GEOLOGICA CARPATHICA, FEBRUARY 2005, 56, 1, 9199
Evidence for the Neogene small-volume intracontinental
volcanism in Western Hungary: K/Ar geochronology of the
Tihany Maar Volcanic Complex
KADOSA BALOGH
1*
and KÁROLY NÉMETH
2
1
ATOMKI, Institute of Nuclear Sciences, Debrecen, Hungary; balogh@atomki.hu
2
Geological Institute of Hungary, Stefánia út 14, 1143 Budapest, Hungary; nemeth_karoly@hotmail.com
*Corresponding author
(Manuscript received June 10, 2003; accepted in revised form March 16, 2004)
Abstract: The Tihany Maar Volcanic Complex (TMVC) consists of several eruptive centres and is made up mostly of
pyroclastic rocks. It belongs to the Bakony-Balaton Highland Volcanic Field (BBHVF), which is an extensive Late
MiocenePliocene alkaline basaltic volcanic field in Western Hungary. The TMVC is the only known location in the
BBHVF where volcanic rocks are in a stratigraphically fixed position near the boundary of the Congeria balatonica
Prosodacnomya Zones. Since 1985 this stratigraphic importance motivated repeated efforts to obtain unquestionable
radiometric data with sufficient accuracy for the volcanic phases. Due to the difficulties of dating basaltic pyroclastic
rocks (detrital contamination, excess argon, argon loss during hydrothermal alteration, high atmospheric argon content,
etc.), this is for the first time a fully acceptable age of 7.92±0.22 Ma has been obtained for the onset of volcanic activity
of the TMVC at the location Monks cave. This age is a key datum for the boundary of Congeria balatonica
Prosodacnomya Zones and it agrees well with the start of alkali basaltic volcanic activity in Central Slovakia. 7.35±0.45 Ma
is obtained for Dióstetõ. The youngest ages, showing the greatest argon loss were measured for the location Gödrös. An
analysis of the isochron diagrams suggests here an interval from 6.24±0.73 Ma to 5.92±0.41 Ma for the time of volcanic
activity. This age sequence is in agreement with volcanological field observation and in spite of some uncertainty of the
younger age limit, it is indicated that volcanism at Tihany was not a single event of the same volcano, but rather a result
of longer lived eruptions from a closely spaced, nested volcanic system.
Key words: Pannonian Basin, K/Ar geochronology, phreatomagmatic, scoria, monogenetic, maar, tuff ring, alkaline
basalt.
Introduction
The Tihany Volcano is a maar volcanic complex (Németh et
al. 2001) and belongs to the Bakony-Balaton Highland Volca-
nic Field (BBHVF) considered to be an extensive Neogene in-
tracontinental alkaline basaltic volcanic field in Western Hun-
gary (Fig. 1) (Jugovics 1968, 1969; Jámbor et al. 1981;
Németh & Martin 1999b). This volcanic field consists of
maars, tuff rings, scoria cones, mesa flows as well as long
(km-scale) lava flows (Martin et al. 2003), all characteristic of
an intracontinental so called monogenetic volcanic field
(Connor & Conway 2000). The volcanic landforms are strong-
ly modified, and often eroded back to the level of crater and/or
vent filling pyroclastic and coherent lava facies (Németh &
Martin 1999a; Németh et al. 2003) giving perfect exposures to
study the sub-surface architecture of small-volume intraconti-
nental volcanoes. The remnant of an unusual maar volcanic
complex (Tihany Maar Volcanic Complex TMVC) on the
eastern margin of the BBHVF consists of several eruptive cen-
tres (Németh et al. 2001). Base surge and fallout deposits were
formed during an initial phreatomagmatic explosion, caused
by interaction of water-saturated sediment (Pannonian sand)
and karst water stored in fractures of Mesozoic and Paleozoic
rock units (Németh et al. 2001). The rapidly ascending intra-
plate alkali basalt magma often carried peridotite lherzolite xe-
noliths as well as pyroxene and olivine megacrysts (at locali-
ties Dióstetõ and Gödrös) accumulated in various lapilli tuff
and tuff breccia units often associated with rock units assigned
to be deposited in the final stage of the volcanic eruptions
(Németh et al. 1999). The phreatomagmatic interaction be-
tween ascending magma and complex sources of ground water
resulted in deeply excavated maars, which functioned as local
sediment traps, where scoriaceous tephra washed into and de-
posited, building up Gilbert-type delta sequences (Németh
2001). Maar volcanoes reconstructed in the western and cen-
tral areas at Tihany Peninsula (Fig. 1) are interpreted to have
been formed due to phreatomagmatic explosions of magma
mixed with water saturated clastic sediments (Németh et al.
2001). The unusual east maar had a special combination of
water source from both the porous media aquifer and fracture-
controlled aquifer, the latter one was probably the dominant
supply (Németh et al. 2001).
On the basis of previous K/Ar determinations (Balogh et al.
1982; Balogh et al. 1986; Balogh 1995; Harangi et al. 1995)
TMVC is though to be one of the oldest volcanic erosion rem-
nants in the BBHVF (Balogh et al. 1986) and therefore it bears
www.geologicacarpathica.sk
92 BALOGH and NÉMETH
special significance in understanding the onset of the Neogene
small-volume intracontinental volcanism in Western Hungary.
The TMVC almost entirely consists of pyroclastic rocks, in
Western Hungary this is a unique feature of the TMVC. The
pyroclastic rocks are lapilli tuffs and tuff breccias with a
phreatomagmatic origin. These rocks form the basal rock units
and they are often capped by volcaniclastic units interpreted to
be a result of remobilization of pyroclastic fragments from a
crater rim surrounding a maar basin (Németh 2001). At Tiha-
ny no significant volume of coherent lava has been preserved,
however, feeder dykes are known from the northern part of the
complex (Németh et al. 2001). The basal phreatomagmatic
rock units overlie Pannonian (Late Miocene) shallow marine
(brakish) fluvio-lacustrine marly sand-silt and clay units with an
erosional contact, however, contacts between volcanic and non-
volcanic units are poorly exposed (Müller & Szónoky 1989).
The pyroclastic rock units overlie the Congeria balatonica
Limnocardium decurum Zone (Lóczy 1913; Jámbor 1980,
1989; Müller 1998). Moreover, in a large (1 m size), angular
sedimentary block embedded in the basal phreatomagmatic
Fig. 1. Simplified geological map of the Tihany Peninsula showing
the distribution of volcanic rocks on the surface. SH Strombolian
and/or Hawaiian style explosive eruptive products, ML undif-
ferentiated maar lake sedimentary rocks, PH undifferentiated
phreatomagmatic pyroclastic rocks, B bedding, IMR in-
ferred position of maar rims, A locality Monks cave (Barát-
lakások), B locality Diós, C locality Gödrös. CM, EM and
WM Central, Eastern and Western Maar.
Fig. 2. Large accidental lithic fragments from the basal phreato-
magmatic pyroclastic rock units derived from the pre-volcanic
shallow marine to fluvio-lacustrine Pannonian (Late Miocene) si-
liciclastic rock units. Note the intact bedding of the clast.
Fig. 3. Cauliflower bomb in a massive, accidental lithic rich
phreatomagmatic lapilli tuff. Such bombs have been selected for
K/Ar age determination since their eruption history related to the
magma/water interaction of uprising magma and ground water,
therefore its age inferred to be the age of the formation of the
maar/tuff ring at Tihany.
pyroclastic unit (Fig. 2) Prosodacnomya carbonifera fossil
has been found (Müller & Magyar 1992) strongly suggesting
that volcanic eruption started near to the boundary of Conge-
ria balatonica and Prosodacnomya Zones. Thus, the age for
the Tihany Volcano is a key datum both for the beginning of
the alkaline basaltic volcanic activity of Western Hungary and
for the determination of the age of the boundary of Congeria
balatonica and Prosodacnomya Zones.
Finer-grained pyroclastic material contains detrital contami-
nation that during eruption did not release the previously ac-
cumulated radiogenic Ar, therefore K/Ar ages measured on it
are very uncertain. Reliable ages could be expected only when
using volcanic blocks, bombs (e.g. cauliflower bombs) and
lapilli (Fig. 3) which are free of contamination. Unfortunately,
excess Ar and also Ar loss have been detected even in these
samples, Ar loss is likely to be caused by hydrothermal alter-
ation. The aim of the present study was to define a more accu-
rate and reliable K/Ar isotopic age for the eruption of the Ti-
K/Ar GEOCHRONOLOGY OF THE TIHANY MAAR VOLCANIC COMPLEX 93
hany Volcano. This purpose has been achieved by applying
the isochron method to fractions produced from a single piece
of rock and using the criteria for checking the reliability of K/Ar
isochron ages by the method elaborated for the neck of o-
moka-hill (Nógrád (Novohrad) Southern Slovakia Volca-
nic Field, on the border between Hungary and Slovakia) and
the refined techniques for producing fractions from basaltic
rocks (Balogh et al. 1994).
Methodology and problems in interpretation of the
data
During the past decades intensive geochronological re-
searches on young alkaline basaltic rocks from the Pannonian
Basin has confirmed that K/Ar data on these rocks mostly give
the correct geological age and the most frequent error is
caused by the presence of excess Ar (Balogh et al. 1981;
Balogh et al. 1986; Borsy et al. 1986). Even when excess Ar
was detected, the real geological age could be obtained by ap-
plying the isochron methods as introduced by McDougall et
al. (1969) and Harper (1970) and analysed by Shafiqullah &
Damon (1974) and Hayatsu & Carmichael (1977). In spite of
the successes of isochron methods it still remained impossible
to measure reliable age on a part of the basaltic rocks. The rea-
sons for the occasional failures were variable and included:
1 It is very difficult to distinguish real isochrons and
mixing lines, especially when samples with similar K con-
tents are plotted in the isochron diagram (Hayatsu & Car-
michael 1977; Bowen 1988). It is usually difficult to collect
samples with highly variable K concentrations from a single
basalt body. Therefore the isochron methods were applied as
suggested by Fitch et al. (1976) using fractions of a single piece
of rock for fitting the isochron. These fractions are not monom-
ineralic, but differ in their mineral composition. The successful
separation, that is the production of fractions with highly differ-
ing K content, depends on the texture of the rock and also on
applying the optimum process for preparing the fractions.
2 Isochrons fitted to samples with remarkable differences
in their K content can still be erroneous, if there is a correla-
tion between the K and excess Ar content of the rocks. This
error can be recognized and corrected by using fractions se-
lected according to their atmospheric argon concentrations
(Balogh et al. 1994).
3 It is very difficult, in most cases impossible, to date py-
roclastic rocks of phreatomagmatic origin such as the volu-
metrically largest rock unit in Tihany. During especially
phreatomagmatic explosive eruption the fragmented lava mix-
es with disrupted accidental lithic clasts or minerals derived
from deposits where magma intruded. The fast explosive erup-
tion and cooling does not allow the resetting of the K-Ar clock
efficiently, a great part of the radiogenic argon will be retained
in the detritus. This unreliability can be recognized by highly
scattering ages and random distribution of the fractions in the
isochron diagrams. In addition, the permeability of the pyro-
clastic rock units promotes water circulation and alteration of
the volcanic material.
Reliable radiometric dating of pyroclastic rocks could be
expected only, if larger blocks or bombs were available for
dating. These samples must have been completely molten be-
fore eruption and free of macroscopic detrital material.
Dating the pyroclastics of the Tihany Volcano has been re-
peatedly attempted since 1985 (Table 1), but these attempts
were only partly successful. Either the poor accuracy or the
questionable geological reliability of age data limited the val-
ue of the previous efforts. Because of these uncertainties only
7 K/Ar ages and an isochron age of 7.56±0.50 Ma for the on-
set of volcanic eruption has been published up to now in a
widely accessible form and with proper discussion (Balogh et
al. 1986), but, in the light of later results, the interpretation
even of these ages has to be revised now.
The published isochron age (7.56±0.50 Ma, Balogh et al.
1986) was not satisfactory: the error was too large and the K
content varied in the samples only from 1.53 % to 1.87 %.
Thus, there was a chance that our isochron is only a mixing
line and the real age is significantly younger. Efforts to deter-
mine an accurate and convincing age for the Tihany Volcano
were renewed after the criteria for testing the reliability of iso-
chron ages were elaborated (Balogh et al. 1994). These new
efforts failed to improve the previous results because a part of
the newly collected samples proved to be altered and con-
tained excess argon in an irregular distribution. The results
were published only in a conference abstract (Harangi et al.
1995) and in a manuscript (Balogh 1995).
Here we present the first set of data that allow to assign a re-
liable and sufficiently accurate age to the first eruption of the
Tihany Volcano. At the same time, this is the best datum for
the onset of alkaline basaltic volcanism of the BBHVF. In the
light of the new result a short evaluation of previous, mostly
unpublished age data will be given too.
K/Ar geochronology of Tihany Volcano
The first K/Ar age determinations of the alkaline basalts in
the BBHVF aimed to determine the relationship between the
age of the volcanism and the timing of the Pannonian shallow
marine to fluvio-lacustrine sedimentation (Balogh et al. 1982,
1986; Borsy et al. 1986). The first K/Ar ages on the Tihany
Volcano were published by Balogh et al. (1986) in a review
summarizing the results and experiences of K/Ar dating the
post-Sarmatian alkaline basalts in the Carpathian basin. The
first 7 K/Ar ages and an isochron age measured on the basal
phreatomagmatic alternating lapilli tuff and tuff units of the
Tihany Volcano, sampled at the Monks cave (Table 1) have
also been published in this paper. The analytical ages varied
from 9.73 Ma to 7.35 Ma and the isochron age was
7.56±0.50 Ma.
K/Ar ages from the BBHVF suggested that the Tihany Vol-
canic Complex may be the earliest manifestation of the Neo-
gene intracontinental monogenetic volcanism in Western
Hungary. These preliminary results confirmed that further re-
finement and discussion of obtained age data from Tihany is
desirable.
The oldest ages in the BBHVF were obtained from a basal-
tic neck (Ragonya at Mencshely, approximately 10 km away
from Tihany, 7.92±0.33 Ma) and on the oldest eruption of the
Tihany Volcano (at Monks cave, 7.56±0.50 Ma) measured
94 BALOGH and NÉMETH
from a volcanic bomb, a block and their rock fractions from
the basal phreatomagmatic lapilli tuff units (Balogh et al.
1986). Unfortunately, these isochron ages were obtained from
rocks and rock fraction with insufficiently differing K concen-
trations, therefore the age of eruption and also the datum for
the beginning of basaltic volcanism of the BBHVF remained
questionable.
These initial measurements were followed by repeated ef-
forts in order to improve the accuracy and increase the reliabil-
ity of ages of the Tihany Volcano. Dating has continued with
measurement of large, often vesicular, fluidally shaped coher-
ent lava bombs interpreted as lava spatters (Németh et al.
1999) from Dióstetõ (Balogh et al. 1985). Three whole rock
samples (No. 1347, 1349, 1350 in Table 1) and 1 fraction de-
fined an isochron age of 7.35±0.45 (0.1) Ma and the K con-
tent in these samples varied from 0.80 to 1.85 % (Fig. 4). The
error of this isochron age is defined by the errors of individual
measurements, and the error given in parentheses is calculated
from the scatter of points around the straight line. The good fit
of points may indicate an overestimation of individual errors,
but could also be only casual. This datum was in line with the
previous results, but the error of age could not be reduced sig-
nificantly. Some other uncertainties also remained: the isoch-
ron was obtained on different samples, so the assumption of
uniform isotopic composition for the initial Ar remained ques-
tionable and, due to the greater atmospheric argon content in
these samples, the error of individual age data was also great-
er, therefore attempts to determine an accurate and reliable age
for the Tihany Volcano have been given up for a time.
Efforts on dating the Tihany Volcano were resumed only af-
ter recognizing the criteria for checking the reliability of K/Ar
isochron ages. Most reliable isochron ages are those measured
on fractions with 1 highly differing K content, 2 similar
and low atmospheric argon concentration (Balogh et al. 1994).
Encouraged by the successes of this method new samples
have been collected from the Monks cave (cauliflower bombs
from phreatomagmatic lapilli tuff), Dióstetõ (lava spatter) and
Gödrös (fluidal vesicular spindle bomb). The last location is
thought to represent the youngest eruption of the Tihany Vol-
cano on the basis of the general volcanic field relationships
(Németh et al. 1999; Németh et al. 2001).
For all samples from the pyroclastic rocks from the Monks
cave, excluding only one sample with obvious excess Ar con-
tent, an isochron age of 7.80±1.07 (0.38) Ma was obtained,
where 1.07 Ma is the deviance and 0.38 Ma is the error. The
poor fit of points to the straight line shows that conditions,
which allow the use of the isochron method, are not met prop-
erly. In order to preclude the possibility of a significantly
younger age, it has been carefully tested, if the isochron age
could be caused by the correlation of excess Ar and K. Frac-
tions with remarkably different K concentrations were pre-
pared from sample No. 3381 from the same locality. The fitted
line indicates again a similarly old age: 7.91±1.01 (0.65) Ma.
The plot of atmospheric Ar against K concentration does not
Fig. 4. Isochron of the coherent lava fragment of spatter cone from
the Dióstetö (C locality in Fig. 1) volcanic sequence.
No.
Dated
fraction
K % 40Ar
rad
10
7
cm
3
/g
40Ar
rad
%
Age
Ma (±1ó) Ref.
Samples from Monks caves
958
w. r.
1.79
5.852
0.52
8.40±0.36 1
958
M
1
1.84
5.524
0.36
7.72±0.38 1
958
M
3
1.87
5.428
0.65
7.46±0.30 1
1000
w. r.
1.75
5.796
0.25
8.51±0.45 1
1000
M
1
1.53
5.797
0.19
9.73±0.75 1
1000
M
2
1.76
5.686
0.20
8.30±0.61 1
1000
D
1
1.79
5.120
0.17
7.35±0.64 1
3378
w. r.
1.72
4.647
0.15
6.92±0.65 4,3
3379
w. r.
1.78
5.970
0.24
8.60±0.55 3,4
3380
w. r.
2.06
6.234
0.48
7.78±0.33 3,4
3382
w. r.
1.81
5.868
0.34
8.31±0.42 3,4
3442
w. r.
1.96
5.646
0.42
7.41±0.34 4
3443
w. r.
1.65
6.078
0.42
9.43±0.43 4
3381
w. r.
2.04
5.379
0.33
6.78±0.35 3,4
3381
D
1
M
2
2.39
6.927
0.33
7.44±0.38 4
3381
D
1
M
4
2.15
6.820
0.82
8.14±0.31 4
3381
D
2
M
1
0.82
3.305
0.16
10.34±0.90 4
3381
D
2
M
2
0.91
3.118
0.41
8.78±0.40 4
3381
D
2
M
3
1.11
3.832
0.52
8.83±0.37 4
3381
D
2
M
4
2.26
6.036
0.35
6.86±0.34 4
3381
D
2
M
5
2.17
7.632
0.38
9.03±0.39 4
Samples from Dióstetõ
1347
w. r.
0.80
2.305
0.12
7.40±0.86 2
1350
w. r.
1.20
3.647
0.053
7.83±1.45 2
1349
w. r.
1.77
5.117
0.21
7.42±0.52 2
1349
M
1
0.22
1.249
0.26
13.00±1.00 2
1349
M
2
1.85
5.394
0.27
7.49±0.44 2
3344
w. r.
1.40
4.914
0.078
9.03±1.64 4,3
3345
w. r.
1.85
5.065
0.124
7.03±0.73 4,3
3346
w. r.
1.77
3.880
0.075
5.64±1.04 4
3347
w. r.
1.55
3.867
0.26
6.39±0.39 4,3
3445
w. r.
1.72
4.427
0.38
6.62±0.31 4
3445
M
3
1.77
4.551
0.24
6.60±0.31 4
3445
D
1
1.62
4.648
0.13
7.37±0.79 4
Samples from Gödrös
3341
w. r.
0.88
1.802
0.21
5.24±0.39 4,3
3342
w. r.
1.52
3.349
0.25
5.64±0.36 4,3
3342
M
1
0.115
1.172
0.074 26.0 ±5.5
4,3
3342
M
2
1.57
3.385
0.358
5.53±0.27 4,3
3342
M
3
1.66
3.720
0.452
5.77±025
4,3
3343
w. r.
1.35
3.191
0.186
6.07±0.47 4,3
Table 1: Previous K/Ar ages on the Tihany Volcano.
References: 1 Balogh et al. 1986, 2 Balogh et al. 1985, 3 Ha-
rangi et al. 1995, 4 Balogh 1995.
K/Ar GEOCHRONOLOGY OF THE TIHANY MAAR VOLCANIC COMPLEX 95
show correlation, which is an argument for the reality of the
age. As an additional test an isochron has been fitted to the
fractions with low (
36
Ar<3×10
9
cm
3
/g) atmospheric argon
content. An age of 7.93±1.00 (0.55) Ma has been obtained.
These investigations proved for the first time the reality of the
old age of Tihany, but the unfavourable character of the sam-
ples collected from the phreatomagmatic pyroclastic rocks pre-
vented the determination of an accurate age (Balogh 1995).
New volcanological observations suggested that lava spatter
deposits at Dióstetõ might be the results of a younger volcanic
phase in the evolution of TMVC than the basal phreatomag-
matic unit (Harangi et al. 1995; Németh et al. 1999; Németh et
al. 2001), this did not confront with the K/Ar ages
(7.35±0.45 Ma) published first for this deposit (Balogh et al.
1986). The new attempt to improve dating of Dióstetõ was un-
successful. The K concentration in the samples collected in
1995 did not show great differences, their atmospheric Ar
content was mostly too large, and the too young ages of sam-
ples No. 3346 and 3347 indicated greater partial loss of ra-
diogenic argon.
Due to its relatively high K concentration and radiogenic ar-
gon enrichment, sample No. 3445 has been selected for frac-
tionation. However, because of the homogeneity of this sam-
ple no fractions with different K concentrations were obtained.
Omitting the sample with too great an excess of Ar content
(No. 1349 M
1
) and those with too much Ar loss (No. 3346 and
3347), the fitted line defines an age of 6.64±0.71 Ma. This
means that newly collected samples, most likely due to their
more altered character, yielded a less valuable age, which was
obtained previously by Balogh et al. (7.35±0.45 Ma, Balogh
et al. 1985).
The youngest phase of volcanic activity is represented by
the basaltic spindle bomb bearing lapilli tuff and tuff breccia
at Gödrös (Fig. 1) (Németh et al. 1999, 2001). The precise dat-
ing of this occurrence would help to establish the duration of
volcanic activity. Mostly young ages have been obtained fall-
ing between 6.07 Ma and 5.24 Ma, however, one fraction
gave too old an age due to the great amount of excess Ar or
detrital contamination. The younger ages define an isochron
age of 5.92±0.41 Ma (Fig. 5a), which is close to the oldest an-
alytical age of 6.07±0.47 Ma. This suggests that at Gödrös
younger ages may be caused by Ar loss. Plotting the 5 young-
er ages in the Ar(rad)-K diagram, an age of 6.24±0.73 Ma is
obtained (Fig. 5b), this also suggests partial loss of radiogenic
Ar at Gödrös, therefore an age older than about 6 Ma can be
accepted for the pyroclastic rocks at Gödrös. However, con-
trasting the ages measured for the locations Monks cave and
Dióstetõ, ages older than 7 Ma (disregarding the highly con-
taminated fraction 3342 M
1
) are missing: this is another ar-
gument for assuming a longer time span for the volcanism of
the TMVC.
Summarizing the previous chronological work on the Tiha-
ny Volcano, it has been proven that volcanic activity started
before 7 Ma (Table 1), however, a more accurate datum was
still missing. This fact, together with the Ar loss characterizing
the samples from the lapilli tuff at Gödrös, also prevented the
estimation of the duration of volcanic activity. A renewed age
determination survey therefore needed to give a more accurate
and reliable datum for the start and the length of the volcanic
activity. The refined techniques of preparing fractions for K/Ar
dating from the coherent basalt lavas as well as the careful
sampling with insignificant excess Ar content with minimum
degree of alteration made a new K/Ar survey promising.
New K/Ar data from Tihany as a refinement of the
old ages
Methods
New dating has been performed on a part of a basalt block
of 100150 mm size collected from the locality Monks cave.
The size fraction of 0.1250.063 mm has been chosen for pro-
Fig. 5.
40
Ar/
36
Ar-K/
36
Ar (a) and
40
Ar(rad)-K diagram (b) of sam-
ples from Gödrös.
96 BALOGH and NÉMETH
Fig. 6. Isochron of the 1000-C sample derived from the phreatomag-
matic rock units of the Monks cave (Barátlakások A locality in
Fig. 1.)
ducing a set of magnetic and density fractions. The size-frac-
tion was first washed, treated with 20% acetic acid to remove
calcium carbonate and washed again. Density fractions were
produced by using tetrabromoethane and diluted methylene
iodide, so that D
1
<2.94g/cm
3
<D
2
<3.05g/cm
3
<D
3
. Three
magnetic fractions (marked with M
i
, where greater i shows
greater magnetic susceptibility) were produced from each den-
sity fractions using first a permanent magnet and running the
samples repeatedly on a magnetic separator.
The unseparated size fraction (w.r.) and 8 D
i
M
j
fractions
were dated, results are shown in Table 2. In line with our gen-
eral observation (Balogh et al. 1986) excess argon causes the
greatest age increase of the densest and least magnetic frac-
tion, in which olivine is concentrated. Figure 6 shows that
D
3
M
1
is the only point not fitting the straight line. Omitting
D
3
M
1
an isochron age of 7.92±0.22 Ma is defined by the rest
of the points with an initial
40
Ar/
36
Ar ratio of 299.8±5.2, so
that a significant amount of excess argon is not indicated. K
content in the used fractions ranges from 0.49 % to 2.65 %,
thus, the possibility that our isochron is only a mixing line is
negligible. K and atmospheric argon do not correlate, there-
fore this isochron age is regarded as the most likely datum for
the eruption of the Tihany Volcano (Fig. 6).
Discussion on the timing of volcanism and the
duration of volcanic activity at Tihany
Three main volcanic stratigraphic units have been identified
at Tihany on the basis of textural characteristics, field relation-
ships as well as areal distribution of volcanic rocks (Németh et
al. 1999, 2001). These rocks represent erosion remnants of vol-
caniclastic rock units, lava spatter accumulation zones as well as
feeder dykes all associated with at least three maars and a
Strombolian scoria cone (Németh et al. 1999, 2001). The identi-
fication of strong negative gravity anomalies in three well-dis-
tinguished areas at Tihany is in good agreement with the loca-
tion of reworked volcaniclastic rock units inferred to be part of
Gilbert-type delta fronts built in a volcanic depression, presum-
ably a maar (Benderné et al. 1965; Németh 2001; Németh et al.
2001) (Fig. 1). In two areas, maar lake Gilbert-type delta front
deposits cover alternating, well-bedded, accidental lithic clast-
rich phreatomagmatic lapilli tuffs and tuffs. These rocks exhibit
textural features for radial transportation direction from the cen-
tral areas of the Tihany Peninsula. The presence of the large
Dated
fraction
K
%
40Ar
rad
10
7
cm
3
/g
40Ar
rad
%
40
Ar/
36
Ar
36
Ar
10
9
cm
3
/g
K/
36
Ar
[%/(10
9
cm
3
/g]
Age
Ma (±1ó)
w. r.
1.89
5.92
28.9
415.6
4.93
0.383
8.03±0.45
D
1
M
1
2.31
7.72
9.24
325.4
25.6
0.090
8.56±1.29
D
1
M
2
2.31
8.19
13.2
340.4
18.2
0.127
9.09±0.97
D
1
M
3
2.65
8.20
44.7
534.4
3.43
0.773
7.95±0.35
D
2
M
2
1.76
5.58
48.3
571.6
2.02
0.872
8.13±0.35
D
2
M
3
2.02
6.28
52.4
620.8
1.93
1.044
7.99±0.33
D
3
M
1
0.449
1.62
29.4
418.6
1.32
0.333
9.48±0.52
D
3
M
2
0.49
1.61
32.0
434.6
1.15
0.429
8.36±0.44
D
3
M
3
1.14
3.78
30.0
422.1
2.98
0.383
8.50±0.46
Table 2: K/Ar ages on fractions of basalt No. 1000-C from Monks cave (Barátlakások), Tihany.
negative anomaly, and the uniform characteristics of this basal
phreatomagmatic unit indicate that there must be a central maar,
erupted first, which produced extensive pyroclastic sheets
around the maar basin (Central Maar) (Németh et al. 2001). Pre-
viously it has been interpreted that the samples collected for the
K/Ar age survey from the Monks cave area have been derived
from this initial unit, so from the stratigraphical point of view
their age must represent the time of onset of volcanism at Tiha-
ny. There are two other gravity anomaly zones and accompa-
nied capping maar lacustrine sequence indicating that there are
two other maars post-dating the central maar (Németh et al.
2001). The presence of a large amount of scoriaceous detritus in
volcanic rock units inferred to represent gravity driven mass
flow deposits in the maar lakes suggests, that a source zone, a
ready to be eroded scoria cone, must have existed close to the
central areas at Tihany. In this stratigraphical framework it is in-
ferred that the oldest maars are the Central and East Maars. The
West Maar is inferred to be the youngest phreatomagmatic vent
(Németh et al. 2001). Magmatic explosive activity post-dates
production of each maar.
K/Ar GEOCHRONOLOGY OF THE TIHANY MAAR VOLCANIC COMPLEX 97
The present K/Ar age survey, which measured cauliflower
bombs, vesicular spindle-shape lava bomb and/or spatter as
well as co-genetic volcanic lithic fragments gave two distinct
age groups in good concert with the volcanic stratigraphical
position of the host rock units; A 7.92±0.22 for Monks
cave and 7.35±0.45 for Dióstetõ, and B fro m
6.24±0.73 Ma to 5.92±0.41 Ma range for the minimum age of
eruption of Gödrös. This age data distribution is in good
agreement with the identified volcanic stratigraphy.
Most of maar-diatreme volcanoes are the phreatomagmatic
equivalent of scoria cones and their lava flows and thus may
have been active for days, weeks, months and perhaps up to
1015 years (Vespermann & Schmincke 2000; Walker 2000;
Lorenz 2003). Similar to scoria cones maar-diatreme volca-
noes grow in size the longer their phreatomagmatic activity
lasts (Kienle et al. 1980; Vespermann & Schmincke 2000;
Lorenz 2003). Short-lived maar-diatreme volcanoes have a
small maar crater and thus can serve as a small depot centre
whereas longer-lived maar-diatreme volcanoes have a larger
maar crater and thus can serve as a larger and deeper depot
centre for a much longer period of time (Lorenz 1986, 2003).
From historic observations scoria cones are active for days,
weeks, months, or years. Paricutin in Mexico erupted for 9
years (20.2.19434.3.1952) (Luhr & Simkin 1993). In 1759
the scoria cone Jorullo erupted in the neighbourhood of Pari-
cutin and after 15 years (17591775) of activity it reached a fi-
nal height of 350 m and its associated lava field reached
1.25 km
2
in size (Luhr & Simkin 1993). The activity of most
scoria cones is over within one year (Luhr & Simkin 1993).
In historic time only a few maar-diatreme volcanoes erupt-
ed, thus there is a very little information on the duration of
maar volcanism. The two Ukinrek Maars erupted in 1977 for 3
and 8 days, respectively (Kienle et al. 1980; Self et al. 1980;
Büchel & Lorenz 1993; Ort et al. 2000). Ukinrek West Maar
erupted for 3 days and finally had a 170 m wide (rim to rim)
and 30 m deep maar crater (Kienle et al. 1980). Its tephra ring
had a maximum thickness of 10 m (Kienle et al. 1980). In
1954 the Nilahue Maar erupted in Chile during almost half a
year (Müller & Veyl 1956; Illies 1959), however, the active
phase of eruption varies according to different authors. The
main eruptive phase ended after 10 days (Illies 1959) produc-
ing a maar crater 300 m in diameter. In summary it can be con-
cluded that maar-diatreme volcanoes, similarly to scoria
cones, may be active for days to months, perhaps in exception-
al cases up to more than 10 years (Lorenz 2003). In this re-
spect the well-distinguished age groups from the newly ob-
tained K/Ar ages indicate that volcanism at Tihany was not a
single event of the same volcano, but rather a result of a longer
lived eruption from a closely spaced, nested volcanic system.
The identified stratigraphical changes in the pyroclastic rock
units at Tihany clearly demonstrate a gradual transition from
phreatomagmatic to magmatic explosive fragmentation of up-
rising alkaline basaltic magma (Németh et al. 2001). These
changes seem to coincide with the age variation identified on
the basis of the new K/Ar age determination survey. However,
the errors of the individual K/Ar age data are too large to dem-
onstrate clearly the separation of volcanic events at Tihany,
though their value is good enough to show that volcanism was
a longer lived event at Tihany, at least longer-lived than the
length of a single volcanic eruption of a common monogenet-
ic intracontinental volcano (tuff ring, maar, scoria cone).
The newly obtained and reconfirmed K/Ar age data from
Tihany clearly suggest that other old (older than 5 Ma) age
data obtained from the eastern part of the BBHVF might also
be true geological ages, and further researches may need to re-
fine the timing of the Neogene volcanism. At Tihany, as the
oldest manifestation of the Neogene intracontinental volcan-
ism in Western Hungary, field evidence as well as the textural
characteristics of the volcaniclastic rock units indicate that
these rocks have been derived from volcaniclastic deposits
that accumulated in subaerial conditions. However, deposi-
tion has occurred in a fluvio-lacustrine basin with a large
quantity of surface water and near-surface water-saturated si-
liciclastic sediments (Németh et al. 2001; Martin et al. 2003).
The age data from juvenile lithic clasts separated from basal
phreatomagmatic lapilli tuff units as well as its capping strati-
graphical position in relationship to the Pannonian (Late Mi-
ocene) siliciclastic units indicate that sedimentation in the
Pannonian Lake must have been ended by this time in this re-
gion. This interpretation is in good concert with other consid-
erations on the basis of paleontological evidence (Magyar et
al. 1999). However, the age of the basal phreatomagmatic
units does not imply that the Pannonian lacustrine sedimenta-
tion lasted up to this date, because the contact between the
basal phreatomagmatic rock units are rather erosional, discon-
tinuous (Németh et al. 2001). The pyroclastic rocks have been
interpreted as representing near vent, and/or intracrater as well
as diatreme filling rock units, which cut through the pre-vol-
canic stratigraphy. Therefore the obtained ages should be
viewed as a minimum value of the finishing stage of the Pan-
nonian lacustrine sedimentation at Tihany.
In respect to evaluation of the value and validity of the newly
obtained age data, it is inferred that sample collection is abso-
lutely critical. In general age determination on pyroclastic rocks
of phreatomagmatic origin is difficult, and carries significant
possibilities of errors. In a volcanic field, where volcanism is
long-lived and/or sills and dykes may have intruded the pre-
volcanic rock formations, a potential error is always present.
The separation, and/or differentiation between syn-volcanic ju-
venile lithic clasts from volcanic accidental lithics from the
same co-genetic feeding system is difficult. This problem may
cause a greater scatter of the ages and older ages may be ob-
tained for certain units than they are. In this work we tried to re-
duce this risk by selecting carefully samples showing clear
signs of cauliflower and/or spindle texture and common textur-
al relationship with other similar clasts in the same beds.
Conclusion
It has been shown that reliable K/Ar ages on basaltic pyro-
clastic rocks can be determined, if 1 blocks and/or bombs
coeval with the eruption are used for dating, 2 the isochron
method is applied to fractions of a single piece of rock (Fitch
et al. 1976) and, 3 if the criteria elaborated for checking the
reliability of isochron ages are fulfilled (Balogh et al. 1994).
An isochron K/Ar age of 7.92±0.22 Ma has been deter-
mined on a whole rock sample and its fractions for the onset
98 BALOGH and NÉMETH
of eruption of the Tihany Maar Volcanic Complex at the local-
ity Monks cave (Barátlakások). This age meets all the re-
quirements worked out checking the reliability of K/Ar isoch-
ron ages. It is a key datum for the boundary of the Congeria
balatonica and Prosodacnomya Zones and it is very close to
the start of alkali basaltic volcanic activity in Central Slovakia
(Koneèný et al. 1999). An isochron age of 7.35±0.45 Ma has
been obtained for the basalt at Dióstetõ. This younger age is in
agreement with volcanological field observation. However,
the isochron is less convincing being fitted mostly to whole
rock samples. An analysis of isochron and analytical ages sug-
gests an interval from 6.24±0.73 Ma to 5.92±0.41 Ma for the
youngest volcanic phase at Gödrös. Although the rejuvenating
effect of post-volcanic hydrothermal alterations is strongest at
this locality, the lack of analytical ages older than 7 Ma (in
contrast to the basalts at Monks cave and Dióstetõ) suggests
that there is a significant age difference (at least a longer time
than it requires to solidify an average feeder dyke of a small-
volume intracontinental volcano) between the phases of volca-
nic eruption of the Tihany Maar Volcanic Complex. This im-
plies that volcanism at Tihany was not a single event of the
same volcano, but rather a result of a longer lived eruption
from a closely spaced, nested volcanic system. Thus Tihany is
another good example from Western Hungary to demonstrate
that small-volume intracontinental volcanoes tend to form
nested volcanic systems where eruption may recur more-less
in the same site after significant time delay similarly to other
sites where such delays have been clearly demonstrated (Mar-
tin & Németh 2002a,b; Martin et al. 2003).
Acknowledgments: Financial support for this research was
provided by the Hungarian Science Foundation (OTKA
T 029897, T 043344 granted to K. Balogh. and OTKA
F 043346 granted to K. Németh). Critical reviews and con-
structive suggestions by Dr. Stanislaw Ha³as and Dr. Ulrike
Martin Journal reviewers as well as Dr. Jaroslav Lexa
Journal Editor are appreciated. Thanks are also due to Dr.
Zoltán Pécskay (ATOMKI, Debrecen), Dr. Pál Müller (MÁFI,
Budapest) for their support in various stages of this research.
References
Balogh K. 1995: K/Ar study of the Tihany Volcano, Balaton High-
land, Hungary. Report on the work supported by the European
Community in the frame of program Integrated Basin Studie.
Institute of Nuclear Research, Hungarian Academy of Sci., De-
brecen, 112.
Balogh K., Miháliková A. & Vass D. 1981: Radiometric dating of
basalts from Southern and Central Slovakia. Západ. Karpaty,
Sér. Geol. 7, 113126.
Balogh K., Jámbor A., Partényi Z., Ravaszné Baranyai L. & Solti G.
1982: K/Ar radiogenic age of Transdanubian basalts. MÁFI
Ann. Rep. on 1980, 243259 (in Hungarian).
Balogh K., Árváné Sós E. & Pécskay Z. 1985: K/Ar dating of mag-
matic rocks. Report on contract No. 4212/85 to the Hungarian
Geological Institute by the Institute of Nuclear Research of
HAS. MS, Archives of Hung. Geol. Inst., Budapest, and the
Inst. of Nucl. Res. of HAS, Debrecen, 121.
Balogh K., Árva-Sós E., Pécskay Z. & Ravasz-Baranyai L. 1986: K/
Ar dating of post-Sarmatian alkali basaltic rocks in Hungary.
Acta Mineral. Petrogr. (Szeged) 28, 7594.
Balogh K., Vass D. & Ravasz-Baranyai L. 1994: K/Ar ages in the
case of correlated K and excess Ar concentrations: A case
study for the alkaline olivine basalt of omoka, Slovak-Hun-
garian frontier. Geol. Carpathica 45, 2, 97102.
Benderné K., Böjtösné V. & Reményi G. 1965: Geological mapping,
geomagnetic and gravimetric studies around the Geomagnetic
Observatory at Tihany. MÁELGI Geofiz. Közlem. 15, 14 (in
Hungarian).
Borsy Z., Balogh K., Kozák M. & Pécskay Z. 1986: Contributions to
the evolution of the Tapolca-basin, Hungary. Acta Geogr. De-
brecina 23, 79104 (in Hungarian).
Bowen R. 1988: Isotopes in the Earth Sciences. Elsevier Applied
Sci., London, New York, 1647.
Büchel G. & Lorenz V. 1993: Syn- and post-eruptive mechanism of
the Alaskan Ukinrek maars in 1977. In: Negendank J.F.W. &
Zolitschka B. (Eds.): Paleolimnology of European Maar Lakes.
Springer-Verlag, Berlin, Heidelberg, 49, 1560.
Connor C.B. & Conway F.M. 2000: Basaltic volcanic fields. In: Sig-
urdsson H. (Ed.): Encyclopedia of Volcanoes. Academic Press,
San Diego, 331343.
Fitch F.J., Miller J.A. & Hooker P.J. 1976: Single whole rock K-Ar
isochrons. Geol. Mag. 111, 110.
Harangi S., Németh K. & Balogh K. 1995: Volcanology and chro-
nology of the Tihany Volcano, Balaton Highland (Pannonian
Basin, Hungary). 10
th
Congress of Regional Committee on
Mediterranean Neogene Stratigraphy, Bucharest, Romania. Ro-
manian J. Stratigr. 76, 1921.
Harper C.T. 1970: Graphical solution to the problem of radiogenic
argon-40 loss from metamorphic minerals. Eclogae Geol. Helv.
63, 119140.
Hayatsu A. & Carmichael C.M. 1977: Removal of atmospheric ar-
gon contamination and the use and misuse of the K-Ar isochron
methods. Canad. J. Earth Sci. 14, 337345.
Illies J.H. 1959: Die Entstehungsgeschichte eines Maares in Süd-
Chile (ein aktualgeologischer Beitrag zum Problem des Maar-
Vulkanismus). Geol. Rdsch. 48, 232247.
Jámbor A., Partényi Z. & Solti G. 1981: Geological characteristics
of the Transdanubian basaltic volcanic rocks. MÁFI Ann. Rep.
on 1979, 225239 (in Hungarian).
Jámbor Á. 1980: Pannonian in the Transdanubian Central Moun-
tains. Ann. Geol. Inst. Hung. 65, 1259.
Jámbor Á. 1989: Review of the geology of the s.l. Pannonian For-
mations of Hungary. Acta Geol. Acad. Sci. Hung. 32, 269324.
Jugovics L. 1968: Basalt- und Basalttuffgebiete Ungarns. MÁFI
Ann. Rep. on 1967, 7582 (in Hungarian).
Jugovics L. 1969: Geological characteristics of the basalt lands at
the Balaton Highland and in the Tapolca Basin. MÁFI Ann.
Rep. on 1968, 223243 (in Hungarian).
Kienle J., Kyle P.R., Self S., Motyka R.J. & Lorenz V. 1980: Unin-
rek Maars, Alaska, 1. April 1977 eruption sequence, petrology,
and tectonic settings. J. Geophys. Res. 7, 1137.
Koneèný V., Lexa J. & Balogh K. 1999: Neogene-Quaternary alkali
basalt volcanism in Central and Southern Slovakia (Western
Carpathians). Geolines 9, 6775.
Lóczy L. sen. 1913: Geological units of the Balaton area and their
stratigraphy. In: Lóczy L. sen. (Ed.): New results of the scien-
tific research of the Balaton. Magy. Királyi Földt. Intéz. (Roy.
Hung. Geol. Inst.), Budapest, I/I, 617 (in Hungarian).
Lorenz V. 1986: On the growth of maars and diatremes and its rele-
vance to the formation of tuff rings. Bull. Volcanol. 48, 265274.
Lorenz V. 2003: Syn- and post-eruptive processes of maar-diatreme
volcanoes and their relevance to the accumulation of post-erup-
tive maar crater sediments. Földt. Kutatás (Quart. J. Geol.
Surv., Hung.), (in press).
Luhr J.F. & Simkin T. 1993: Paricutin. The volcano born in a Mexi-
K/Ar GEOCHRONOLOGY OF THE TIHANY MAAR VOLCANIC COMPLEX 99
can cornfield. Geosciences Press, Phoenix, 1427.
Magyar I., Geary D. & Müller P. 1999: Paleogeographic evolution
of the Late Miocene Lake Pannon in Central Europe. Palaeo-
geogr. Palaeoclimatol. Palaeoecol. 147, 151167.
Martin U. & Németh K. 2002a: Magma wet sediment interaction
in a crater lake of a tuff ring, developed in a pyroclastic mound
dammed valley: Kissomlyó volcano (Western Hungary). Proc.
Amer. Geophys. Union Chapman Conference on Explosive
Subaqueous Volcanism, Dunedin, New Zealand, January 21
25, 2002, 137.
Martin U. & Németh K. 2002b: Peperitic lava lake-fed intravent sills
at Ság-hegy, western Hungary: a complex interaction of wet te-
phra ring and lava in a phreatomagmatic volcanic complex. In:
Breitkreuz C., Mock A. & Petford N. (Eds.): First International
Workshop: Physical Geology of Subvolcanic Systems Lac-
coliths, Sills, and Dykes (LASI). Wiss. Mitt. Inst. Geol.
(Freiberg) 20, 3334.
Martin U., Auer A., Németh K. & Breitkreuz C. 2003: Mio/Pliocene
phreatomagmatic volcanism in a fluvio-lacustrine basin in
western Hungary. Geolines 15, 7581.
McDougall J., Pollack H.A. & Stipp J.J. 1969: Excess radiogenic ar-
gon in young subareal basalts from Auckland volcanic field,
New Zealand. Geochim. Cosmochim. Acta 33, 14851520.
Müller G. & Veyl G. 1956: The birth of Nilahue, a new maar type
volcano at Rininahue, Chile. 20th Int. Geol. Congress Report
(Congreso Geologico Internacional), Seccio I Vulcanologia
del Cenozoico, Mexico City, 375396.
Müller P. & Szónoky M. 1989: Faciostratotype Tihany-Feherpart
(Hungary), (Balatonica Beds by Lorenthey, 1905). In: Ste-
vanovic P., Nevesskaya L.A., Marinescu F.A.S. & Jámbor Á.
(Eds.): Chronostratigraphie und Neostratotypen, Neogen der
Westliche (Zentrale) Paratethys 8, Pontien. JAZU and SANU,
Zagreb-Beograd, 427436.
Müller P. & Magyar I. 1992: Stratigraphical importance of Proso-
dacnomy bearing Pannonian s.l. sediments from Kötcse. Földt.
Közl. (Bull. Hung. Geol. Soc.) 122, 138 (in Hungarian).
Müller P. 1998: Stratigraphy of the Pannonian sediments. In: Bérczi
I. & Jámbor Á. (Eds.): Stratigraphy of geological units of Hun-
gary. MOL Rt & MÁFI, Budapest, 485493 (in Hungarian).
Németh K. 2001: Deltaic density currents and turbidity deposits re-
lated to maar crater rims and their importance for paleogeo-
graphic reconstruction of the Bakony-Balaton Highland Volca-
nic Field (BBHVF), Hungary. In: Kneller B., McCaffrey B.,
Peakall J. & Druitt T. (Eds.): Sediment transport and deposition
by particulate gravity currents. Blackwell Sciences, Oxford,
Spec. Publs. Int. Ass. Sediment, 261277.
Németh K. & Martin U. 1999a: Late Miocene paleo-geomorpholo-
gy of the Bakony-Balaton Highland Volcanic Field (Hungary)
using physical volcanology data. Z. Geomorphol. N.F. 43,
417438.
Németh K. & Martin U. 1999b: Large hydrovolcanic field in the Pan-
nonian Basin: general characteristics of the Bakony-Balaton
Highland Volcanic Field, Hungary. Acta Vulcanol. 11, 271282.
Németh K., Martin U. & Harangi S. 1999: Miocene maar/diatreme
volcanism at the Tihany Peninsula (Pannonian Basin): The Tih-
any Volcano. Acta Geol. Hung. 42, 349377.
Németh K., Martin U. & Harangi S. 2001: Miocene phreatomagmat-
ic volcanism at Tihany (Pannonian Basin, Hungary). J. Volca-
nol. Geothermal Res. 111, 111135.
Németh K., Martin U. & Csillag G. 2003: Erosion rate calculation
based on eroded monogenetic alkaline basaltic volcanoes of the
Mio/Pliocene Bakony-Balaton Highland Volcanic Field, Hun-
gary. Geolines 15, 9397.
Shafiqullah M. & Damon P.E. 1974: Evaluation of K-Ar isochron
methods. Geochim. Cosmochim. Acta 38, 13411358.
Ort M.H., Wohletz K., Hooten J.A., Neal C.A. & McConnel V.S.
2000: The Ukinrek maars eruption, Alaska, 1977: a natural lab-
oratory for the study of phreatomagmatic processes at maars.
Terra Nostra 2000, 6, 396400.
Self S., Kienle J. & Huot J.-P. 1980: Ukinrek Maars, Alaska, II.
Deposits and formation of the 1977 Crater. J. Geophys. Res.
7, 3965.
Vespermann D. & Schmincke H.-U. 2000: Scoria cones and tuff
rings. In: Sigurdsson H., Houghton B.F., McNutt S.R., Rymer
H. & Stix J. (Eds.): Encyclopedia of volcanoes. Academic
Press, San Diego, 683694.
Walker G.P.L. 2000: Basaltic volcanoes and volcanic systems. In:
Sigurdsson H., Houghton B.F., McNutt S.R., Rymer H. & Stix
J. (Eds.): Encyclopedia of volcanoes. Academic Press, San Di-
ego, 283290.