, Zn, Rb, Sr, Zr, Sb, Cs,
Ba, Ce, Gd, Eu, Tb, Tm, Yb, Hf, Ta,
Hg, Th
45–70 days 70,000 Ni, Rb, Sr, Zr, Cs, Nd, Gd, Dy, Tm,
Hg
Fig. 1 RM activation rates (counts s
-1
lg
-1
) of the radionuclides
observed within the short irradiation not affected by interferences.
Average value (solid line) and standard deviation (dotted lines) are
also reported. Each standard was measured at different times. The
considered c-energies are 1368.3 keV (Na), 3084.4 keV (Ca),
320.1 keV (Ti), 1434.1 (V), 1810.8 (Mn). 1corresponds to NIST
1645, 2to NIST 2704, 3to NIST 2709, 4and 5to NIST 2710, 6to
USGS AGV2, 7to USGS BCR2
J Radioanal Nucl Chem
123

grain size distribution and mineral crystallization in the
tephra particles (see ‘‘Geochemistry and grain size analy-
sis’’ section). Sb, the element presenting the highest
variability among the different measurements, is a mod-
erately volatile and mobile element. It can be deposited on
the surface of ash during the cooling of the eruptive plume,
forming a soluble coating [40]. Its concentration in the bulk
material, together with the other volatile elements, could be
influenced by these processes, since the soluble salts pre-
sent in the coating can be easily removed. Unfortunately
other highly volatile elements (Cu, Br, Cd, Te, Hg, Tl, Pb,
Bi) were not measured by INAA or were not measured in
the other considered works, thus a further investigation on
this process is not possible. In the case of Br, Au and Hg
this is the first measurement in the Eyjafjallajo
¨kull tephra.
Despite the different techniques and the different sam-
pling sites, the comparison shows that for most of the
elements the composition of the tephra is very similar.
These data point to a quite homogenous geochemical
composition of the tephra produced during the 2010 Ey-
jafjallajo
¨kull eruption.
Geochemistry and grain size analysis
A first geochemical picture of the sample is given in Fig. 2,
where the composition of major, rare earth and incom-
patible elements is presented. All concentrations are nor-
malized to MORB; also OIB and UCC reference
compositions are reported. A main feature, recognizable in
each graph, is the relative enrichment in highly incom-
patible elements. Observing Fig. 2a, relative to major ele-
ments, the most enriched element is K, actually the most
incompatible element among major ones. The other major
elements are similar to MORB reference values. Also
Fig. 2b, c present similar evidence. Observing REE com-
position it is possible to note the high enrichment in light
REE with respect to heavy REE. This pattern can be ex-
plained considering incompatibility, since light REE have
higher ionic radii than heavy REE and are thus more in-
compatible [28]. In Fig. 2c a selection of LILE, HFSE and
REE is presented. The elements were ordered from Cs to
Yb, from the most incompatible to the less one, following
the order proposed in [25] for basaltic rocks. The higher
normalized concentrations are relative to highly incom-
patible elements on the left side of the graph, they decrease
progressively moving toward less incompatible elements
on the right. From La to Yb (with the exception of Sb) the
tephra presents higher concentration in incompatible ele-
ments also with respect to UCC, considered the principal
reservoir of incompatible elements in the earth system [26].
These evidences suggest that incompatibility is a major
driver in relation to trace element composition in the 2010
Eyjafjallajo
¨kull tephra. Their concentration is highly cor-
related to the incompatibility degree. The only elements
which are not in accordance to the order defined in [25] are
W, Sr and Sb. In the case of Sb the formation of volatile
Table 2 Elemental composition determined through INAA
Element Sample concentrations (ppm)
Bulk \63 lm\20 lm
Na* 3.8 ±0.2 3.90 ±0.15 3.5 ±0.2
Mg* 1.5 ±0.3 1.5 ±0.3 1.5 ±0.3
Al* 8.0 ±0.9 8.1 ±0.9 7.3 ±0.8
K* 1.6 ±0.2 1.5 ±0.2 1.5 ±0.2
Ca* 4.0 ±0.4 3.6 ±0.5 3.0 ±0.3
Sc 18 ±1 17.2 ±0.5 17.5 ±1.0
Ti* 1.0 ±0.1 1.0 ±0.1 1.0 ±0.1
V89±480±589±5
Cr 33 ±221±117±1
Mn* 0.19 ±0.02 0.19 ±0.02 0.17 ±0.02
Fe* 8.0 ±0.4 7.5 ±0.4 8.10 ±0.45
Co 17.8 ±0.7 15.6 ±0.9 18.1 ±0.8
Ni 70 ±457±260
±4
Zn 195 ±15 190 ±20 200 ±10
As 1.4 ±0.3 1.7 ±0.3 1.9 ±0.2
Br 6.3 ±0.5 7.6 ±0.5 7.1 ±0.4
Rb 48 ±250±250±2
Sr 400 ±20 350 ±30 395 ±15
Zr 500 ±30 500 ±20 515 ±20
Mo 4.7 ±0.8 5.4 ±0.9 4.6 ±0.9
Sb 0.38 ±0.04 0.30 ±0.03 0.35 ±0.04
Cs 0.61 ±0.06 0.75 ±0.08 0.73 ±0.04
Ba 500 ±100 480 ±90 510 ±70
La 53 ±355±255±2
Ce 125 ±6 124 ±4 130 ±6
Nd 65 ±569±267±3
Sm 13.5 ±1.5 14 ±114±1
Eu 5.3 ±0.5 5.1 ±0.6 5.0 ±0.6
Gd 12.2 ±0.6 13 ±1 11.3 ±0.8
Tb 2.5 ±0.2 2.3 ±0.2 2.5 ±0.3
Dy 11 ±2 11.0 ±0.8 12 ±1
Ho 1.8 ±0.2 2.0 ±0.4 2.0 ±0.4
Tm 1.6 ±0.2 1.5 ±0.2 1.7 ±0.2
Yb 6.9 ±0.5 6.9 ±0.4 6.7 ±0.6
Lu 1.1 ±0.1 1.2 ±0.2 1.1 ±0.1
Hf 12 ±212±212±2
Ta 4.0 ±0.2 3.8 ±0.2 4.0 ±0.2
W 1.2 ±0.6 1.4 ±0.5 1.5 ±0.4
Au 0.005 ±0.002 0.007 ±0.001 \0.0007
Hg 0.18 ±0.02 0.16 ±0.01 0.16 ±0.01
Th 7 ±16±1 7.5 ±1.0
U 1.8 ±0.1 1.9 ±0.1 1.81 ±0.09
Concentrations are expressed in ppm, except for major elements
(marked by an asterisk), whose concentrations are reported in %
J Radioanal Nucl Chem
123

and soluble compounds during the eruption could be a
possible explanation for the anomaly (see ‘‘Concentrations
and comparison with published data’’ section), but this is
not the case for W and Sr.
The elevated concentrations of incompatible elements,
well above MORB and OIB concentrations, are indicative
of the processes involved in the 2010 Eyjafjallajo
¨kull
eruption. It was observed that the geochemical features of
the eruption presented a complex evolution [2,18,19].
Observations based on the major elements composition of
the volcanic products showed that the eruption was prob-
ably triggered by a deep primitive basalt injection which
interacted with a second magmatic intrusion, remnant of
previous volcanic activity and with a different composition.
This in accordance with the trace element composition
presented in Fig. 1: the enrichment in incompatible ele-
ments points to the influence of an evolved magmatic
source whose composition is well distinct from a basaltic
one, like MORB.
Concerning grain size analysis, for most of the elements
no compositional differences can be observed comparing
the three grain-size classes, revealing a homogenous
composition of the material. The only elements showing
significative differences are Cr, Co, and Sr (Fig. 3). The
term ‘‘significative’’ is used here to indicate concentrations
relative to the three grain size classes showing differences
exceeding the uncertainties. The most evident fractionation
pattern is observable for Cr, which appears enriched in the
bulk sample. A progressive depletion of this element is
observable in the \63 lm fraction (-36 % with respect to
bulk) and in the\20 lm one (-48 % with respect to bulk).
This could be caused by a partial fractionation of Cr in
chromite which was observed in single tephra grains. In
[18] it is shown that chromite crystals formed in some
coarse ash grains (0.1–1.0 mm), thus leading to a relative
Fig. 2 Major (a), rare earth (b) and incompatible (c) element
composition of the tephra produced during the first explosive phase
of the 2010 Eyjafjallajo
¨kull eruption (15–19 April). Concentrations
determined within this work by INAA are reported with error bars,
but most of them are not visible since their extension doesn’t exceed
the marker size. The bulk composition determined by INAA was
compared to other bulk measurements carried out with other
techniques: [13] ICP-OES, [18] ICP-OES and ICP-MS, [19] EDS
and ICP-MS, [21] EDS and ICP-MS, [22] wet analysis for major
elements and ICP-MS. Also OIB and UCC are reported. All
concentrations are normalized to MORB. In (a) and (b) elements
are listed following atomic numbers, in the case of (c) they were listed
following the incompatibility order proposed in [25]
Fig. 3 The concentrations of the elements which show a significative
grain-size fractionation with the associated errors. ‘‘b’’ stands for the
bulk sample, ‘‘63’’ for the\63 lm fraction and ‘‘20’’ for the\20 lm
fraction
J Radioanal Nucl Chem
123

enrichment of this element in the bulk fraction. For Co and
Sr a similar pattern is observed: a relatively depleted
\20 lm fraction (-12 % with respect to bulk for both of
them).
Defining a geochemical fingerprint
Considering the presented data we tried to define a geo-
chemical fingerprint of the tephra produced within this event.
Four element ratios were used to highlight the geochemical
composition of the material and to distinguish it from ordi-
nary mineral dust typically deposited on Alpine and Green-
land glaciers. The ratios are Cs/Ta and Eu/Sc (Fig. 4a), Eu/
La and Hf/Th (Fig. 4b). They were selected considering the
main differences observed between the 2010 Eyjafjallajo
¨kull
tephra, UCC and samples of dust collected in the source areas
for the Alps and Greenland, i.e. Saharan regions for the Alps
[41], cold deserts and loess deposits in North-Eastern Asia
for Greenland [42]. In Fig. 4also other tephra are repre-
sented; they were collected at different sites of Greece and
Turkey [35,36]. The comparison shows that the composition
of the Eyjafjallajo
¨kull tephra is well distinct from the other
dust and tephra samples in Fig. 4. Unfortunately it was not
possible to carry out a more detailed comparison because
trace element geochemical characterization of bulk tephra is
not well established. Most of the published works on this
topic regard major elements geochemistry and single grain
analysis. But the preliminary results presented within this
work suggest that the application of INAA to tephra bulk
samples could give an important effort in this context, also
considering a future application of INAA to ice core samples
[43], where concentration of dust reaches very low concen-
tration and the visual identification of the tephra layer is
difficult.
Conclusions
A chemical characterization of the tephra erupted during
the 2010 eruption of Eyjafjallajo
¨kull (Iceland, April 14th–
May 22nd) determined through INAA is presented. The
composition was compared to other data previously pub-
lished, revealing a good agreement for all the measure-
ments taken into account. This confirms the applicability of
different techniques to this kind of bulk samples, where
INAA proved to be the only one able to determine major
and trace elements at the same time, with the only excep-
tion of Si. Trace element composition points to the influ-
ence of an evolved magmatic source rich in incompatible
elements, well distinct from the primitive basaltic one
which is thought to have triggered the eruption. Comparing
our data with literature data and analyzing the composition
of different grain sizes it was possible to assess the com-
positional homogeneity of the tephra erupted during the
first explosive phase of the eruption. The only significative
exception is represented by Cr. According to the homo-
geneity of the tephra and to the wide dispersion of the
tephra across Europe and the North Atlantic area, this event
could be used as a reference layer in the Alpine and
Greenland glaciers, where the plume is known to have
arrived. Incompatible element analysis allowed to define a
geochemical fingerprint of the tephra erupted by Eyjafjal-
lajo
¨kull, based on four elemental ratios: Eu/Sc, Cs/Ta, Hf/
Th and Eu/La. Using these features it will be possible to
distinguish the tephra from the ordinary mineral dust nor-
mally deposited on the Alpine and Greenland glaciers.
Acknowledgments A sincere thanks to the staff of LENA labora-
tory of the Pavia University for their assistance during the irra-
diations. We would also thank Arny Sveinbjornsdottir and Gudrun
Larsen for their precious tephra samples and Biancamaria Narcisi for
her useful advice.
References
1. Sturkell E, Einarsson P, Sigmundsson F, Hooper A, O
´feigsson
BG, Geirsson H, O
´lafsson H (2010) Dev Quat Sci 13:5–21
2. Gudmundsson MT, Thordarson T, Ho
¨skuldsson A, Larsen G,
Bjo
¨rnsson H, Prata FJ, Oddsson B, Magnu
´sson E, Ho
¨gnado
´ttir T,
Fig. 4 Cs/Ta–Eu/Sc and Eu/La-Hf/Th element ratios of the 2010
Eyjafjallajo
¨kull tephra (data from this work and [13,18,19,21,22]),
compared to Saharan dust samples [44,45], Asian dust samples
(Taklmakan Desert dust and Chinese loess [46,47]) and tephra
samples from the Mediterranean area [35,36]. The UCC reference is
also reported
J Radioanal Nucl Chem
123

Petersen GN, Hayward CL, Stevenson JA, Jo
´nsdo
´ttir I (2012) Sci
Rep. doi:10.1038/srep00572
3. The impact of Eyjafjallajokull’s volcanic ash plume, IATA, www.
iata.org/whatwedo/Documents/economics/Volcanic-Ash-Plume-
May2010.pdf. Accessed 04 Mar 2015
4. Rust AC, Cashman KV (2012) J Geophys Res. doi:10.1029/
2011JB008494
5. Petersen GN (2010) Weather 65(8):203–207
6. Petersen GN, Bjornsson H, Arason P (2012) J Geophys Res-
Atmos. doi:10.1029/2011JD016762
7. Bukowiecki N, Zieger P, Weingartner E, Juranyi Z, Gysel M,
Neininger B, Schneider B, Hueglin C, Ulrich A, Wichser A,
Henne S, Brunner D, Kaegi R, Schwikowski M, Tobler L,
Wienhold FG, Engel I, Buchmann B, Peter T, Baltensperger U
(2011) Atmos Chem Phys. doi:10.5194/acp-11-10011-2011
8. Stevenson JA, Loughlin S, Rae C, Thordarson T, Milodowski
AE, Gilbert JS, Harangi S, Lukacs R, Hojgaard B, Arting U,
Pyne-O’Donnel S, MacLeod A, Whitney B, Cassidy M (2012) J
Geophys Res. doi:10.1029/2011JB008904
9. Sigmundsson F, Hreinsdo
´ttir S, Hooper A, A
´rnado
´ttir T, Pedersen
R, Roberts MJ, O
´skarsson N, Auriac A, Decriem J, Einarsson P,
Geirsson H, Hensch M, O
´feigsson BG, Sturkell E, Svein-
bjo
¨rnsson H, Feigl KL (2010) Nature 468:426–430
10. Edwards B, Magnu
´sson E, Thordarson T, Gud
¯mundsson MT,
Ho
¨skuldsson A, Oddsson B, Haklar J (2012) J Geophys Res.
doi:10.1029/2011JB008985
11. Magnu
´sson E, Gudmundsson MT, Roberts MJ, Sigurðsson G,
Ho
¨skuldsson F, Oddsson B (2012) J Geophys Res. doi:10.1029/
2012JB009250
12. Kaminski E, Tait E, Ferrucci F, Martet M, Hirn B, Husson P
(2011) J Geophys Res. doi:10.1029/2011JB008297
13. Gislason SR, Hassenkam T, Nedel S, Bovet N, Eiriksdottir ES,
Alfedsson HA, Hem CP, Balogh ZI, Dideriksen K, Oskarsson N,
Sigfusson B, Larsen G, Stipp SLS (2011) PNAS 108:7307–7312
14. Gislason SR, Alfredsson HA, Eiriksdottir ES, Hassenkam T,
Stipp SLS (2012) Appl Geochem 26:S188–S190
15. Dellino P, Gudmundsson MT, Larsen G, Mele D, Stevenson JA,
Thordarson T, Zimanowski B (2012) J Geophys Res. doi:10.
1029/2911JB008726
16. Bagnato E, Aiuppa A, Bertagnini A, Bonadonna C, Cioni R,
Pistolesi M, Pedone M, Hoskuldsson A (2013) Geochim Cos-
mochim Acta 103:138–160
17. Olsson J, Stipps SLS, Makovicky E, Gislason SR (2014) Chem
Geol 384:135–148
18. Sigmarsson O, Vlastelic I, Andreasen R, Bindman I, Devidal JL,
Moune S, Keiding JK, Larsen G, Hoskuldsson A, Thordarson T
(2011) Solid Earth 2:271–281
19. Borisova AY, Toutain JP, Stefansson A, Gouy S, De Parseval P
(2012) J Geophys Res. doi:10.1029/2012JB009213
20. Moune S, Sigmarsson O, Schiano P, Thordarson T, Keiding JK
(2012) J Geophys Res. doi:10.1029/2011JB008718
21. Schleicher N, Kramar U, Dietze V, Kaminski U, Norra S (2012)
Atmos Environ 48:113–121
22. Navratil T, Hladil J, Strnad L, Koptikova L, Skala R (2013)
Aeolian Res 9:191–202
23. Presnall DC, Hoover JD (1987) High pressure phase equilibrium
constraints on the origin of mid–ocean ridge basalts. In: Mysen
BO (ed) Magmatic processes: physicochemical principles. Geo-
chemical Society, Richardson, pp 75–90
24. Hofmann AW (1988) Earth Planet Sci Lett 90:297–314
25. Sun SS, McDonough WF (1989) Geol Soc Spec Publ 42:313–345
26. Rudnick RL, Gao S (2003) In: Rudnick RL (ed) Treatise on
Geochemistry, vol 3. Elsevier, Oxford
27. Hoffmann AW (1997) Nature 385:219–228
28. Weaver BL (1991) Earth Planet Sci Lett 104:381–397
29. Petit JR, Wegner A, Narcisi B, Svensson A, Bigler M, Steffensen
JP (2013) The NEEM record of aeolian dust: contributions from
Coulter counter measurements. Geophysical research abstracts,
vol 15, EGU 2013–6255, EGU General Assembly 2013
30. Dumont M, Brun E, Picard G, Michou M, Libois Q, Petit JR,
Geyer M, Morin S, Josse B (2014) Nat Geosci. doi:10.1038/
ngeo2180
31. Lowe DJ (2011) Quat Geochronol 6:107–153
32. Schwikowski M, Doscher A, Gaggeler HW, Schotterer U (2002)
Tellus B 51:938–951
33. Koeberl C (1993) J Radioanal Nucl Chem 168:47–60
34. Joron JL, Treuil M, Raimbault L (1997) J Radioanal Nucl Chem
216:229–235
35. Bichler M, Duma B, Huber H (2004) J Radioanal Nucl Chem
262:57–65
36. Landmann G, Steinhauser G, Sterba JH, Kempe S, Bichler M
(2011) Appl Radiat Isotopes 69:929–935
37. Borio di Tigliole A, Cammi A, Clemenza M, Memoli V, Pat-
tavina L, Previtali E (2010) Prog Nucl Energy 52:494–502
38. Borio di Tigliole A, Cammi A, Chiesa D, Clemenza M, Manera
S, Nastasi M, Pattavina L, Ponciroli R, Pozzi S, Prata M, Previtali
E, Salvini A, Sisti M (2014) Prog Nucl Energy 70:249–255
39. Baccolo G, Baroni C, Clemenza M, Delmonte B, Maggi V, Motta
A, Nastasi M, Previtali M, Salvatore MC (2014) J Radioanal Nucl
Chem 299:1615–1623
40. Moune S, Gauthier PJ, Gislason SR, Sigmarsson O (2006) Geo-
chim Cosmochim Acta 70:461–479
41. De Angelis M, Gaudichet A (1991) Tellus B 43:61–75
42. Bory AJM, Biscaye PE, Svensson A, Grousset FE (2002) Earth
Planet Sci Lett 196:123–134
43. Baccolo G, Clemenza M, Delmonte B, Maffezzoli N, Maggi V,
Previtali E, Salvini A (2014), submitted
44. Moreno T, Querol X, Castillo S, Alastuey A, Cuevas E, Her-
rmann L, Mounkaila M, Elvira J, Gibbons W (2006) Chemo-
sphere 65:261–270
45. Castillo S, Moreno T, Querol X, Alastuey A, Cuevas E, Her-
rmann L, Mounkaila M, Gibbons W (2008) J Arid Environ
72:1034–1045
46. Liu CQ, Masuda AOA, Yabuki S, Zhang J, Fan ZL (1993) Chem
Geol 106:359–374
47. Yang X, Zhu B, White PD (2007) Quat Int 175:71–85
J Radioanal Nucl Chem
123