www.geologicacarpathica.sk
GEOLOGICA CARPATHICA, DECEMBER 2009, 60, 6, 485—494 doi: 10.2478/v10096-009-0035-y
Introduction
Knowledge of reservoir architecture is one of the basic re-
quirements for hydrocarbon development and production.
The common methodology used during the initial phase of
such prospection is lithofacies analysis of sediments exposed
on the surface followed by interpretation and modelling of
depositional environments. However, this attempt is often
hampered by poor rock exposures and/or dense vegetation
resulting in a discontinuous record of sedimentary succes-
sions.
One of the methods used as an aid for interpretation of dep-
ositional deep-water environments, is statistical evaluation of
bed thickness. Because the thickness of a turbidite bed is de-
termined by the shape of the bed and the distance from the
source, initial sediment volume, grain size and flow type
(Talling 2001), the statistical evaluation may provide informa-
tion on processes controlling turbidite feeding, depositional
processes, distribution and geometry (e.g. Rothman et al.
1995; Malinverno 1997; Carlson & Grotzinger 2001). More-
over, the data may serve as an specifying factor for interpreta-
tion when “classical” facies analysis cannot be fully
developed due to the scarcity of exposures.
The problem of frequency distribution of turbidite bed
thicknesses, has been the object of study since the 1960’s.
Generally, four types of distribution – truncated Gaussian,
exponential, lognormal and power law distributions were ap-
plied to bed thickness distributions in turbidite deposits (e.g.
Ricchi Lucchi & Valmori 1980; Muto 1995; Drummond &
Wilkinson 1996; Malinverno 1997; Talling 2001; Sinclair &
Quantitative approach in environmental interpretations of
deep-marine sediments (Dukla Unit, Western Carpathian
Flysch Zone)
MARTA PREKOPOVÁ
1
and JURAJ JANOČKO
1
1
Institute of Geosciences, Technical University of Košice, Letná 9, 04001 Slovak Republic;
marta.prekopova@tuke.sk; juraj.janocko@tuke.sk
(Manuscript received July 2, 2008; accepted in revised form June 25, 2009)
Abstract: In structurally complicated terranes with outcrops limited in number and extent, additional methods for
interpreting depositional environments are required. Statistical analysis of bed thicknesses, in addition to conventional
sedimentological analysis, is a quantitative way to refine environmental interpretations, interpretations that can be
useful in predicting reservoir architecture. We analysed Paleogene deep-water sediments belonging to the Cisna, Sub-
Menilite, and Menilite Formations of the Dukla Unit, Outer Carpathian Flysch Zone and, using two independent quan-
titative methods, tried to define their depositional environments. As a first approach we used Carlson & Grotzinger’s
model (2001), which suggests power law distribution of turbidite bed thicknesses. The second one is the lognormal
mixture model of Talling (2001). Based on a quantitative approach, we suggest deposition of the lowermost Cisna
Formation in the channel-levee environment. The overlying sediments of the Sub-Menilite Formation were deposited in
a more distal, probably outer lobe environment. The uppermost Menilite Formation is interpreted as deposits from an
outer lobe/basin plain environment.
Key words: Western Carpathian Flysch Zone, Dukla Unit, statistics, depositional environment, turbidites.
Cowie 2003). Among these, lognormal and power law distri-
bution seem to be most appropriate for describing the turbidite
bed thickness variation (see discussion in Sylvester 2007).
The power law distribution is often interpreted as a sign of
the tendency of large dissipative systems to develop a state of
criticality and generate events of all sizes (Sylvester 2007).
For example, it has been linked to self-organization of conti-
nental slopes to a critical state or to generation of slope fail-
ures by earthquakes (Rothman & Grotzinger 1996; Talling
2001). The distribution of the turbidite beds is often ascribed
to the so-called segmented power law distribution that is char-
acterized by several straight-segment intervals on an ex-
ceedence probability plot (Winkler & Gawenda 1999; Carlson
& Grotzinger 2001; Sinclair 2003). However, Sylvester
(2007) argues that the segmented power law trend is not a
simple mixture of the power law population. Carlson &
Grotzinger (2001) assume that power law distribution may be
the primary input signal for some systems. Departure of the
measured distributions from the power law model is indicative
of different processes such as erosion and amalgamation, and
thus it can help in determining depositional environment. The
proximal part of the turbidite system represents areas with in-
creased erosion and amalgamation and frequency distributions
of turbidite beds have considerably curved course. Basinward,
in more distal areas of the turbidite system, the effects of ero-
sion and amalgamation are moderate and the turbidite beds
distribution displays a linear trend with departures at the ends
of the distribution. In the most distal areas of the system (outer
fan, basin plain), the flows are unconfined, the effects of ero-
sion and amalgamation are minimal, and thus the distribution
486
PREKOPOVÁ and JANOČKO
is expressed by a line segment typical for the power law distri-
bution (see Table 2 in Carlson & Grotzinger 2001).
The lognormal distribution applies to many geological pro-
cesses and has been found useful in bed thickness analysis of
turbidites elsewhere (e.g. Ricchi Lucchi & Valmori 1980).
Talling (2001) points out that frequency distribution of turbid-
ite thickness comprises the sum of a series of lognormal fre-
quency distributions, associated with a basal Bouma division.
If only thin- or thick-bedded turbidites are present, the fre-
quency distribution follows a lognormal distribution. Combi-
nation of lognormal distributions produces a stepped trend on
the probability plot for the entire bed population. This stepped
trend is equivalent to the segmented power law trend seen in
cumulative plots of bed thickness. Sylvester (2007) also pre-
fers description of turbidite bed thickness data by a lognormal
mixture model and emphasizes the importance of variability
of bed thicknesses (thin and thicker beds) for interpretation of
deposition setting.
The main objective of this study is to apply two different
statistical approaches for interpreting depositional environ-
ments, and based on comparison with the sedimentological in-
terpretation to discuss how these approaches may contribute
to the environmental analysis.
The analysed sediments occur in extensive outcrops flanking
the water-dam Starina in Eastern Slovakia, Western Carpathians
(Figs. 1, 2). After brief characterization of the geological
structure, we describe the sedimentary successions, based on
measured sections. Finally, the obtained data are used to inter-
pret the depositional environments.
Geological setting
The Outer Carpathians (Fig. 1) record the development of
an orogenic wedge from a remnant ocean basin to a collision-
related foreland basin (e.g. Oszczypko 1999). The closure of
the Alpine Tethys during the Late Cretaceous and collision of
the Carpathian orogenic wedge with the European passive
margin at the transition from Eocene to Oligocene caused the
transformation of remnant ocean basin to foreland basin with
dominant turbiditic and pelagic sedimentation. The modern
configuration of the Outer Carpathian Flysch Zone is a conse-
Fig. 1. Location of the study area in the Carpathian mountain system, and individual nappes of the Outer Carpathian Flysch Zone. The stud-
ied outcrops of Paleogene flysch sediments are marked by a heavy line in the close-up map (after Kováč et al. 1998 and Koráb & Ďurkovič
1978, modified).
487
DEEP MARINE SEDIMENTS – ENVIRONMENTAL INTERPRETATIONS (WESTERN CARPATHIANS)
quence of complex tectonic processes including folding,
thrusting and rotation (e.g. Janočko & Elečko 2002; Golonka
2003; Osczypko 2006).
The study area lies in the Dukla Unit that extends from SE
Poland to NE Slovakia and adjacent parts of Ukraine (Fig. 1).
The Dukla Unit, sandwiched between the Silesian Nappe in
the north and Magura Nappe in the south, contains Cretaceous
and Paleogene flysch sediments (mostly rhythmically inter-
bedded shales and sandstones) and has a complicated fold-and-
thrust structure. The interpretation of the paleogeographical
position of the Dukla Basin is still ambiguous and subject to
debate (e.g. Koráb & Ďurkovič 1978; Slaczka & Walton 1992;
Malata 2006). The paleoflow direction reflects changing
source areas and topography at the active plate margin. While
the prevailing paleoflow during the Paleocene and Eocene was
from the E, NE and SE (e.g. Bąk & Wolska 2005), the paleo-
flow changed to the opposite direction during the Oligocene,
probably as a result of emerging new source areas formed dur-
ing subduction.
The sediments in the study area are affiliated to the Cisna,
Sub-Menilite and Menilite Formations. The entire succession
is capped by the Cergow Sandstone that is a lithologically
contrasting member of the Menilite Formation (Figs. 2 and 3).
The Paleocene Cisna Formation contains medium- to thick-
bedded sandstones alternating with shales. The Paleocene to
Middle Eocene Sub-Menilite Formation is represented by
thin-bedded fine-grained sandstones rhythmically alternating
Fig. 2. Detailed location of stud-
ied sedimentary profiles in the
Dukla Unit.
Fig. 3. Composite sedimentary profile showing stratigraphy of sedi-
ments in the studied outcrops. The lowermost Cisna Formation (Pa-
leocene) contains medium- to thick-bedded sandstones alternating
with shales. The overlying Sub-Menilite Formation (Paleocene to
Middle Eocene) is represented by thin-bedded fine-grained sand-
stones rhythmically alternating with shales. The uppermost Menilite
Formation (Upper Eocene—Lower Oligocene) consists of thin-bed-
ded shales siltier or sandier at the base, occasionally alternating
with thin-bedded, fine-grained sandstones. The entire succession is
capped by the Cergow Sandstone (Lower Oligocene) representing a
member of the Menilite Formation. This is composed of medium- to
thick-bedded sandstones interlayered with alternating, thin-bedded
shales and sandstones and/or typical menilite shales.
488
PREKOPOVÁ and JANOČKO
with shales. The Menilite Formation of Oligocene age may be
divided into two superimposed subunits in the studied area:
the lower one consists of thin-bedded shales with silty or
sandy bases occasionally alternating with thin-bedded, fine-
grained sandstones. The overlying Cergow Sandstone, defined
as a member of the Menilite Formation, is composed of medi-
um- to thick-bedded sandstones alternating with thin-bedded
rhythmical flysch and/or typical menilite shales (Fig. 3).
Methods
Vertical bed thickness measurements were carried out in 17
profiles (Fig. 2) and combined into 3 composite vertical sec-
tions. The profiles were measured bed by bed and graphically
displayed in logs. The first section is reconstructed from 13
sedimentary profiles with a total thickness of 412 m. Three
sedimentary profiles with a total length of 41 m combine in
the second vertical section, while the third section consists of
one 79 m long sedimentary profile (Figs. 2 and 3). Each sand-
stone bed was measured from its base to the overlying mud-
stone or sandstone bed. Every bed thicker than 1 cm was
measured. In the case of amalgamated beds, if the sandstone
beds were distinguishable, each bed was measured separately.
Individual Bouma divisions were measured separately. Sand-
stone thickness data were also displayed in histograms.
The type of thickness distribution can be recognized using
diagrams with different scales on the horizontal axis. In the
first type of diagram, the sandstone bed thickness (h) is dis-
played linearly on the horizontal axis (LN plot) while in the
second type the logarithms of bed thicknesses (log h) are de-
picted (LL plot). The vertical axis is the same for both types of
diagrams: the logarithm of the number of beds thicker than the
measured bed (N
≥ h). The resulting frequency distribution
exhibits the characteristics of one of the three possible distri-
butions – lognormal, exponential, or power law distribution
(Fig. 4).
After displaying turbidite thicknesses either on LL or on LN
plot it is not always clear what type of distribution is repre-
sentative for the given dataset. Therefore we also constructed
logarithmic probability plots with bed thickness h (cm) on
the horizontal axis and with percentages of beds thinner than
h (%) presented on vertical axes.
The Kolmogorov-Smirnov test and quantile-quantile plot
were used for testing the lognormal distribution of our data.
Description of sedimentary profiles and bed
statistics
440 m of sediments, encompassing 17 logs and divided into
three sections were measured bed-by-bed. The sediments of
the Cisna Formation reach 400 m in thickness in the studied
area; only 125 m could be measured because of discontinuity
of outcrops. The total of the measured beds of the Sub-Meni-
lite Formation is 1870 m. The sedimentary log of the Menilite
Formation encompasses 225 m of sediments (Fig. 3).
For the statistical analysis 11 logs were used. The remaining
6 logs belonging to the Sub-Menilite Formation were left out
either because they are dominated by shales with minor thin
sandstone beds or they are too short for statistical analysis.
However, these logs still provide important data for the over-
all picture of the sedimentological evolution of the area. In the
following section, we describe the analysed sedimentary suc-
cessions from base to top (see Fig. 3).
Cisna Formation
(Fig. 5)
The Early Paleocene Cisna Formation is represented by logs
13 and 12, with a total thickness of 125 m (Fig. 5). The sedi-
mentary profile is dominated by intervals of thick-bedded,
amalgamated sandstones alternating with intervals of thin- to
medium-bedded sandstones with shales. The sandstone/shale
ratio of the entire succession is 8 : 1. The medium-grained grey
sandstones are predominantly massive and occasionally paral-
lel laminated and the dark grey shales are parallel laminated.
The sandstones are arranged in sharply-based beds. Locally,
shallow erosive channels are developed.
The bed thicknesses of 156 sedimentary beds vary from 1 to
300 cm, with a mean shale thickness of 11 cm and mean sand-
stone thickness of 26 cm (Fig. 5). The prevalence of thicker
beds is demonstrated in the frequency histogram of sandstone
bed thickness showing 72 % of beds thicker or equal to 10 cm.
Beds of 10 cm thickness are the most frequent.
The thick-bedded, amalgamated, massive, and medium-
grained sandstones may represent deposits of concentrated
flows (Mulder & Alexander 2001) in the channelized part of
the turbidite system. As these flows develop, progressive fluid
entrainment and dilution increase their turbulence and cause
them to transform into turbidity flows, driving the deposition
of the thin-bedded massive or parallel laminated sandstones
capped by shales.
The cumulative distribution of turbidite sandstone beds ex-
pressed by diagram with logarithmic scales on both horizontal
and vertical axes (LL plot) and logarithmic values only on the
vertical axis (LN plot) is displayed in Fig. 5. The LL plot
Fig. 4. Various turbidite bed-thickness distributions (after Sinclair
& Cowie 2003, modified).
489
DEEP MARINE SEDIMENTS – ENVIRONMENTAL INTERPRETATIONS (WESTERN CARPATHIANS)
Fig. 5. Sedimentary profile of the Cisna Formation and histogram of bed thicknesses. The cumulative distribution of turbidite sandstone
beds in the LL and LN plots shows deviations from the power law distribution at both ends of the curve. The distribution of bed thickness
of all sandstones is expressed by linear trend line on logarithmic probability plot, thus indicating lognormal distribution of our data.
shows a concave curve while LN plot follows a convex
curve. For a more complete evaluation, we also display the
data as a logarithmic probability plot where the distribution
of bed thickness of all measured sandstones is expressed by a
linear trend. The Ta, Tb and Td divisions display a straight
line as well.
After displaying our data on the LL, LN and logarithmic
probability plots we suggest a lognormal distribution of tur-
bidite thickness for the Cisna Formation as confirmed by
Kolmogorov-Smirnov test (KS test), and quantile-quantile
plot (q-q plot, Fig. 8). From the Bouma divisions, the Ta divi-
sion has high variability similarly to all sandstone bed lines.
Curves representing Tb, Tc, Td and Te divisions show mark-
edly lower bed thickness variability.
Comparing similarly distributed bed thicknesses from the
Tarcu Sandstone (Eastern Carpathians), which is interpreted
490
PREKOPOVÁ and JANOČKO
as channel-and-levee complex (Sylvester 2007), we think
that the lower variability of Tb, Tc, Td, Te may suggest
levee deposits above channelized sandstone characterized by
Ta division with higher variability. This interpretation is
supported by our results based on facies analysis.
According to Carlson & Grotzinger’s (2001) methodolo-
gy, which argues for power law distribution of turbidites, the
absence or faint linear segment may indicate segmented
power law distribution. In this case the shape of the analysed
distribution in LL and LN plots implies deposition of sedi-
ments in environment characterized by higher influence of
erosion and amalgamation – such as upper fan, channelized
part of depositional lobe or channel-lobe transition.
Sub-Menilite Formation
(Fig. 6)
The Paleocene to Middle Eocene Sub-Menilite Formation
overlies the Cisna Formation and consists of rhythmically al-
ternating, sharply based medium- to thin-bedded sandstones
and shales at the base of the formation overlain by an inter-
val of thick-bedded sandstones with a thinning-upward trend
(log 11, Figs. 2 and 6). The sandstone is fine- and medium-
grained, with sporadic coarse-grained intervals. The most
frequent thickness of beds is 6 cm but amalgamated sand-
stones are up to 140 cm thick. Massive structure and occa-
sional normal grading prevail in the coarse- and
medium-grained sandstones, while the fine-grained sand-
stones are mostly ripple-cross laminated and parallel lami-
nated with rare convolute bedding (Fig. 6). Upward the
whole succession becomes finer grained with mean thickness-
es of both sandstone or mudstone decreasing to ca. 2—4 cm.
Sharp bed soles exhibit traces characteristic of the Nereites
ichnofacies.
The main difference between the Sub-Menilite and Cisna
Formations is the increase in shale content resulting in lower
sandstone/mudstone ratio 2 : 1 in the whole succession and
even less than 1 : 1 in the section above the basal thick sand-
stones (Fig. 6). In contrast to the Cisna Formation, the per-
centage of beds thinner than 10 cm increases to 89 %, while
10 cm and thicker beds represent only 11 % of the whole bed
population. The most frequent are beds with thicknesses of
about 1 cm.
The vertical trend of the sediments (Fig. 6) suggests shift
of the depocentre during their deposition to areas less affect-
ed by erosion (e.g. distal part of lobes). The trend may also
be interpreted as representing overbank deposits (typical
rhythmically alternating thin beds of sandstones and shales);
however, lack of associated channel sediments and erosion
features favour the outer lobe interpretation.
The frequency distribution of bed thickness on the LL plot
exhibits linear segment bounded with rollover on the both
ends. Departure is more significant at the thin bed end and it
occurs in beds thinner than 3 cm and thicker than 130 cm.
On the LN plot, the distribution has a convex shape (Fig. 6).
The distribution of bed thicknesses of all sandstones on the
logarithmic probability plot displays a slight deviation from
a linear trend. Plotted individual Bouma divisions show a
similarity between the distributions of Ta division and all
sandstone beds.
The distribution of bed thickness data based on the LL, LN
and logarithmic probability plots cannot be interpreted un-
equivocally. The goodness-of-fit of the data using KS test did
not confirm lognormal distribution and the data also markedly
deviate from the modelled distibution on q-q plot (Fig. 8) The
higher variability of Tc, Td and Te divisions (Fig. 6) com-
pared to that in the Cisna Formation may indicate a more dis-
tal depositional environment, probably lobe fringe. Similar
relationships are found in data from the Marnoso-Arenacea
Formation (Sylvester 2007) and it is also in accordance with
the interpretation based on facies analysis.
Assuming Carlson & Grotzinger’s (2001) hypothesis that
turbidites without signs of erosion and amalgamation have
power law distribution, the data from the Sub-Menilite Forma-
tion would suggest deposition in lower fan, lobe fringe or ba-
sin plain environments.
Menilite Formation
(Fig. 7)
The Menilite Formation mainly consists of brown, grey and
black, laminated shales and occasional sandstones. The shales
often contain “menilites” – black, silicious shales with the
origin related to decreased supply of terrigenous material and
anoxic regime caused by major paleogeographical changes
during the Eocene-Oligocene boundary (e.g. Puglisi 2006).
The formation is deposited above the Globigerina marl hori-
zon that is typical for the Eocene-Oligocene boundary
(Leszczyński 1996). A part of the formation, defined as a
member, is called the Cergow Sandstone and is composed of
thick sandstone beds separated by dark grey and black,
“menilitic” shales (Fig. 7).
In the study area the formation is represented by a 60 m
thick interval of grey, brown and black shales overlain by the
129 m thick Cergow Sandstone (Fig. 7). The monotonous
shale succession is interrupted by several beds of massive
sandstone that indicate rare incursions of flows with high
competence. The lowermost part of the Cergow Sandstone is
characterized by thick-bedded, amalgamated sandstones sepa-
rated by thin, parallel laminated and massive dark shales. The
bases of the sandstones are sharp and loaded and the sand-
stones occasionally contain rip-up clasts in the higher part of
the beds. Frequent flute and rill marks indicate main paleo-
flow direction from WWN to EES. Upsection, several up-
ward-thinning cycles in the medium to thick sandstone beds
alternating with shales are developed. The sandstones are pre-
dominantly fine- and medium-grained, massive, planar paral-
lel and ripple cross-laminated, occasionally convoluted. The
cycles are separated by 3 to 6 m thick intervals of typical
black “menilitic” shales.
The entire Menilite Formation contains 509 sandstone beds
of which the 1 cm thick beds are the most frequent. Although
this compares well with the Sub-Menilite Formation, the beds
thicker or equal to 10 cm increase to 21 % and beds thinner
than 10 cm decrease to 79 % here (Fig. 7). Thick accumula-
tion of dark and black shales with menilites suggests quiet
depositional conditions with restricted circulation. Prevailing
suspension sedimentation was abruptly changed to turbidite
deposition resulting in sedimentation of the thick Cergow
Sandstone.
491
DEEP MARINE SEDIMENTS – ENVIRONMENTAL INTERPRETATIONS (WESTERN CARPATHIANS)
Fig. 6. Sedimentary profile showing the sediments of the Sub-Menilite Formation. The bed-thickness histogram strongly suggests domi-
nance of thin-bedded beds. The frequency distribution of the bed thickness in LL plot shows slight deviation from power law distribution.
The distribution on the LN plot displays a convex form that may indicate both power law and lognormal distributions. The logarithmic
probability plot of turbidite thickness of the Sub-Menilite Formation differs from the probability plot of the Cisna Formation (see Fig. 5).
The distribution of bed thickness of all sandstones displays slight deviation from lognormal distribution. Note the resemblance between dis-
tributions of Ta division and all sandstone beds.
492
PREKOPOVÁ and JANOČKO
Fig. 7. The uppermost Menilite Formation topped with the Cergow Sandstone Member in its upper part. The bed-thickness frequency dis-
tribution on the LL plot and LN plots indicates either power law or lognormal distributions. The logarithmic probability plot of the Menilite
Formation resembles the probability plot of the Sub-Menilite Formation (see Fig. 6). The linear trend of distribution of all sandstone divi-
sions again indicates a lognormal distribution of turbidite thicknesses.
493
DEEP MARINE SEDIMENTS – ENVIRONMENTAL INTERPRETATIONS (WESTERN CARPATHIANS)
The LL plot of bed thickness implies power law distribution
contrary to the logarithmic probability plot that rather shows
lognormal distribution. However, the lognormal distribution
did not pass the KS test. The analysed bed thickness distribu-
tion also slightly deviates from the theoretical lognormal dis-
tribution on q-q plot (Fig. 8).
The variability of Tb, Tc, Td and Te Bouma divisions close-
ly resembles divisions from the Sub-Menilite Formation.
However, higher variability of Te divisions indicates occur-
rence of thicker fines thus suggesting weaker erosion. This
may be represented by the outer fan/basin floor environment.
The cumulative distribution of beds displayed on the LL plot
has a slightly concave shape (Fig. 7). The rollover begins at
the thin beds end with the beds thinner than 4 cm, and at the
thick beds end with beds thicker than 120 cm. The curve in the
LN plot shows a convex shape. The logarithmic probability
plot has a linear trend.
Application of Carlson and Grotzinger’s approach to bed
thicknesses having power law distribution, results in interpre-
tation of a depositional environment typical of weak erosion.
Such an environment may be represented by the distal parts of
a turbidite system.
Discussion and conclusions
More than 440 m of Paleocene to Oligocene sediments as-
signed to the Cisna, Sub-Menilite and Menilite Formations of
the Dukla Unit were measured bed-by-bed in order to obtain
data for statistical analysis, which should serve as an addition-
al tool for interpretation of the sedimentary environment. The
entire analysed sedimentary succession consists of thick-
bedded massive sandstones alternating with shales (Cisna
Formation) overlain by thin-bedded sandstones and shales
(Sub-Menilite Formation). The succession passes upward to
thick shales capped by massive sandstones of the Menilite
Formation. Based on the facies analysis we interpreted depo-
sition of the Cisna Formation in a proximal fan (channelized
section) while the sediments of the Sub-Menilite Formation
were deposited in the outer lobe. The uppermost deposits –
the Menilite Formation are interpreted as outer fan/basin floor
and lobe deposits, respectively.
The obtained bed thickness data were used for construction
of histograms, LL, LN and logarithmic probability plots. For
evaluation of our data we used interpretational methods as-
suming a) lognormal distribution and b) power law distribu-
tion of turbidite bed thicknesses.
Bed thickness distribution for the Cisna Formation shows
lognormal distribution with Ta division having high and Tb,
Tc, Td, Te divisions having much smaller variabilities
(Fig. 5). According to our interpretation based on facies analy-
sis, these bed thickness characteristics reflect deposition in a
channel and levee environment. Similar variability of Bouma
divisions was identified in the Tarcău Sandstone (Sylvester
2007) and also interpreted as a channel-levee complex. The
distributions of the Sub-Menilite and Menilite Formations
show only slight divergence from the lognormal distribution.
The bed thickness distribution of the Sub-Menilite Formation,
facially interpreted as outer lobe sediments, suggests that for
this environment higher variability of Tb, Tc, Td and Te divi-
sions (Fig. 6) is typical. The variability of these divisions from
the Menilite Formation, deposited in an outer fan/basin plain
environment, is similar (Fig. 7). The only difference is consider-
ably higher variability of Te division. The high thickness of Te
division probably reflects an environment with negligible ero-
sion that is in accord with our sedimentological interpretation.
If we assume power law distribution of bed thicknesses
(Carlson & Grotzinger 2001), the deviation of our data from
power law distributions on LL and LN plots indicates ero-
sion and amalgamation. The greatest deviation is observed
on the plots of the Cisna Formation suggesting its deposition
in the proximal part of the turbiditic system. The distribution
of the Sub-Menilite Formation sediments slightly deviates
Fig. 8. Quantile-quantile plots of all studied formations showing the
fit of the distributions to the assuming lognormal model.
494
PREKOPOVÁ and JANOČKO
from the power law distribution implying a more distal envi-
ronment with lesser erosion. The uppermost sediments of the
Menilite Formation only depict a slight deviation on both
sides of the bed thickness distribution curve with the best ap-
proximation to the power law distribution with regard to the
older formations. This indicates deposition in an environment
with minimum erosion and corroborates outer fan/basin floor
environments inferred from the sedimentological analysis.
The lognormal distribution of bed thicknesses from individu-
al formations were tested by the Kolmogorov-Smirnov test,
which only confirmed this distribution for the Cisna Formation.
The data from this formation also show the best approximation
to the modelled lognormal distribution on the quantile-quantile
plot in contrast to the data from the Sub-Menilite and Menilite
Formations. However, the step-like trend visible in the all sand-
stone curve on logarithmic probability plots of these two forma-
tions (Figs. 6 and 7) may suggest a polymodal bed thickness
population. Thus, testing of individual populations for lognor-
mal distribution, which may form components of lognormal
mixture model should be a subject of the next study.
The applied methodology may represent an important part
of depositional environment analysis in the areas with limited
outcrops or subsurface data, such as the Outer Flysch Zone of
the Carpathians. Our interpretation was made on the basis of
comparison with similar analyses elsewhere (Tarcău Sand-
stones, Eastern Carpathians; Marnoso-Arenacea Formation,
Apennines; Cloridorme Formation, Quebec, etc.) and was
confirmed by the results of facies analysis. It is interesting that
the interpretations from both the applied methods yielded sim-
ilar results proposing the need for more case studies in order to
select the most reliable method. We agree with Sylvester
(2007) who pointed out that it is unlikely that a few numbers
or curves derived from bed thickness data alone can give gen-
eral guidance about depositional setting, degrees of confine-
ment, erosion, bypass and other important characteristics of
the turbidite thickness. However, even if the interpretation
based on quantitative methods is not always unambiguous, to-
gether with other sedimentological methods it contributes to
more reliable interpretations.
Acknowledgments: The paper was written as a part of the
Project VEGA No. 1/3061/06 and Operational Programme
Research and Development for Project 26220220031 co-fi-
nanced by European Regional Development Fund. We are
grateful to Z. Sylvester and W. Winkler, who reviewed,
commented and considerably improved the manuscript.
References
Bąk K. & Wolska A. 2005: Exotic orthogneiss pebbles from Paleocene
flysch of the Dukla Nappe (Outer Eastern Carpathians, Poland).
Geol. Carpathica 56, 3, 205—221.
Carlson J. & Grotzinger P. 2001: Submarine fan environment in-
ferred form turbidite thickness distributions. Sedimentology 48,
1331—1351.
Drummond C.N. & Wilkinson B.H. 1996: Stratal thickness frequencies
and the prevalence of orderedness in stratigraphic sections. J.
Geol. 104, 1—18.
Golonka J. 2003: Plate tectonics of the circum-Carpathian area and the
ultra deep drilling proposal. In: Golonka J. & Lewandowski M.
(Eds.): Geology, geophysics, geothermics and deep structure of
the West Carpathians and their basement. Publ. Inst. Geophys.,
Polish Acad. Sci., Monograph. 28, 363, 113—125.
Janočko J. & Elečko M. 2003: Tectono-sedimentary evolution of West-
ern Carpathian Tertiary Basins. Miner. Slovaca 3—4, 35, 181—254.
Janočko J., Elečko M., Karoli S., Konečný V., Kováč M., Nagy A.,
Vass D., Jacko S., Jr. & Kaličiak M. 2003: Sedimentary evolution
of Western Carpathian Tertiary basins. In: Janočko J. & Elečko M.
(Eds.): Tectono-sedimentary evolution of Western Carpathian
Tertiary Basins. Miner. Slovaca 3—4, 35, 181—254.
Koráb T. & Ďurkovič T. 1978: Geology of Dukla Unit (Flysch of the
Eastern Slovakia). (Geológia Dukelskej jednotky (flyš východ-
ného Slovenska)). ŠGÚDŠ, Bratislava, 1—194 (in Slovak).
Kováč M., Nagymaros A., Oszczypko N., Ślączka A., Csontos L.,
Marunteanu M., Matenco L. & Márton M. 1998: Palinspastic recon-
struction of the Carpathian-Pannonian region during the Miocene.
In: Rakús M. (Ed.): Geodynamic development of the Western Car-
pathians. Geol. Surv. Slovak Rep., Bratislava, 189—217.
Leszczyński S. 1996: Origin of lithological variation in the sequence of
the Sub-Menilite Globigerina marl at Znamirowice (Eocene-Oli-
gocene transition, Polish Outer Carpathians). Ann. Soc. Geol. Pol.
66, 245—267.
Malata T. 2006: Tectonic evolution of the Dukla Subbasin. In: Osz-
czypko N., Uchman A. & Malata E. (Eds.): Paleotectonic evolu-
tion of the Outer Carpathian basins and Pieniny Klippen Belt. Inst.
Geol. Soc., Jagellonian Univ., Krakow, 127—132.
Malinverno A. 1997: On the power law size distribution of turbidite
beds. Basin Research 9, 263—274.
Mulder T. & Alexander J. 2001: The physical character of subaqueous
sedimentary density flows and their deposits. Sedimentology 48,
269—299.
Muto T. 1995: The Kolmogorov model of bed-thickness distribution:
an assessment based on numerical simulation and field-data analy-
sis. Terra Nova 7, 417—423.
Oszczypko N. 1999: From remnant oceanic basin to collision-related
foreland basin – a tentative history of the Outer Western Car-
pathians. Geol. Carpathica, Spec. Issue 50, 161—163.
Oszczypko N. 2006: Late Jurassic-Miocene evolution of the Outer
Carpathian fold-and-thrust belt and its foredeep basin (Western Car-
pathians, Poland). Geol. Quart. 50, 1, 169—194.
Puglisi D., Badescu D., Carbone S., Corso S., Franchi R., Gigliuto L.G.,
Loiacono F., Miclaus C. & Moretti E. 2006: Stratigraphy, petrog-
raphy and paleogeographic significance of the Early Oligocene
“menilite facies” of the Tarcău Nappe (Eastern Carpathians, Ro-
mania). Acta Geol. Pol. 56, 1, 105—120.
Ricchi-Lucchi F. 2003: Turbidites and foreland basins: an Apenninic
perspective. Mar. and Petroleum Geology 20, 727—732.
Ricchi Lucchi F. & Valmori E. 1980: Basin-wide turbidites in a Mi-
ocene, over-supplied deep-sea plain: a geometrical analysis. Sedi-
mentology 27, 3, 241—270.
Rothman D.H. & Grotzinger J.P. 1995: Scaling properties of gravity-
driven sediments. Nonlinear Processes in Geophysics 2, 178—185.
Sinclair H.D. & Cowie P.A. 2003: Basin-floor topography and the scal-
ing of turbidites. J. Geol. 3, 277.
Slaczka A. 2005: Bukowiec Ridge: A cordiliera in front of the Dukla
Basin (Outer Carpathians). Miner. Slovaca 37, 255—256.
Slaczka A. & Walton E.K. 1992: Flow characteristics of Metresa: an
Oligocene seismoturbidite in the Dukla Unit, Polish Carpathians.
Sedimentology 39, 3, 383—392.
Sylvester Z. 2007: Turbidite bed thickness distribution: methods and
pitfalls of analysis and modelling. Sedimentology 54, 4, 847—870.
Talling P.J. 2001: On the frequency distribution of turbidite thickness.
Sedimentology 48, 1297—1329.
Winkler W. & Gawenda P. 1999: Distinguishing climatic and tectonic
forcing of turbidite sedimentation, and the bearing on turbidite bed
scaling: Palaeocene—Eocene of northern Spain. J. Geol. Soc. 156,
791—800.