Harangi JVGR 2005 143 PDF

Journal of Volcanology and Geothermal Research 143 (2005) 237 – 257
www.elsevier.com/locate/jvolgeores
Correlation and petrogenesis of silicic pyroclastic rocks in the

Northern Pannonian Basin, Eastern-Central Europe: In situ trace
element data of glass shards and mineral chemical constraints
Szabolcs Harangia,*, Paul R.D. Masonb, Réka Lukácsa,c
a
Department of Petrology and Geochemistry, Eötvös University, H-1117 Budapest, Pázmány Péter sétány 1/C, Hungary
b
Vening Meinesz Research School of Geodynamics, Department of Earth Sciences, Utrecht University, Budapestlaan 4,
3584 CD Utrecht, The Netherlands
c
Institute of Earth Sciences, Geozentrum, University of Vienna, Althanstrasse 14, A-1090 Vienna, Austria
Received 8 September 2003; accepted 19 November 2004
Abstract
The Neogene volcanism of the Carpathian–Pannonian Region, Eastern-Central Europe started with repeated explosive
eruption of silicic magmas. Volcanic products consist mostly of non-welded and partially to densely welded ignimbrites, which
cover large areas in the Pannonian Basin. Since this volcanism occurred during a long time interval, from 21 Ma to 13.5 Ma
ago, these pyroclastic deposits have great stratigraphic importance, as well as providing valuable information about petrogenetic
processes during the formation of the back-arc basin area. In this study, we used in situ trace element data of glasses (glass
shards, pumice and fiamme glass) determined by laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS),
combined with microprobe geochemical data of the main mineral phases (plagioclase, biotite), to correlate the scattered
outcrops of the ignimbrite units. Based on these data, we distinguished four ignimbrite units in the Bükkalja Volcanic Field,
Northern Pannonian Basin. Each of these units is characterized by specific geochemical fingerprints. Thorium, Nb, Y and the
rare earth elements are the most effective discriminator elements in glasses. The modal composition of mineral phases
(occurrence or lack of certain minerals) and chemistry of plagioclases and biotites are also good correlation tools, especially the
Fe, Mg and Ti contents of biotites. We suggest that these correlation criteria, particularly the grain-specific LA-ICP-MS data,
could be effectively used also in other ignimbrite fields and in wider tephrochronological studies. The in situ trace element
composition of glasses, representing the liquid part of the erupted magma, can be also used to constrain the petrogenesis of the
rhyolitic magmas. Trace element ratios such as La/Nb, La/Y and Th/Nb suggest the importance of minor (e.g., hornblende and
ilmenite) and accessory (e.g., zircon, allanite) minerals controlling the composition of the erupted melt. Rhyolitic magmas
probably evolved from metaluminous andesitic parental melts via fractional crystallization. Syn-eruptive magma mingling was
* Corresponding author. Tel.: +36 12090555x8355; fax: +36 13812108.

E-mail address: [email protected] (S. Harangi).
0377-0273/$ - see front matter D 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jvolgeores.2004.11.012
238 S. Harangi et al. / Journal of Volcanology and Geothermal Research 143 (2005) 237–257
detected in the genesis of the Middle Ignimbrite Unit, based on the strong intra-sample geochemical variation both in the glasses
and in the phenocrysts.
D 2004 Elsevier B.V. All rights reserved.
Keywords: LA-ICP-MS; ignimbrite; glass shard; silicic volcanism; tephrochronology; petrogenesis; Pannonian Basin
1. Introduction (Froggatt, 1983, Stokes and Lowe, 1988, Shane and

Froggatt, 1991, 1994; Stokes et al., 1992; Lowe et al.,
Ash fall and pyroclastic flow deposits (e.g., 1999). Trace element chemistry of glass shards is,
ignimbrites) derived from eruption of silicic magmas however, a more powerful fingerprint of the erupted
often cover large areas and can extend for distances of volcanic products, because trace elements show larger
several hundreds of kilometres. These volcanic variability than the major elements. However, this
products are formed during hours to weeks, i.e. during method has not been widely used in tephrochrono-
instantaneous geologic events, and consequently they logical studies, because of the laborious preparation
provide unique chronological marker horizons in procedure of separation of glass shards from the bulk
stratigraphic sections on a regional scale. Thorarins- rock (De Silva and Francis, 1989). In a review paper
son (1944, 1974), correlating the tephra horizons in of tephrochronological studies in New Zealand, Shane
Iceland, introduced the term tephrochronology, which (2000) concluded that grain-specific techniques such
has become an important field of Quaternary science as laser-ablation inductively coupled plasma mass
(Westgate and Gorton, 1981; Sarna-Wojcicki et al., spectrometry (LA-ICP-MS) could become particularly
1984, 1987; Bogaard and Schmincke, 1985; Froggatt important in the future.
and Lowe, 1990; Shane, 2000). The first application of LA-ICP-MS for tephro-
Pyroclastic deposits can, however, be subsequently chronological purposes was conducted by Westgate et
easily eroded, broken apart and dispersed, making al. (1994), but they used separated multiple glass
field studies difficult. Correlation of pyroclastic flow shard grains for analysis. More recently, the rapid
deposits (e.g., ignimbrites) could be particularly development of this technique (Sylvester, 2001) has
problematic due to additional effects such as vertical enabled the analysis of a large number of trace
and horizontal zonation in physical and geochemical elements even in single glass shards in situ on a scale
properties (Smith, 1960; Hildreth, 1981; Hildreth and of 30–40 Am with very low (sub Ag g 1) detection
Mahood, 1985). Hildreth and Mahood (1985) dis- limits (Pearce et al., 1999, 2002). This method allows
cussed a number of possible correlation tools for ash- the routine analysis of individual glass shards with
flow tuffs and suggested that the composition of minimal sample preparation, i.e. from standard thin
juvenile material and lithoclasts can be especially sections (but with about 100 Am thickness) and
useful in addition to high-precision geochronological meanwhile the petrographic characteristics of the
and paleomagnetic measurements. De Silva and sample can be also studied. The in situ trace element
Francis (1989) provided case studies for correlation analysis of single glass shards provides additional
of large ignimbrite sheets in the Central Andes and benefits, such as checking the intra-sample chemical
concluded that phenocryst chemistry (e.g. Fe, Mg, Mn homogeneity and as a result providing an insight into
and Ti content of biotite, hornblende and orthopyrox- the magma chamber processes and constraining the
ene) could be the most effective method for finger- petrogenesis of silicic magmas (Bryant et al., 1999).
printing ignimbrite units. Shane et al. (2003) In this paper, we present a case study from the
emphasized also the effectiveness of biotite composi- Northern Pannonian Basin, Eastern-Central Europe to
tion in the correlation of Quaternary tephra beds. The show that in situ trace element data of glass shards from
major element composition of glass shards, which ignimbrites combined with mineral chemical data
represents the liquid part of the erupted magma, has could be used effectively to correlate scattered silicic
also been widely used to correlate Quaternary tephras pyroclastic occurrences, with improved resolution over
S. Harangi et al. / Journal of Volcanology and Geothermal Research 143 (2005) 237–257 239
existing palaeomagnetic and K–Ar geochronological 1993). The Pannonian Basin was formed by litho-
techniques. In addition, these data can be used also to spheric extension during the Early to Middle Miocene
constrain the genesis of the silicic magmas. due to the combined effect of collapse of over-
thickened crust and slab pull in the retreating
subduction zone beneath the Outer Carpathians
2. Geological background (Horváth, 1993; Fodor et al., 1999; Tari et al.,
1999). The formation and tectonic evolution of this
The Carpathian–Pannonian Region in Eastern region were accompanied by widespread volcanic
Central Europe (Fig. 1) consists of an arcuate activity (Szabo et al., 1992; Seghedi et al., 1998;
orogenic belt (the Carpathians) and an associated Harangi, 2001). Neogene volcanism started with
back-arc basin (the Pannonian Basin) formed as a part explosive eruptions of silicic magmas in the Early
of the orogenic systems of the Alpine–Mediterranean Miocene (at about 21 Ma; Pécskay et al., 1995). This
region (Horváth and Berckhemer, 1982, Wortel and silicic volcanism lasted for about 7 Ma and can be
Spakman, 2000). Subduction of oceanic lithosphere of divided into a pre-extensional and a syn-extensional
the Magura basin commenced in the Palaeogene and stage. The latter one was contemporaneous with the
terminated gradually from west to east during the early phase of the calc-alkaline andesitic to dacitic
Middle Miocene to Quaternary (Royden and Burch- volcanism at the Northern Pannonian Basin. Calc-
fiel, 1982; Csontos et al., 1992; Tomek and Hall, alkaline volcanic activity started at 16.5 Ma along the
Fig. 1. Location of the study area in the Carpathian–Pannonian Region, Eastern-Central Europe. Carp—Carpathians, CSVF—Central Slovakian
Volcanic Field.
Northern Pannonian Basin with eruption of garnet- of these rocks are buried by Late Miocene to
bearing andesites and rhyodacites (Harangi et al., Quaternary sediments (Fig. 1) due to strong post-
2001) followed by mostly andesites and dacites rift subsidence. Outcrops of the silicic pyroclastic
during the next volcanic phases up to 9 Ma (Konečný rocks are found mostly in the Northern Pannonian
et al., 1995; Harangi, 2001). In the eastern parts of the Basin, especially on the southern slopes of the
Carpathian–Pannonian Region, calc-alkaline volcan- uplifted Bükk Mts. (Bükkalja Volcanic Field, BVF;
ism started later, at about 14 Ma and shows a gradual Figs. 1 and 2), where numerous scattered outcrops
younging towards the southeast with the final stages cover the whole sequence of the silicic volcanism
of magmatism occurring only at 13 ka (Mason et al., from 21 Ma to 13.5 Ma (Márton and Pécskay, 1998;
1996, 1998). Eruptions of alkaline mafic rocks Szakács et al., 1998). Most of the exposed volcanic
followed the calc-alkaline volcanism and they form rocks are remarkably fresh, making this a key area to
monogenetic volcanic fields towards the inner parts of study the nature of the silicic pyroclastic units.
the Pannonian Basin (Embey-Isztin et al., 1993; However, correlation of the pyroclastic rocks derived
Embey-Isztin and Dobosi, 1995). from the same volcanic eruption phase is very
The Miocene silicic volcanism resulted in rhyo- difficult even in this area where the rocks are well
dacitic to high-Si rhyolitic volcanic products (pre- exposed, due to the lack of direct contact between
dominantly ignimbrites and phreatomagmatic ash fall units and with fossiliferous sedimentary beds, the
deposits; Pantó, 1963; Szakács et al., 1998) covering very similar macroscopic features in the field, and
a large area in the Pannonian Basin. However, most the insufficient accuracy and precision of existing K–
Fig. 2. Simplified geological map of the Bükkalja Volcanic Field showing the areal distribution of the main ignimbrite units (modified from
Szakács et al., 1998) and the localities of the analyzed samples.
Ar age data (Márton and Pécskay, 1998). Never- Variably welded pyroclastic flow deposits occur
theless, three main units were distinguished based on both in the LIU and MIU. The welded ignimbrite of
paleomagnetic data, because two major rotation the MIU appears to follow a southwest to northeast
events occurred in this area during the Miocene trending narrow belt in the BVF, whereas that of the
(Márton and Fodor, 1995; Márton and Pécskay, LIU can be found in scattered localities (Fig. 2). No
1998). This subdivision (Lower, Middle and Upper contacts between the non-welded and welded ignim-
Ignimbrite Units; LIU, MIU and UIU, respectively; brites have been found in the field. The welded
Fig. 2) can serve as a starting point in the correlation pyroclastic rocks are fiamme-bearing lapilli tuffs
of the pyroclastic sequence. (Fig. 3C and D) with a range of colours (grey,
brown and red) and varying sizes of fiamme (few
millimeters to 20 cm in length). The fiamme consist
3. Volcanology and petrology of a fresh glassy groundmass occasionally with
visible structural remnants of the original pumices
The Miocene volcanic rocks in the BVF cover a and glass shards. In some localities, flattened glass
surface area of approximately 300 km2, with a shards can be still observed in the matrix (Fig. 3F).
continuation at depth towards the south. The volcanic Some of the fiamme show incipient perlitic structure.
sequence reaches up to 750 m in thickness, based The quantity of phenocrysts in the fiamme is
upon borehole data (Szakács et al., 1998). Most of the strongly variable from a few percent in the LIU to
volcanic rocks are pyroclastic flow deposits (ignim- 50% in the MIU. The two welded pyroclastic units
brites with minor scoria-flows) with both non-welded also differ in their mineralogy. Although quartz,
and welded facies (Varga, 1981; Capaccioni et al., plagioclase and biotite are common in both units,
1995; Szakács et al., 1998). In addition, phreatomag- orthopyroxene is found only in the MIU rocks.
matic pyroclastic fall layers can be found throughout Ilmenite and allanite are characteristic accessory
the area as accretionary lapilli-bearing tuffs. No lava mineral phases in the MIU, whereas zircon is more
flows associated with the silicic pyroclastic suite have frequent in the LIU rocks. The matrix contains
been identified in this area. similar mineral assemblage as the fiamme.
The most common volcanic lithofacies is the non- In addition to the dominant ignimbrite deposits, a
welded pumiceous pyroclastic flow deposit (ignim- scoria-flow deposit occurs also in the MIU overlying
brite), which is the dominant rock type of the LIU and the fiamme-bearing pyroclastic beds. It contains a
UIU. They are poorly sorted block-bearing lapilli tuffs mixed juvenile clast population with dark scoria
(Fig. 3A and B). The maximum size of the pumices fragments, pumices and composite clasts (Czuppon et
can reach 40–50 cm. The pumices are phenocryst poor al., 2001; Harangi et al., 2002). Segregation pipes are
(generally b5%), consisting of quartz, plagioclase, commonly observed in this deposit. Accretionary
biotite and occasional sanidine or hornblende. Zircon, lapilli-bearing tuffs are also occasionally found in the
ilmenite and allanite occur as accessory mineral BVF. They are crystal-poor vitric tuffs containing
phases with variable amounts. The lithoclasts (5– cuspate and platy glass shards and loose phenocrysts of
10%) have dominantly igneous origin (basaltic quartz, plagioclase and biotite with rare hornblende.
andesite, andesite and rhyolite). Most of the lithoclasts
are porphyritic, containing plagioclase and orthopyr-
oxene phenocrysts with minor amounts of clinopyr- 4. Analytical techniques
oxene, hornblende, biotite or quartz, sitting in a glassy
groundmass. The matrix of the non-welded ignim- The major element composition of the glasses
brites is composed of cuspate, pumiceous to platy (glass shards, pumices and fiamme) and the main
glass shards and loose crystals (Fig. 3E) with a similar mineral phases were determined using a CAMECA
mineral assemblage to the pumices. This lithofacies SX100 electron microprobe with a 15 kV voltage and
shows very similar volcanological character both in a 20 nA beam current at the Institute of Petrology,
the LIU and UIU, making it difficult to distinguish the University of Vienna, Austria. Further analysis was
flow units in the field. carried out using a JEOL-733 Superprobe equipped
Fig. 3. Volcanological and petrographical characters of the ignimbrites of the Bükkalja Volcanic Field. (A) Non-welded ignimbrite at Szomolya
(LIU); (B) non-welded ignimbrite at Tibolddaróc (UIU); (C) fiamme-bearing, partially welded ignimbrite at Pünkfsdhegy, Demjén (LIU); (D)
fiamme-bearing welded ignimbrite at Vén-hegy, Bogács (MIU); (E) typical microscopic texture of the unwelded ignimbrites (Mocsolyás, LIU);
(F) typical microscopic texture of the partially welded ignimbrites (Túr-bucka, LIU). LA-ICP-MS analyses were carried out on single glass
shards and on the glasses of fiamme.
with an Oxford Instrument ISIS energy dispersive other. During glass analysis we used a defocused
system (Birkbeck College, University of London) beam (5 Am) to minimise the alkali loss.
with a 15 kV accelerating voltage for 100 s. Data from We determined 25 trace elements in 221 glass
both laboratories were checked by analyzing the same shards and fiamme from 25 samples of different
samples and the results were within 1–5% of each localities of the BVF by LA-ICP-MS using a 193 nm
diameter of 40 Am (Fig. 4). The signal recorded by

the ICP-MS during ablation was carefully checked for
inclusions and compositional boundaries to ensure
that only data for the glass shards were integrated
(Fig. 5). Quantitative concentrations were calculated
using NIST SRM 612 as a calibration standard
(Pearce et al., 1997) with Ca (previously determined
by electron microprobe analysis) as an internal stand-
ard element. The USGS reference glass BCR-2G was
continuously measured throughout the analysis of the
glasses and the results were within 5–10% of
recommended values. Detection limits were typically
in the range 0.01–1 Ag g 1 and internal precision was
b5% RSD (1r) for concentrations above 1 Ag g 1 and
Fig. 4. Ablation crater with a width of 40 Am in a single glass shard
(Harsány, UIU). b15% RSD (1r) below 1 Ag g 1.
ArF excimer laser ablation system (MicroLas GeoLas 5. Geochemistry

200Q) in combination with a quadrupole ICP-MS
(Micromass Platform ICP) at Utrecht University 5.1. Major element composition of glasses and the
(Mason and Kraan, 2002). Polished thin sections with main minerals
about 100 Am thickness were used for analysis. Glass
shards were identified with a Zeiss petrographic The major element compositions of the glasses and
microscope attached to the LA-ICP-MS instrument. the main mineral phases of representative samples are
Ablation was performed at a fixed point on the sample presented in Tables 1 and 2. The glasses (glass shards,
with an irradiance of 0.2 GW cm 2, a laser pulse pumices and fiamme) of the BVF pyroclastic rocks
repetition rate of 10 Hz and an ablation crater are remarkably fresh based on the microscopic
Fig. 5. Time-resolved signals for selected isotopes during LA-ICP-MS analysis of two glass shards with different thickness (sample N20;
Harsány, UIU). Note the sharp change in the signal at the grain boundary. For calculation the concentration of trace elements, only the signals
coming from the glass shard were integrated.
244
Table 1
Major (in wt.%) and trace element (in ppm) concentrations of glasses from representative samples of the Bükkalja Volcanic Field
Unit LIU MIU UIU-W UIU-E Tar, NW-Mátra
S. Harangi et al. / Journal of Volcanology and Geothermal Research 143 (2005) 237–257
Sample U13L-4-2 BKA10-5 N1-1-2 N37-1 N14-12 N25-2 U15L-4 V1-4-1 V1-26 V1-5 V1-8 N13-2 BKA17-1 TD2-9 TD2-6-2 N20-2-1 TAR-11 TAR-12
Glass gs gs fi gs gs gs fi fi fi gs gs gs gs gs gs gs gs gs
type
Locality Kács Túr-bucka Túr-bucka Szomolya Demjén Mocsolyás Tibolddaróc Bogács, Bogács, Bogács, Bogács, Demjén Bogács Tibolddaróc Tibolddaróc Harsány Tar Tar
Vén-hill Vén-hill Vén-hill Vén-hill
SiO2 77.05 77.30 77.05 77.23 78.03 77.31 71.88 76.01 76.55 75.29 77.44 77.73 77.88 78.41 77.71 77.24 77.50 77.38
TiO2 0.07 n.d. n.d. 0.12 n.d. n.d. 0.39 0.09 0.11 0.18 n.d. 0.06 0.06 0.04 0.04 0.06 0.09 0.07
Al2O3 12.77 12.94 13.37 12.86 12.69 13.20 15.59 13.99 13.54 13.98 12.79 12.69 12.62 12.90 12.87 12.99 12.90 12.74
FeOt 1.00 0.67 0.43 1.40 0.64 0.46 1.79 0.51 0.61 0.54 0.35 0.27 1.08 0.84 0.84 0.71 1.19 1.13
MgO 0.10 n.d. n.d. 0.11 0.06 0.06 0.58 n.d. 0.06 n.d. n.d. n.d. 0.06 n.d. 0.05 0.06 0.09 0.10
CaO 1.00 0.70 0.64 1.27 0.94 0.95 1.93 0.95 0.94 1.35 0.03 0.87 1.05 0.62 0.63 0.65 1.09 1.15
Na2O 2.45 2.48 2.70 2.28 2.01 2.91 1.72 2.01 2.12 2.06 1.63 2.12 1.89 2.78 2.42 3.70 2.26 2.10
K2O 5.55 5.91 5.81 4.74 5.63 5.10 6.12 6.45 6.08 6.60 7.75 6.27 5.36 4.41 5.45 4.60 4.88 5.33
LOI* 4.36 3.97 4.61 4.95 4.98 4.35 4.69 5.85 5.38 5.32 5.08 5.04 6.82 5.18 4.91 3.93 4.44 4.43
Rb 152 180 186 216 153 151 181 320 378 185 364 133 200 179 205 162 164 178
Sr 67 68 66 87 90 69 181 67 67 113 9 66 84 31 34 36 83 88
Y 14.6 17.8 17.4 22.3 18.2 16.5 40.2 35.6 35.1 26.3 38.6 6.6 8.2 39.7 38.4 31.6 8.7 9.1
Zr 51 60 54 70 75 52 373 179 127 141 63 57 63 66 92 65 74 86
Nb 7.8 8.4 8.5 11.4 9.2 9.1 12.7 16.7 16.2 10.9 13.3 6.1 6.9 14.0 16.8 11.6 6.6 6.9
Ba 744 789 752 912 741 610 1074 1252 1085 794 461 510 715 723 705 716 523 585
La 20.8 24.8 18.2 25.5 27.6 21.2 50.2 71.6 57.7 53.8 22.0 21.0 26.4 21.6 22.7 20.1 19.9 24.0
Ce 42.0 49.9 37.4 47.9 51.2 39.8 94.8 140.9 124.9 95.8 50.1 32.3 41.9 46.1 49.3 38.3 34.3 37.7
Pr 4.00 4.80 3.60 5.30 5.17 4.00 10.92 13.47 12.00 10.58 5.90 3.09 3.58 4.90 5.35 4.78 3.08 3.19
Nd 13.10 15.50 12.40 16.80 18.70 14.80 39.72 47.60 43.88 34.95 25.30 8.64 9.95 19.37 22.27 16.68 10.01 10.67
Sm 3.40 3.35 2.55 3.95 3.41 2.80 8.16 8.80 9.35 6.41 7.30 1.33 1.29 4.84 5.15 4.43 1.65 1.65
Eu 0.38 0.50 0.37 0.57 0.44 0.53 1.56 0.92 0.68 0.98 0.20 0.20 0.29 0.34 0.33 0.37 0.31 0.31
Gd 2.70 2.70 1.50 3.60 3.19 3.10 8.04 8.40 9.66 6.38 7.35 1.60 1.34 5.95 5.09 5.48 1.54 1.65
Dy 2.50 3.00 2.90 3.30 2.97 2.50 7.02 7.20 7.51 4.28 7.73 1.12 1.34 5.95 6.49 5.60 1.43 1.54
Er 1.45 1.85 1.45 2.10 1.87 1.55 4.14 3.67 3.82 2.44 4.45 0.72 0.84 3.97 3.82 3.44 0.80 0.88
Yb 1.80 1.85 2.05 2.60 1.93 1.75 4.50 3.60 3.85 3.00 4.50 0.91 0.73 4.08 4.14 3.50 0.91 1.32
Lu 0.25 0.30 0.27 0.36 0.31 0.23 0.72 0.48 0.49 0.37 0.70 0.18 0.22 0.61 0.64 0.49 0.23 0.24
Hf 2.20 2.60 1.80 2.80 2.53 2.10 8.52 5.33 5.17 4.35 4.60 1.81 2.24 3.27 2.93 2.57 2.64 2.31
Ta 0.90 1.10 1.18 1.50 0.81 1.10 0.96 1.28 1.42 0.98 2.30 0.81 1.01 1.63 1.53 1.27 0.75 0.85
Pb 26.6 33.6 28.6 34.2 27.6 25.9 28.1 26.8 25.4 12.6 11.8 22.6 31.5 24.5 25.8 21.7 31.5 31.7
Th 15.9 17.8 16.4 21.0 17.6 16.8 17.5 23.5 24.1 21.3 20.6 17.3 22.5 15.4 15.4 13.2 17.4 21.1
U 5.30 5.60 6.10 7.70 7.59 4.40 4.44 5.47 7.20 5.25 7.05 5.23 5.81 5.72 5.22 4.20 6.49 6.71
Major elements were determined by microprobe and were recalculated to 100% on anhydrous basis. LOI* denoted the difference of the measured total from 100%. The trace elements were determined by laser ablation ICP-MS. LIU—lower
ignimbrite unit; MIU—middle ignimbrite unit; UIU—upper ignimbrite unit (W—west; E—east); gs—glass shard; fi—fiamme. The localities are shown in Figs. 1 and 2.
Table 2
Representative electron microprobe analyses of biotites and plagioclases of the Miocene ignimbrites of the Bükkalja Volcanic Field (for abbreviations see Table 1)
S. Harangi et al. / Journal of Volcanology and Geothermal Research 143 (2005) 237–257
Biotites
Unit LIU MIU UIU-E UIU-W
Locality Demjén Túr-bucka Mészhegy Vénhegy, Vénhegy, Tibold-daróc Harsány Tibold-daróc Harsány Demjén
Bogács Bogács
SiO2 36.31 36.12 36.4 34.78 34.65 35.32 34.49 34.39 34.43 34.5
TiO2 3.45 3.39 3.56 4.29 4.6 4.7 3.61 3.89 3.66 4.73
Al2O3 14.37 13.99 13.8 13.93 14.35 13.69 13.23 13.85 13.46 13.7
FeO 23.71 24.08 22.84 27.89 26.37 24.51 28.64 29.33 28.6 22.84
MgO 0.23 0.19 0.27 0.27 0.16 0.16 0.38 0.25 0.4 0.24
CaO 9.37 9.3 9.07 4.76 6.97 8.68 5.65 4.89 5.75 9.5
Na2O 0.42 0.54 1.06 1.07 0.57 0.64 0.51 0.39 0.45 0.42
K2 O 8.51 8.65 8.58 8.63 8.62 8.52 8.56 8.56 8.71 8.64
TOTAL: 96.37 96.26 95.58 95.62 96.29 96.22 95.07 95.55 95.46 94.57
Plagioclases
Unit LIU MIU UIU-W UIU-E
Locality Szomolya Túr-bucka Mészhegy Demjén Vénhegy, Vénhegy, Vénhegy, Vénhegy, Demjén Demjén Harsány Harsány Tibold-daróc
Bogács Bogács Bogács Bogács
SiO2 56.72 58.48 55.17 56.80 53.50 50.59 57.00 46.30 53.23 57.12 63.44 58.35 63.69
Al2O3 27.22 26.36 28.86 27.46 29.22 30.81 26.89 33.83 29.87 27.96 23.25 26.20 23.50
FeO 0.18 0.12 0.07 0.12 0.18 0.21 0.20 0.30 0.37 0.24 0.14 0.14 0.05
CaO 9.95 8.61 10.31 9.68 12.24 14.29 9.40 17.73 12.60 9.09 4.74 8.41 3.40
Na2O 5.49 6.65 4.73 6.06 4.30 3.28 5.71 1.39 4.52 5.90 6.92 6.38 7.76
K2 O 0.35 0.41 0.35 0.38 0.25 0.16 0.36 0.04 0.21 0.54 0.87 0.40 1.13
TOTAL: 99.90 100.63 99.49 100.50 99.69 99.34 99.56 99.59 100.80 100.85 99.36 99.88 99.53
An 49.02 40.75 53.46 45.88 60.24 70.00 46.63 87.37 59.93 44.54 25.88 41.16 18.10
Ab 48.92 56.95 44.38 51.98 38.30 29.07 51.25 12.40 38.89 52.32 68.47 56.51 74.75
Or 2.06 2.30 2.16 2.14 1.46 0.93 2.12 0.23 1.18 3.14 5.65 2.33 7.15
245
(Fig. 6; SiO2=75–78.5 wt.%, K2O=4–8 wt.%, K2O/

Na2O=1.5–5). Glasses from the LIU and UIU are
indistinguishable based on their major elements,
whereas the MIU glasses have usually less SiO2 and
slightly higher K2O/Na2O ratio.
Plagioclases and biotites are common minerals in
each pyroclastic unit of the BVF. Most of the
plagioclase crystals are fairly homogeneous, while
the others show normal or occasional reverse zoning.
They show a wide range of composition from An18 to
An90 (Fig. 7). This large compositional variation
characterizes especially the plagioclases of the MIU
(An22–90), whereas those of the LIU and UIU show
more restricted anorthite content (An28–58 and An20–60,
Fig. 6. SiO2 vs. K2O diagram showing major element compositional respectively). Biotites are also ubiquitous, although
variation of the glasses of the Bükkalja ignimbrites. Note the some of them are altered. The MgO content of the fresh
overlap of the LIU and UIU glasses.
biotites is in the range between 4.5 and 9 wt.%, whereas
the TiO2 content varies between 3 to 5 wt.% (Fig. 6).
observations in spite of the relatively old ages (13.5– The largest and most continuous compositional varia-
21 Ma) of the deposits. The LOI content of the tion is found, as for plagioclase, in the MIU. The
glasses, calculated from the deficiency in the total of biotites of the LIU have fairly uniform chemistry
the microprobe analysis from 100%, is mainly in the (MgO=8.5–9.5 wt.%, TiO2=3.2–3.8 wt.%), whereas
range between 3 and 6 wt.% irrespective of their types those of the UIU appear to show a bimodal composi-
(glass shard, pumice or fiamme glass). In the tion. Biotites from the UIU ignimbrites occurring at the
interpretation of the major element data of glasses, eastern part of the Bükkalja Volcanic Field, have
we recalculated them to 100% on anhydrous basis. characteristically low MgO (5.3–6 wt.%) and low
The glasses have a relatively homogeneous, rhyolitic TiO2 (3.4–3.7 wt.%), whereas biotites from the western
major element character and are strongly potassic part of the Bükkalja have typically higher MgO (8.8–
Fig. 7. Compositional variation of plagioclases and biotites from different ignimbrite units of the BVF. The UIU is divided into two groups, a
western and an eastern one (UIU-W and UIU-E, respectively) based on the difference in the mineral chemical data.
Fig. 8. Primitive mantle (Sun and McDonough, 1989) normalized trace element patterns of glasses from single samples (U13L: Kács; V1: Vén-
hegy, Bogács; Td2: Tibolddaróc; N13: Demjén). Note the relatively homogeneous character of the LIU and UIU samples, but the strong
compositional heterogeneity of glasses in the sample V1.
Fig. 9. Comparison of primitive mantle (Sun and McDonough, 1989) normalized trace element patterns of glasses and bulk pumice/fiamme
from the same sample.
9.8 wt.%) and higher TiO2 (4–4.7 wt.%). Orthopyrox- 5.2. Trace element composition of glasses
enes are confined to the pyroclastic deposits in the
MIU. Both homogeneous and normally zoned crystals Results of the LA-ICP-MS trace element analyses
occur and have the composition of ferroan enstatite of representative samples are given in Table 1. Each
(En49–82 and Wo1–3). analysis was inspected individually, based on the
Fig. 10. Variation of incompatible trace elements and trace element ratios of glasses. Note the separation of main ignimbrite units in these
diagrams. Symbols as in Fig. 7.
time-resolved signals of several elements (Fig. 5) to ([La/Yb]cn=3.5 on average, whereas it is 13 to 15 in

remove any spurious data derived either from a the UIU samples at the western BVF; Fig. 10). Trace
microcryst or from the matrix surrounding the glass element characteristics of the LIU glasses lie between
shard. The analyzed glasses do not show significant these two end-member types (Eu/Eu*=0.3–0.6, [La/
geochemical variation in single samples of the LIU Yb]cn=6.5 to 10).
and UIU (Fig. 8). However, there is great composi-
tional heterogeneity within and between single sam-
ples derived from the MIU (Fig. 8). For example, 6. Discussion
glasses in the sample V1 show considerable variation
in Ba, Sr, Zr and rare earth element (REE) contents. In 6.1. Correlation of pyroclastic units
contrast, no compositional differences can be found
between the glass shards and the fiamme in the The composition of juvenile components (phenoc-
partially welded (i.e. where the glass shards are still rysts and glasses) in pyroclastic deposits reflects the
visible, although they are flattened; Fig. 3F) ignim- chemistry of the erupted magma. A change in the
brite samples of the LIU. No significant differences in composition of the magma erupted in different
composition were observed between the glass shards episodes of the silicic volcanism should thus be
and the bulk pumice blocks, or between the fiamme reflected in the chemistry of the phenocrysts and the
glass and the bulk fiamme in the same sample (Fig. 9). glasses. Indeed, the modal composition of the
Bulk pumices have a slightly higher Sr, Ba, Eu, Zr and phenocrysts (occurrence and lack of certain minerals,
light REE content than the glass shards, reflecting as well as change in the relative amounts of them) and
their enclosed phenocrysts and accessory mineral the chemical variability of plagioclases and biotites
assemblage (plagioclase, biotite, zircon). Pearce et (Fig. 7) appear to be a possible correlation tool in the
al. (2002) observed the same difference in the trace Bükkalja Volcanic Field. However, the best candidate
element contents of bulk pumices and single glass to characterize the erupting magma is the quenched
shards in the Minoan tephra layers. The bulk fiamme fragment of the liquid part of the magma, i.e. the glass
contains slightly lower concentrations of incompatible shards. Pyroclastic eruptive units can thus be corre-
elements than the fiamme glass, a feature that can be lated by determining the composition of the glasses
explained by the abundant orthopyroxene (it is (Westgate and Gorton, 1981; Sarna-Wojcicki et al.,
relatively poor in most trace elements) in the fiamme. 1984; Westgate et al., 1994, 1998; Pearce et al., 1999,
The primitive mantle normalized trace element 2002), providing that alteration, compositional zona-
patterns of the glasses (Fig. 8), show, in general, a tion and mixing are taken into account (Hildreth and
strong enrichment in the large ion lithophile elements Mahood, 1985).
(LILE; e.g., Cs, Rb, Ba) and Pb, as well as in Th and
U, whereas they are relatively depleted in Nb, Sr and 6.1.1. Correlation by petrography and mineral
Eu. However, the magnitude of positive and negative chemistry
anomalies and the total concentration of the incom- The modal composition of the phenocrysts pro-
patible trace elements show a systematic variation. vides a possible tool to distinguish the partially to
The highest incompatible trace element content is densely welded ignimbrites in this area. Orthopyrox-
shown by the fiamme of the MIU samples (Fig. 8). It ene can be found only in the fiamme-bearing
is remarkable that glasses of the pyroclastic rocks of ignimbrites of the MIU, while it is lacking in the
the UIU occurring at the western and eastern parts of welded ignimbrites of the LIU. Amphibole (horn-
the BVF are distinct in their trace element composi- blende) is present also in the pyroclastic rocks of the
tions (Fig. 10). The samples from the western parts MIU, but appears in some non-welded ignimbrites
show only a small negative Eu-anomaly (Eu/ (UIU at the western parts of the Bükkalja Volcanic
Eu*=0.4–0.65) and a strong depletion in the heavy Field, denoted UIU-W), too. Among the feldspars,
REE, whereas those from the eastern part have a plagioclase is the dominant mineral throughout the
strong negative Eu-anomaly (Eu/Eu*=0.2–0.4) and whole volcanic sequence, but occasionally sanidine
relatively minor light REE/heavy REE fractionation can be also found, mostly in the non-welded
ignimbrites of the UIU at the eastern part of the 6.1.2. Correlation by glass chemistry
Bükkalja (denoted UIU-E). Thus, a rough distinction The glasses (glass shards, pumice and fiamme
can be made using the phenocryst assemblage, glass) of the Bükkalja ignimbrites are usually fresh,
although the accuracy of this method is limited. i.e. they do not show post-eruptive recrystallization.
Plagioclase and biotite occur in each pyroclastic Therefore, their composition still reflects the chem-
deposit and they show wide variation in composition istry of the liquid part of the erupted magma. Most of
(Fig. 7). Biotites, however, show often incipient to the glasses are high-Si potassic rhyolite (SiO2N75
strong alteration that makes the interpretation of the wt.%, K2ON4 wt.%; recalculated to 100%, on
microprobe data difficult and a limitation of the use of anhydrous basis), whereas the MIU contains dacitic
this mineral. Composition of plagioclases varies from glasses, too. Unfortunately, the major element com-
An=18 to 90 mol% and shows some systematic position of the glasses is very similar (Fig. 6), thus
differences (Fig. 7). Plagioclases from the MIU these data cannot be used to distinguish the pyro-
ignimbrites have a very wide compositional range clastic units in Bükkalja. Trace elements are usually
(An25–90), whereas they show more restricted ranges decoupled from the major elements and show greater
in the samples of the LIU and UIU. Although they and more discriminative variation in volcanic glasses
overlap each other in the andesine field (An30–50), (Sarna-Wojcicki et al., 1984; De Silva and Francis,
plagioclases in the UIU-E pyroclastic rocks often 1989; Westgate et al., 1998; Clift and Blusztajn, 1999;
show a more sodic composition (An18–30), whereas Shane, 2000). The grain-specific LA-ICP-MS techni-
they shift towards the more calcic end-member in the que enables the rapid analysis of single glass shards
UIU-W samples. The most unambiguous result in from thin sections and provides more reliable trace
distinguishing pyroclastic units in this region can be element data for glasses than other techniques which
obtained using the composition of fresh biotites (Fig. use bulk glass separates for analysis (Pearce et al.,
7). De Silva and Francis (1989) and Chesner (1998) 1999, 2002).
could also successfully use the microprobe determi- However, first the results have to be checked as to
nation of Fe, Mg, Ti and Mn for biotite to discriminate whether the final data are derived exclusively from the
ignimbrite units in the Central Andes and within the glass or the laser penetrated microcrysts or any
Toba Tuffs, Indonesia, respectively. The biotites in the material beneath the glass shard. The signals recorded
Bükkalja ignimbrites are usually more iron-rich by the ICP-MS during ablation can indicate these
(FeOtot=22–31 wt.%) than those in the ignimbrites contaminations sensitively (Günther, 2001). The Fig.
of the Andes and in the Toba Tuffs. In the MgO vs. 5A shows an example, when only a very thin glass
TiO2 diagram, the three pyroclastic units distinguished shard was analyzed. After approximately 30 seconds,
by palaeomagnetic data (i.e., LIU, MIU and UIU) are the signal of 44Ca and 88Sr isotopes increased,
clearly separated from each other. Furthermore, the implying the presence of feldspar beneath the glass
UIU can be subdivided into two groups, i.e. the non- shard. Nevertheless, integration of only the early
welded ignimbrites at the western and eastern parts of range of data could give still reliable result for the
the Bükkalja Volcanic Field (UIU-W and UIU-E). It is trace element composition of the glass shard. The Fig.
remarkable that the MIU biotites show a wide range in 5B presents another example from the same sample,
FeO and MgO contents, similarly to the wide range of where signals were coming entirely from the glass
An-content of plagioclases in these rocks. On the during almost the whole counting time.
contrary, biotites of the other three ignimbrites units Glasses of the Bükkalja ignimbrites show charac-
show fairly homogeneous, but distinct, composition. teristic trace element fingerprints for each units (Figs.
In summary, the petrographic features (i.e., occur- 8 and 10). The discriminator elements are Nb, Y, Sm,
rence or lack of certain minerals) and the composition Eu and the heavy REE, whereas the MIU samples are
of the phenocrysts phases, especially that of biotites, characterized by distinctively high concentration of
can be applied as correlation tools for the Bükkalja light REE. The degree of the negative Eu-anomaly, as
ignimbrites. However, obtaining sufficient amount of expressed by Eu/Eu*, shows also systematic variation.
data (e.g., avoiding altered biotites) can be time The UIU-E glasses have the most pronounced
consuming. negative Eu-anomaly (Eu/Eu*b0.4). Combining these
elements, an effective discrimination between the these pyroclastic flow deposits may have originated
main pyroclastic units of Bükkalja can be achieved from different volcanic eruptions, the erupted silicic
based on the La/Y, Nb/Y, La/Nb and Th/Nb ratios magmas should have undergone very similar petro-
(Figs. 10 and 12). Using these geochemical parame- genetic processes. New palaeomagnetic and K/Ar data
ters, pyroclastic rocks from scattered outcrops can be (Márton and Pécskay, 2002, personal communication)
readily correlated over distances of several tens of km. imply also a temporal correlation between these
The trace element fingerprints of individual glasses ignimbrites. Based on the sum of this evidence, it
support also the distinction of two units within the appears likely that the Tar ignimbrite belongs to the
UIU as was suggested also by the chemistry of same volcanic eruption that resulted in the UIU-W
biotites. Thus, a more refined correlation could be pyroclastic flow deposits. This suggests a need for a
obtained using geochemical characteristics of the reconsideration of the stratigraphic classification of
juvenile components of the ignimbrites than using the widespread Miocene silicic volcanic rocks in the
geochronological and palaeomagnetic data. Based on Pannonian Basin. In such a correlation study, the LA-
the trace element characteristics of the glasses, we ICP-MS analysis of glasses (glass shards, and silicate
could refine the areal distribution of the silicic melt inclusions in phenocrysts) combined with min-
pyroclastic units in the BVF (Fig. 2), slightly eral chemical data would be a powerful tool.
modifying the previous geological maps of this region
(Szakács et al., 1998). 6.2. Petrogenetic implications
In Fig. 11, we present an example of how this
geochemical correlation can be extended over larger The variation in trace element composition of
distances from the BVF. Approximately 40 km west glasses and in the chemistry of minerals implies
from the outcrops of the UIU-W pyroclastic rocks, a differences in the petrogenetic evolution of the
thick non-welded ignimbrite can be found NW of the rhyolitic magmas in the Bükkalja Volcanic Field. This
Middle Miocene andesitic Mátra stratovolcano (Fig. rhyolite volcanism was partly contemporaneous with
1). This locality at the village of Tar was defined as calc-alkaline volcanic activity at the Northern Pan-
the type-locality of the Middle Ignimbrite Unit (or nonian Basin. Explosive and widespread silicic
dMiddle Rhyolite Tuff HorizonT; Hámor et al., 1987) magmatism often precedes or is coeval with calc-
in the Pannonian Basin. However, the trace element alkaline andesitic to dacitic magmatism in convergent
pattern of the glasses of this ignimbrite is indistin- continental margin and continental back-arc basin
guishable from that characterizing the UIU-W ignim- tectonic settings (De Silva, 1989; Gans et al., 1989;
brites in the BVF. This strong geochemical similarity Davidson et al., 1991; Johnson, 1991; Feeley and
is supported also by the mineral chemical data, i.e. Grunder, 1991; Siebel et al., 2001). The andesitic and
composition of plagioclases and biotites. Although rhyolitic magmas could be generated independently,
Fig. 11. Comparison of chondrite (Nakamura, 1974) normalized rare-earth element patterns of glasses of non-welded ignimbrites from the UIU-
W [(2) Demjén, (3) Bogács] and from Tar, NW-Mátra (1). The Tar ignimbrite was considered previously as the type-locality of MIU (Hámor
et al., 1987). However, the strong geochemical similarity with the UIU-W ignimbrites suggests a reconsideration of its stratigraphic classification.
but can also be genetically related. In general, two fore they can be used to constrain the source of the
end-member models are proposed for the origin of Si- magmas and/or to reveal the fractionation of minor
rich magmas that have a close relationship with and accessory minerals such as hornblende, biotite,
geodynamic evolution. Silicic magmas can be formed ilmenite, zircon and allanite. In this plot (Fig. 12), the
by anatexis of continental crust (Reid, 1983; Huppert four main ignimbrite groups are clearly separated,
and Sparks, 1988; Ruiz et al., 1988; Francis et al., suggesting a different petrogenetic evolution. The
1989), a process that occurs predominantly in areas analyzed glasses of the Bükkalja ignimbrites are
with thick continental crust and typically produces rhyolites with high SiO2 content (N75 wt.%). There-
large-volume ignimbrite sheets. Alternatively, silicic fore, the differences in the trace element ratios cannot
magmas can be generated by fractional crystallization be explained by different degrees of magmatic differ-
of mantle-derived mafic or andesitic melts, combined entiation alone. Instead, different ratios of highly
with crustal assimilation (Bacon and Druitt, 1988, incompatible trace elements could reflect composi-
Musselwhite et al., 1989; Johnson, 1991; McCulloch tional differences in their parental magmas and limited
et al., 1994; Siebel et al., 2001). Eruption of such fractionation of specific mineral phases. The glasses
small- or medium-volume ignimbrites is usually of LIU and UIU define a linear trend in the La/Nb vs.
thought to take place in regions with thinner crust Th/Nb plot spreading from the lower end of the
(Wark, 1991). The Miocene silicic ignimbrites of the Miocene calc-alkaline andesite field towards the
Northern Pannonian Basin were formed roughly higher La/Nb and Th/Nb ratios. Fractionation of
coeval with the formation of a back-arc basin area hornblende, biotite and ilmenite could be responsible
by lithospheric extension. The LIU can be considered for this shift, whereas crystallization of zircon and
as pre-extensional (N18 Ma), whereas the MIU and allanite move the data points in the opposite direction.
UIU are syn-extensional (18–13 Ma) volcanic suites. Petrographic observations are consistent with this
Thus, they were generated during a period with a scenario, because the pyroclastic rocks of the UIU-W
transition from a thicker (N35 km) to thinner (b30 km; contain hornblende and ilmenite, whereas allanite and
Tari et al., 1999) crust. zircon are common in the UIU-E rocks. The spread
The primitive mantle-normalized trace element
patterns of the glasses of the ignimbrites (Fig. 8)
show some common features, such as enrichment in
the LIL elements and Pb and a relative depletion in
Ba, Nb, Sr and Eu. Depletion in Ba, Sr and Eu could
reflect extensive feldspar fractionation. The relatively
low concentration of high field strength elements
(HFSE; e.g., Nbb15 ppm, Tab1.5 ppm and Zrb100
ppm) and trace element ratios such as Ba/La (20 to 40)
and La/Nb (1 to 5) suggest a subduction-related
component in their genesis. However, all of these
geochemical features could also be inherited in the
continental crust, providing the source of the silicic
magmas. Previous models usually invoked a crustal
anatectic origin of the Miocene silicic magmas in the Fig. 12. La/Nb vs. Th/Nb plot for the glasses of the Bükkalja
ignimbrites (symbols as in Fig. 7). Note the clear separation of the
Pannonian Basin (e.g., Downes, 1996; Lexa and four ignimbrite units. Average crustal (LC—lower crust, UC—
Konečný, 1998; Póka et al., 1998), although Harangi upper crust) values are from Taylor and McLennan (1985). Field of
(2001) pointed out that variation in radiogenic isotope calc-alkaline andesites from the Northern Pannonian Basin: Harangi
ratios of the pumices could only be explained by et al. (2001) and Harangi and Downes, unpublished data.
involvement of both mantle and crustal components in Fractionation vectors of different minerals (zr—zircon, al—allanite,
am—amphibole, il—ilmenite, bi—biotite) are based on the distri-
their genesis. Incompatible trace element ratios such bution coefficients published by Mahood and Hildreth (1983),
as La/Nb and Th/Nb ratios are not influenced by the Bacon and Druitt (1988) and Ewart and Griffin (1994). Numbers
fractionation of feldspars and pyroxenes, and there- denote the amount of fractionated mineral in weight percent.
within the data of the LIU rocks could be primarily different magma batches (Harangi et al., 2002). Strong
due to fractionation of biotite and zircon. This linear intra-sample variations of trace element composition
trend may imply also fairly similar parental magmas. in glasses support this suggestion. The fiamme and
However, radiogenic isotope ratios of the pumices glass shards in these rocks could represent magma
(Harangi, 2001; Seghedi et al., 2004) do not support a batches that have undergone different degrees of
common parental magma. Pumices of the UIU-W and differentiation. The chemistry of the main mineral
UIU-E ignimbrites are characterized by relatively low phases (e.g., plagioclase, orthopyroxene) suggests
initial 87Sr/86Sr ratios (0.707–0.708) and high initial derivation of the rhyolitic melts from metaluminous
143
Nd/144Nd ratios (0.5124–0.5125) compared with andesitic parental magma. Plagioclases often show
those of the LIU (0.709–0.710 and 0.51220–0.51225, high An-content (N75 mol%), whereas orthopyrox-
respectively). This suggests an involvement of vari- enes are relatively Mg-rich (En=49–82 mol%; Lukács
able amounts of mantle and crustal components in the et al., 2002) and have CaO=0.5–1.7 wt.% and
genesis of the silicic volcanic rocks of the BVF and Al2O3=0.6–2.2 wt.% that are consistent with metal-
similar sources for the silicic magmas of the two UIU uminous host magma affinity based on the discrim-
ignimbrite groups. The difference in trace element ination scheme of Maeda et al. (1991). The overlap of
composition between the two UIU groups can be the trace element ratios of the MIU glasses and the
explained primarily by the different fractionation calc-alkaline andesites close to the ignimbrite field
paths of their magmas. Amphibole fractionation in could support this conclusion. The Miocene silicic
addition to separation of plagioclase and orthopyrox- volcanism was contemporaneous with the formation
ene characterized the evolution of the UIU-W of the back-arc basin area, i.e. with the continuous
magmas, whereas the UIU-E rhyolites are more thinning of the lithosphere. The magmagenesis of the
differentiated, where plagioclase, biotite, sanidine rhyolitic magmas in the Northern Pannonian Basin,
and accessory minerals such as allanite and zircon i.e. the involvement of mantle and crustal component,
controlled the chemical evolution of the melt. It is is consistent with this geodynamic situation.
remarkable that these different magma evolution
processes occurred spatially very close to one another
roughly at the same time (about 13.5 Ma), implying 7. Conclusions
the existence of separated silicic magma chambers
beneath the Northern Pannonian Basin during the Late Our new detailed geochemical study on the
Miocene. The compositional homogeneity of the LIU Miocene ignimbrite suite of the Northern Pannonian
and UIU ignimbrites, irrespective of the stratigraphic Basin suggests that the in situ trace element content
position of the samples, suggests that the eruption of of glasses, along with compositions of the main
silicic magmas could have occurred from a homoge- mineral phases, could be a powerful discrimination
neous part of the magma chambers. tool in the correlation of silicic pyroclastic rocks
The glasses of the MIU shift away toward the over distances of several tens of kilometres. The LA-
higher La/Nb ratios from the LIU-UIU trend. This ICP-MS analysis of glass shards with a size of only
cannot be explained by fractionation of any mineral a few tens of micrometers is a rapid technique
phases from a magma common with LIU or UIU, but providing the concentration of a large number of
rather suggests a different origin for the MIU silicic trace elements with low detection limits and with
magma. One of the most characteristic features of the minimal sample preparation. The in situ trace
MIU rocks is the strong geochemical heterogeneity in element data of glasses reflect the composition of
the compositions of both the phenocrysts (plagioclase, the liquid part of the erupted magma and therefore
biotite) and the glasses of fiamme and glass shards provide a better geochemical fingerprint for the
even in single samples. Furthermore, in the upper part pyroclastic deposits than the bulk pumices, which
of this volcanic suite andesitic scoriae and rhyolitic contain also variable amounts of phenocrysts and
pumices, as well as banded pumices, can be found possibly also xenocrysts. Therefore, these data can
(Czuppon et al., 2001). All of these features imply a be used effectively in correlation studies, providing a
zoned magma chamber and syn-eruptive mingling of more refined correlation of scattered silicic pyro-
clastic deposits than the geochronological and palae- central and northern Europe. Geol. Soc. Amer. Bull. 96,
omagnetic data. 1554 – 1571.
Bryant, C.J., Arculus, R.J., Eggins, S.M., 1999. Laser ablation-
In addition, trace element abundances of the glasses inductively coupled plasma-mass spectrometry and tephras: a
can also constrain the petrogenesis of the host silicic new approach to understanding arc-magma genesis. Geology
magmas. Based on the trace element characteristics of 27, 1119 – 1122.
the glass shards and fiamme and the chemical Capaccioni, B., Coradossi, N., Harangi, R., Harangi, S.,
Karátson, D., Sarocchi, D., Valentini, L., 1995. Early
composition of phenocrysts, we have identified at least
Miocene Pyroclastc Rocks of the Bükkalja Ignimbrite Field
four silicic magmatic systems that were active beneath (North Hungary)—A Preliminary Stratigraphic Report. In:
the Northern Pannonian Basin during a continuous Downes, H., Vaselli, O. (Eds.), Neogene and related
thinning of the crust and the lithosphere in the Miocene volcanism in the Carpatho–Pannonian Region, Acta Vul-
(21 to 13.5 Ma). The trace element composition of the canol., vol. 7, pp. 119 – 124.
erupted melts was controlled mostly by crystallization Chesner, C.A., 1998. Petrogenesis of the Toba Tuffs, Sumatra,
Indonesia. J. Petrol. 39 (3), 397 – 438.
of minor and accessory minerals, such as hornblende, Clift, P., Blusztajn, J., 1999. The trace element characteristics of
ilmenite, allanite and zircon. The ignimbrite precursor aegean and aeolian volcanic arc marine tephra. J. Volcanol.
magmas, especially those of MIU, might have sim- Geotherm. Res. 92, 321 – 347.
ilarities with the Middle Miocene calc-alkaline ande- Csontos, L., Nagymarosy, A., Horváth, F., Kovác, M., 1992. Tertiary
sitic magmas erupted to the north of the ignimbrite field evolution of the Intra-Carpathian area: a model. In: Ziegler, P.A.
(Ed.), Geodynamics of Rifting, vol. I. Case Studies on Rifts:
in the Northern Pannonian Basin (dWestern Carpathian Europe and Asia, Tectonophysics, vol. 208, pp. 221 – 241.
ArcT), suggesting a common origin. Syn-eruptive Czuppon, Gy., Harangi, Sz., Ntaflos, T., Lukács, R., Szabó, Cs.,
magma mingling resulted in strong intra-sample Koller, F., 2001. Mixed andesite-rhyolite ignimbrite from the
compositional variation in the MIU pyroclastic suite. Miocene Bükkalja Ignimbrite Volcanic Field, Northern Hun-
At the end of the silicic volcanic period, eruption of gary: evidence for magma mixing. Mitt. Osterr. Mineral. Ges.
146, 61 – 63.
different kinds of rhyolitic magmas occurred contem- Davidson, J.P., Harmon, S.R., Wfrner, G., 1991. The source of
poraneously and spatially very close to one another in central Andean magmas: some considerations. In: Harmon,
the Bükkalja Volcanic Field. R.S., Rapela, C.W. (Eds.), Andean Magmatism and its Tectonic
Setting, Geol. Soc. Am. Spec. Pap., vol. 265, pp. 233 – 243.
De Silva, S.L., 1989. The Altiplano–Puna volcanic complex of the
central Andes. Geology 17, 1102 – 1106.
Acknowledgements De Silva, S.L., Francis, P.W., 1989. Correlation of large ignim-
brites—two case studies from the Central Andes of Northern
We gratefully acknowledge the kind assistance of Chile. J. Volcanol. Geotherm. Res. 37, 133 – 149.
Drs. Andy Beard and Theo Ntaflos during the Downes, H., 1996. Neogene magmatism and tectonics in the
Carpatho–Pannonian region. In: Decker, K. (Ed.), PANCARDI
microprobe work. This study was carried out during Workshop 1996, Dynamics of the Pannonian–Carpathian–Dina-
Sz.H.’s stays in Utrecht supported by a Europrobe ride System, Mid. Ges. Geol. Bergbaustud. Österr., vol. 41,
grant and a NATO fellowship, respectively. The pp. 104 – 105.
Utrecht ICP-MS laboratory has received funding from Embey-Isztin, A., Dobosi, G., 1995. Mantle source characteristics for
the Netherlands Organization of Scientific Research miocene–pleistocene alkali basalts, Carpathian–Pannonian
Region: a review of trace elements and isotopic composition.
(NWO/GOA). Constructive suggestions given by Drs. In: Downes, H., Vaselli, O. (Eds.), Neogene and Related
R.M. Briggs and S. De Silva helped to clarify our Volcanism in the Carpatho–Pannonian Region, Acta Vulcanol.,
ideas presented in this paper. vol. 7, pp. 155 – 166.
Embey-Isztin, A., Downes, H., James, D.E., Upton, B.G.J.,
Dobosi, G., Ingram, G.A., Harmon, R.S., Scharbert, H.G.,
1993. The petrogenesis of pliocene alkaline volcanic rocks
References from the Pannonian Basin, Eastern Central Europe. J. Petrol.
34, 317 – 343.
Bacon, C.R., Druitt, T.H., 1988. Compositional evolution of the Ewart, A., Griffin, W.L., 1994. Application of proton-microprobe
zoned calc-alkaline magma chamber of Mount Mazama, Crater data to trace-element partitioning in volcanic rocks. Chem.
Lake, Oregon. Contrib. Mineral. Petrol. 98, 224 – 256. Geol. 117, 251 – 284.
Bogaard, v.d.P., Schmincke, H.-U., 1985. Laacher see tephra: a Feeley, T.C., Grunder, A.L., 1991. Mantle contribution to the
widespread isochronous late quaternary tephra layer in evolution of middle tertiary silicic magmatism during early
stages of extension: the Egan Range volcanic complex, east- Johnson, C.M., 1991. Large-scale crust formation and lithosphere
central Nevada. Contrib. Mineral. Petrol. 106, 154 – 169. modification beneath middle to late cenozoic calderas and
Fodor, L., Csontos, L., Bada, G., Gyfrfi, I., Benkovics, L., 1999. volcanic fields, Western North America. J. Geophys. Res. 96,
Tertiary tectonic evolution of the Pannonian basin system and 13485 – 13507.
neighbouring orogens: a new synthesis of palaeostress data. In: Konečný, V., Lexa, J., Hojstricová, V., 1995. The Central Slovakia
Durand, B., Jolivet, L., Horváth, F., Séranne, M. (Eds.), The Neogene volcanic field: a review. In: Downes, H., Vaselli, O.
Mediterranean Basins: Tertiary Extension Within the Alpine (Eds.), Neogene and Related Magmatism in the Carpatho–
Orogen, Geol. Soc. London, Spec. Pub., vol. 156, pp. 295 – 334. Pannonian Region, Acta Vulcanol., vol. 7, pp. 63 – 78.
Francis, P.W., Sparks, R.S.J., Hawkesworth, C.J., Thorpe, R.S., Lexa, J., Konečný, V., 1998. Geodynamic aspects of the neogene to
Pyle, D.M., Tait, S.R., 1989. Petrology and geochemistry of quaternary volcanism. In: Rakús, M. (Ed.), Geodynamic
volcanic rocks of the Cerro Galan caldera, N.W. Argentina. Development of the Western Carpathians, Geol. Surv. Slovak
Geol. Mag. 126, 515 – 547. Republik, Bratislava, pp. 219 – 240.
Froggatt, T.C., 1983. Toward comprehensive upper quaternary tephra Lowe, D.J., Newnham, R.M., Ward, C.M., 1999. Stratigraphy and
and ignimbrite stratigraphy in New Zealand using electron chronology of a 15 ka sequence of multi-sourced silicic tephras
microprobe analysis of glass shards. Quat. Res. 19, 188 – 200. in a montane peat bog, eastern North Island, New Zealand. N.Z.
Froggatt, T.C., Lowe, D.J., 1990. A review of late quaternary silicic J. Geol. Geophys. 42, 565 – 579.
and some other tephra formations from New Zealand: their Lukács, R., Czuppon, G., Harangi, S., Szabó, C., Ntaflos, T., Koller,
stratigraphy, nomenclature, distribution, volume and age. N.Z. J. F., 2002. Silicate melt inclusions in ignimbrites, Bükkalja
Geol. Geophys. 33, 89 – 109. Volcanic Field, Northern Hungary—texture and geochemistry.
Gans, P.B., Mahood, G.A., Schermer, E., 1989. Synextensional Acta Geol. Hung. 45 (4), 341 – 358.
magmatism in the Basin and Range Province: a case study Maeda, J., Shimura, T., Arita, K., Osanai, Y., Maruta, M., Bamba,
from the eastern Great Basin. Spec. Pap.-Geol. Soc. Am. 233 M., Suetake, S., 1991. Chemical features of orthopyroxene in
1–53. peraluminous igneous rocks. Am. Mineral. 76, 1674 – 1682.
Günther, D., 2001. Quantitative fluid inclusion analysis using a 193 Mahood, G.A., Hildreth, E.W., 1983. Large partition coefficients for
nm excimer laser ablation system coupled to ICP-MS. In: trace elements in high-silica rhyolites. Geochim. Cosmochim.
Sylvester, P. (Ed.), Laser-ablation-ICPMS in the Earth Sciences. Acta 47, 11 – 30.
Principles and Applications, Mineralogical Associations of Márton, E., Fodor, L., 1995. Combination of paleomagnetic and
Canada Short Course Series, vol. 21, pp. 47 – 61. stress data—a case study from North Hungary. Tectonophysics
Hámor, G., Balogh, K., Ravaszné Baranyai, L., 1987. Az észak- 242, 99 – 114.
magyarországi harmadidõszaki formációk radiometrikus kora. Márton, E., Pécskay, Z., 1998. Complex evaluation of paleo-
Ffldt. Intéz. Évi Jel. 1986-ról, 61 – 76. magnetic and K/Ar isotope data of the miocene ignimbritic
Harangi, Sz., 2001. Neogene to quaternary volcanism of the volcanics in the Bükk Foreland, Hungary. Acta Geol. Hung.
Carpathian–Pannonian Region—a review. Acta Geol. Hung. 41, 467 – 476.
44, 223 – 258. Mason, P.R.D., Kraan, W.J., 2002. Attenuation of spectral
Harangi, Sz., Downes, H., Kósa, L., Szabó, Cs., Thirlwall, M.F., interferences during laser ablation inductively coupled plasma
Mason, P.R.D., Mattey, D., 2001. Almandine garnet in calc- mass spectrometry (LA-ICP-MS) using an rf only collision and
alkaline volcanic rocks of the Northern Pannonian Basin reaction cell. J. Anal. At. Spectrom. 17, 858 – 867.
(Eastern-Central Europe): geochemistry, petrogenesis and geo- Mason, P.R.D., Downes, H., Thirlwall, M., Seghedi, I., Szakács, A.,
dynamic implications. J. Petrol. 42, 1813 – 1843. Lowry, D., Mattey, D., 1996. Crustal assimilation as a major
Harangi, Sz., Lukács, R., Czuppon, Gy., Szabó, Cs., 2002. Magma petrogenetic process in east Carpathian Neogene to Quaternary
mixing in a compositionally layered magma chamber: a silicate continental margin arc magmas. J. Petrol. 37, 927 – 959.
melt inclusion study. Workshop on Volcanic Systems, Seiano, Mason, P.R.D., Seghedi, I., Szakács, A., Downes, H., 1998.
Italy, Proceedings, pp. 101 – 106. Magmatic constraints on geodynamic models of subduction
Hildreth, W., 1981. Gradients in silicic magma chambers: in the Eastern Carpathians, Romania. Tectonophysics 297,
implications for lithospheric magmatism. J. Geophys. Res. 86, 157 – 176.
10153 – 10192. McCulloch, M.T., Kysler, T.K., Woodhead, J.D., Kinsley, L., 1994.
Hildreth, W., Mahood, G., 1985. Correlation of ash-flow tuffs. Geol. Pb–Sr–Nd–O isotopic constraints on the origin of rhyolites from
Soc. Amer. Bull. 96, 968 – 974. the Taupo Volcanic Zone of New Zealand: evidence for
Horváth, F., 1993. Towards a mechanical model for the formation of assimilation followed by fractionation from basalt. Contrib.
the Pannonian basin. Tectonophysics 226, 333 – 357. Mineral. Petrol. 115, 303 – 312.
Horváth, F., Berckhemer, H., 1982. Mediterranean back arc basins. Musselwhite, D.S., DePaolo, D.J., McCurry, M., 1989. The
In: Berckhemer, H., Hsü, K. (Eds.), Alpine-Mediterranean evolution of a silicic magma system: isotopic and chemical
Geodynamics, Amer. Geophys. Union Geol. Soc. Amer., Geo- evidence from the Woods Mountains Volcanic Center, eastern
dynamic Series, vol. 7, pp. 141 – 173. California. Contrib. Mineral. Petrol. 101, 19 – 29.
Huppert, H.E., Sparks, R.S.J., 1988. The generation of granitic Nakamura, N., 1974. Determination of REE, Ba, Mg, Na and K in
magmas by intrusion of basalt into continental crust. J. Petrol. carbonaceous and ordinary chondrites. Geochim. Cosmochim.
29, 599 – 624. Acta 38, 757 – 775.
Pantó, G., 1963. Ignimbrites of Hungary with regard to the genetics Shane, P.A.R., Froggatt, P.C., 1991. Glass chemistry, paleomagnet-
and classification. Bull. Volcanol. 25, 175 – 181. ism, and correlation of middle pleistocene tuffs in southern
Pearce, N.J.G., Perkins, W.T., Westgate, J.A., Gorton, M.P., North Island, New Zealand, and Western Pacific. N.Z. J. Geol.
Jackson, S.E., Neal, C.R., Chenery, S.P., 1997. A compilation Geophys. 34, 203 – 211.
of new and published major and trace element data for NIST Shane, P.A.R., Froggatt, P.C., 1994. Discriminant function analysis
SRM 610 and NIST SRM 612 glass reference materials. of glass chemistry of New Zealand and North American tephra
Geostand. Newsl. 21, 115 – 144. deposits. Quat. Res. 41, 70 – 81.
Pearce, N.J.G., Westgate, J.A., Perkins, W.T., Eastwood, W.J., Shane, Shane, P.A.R., Smith, V., Nairn, I.A., 2003. Biotite composition as a
P., 1999. The application of laser ablation ICP-MS to the analysis tool for the identification of quaternary tephra beds. Quat. Res.
of volcanic glass shards from tephra deposits: bulk glass and 59, 262 – 270.
single shard analysis. Glob. Planet. Change 21, 151 – 171. Siebel, W., Schnurr, W.B.W., Hahne, K., Kraemer, B., Trumbull,
Pearce, N.J.G., Eastwood, W.J., Westgate, J.A., Perkins, W.T., 2002. R.B., Bogaard, P., Emmermann, R., 2001. Geochemistry and
Trace-element composition of single glass shards in distal isotope systematics of small- to medium-volume Neogene–
Minoan tephra from SW Turkey. J. Geol. Soc. (Lond.) 159, Quaternary ignimbrites in the southern central Andes: evidence
545 – 556. for derivation from andesitic magma sources. Chem. Geol. 171,
Pécskay, Z., Lexa, J., Szakács A., Balogh, K., Seghedi, I., Konečny, 213 – 237.
V., Kovács, M., Márton, E., Kaliciak, M., Széky-Fux, V., Póka, Smith, L.R., 1960. Zones and zonal variations in welded ash flows.
T., Gyarmati, P., Edelstein, O., Rosu, E., Zec, B., 1995. Space U. S. Geol. Surv. Prof. Pap. 354-F, 149 – 159.
and time distribution of Neogene–Quatenary volcanism in the Stokes, S., Lowe, D.J., 1988. Discriminant function analysis of
Carpatho–Pannonian Region. In: Downes, H., Vaselli, O. (Eds.), late quaternary tephras from five volcanoes in New Zealand
Neogene and Related Volcanism in the Carpatho–Pannonian using glass shard major element chemistry. Quat. Res. 30,
Region, Acta Vulcanologica, pp. 15 – 28. 270 – 283.
Póka, T., Zelenka, T., Szakács, A., Seghedi, I., Nagy, G., Simonits, Stokes, S., Lowe, D.J., Froggatt, P.C., 1992. Discriminant
A., 1998. Petrology and geochemistry of the miocene acidic function analysis and correlation of late quaternary tephra
explosive volcanism of the Bükk Foreland; Pannonian Basin, deposits from Taupo and Okataiana volcanoes, New Zealand,
Hungary. Acta Geol. Hung. 41, 437 – 466. using glass shard major element composition. Quat. Int. 13/14,
Reid, F.W., 1983. Origin of the rhyolitic rocks of the Taupo 103 – 120.
Volcanic Zone. J. Volcanol. Geotherm. Res. 15, 315 – 338. Sun, S.S., McDonough, W.F., 1989. Chemical and isotopic
Royden, L.H., Horváth, F., Burchfiel, B.C., 1982. Transform systematics of oceanic basalts: implications for mantle
faulting, extension and subduction in the Carpathian–Pannonian composition and processes. In: Saunders, A.D., Norry, M.J.
region. Geol. Soc. Amer. Bull. 93, 717 – 725. (Eds.), Magmatism in the Oceanic Basins, Geol. Soc. London
Ruiz, J., Patchett, P.J., Arculus, R.J., 1988. Nd–Sr isotope Spec. Publ., vol. 42, pp. 313 – 345.
composition of the lower crustal xenoliths, evidence for the Sylvester, P. (Ed.), 2001. Laser-ablation-ICPMS in the Earth
origin of mid-Tertiary felsic volcanics in Mexico. Contrib. Sciences. Principles and Applications, Mineralogical Associa-
Mineral. Petrol. 99, 36 – 42. tions of Canada Short Course Series, vol. 21.
Sarna-Wojcicki, A.M., Bowman, H.R., Meyer, C.E., Russell, P.C., Szabo, C., Harangi, S., Csontos, L., 1992. Review of neogene and
Woodward, M.J., McCoy, G., Rowe, J.J., Baedecker, P.A., quaternary volcanism of the Carpathian–Pannonian region.
Asaro, F., Michael, H., 1984. Chemical analyses, correlations, Tectonophysics 208 (1–3), 243 – 256.
and ages of Upper Pliocene and Pleistocene ash layers of east- Szakács, A., Zelenka, T., Márton, E., Pécskay, Z., Póka, T., Seghedi,
central and southern California. U. S. Geol. Surv. Prof. Pap. I., 1998. Miocene acidic explosive volcanism in the Bükk
1293, 1 – 40. Foreland, Hungary: identifying eruptive sequences and search-
Sarna-Wojcicki, A.M., Morrison, S.D., Meyer, C.E., Hillhouse, ing for source locations. Acta Geol. Hung. 41, 413 – 435.
J.W., 1987. Correlations of upper cenozoic tephra layers Tari, G., Dfvényi, P., Horváth, F., Dunkl, I., Lenkey, L.,
between sediments of the Western United States and Eastern Stefanescu, M., Szafián, P., Tóth, T., 1999. Lithospheric
Pacific Ocean and comparison with biostratigraphic and structure of the Pannonian basin derived from seismic, gravity
magnetostratigraphic age data. Geol. Soc. Amer. Bull. 98, and geothermal data. In: Durand, B., Jolivet, L., Horváth, F.,
207 – 223. Séranne, M. (Eds.), The Mediterranean Basins: Tertiary
Seghedi, I., Balintoni, I., Szakacs, A., 1998. Interplay of tectonics Extension Within the Alpine Orogen, Geol. Soc. London, Spec.
and neogene post-collisional magmatism in the intracarpathian Publ., vol. 156, pp. 215 – 250.
region. Lithos 45 (1–4), 483 – 497. Taylor, S.R., McLennan, S.M., 1985. The Continental Crust: its
Seghedi, I., Downes, H., Szakacs, A., Mason, P.R.D., Thirlwall, composition and Evolution. Blackwell Sci. Publ., Oxford.
M.F., Rosu, E., Pecskay, Z., Marton, E., Panaiotu, C., 2004. Thorarinsson, S., 1944. Tefrokronologiska studier pa Island. Geogr.
Neogene–Quaternary magmatism and geodynamics in the Ann., 1 – 203.
Carpathian–Pannonian region: a synthesis. Lithos 72 (3–4), Thorarinsson, S., 1974. The terms tephra and tephrochronology. In:
117 – 146. Westgate, J.A., Gold, C.M. (Eds.), World Bibliography and
Shane, P.A.R., 2000. Tephrochronology: a New Zealand case study. Index of Quaternary Tephrochronology. University of Alberta,
Earth-Sci. Rev. 49, 223 – 259. Alberta, pp. xvii – xviii.
Tomek, C., Hall, J., 1993. Subducted continental margin image in laser ablation inductively coupled plasma mass spectrometry:
the Carpathians of Czechoslovakia. Geology 21, 535 – 538. application to quaternary tephro-chronological studies. Appl.
Varga, Gy., 1981. Újabb adatok az fsszesült tufatelepek és Geochem. 9, 323 – 335.
ignimbritek ismeretéhez. MÁFI Évi Jel. 1979-rIl, 499 – 509. Westgate, J.A., Shane, P.A.R., Pearce, N.J.G., Perkins, W.T.,
Wark, D.A., 1991. Oligocene ash flow volcanism, Northern Sierra Korisettar, R., Chesner, C.A., Williams, M.A.J., Acharyya,
Madre Occidental: role of mafic and intermediate-composition S.K., 1998. All toba tephra occurrences across Peninsular India
magmas in rhyolite genesis. J. Geophys. Res. 96, 13389 – 13411. belong to the 75,000 yr B. P. eruption. Quat. Res. 50, 107 – 112.
Westgate, J.A., Gorton, M.P., 1981. Correlation techniques in tephra Wortel, M.J.R., Spakman, W., 2000. Subduction and slab detach-
studies. In: Self, S., Sparks, R.S.J. (Eds.), Tephra Studies. ment in the Mediterranean–Carpathian Region. Science 290,
Reidel, Dordrecht, pp. 73 – 94. 1910 – 1917.
Westgate, J.A., Perkins, W.T., Fuge, R., Pearce, N.J.G., Wintle,
A.G., 1994. Trace element analysis of volcanic glass shards by

Harangi JVGR 2005 143 PDF

Uploaded by

Copyright:

Available Formats

Harangi JVGR 2005 143 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Harangi JVGR 2005 143 PDF

Uploaded by

Copyright:

Available Formats

Journal of Volcanology and Geothermal Research 143 (2005) 237 – 257

Correlation and petrogenesis of silicic pyroclastic rocks in the

* Corresponding author. Tel.: +36 12090555x8355; fax: +36 13812108.

1. Introduction (Froggatt, 1983, Stokes and Lowe, 1988, Shane and

diameter of 40 Am (Fig. 4). The signal recorded by

ArF excimer laser ablation system (MicroLas GeoLas 5. Geochemistry

(Fig. 6; SiO2=75–78.5 wt.%, K2O=4–8 wt.%, K2O/

time-resolved signals of several elements (Fig. 5) to ([La/Yb]cn=3.5 on average, whereas it is 13 to 15 in

You might also like