A knowledge of seismic wave velocities in the sedimentary cover is of great importance for interpreting reflection and refraction seismic data, deep seismic soundings and regional and global seismic tomogra-phy. This is particularly true for regions characterized by significant thicknesses and a complex sedimentary cover structure. This paper pre-sents the results of an analysis of seismic P-wave velocities in the sedi-mentary cover of Poland, a complex area of juxtaposition of major tectonic units: the Precambrian East European Craton, the Palaeozoic Platform of Central and Western Europe, and the Alpine orogen repre-sented by the Carpathian Mountains. Based on vertical seismic profiling data from 1188 boreholes, the dependence of velocity versus depth was determined for regional geological units and for successions from the Tertiary and Quaternary to the Cambrian. The data have been approxi-mated by polynomials, and velocity-depth formulas are given down to 6000 m depth. The velocities in the sedimentary cover have been com-pared with those from other areas in Europe.
Key words: vertical seismic profiling in boreholes, seismic velocity analysis, sedimentary cover, East European Craton, Palaeozoic Platform, Trans-European Suture Zone, Carpathians.
M. GRAD and M. POLKOWSKI
986
1. INTRODUCTION
The area of Poland comprises several major tectonic units of different tec-tonic age (Pożaryski et al. 1982, Znosko 1975, 1979, Berthelsen 1992, 1998). These include the Precambrian East European Craton (EEC), the Palaeozoic Platform of Central and Western Europe (PP), and the Alpine orogen represented by the Carpathian Mountains (Fig. 1a). Such a variety of basement units and their histories are reflected in a complex and differenti-ated structure of the sedimentary cover in this area. A good knowledge of seismic wave velocities in such a complex area is of great importance for interpreting reflection and refraction seismic profiles, deep seismic sounding data and regional and global seismic tomography.
The average depth of the crystalline basement of the EEC is in the order of 2 km. In NE Poland, the basement depth is only 0.3-1 km and increases towards the southwest to reach 7-8 km along the margin of the EEC (Skorupa 1974, Młynarski 1984). In the Trans-European Suture Zone (TESZ) in NW Poland, the thickness of Permian–Mesozoic, and presumably pre-Permian strata, reaches 18-20 km, which was found by wide-angle re-flection and refraction (WARR) experiments (e.g., Guterch et al. 1994, Grad et al. 1999, 2003, 2008, Janik et al. 2002). South of the TESZ the basement depth decreases, being 2-4 km (Toporkiewicz 1984). In the Sudetes (SW Poland), which form the northeastern margin of the Bohemian Massif, the Palaeozoic consolidated basement is observed as being close to the Earth’s surface, at depths of about 1 km. It should be noted that velocities in the basement beneath the Sudetes are relatively low (Vp ~ 5.9 km/s) compared to those observed for the crystalline basement of the Precambrian EEC (Vp ~ 6.1-6.2 km/s). In the Carpathian region, the crystalline basement descends to the south, reaching its maximum depth in the eastern part of the Polish Carpathians. In the Kuźmina 1 well, the deepest borehole in Poland (7541 m), a Proterozoic crystalline basement was found at a depth of 7390 m. Further to the south, the depth of the Proterozoic basement reaches 15-20 km, as determined by WARR experiments (e.g., Grad et al. 2006, Środa et al. 2006, Janik et al. 2011). Similar depths were found in the Carpa-thians using results of magnetotelluric soundings, in which a high resistivity horizon was interpreted as being the top of the Precambrian crystalline basement (Stefaniuk and Klityński 2007).
Such large depths are still inaccessible even by the deepest wells. In order to describe the distribution of seismic velocities within the entire sedi-mentary cover, down to the crystalline basement, the results of deep seismic studies should be used, particularly deep seismic refraction (WARR) ones.
The complex structure of the sedimentary cover of Poland is reflected in the division of this area into regional and local units. According to
SEISMIC WAVE VELOCITIES IN THE SEDIMENTARY COVER OF POLAND
987
Sokołowski (1968), four regional units (A-D) could be sub-divided into local sub-units (Fig. 1b): A – East European Craton; B – Lowland (B1 – marginal synclinorium, B2 – Pomorze–Kujawy anticlinorium, B3 – Szczecin–Łódź synclinorium, B4 – northern fore-Sudetic monocline); C – Folded area (Ca – Sudety Mts. and fore-Sudetic block, Cb – Upper Silesian block, Cc – southern fore-Sudetic monocline, Cd – Miechów synclinorium, Goleniów anticlinori-um, Holy Cross anticlinorium; Ce – San elevation, Cf – Lublin synclinorium); D – Carpathians (Da – Outer Carpathians, Db – Silesian unit, Dc – Magura unit and Inner Carpathians).
A location map of boreholes with vertical seismic profiling (VSP) used in this study is shown in Fig. 1. As can be seen from this figure, the coverage
Fig. 1: (a) The study area (yellow frame) on the background of the main tectonic units of Central Europe. BM – Bohemian Massif, Carp – Carpathians, EEC – East European Craton, PP – Palaeozoic Platform, TESZ – Trans-European Suture Zone; (b) Location map of boreholes used in this study on the background of the geologi-cal division of Poland, simplified from Sokołowski (1968). A – East European Craton; B – Lowland: B1 – marginal synclinorium, B2 – Pomorze–Kujawy anticlino-rium, B3 – Szczecin–Łódź synclinorium, B4 – northern fore-Sudetic monocline; C – Folded area: Ca – Sudetes and fore-Sudetic block, Cb – Upper Silesian block, Cc – southern fore-Sudetic monocline, Cd – Miechów synclinorium, Goleniów anti-clinorium, Holy Cross anticlinorium; Ce – San elevation, Cf – Lublin synclinorium; D – Carpathians: Da – Outer Carpathians, Db – Silesian unit, Dc – Magura unit and Inner Carpathians. The red dot in SE Poland shows the location of the deepest well Kuźmina 1 borehole, 7541 m. Histogram (c) shows distribution of boreholes amount n with depth z; N = 1188 is a total number of boreholes used in this study. Colour version of this figure available in electronic edition only.
M. GRAD and M. POLKOWSKI
988
by boreholes is not uniform. The marginal synclinorium (B1), the Lublin synclinorium (Cf), the fore-Sudetic monocline (B4 and Cc) and the Carpathi-ans (D), i.e., areas that have been explored for minerals and hydrocarbons, are best covered with wells with VSP data. In contrast, the East European Craton (A), Sudety Mts. and fore-Sudetic block (Ca) are poorly covered. Accordingly, we adopted a simplified version of Sokołowski’s (1968) map, joining small second-order units and those with poor borehole coverage into 14 units, as shown in Fig. 1.
No comprehensive analysis of vertical seismic profiling (VSP) data for the whole of Poland has been available until now. A regional compilation for some regions do exist, including the fore-Sudetic monocline (B4 and Cc in Fig. 1b; Śliwiński 1965), selected parts of marginal synclinorium (B1; Ptak 1966, Żaruk 1971, Kamińska and Zagórski 1978, Świtek 1983) and for the East European and Palaeozoic platforms (A, B3, B4, Cc; Grad 1987, 1991, Grad et al. 1991). The goal of this paper is to complement previous results and to provide velocity-depth relationships in the sedimentary cover for all the periods and regions in a systematic way. Such a knowledge of seismic wave velocities in such a complex area is of great importance for interpret-ing seismic profiles and seismic tomography, particularly for areas charac-terized by substantial thickness of sedimentary cover. In the future, it could be valuable for creating a 3-dimensional seismic velocity model of the entire crust in Poland.
2. VERTICAL SEISMIC PROFILING (VSP) BOREHOLE DATA
For this study, VSP data from 1188 boreholes in Poland has been used. Together with the previously used old data from the East European Platform (261 boreholes; Grad 1987) and Palaeozoic Platform (401 boreholes; Grad 1991) we also collected an additional new data set (526 boreholes) to cover the southern Poland area, particularly the Carpathians, as well as included new data from both platforms. The location of all boreholes and a histogram of their depths of penetration are shown in Fig. 1b, c. Most of the boreholes reached a depth of 1000-3500 m. The deepest Polish well, Kuźmina 1, is 7541 m deep with a vertical seismic profiling acquired to a depth of 7520 m.
The analyzed VSP data were obtained from routine velocity logging dur-ing many decades (the oldest profiling was completed in 1952, the newest in 2003). Velocities in individual boreholes were analyzed in terms of succeed-ing geological periods, from the Tertiary and Quaternary to the Cambrian. For old data from the East European Platform and the Palaeozoic Platform, we collected layer velocities, in the form of constant velocity from the top to bottom of the layer (Fig. 2a). In the cases when velocities were highly
SEISMIC WAVE VELOCITIES IN THE SEDIMENTARY COVER OF POLAND
989
TQ
K
J
T
P
C
TQ
K
J
T
P
Δx
Δt
ΔtΔxV = lay Vlay
(a) (b) travel time layer velocity travel time interval velocity
dept
h
dept
h
Vint
Fig. 2. The sketch illustrating vertical seismic profiling borehole data used in this study. (a) Layer velocities determined from average travel time, previously used for the East European Platform (261 boreholes; Grad 1987) and Palaeozoic Platform (401 boreholes; Grad 1991). (b) Interval velocities determined from travel time measured each 20 m depth; additional new data set (526 boreholes) collected to cov-er the southern Poland area, particularly the Carpathians, as well as new data from both platforms. Colour version of this figure available in electronic edition only.
diversified within one stratigraphic unit, a more detailed division with depth was adopted. For the newly analyzed data we used interval velocities calcu-lated from vertical travel time in digital form (Fig. 2b). Usually vertical travel time was obtained from three independent vertical profilings sampled every 20 m. In all the figures presenting the velocity-depth distribution, the new data can be recognized as a short dash, while long bars represent the old data.
3. VELOCITY-DEPTH ANALYSIS FOR GEOLOGICAL UNITS
Initially, the dependences of velocity versus depth were determined for regional geological units (A, B1, ... Dc in Fig. 1). A comparison of data is shown in Fig. 3. Original data, layer and interval velocities, are shown as navy-blue vertical short dashes and long bars. The scattering of data is very big, even within the same geological unit, being for the same depth in the order of 1000-2000 m/s. In a few cases (mostly interval velocities), veloci-ties larger than 7000 m/s were removed from the data base (values unrealis-tic for sedimentary rocks, possibly errors in measurements). Although the scattering of the data for each geological unit is significant, a clear increase of velocity with depth is observed. Moving average values of velocity calcu- lated for each 50 m depth interval (red lines in Fig. 3) are unexpectedly
M. GRAD and M. POLKOWSKI
990
SEISMIC WAVE VELOCITIES IN THE SEDIMENTARY COVER OF POLAND
991
stable, with a scattering of few hundred meters only. The scattering of aver-age values of velocity for a larger depth is bigger because the small amount of available data. Average velocities have been approximated by second order polynomials; the corresponding curves (black lines) and velocity-depth formulas are given for each geological unit in Fig. 3. Horizontal dashed lines give the depth range of the formula application, with the corresponding depth and velocity. The last box in this figure shows all the data from 1188 boreholes (All), with moving average values of velocity calculated for a 50 m depth window (red line) with approximation by a third order poly-nomial: V(z) = a + bz + cz2 + dz3 . (1)
Here, and in all other cases, the depth z = 0 corresponds to the Earth’s sur-face. The averaged velocity-depth curve down to 6000 m depth calculated using the above formula for all collected data is shown in Fig. 3 by a green line, together with a ±500 m/s corridor (highlighted in green). This average line with a ±500 m/s corridor is also shown for comparison for each geologi-cal unit in Fig. 3. The ±500 m/s width band was chosen arbitrarily for visual-izing a typical data scattering. Apart from the values of coefficients for interpolated polynomials, standard deviations for coefficients are given for each unit.
The average velocity-depth relationship for the entire territory of Poland could be used as reference point for individual geological units. Velocities in central Poland (Pomorze–Kujawy anticlinorium (B2), Szczecin–Łódź synclinorium (B3) and northern fore-Sudetic monocline (B4)) are almost identical to the average line. Slightly lower velocities are observed for craton (A) and marginal synclinorium (B1). Velocities slightly higher than the coun-try average are observed for southern Poland (units Ca, Cb, Cc, Cd and Cf). In SE Poland, the Carpathian sediments are dominated by flysch. Interestingly, a velocity increase towards the Inner Carpathians is observed. In the unit sequence Ce, Da, Db and Dc, the velocity increases systematically by about 250 m/s (for example, at a depth of 2000 m, average velocities are 3598, 3886, 4317 and 4430 m/s, respectively).
Fig. 3. P-wave seismic velocities of the sedimentary cover measured in boreholes in the area of Poland. Navy-blue vertical lines are the layer and interval velocities. Red lines are moving average values of velocity calculated for each 50 m depth interval, and the black line is their interpolation by polynomial. Dashed lines show the depth interval for application of interpolation. Division into geological units corresponds to Fig. 1 (A, B1, ... Dc) and to the whole data used (All). The green line is an average velocity-depth dependence for the whole area and, together with the ±500 m/s band, is shown for comparison in the background for all geological units. See text for more explanation. Colour version of this figure available in electronic edition only.
M. GRAD and M. POLKOWSKI
992
4. VELOCITY-DEPTH ANALYSIS FOR GEOLOGICAL PERIODS
The scattering of velocities within the same geological unit is significant, in the order 1000-2000 m/s. One reason for this scattering could be that at the same depth rocks of different ages could be present at different localities. Even in the same geological units, particularly those with very complicated structure, the depth range of individual periods could be very different. Another reason of the data scattering could be the different lithology and physical properties of rocks of the same age, which was not taken into account in this study. So, in the second step of our analysis the dependences of velocity versus depth were determined for geological periods from the Tertiary and Quaternary to the Cambrian. Similarly to the previous section, the velocities averaged for each 50 m have been approximated by polynomi-als, of first or second order, to best fit the data (Figs. 4-6).
Velocities for the Tertiary and Quaternary (TQ) sediments in Lowland deposits are shown in Fig. 4. For most of the Polish territory, TQ sediments are not thick, and their total thickness does not exceed 450 m. On average, the velocity increases from 1787 m/s at the surface to 1964 m/s at 300 m depth. So, the velocities in the TQ complex and their scattering are relatively small. Therefore, they were analyzed jointly for all boreholes and one aver-age formula could by applied for the whole Lowland. North of the Carpathi-ans, in the Carpathian Foredeep, the thickness of TQ complex reaches a value of about 3000 m (for units Cb, Cd and Ce; see Fig. 4) and a relatively high velocity, of about 3700 m/s.
Fig. 4. Velocity-depth distribution for the Tertiary–Quaternary for Lowland units: A, B1, B2, B3, B4, Ca, Cc, and Cf ; and southern Poland units: Cb , Cd , and Ce. Note dif-ferent scales of graphs. Explanations as in Fig. 3. Colour version of this figure avail-able in electronic edition only.
SEISMIC WAVE VELOCITIES IN THE SEDIMENTARY COVER OF POLAND
993
Fig. 5. Velocity-depth distribution for the Carpathians (area D) for the Tertiary–Quaternary (TQ), Cretaceous, Jurassic to Cambrian and all Carpathian sediments. For comparison, the TQ Lowland line shows velocity-depth distribution for the Tertiary–Quaternary for Lowland units (cf. Fig. 4). Explanations as in Fig. 3. Colour version of this figure available in electronic edition only.
The VSP data from the Carpathians (area D) are dominated by thick TQ and Cretaceous complex, while deeper sediments (Jurassic to Cambrian) were reached in a small number of boreholes. Velocities for them are shown in Fig. 5. The Carpathian TQ velocities are about 500 m/s higher than the TQ velocities for the Lowland, as can be seen from a comparison with Fig. 4.
A collection of charts with velocity analysis for periods from Cretaceous to Cambrian outside of the Carpathians is shown in Fig. 6. Also in this case, the average line with ±500 m/s band is shown in the background for com-
M. GRAD and M. POLKOWSKI
994
Fig. 6. Velocity-depth distribution for the whole study area for geological periods from the Cretaceous to the Cambrian. Explanations as in Fig. 3. Colour version of this figure available in electronic edition only.
parison. The scattering of velocities for the same geological period is still significant, but the data seem to be less dispersed, with a clear velocity in-crease with depth in most cases. For the Cretaceous, the linear velocity-depth curve is satisfactory to describe the data down to 2846 m. For the Jurassic, Triassic and Carbonifereous, the second order polynomials practically follow the average line down to depths of 3679, 4913, and 5580 m, respectively. The exceptions are the Permian and Devonian rocks for which an increase of velocity with depth is not significant. In both cases, velocities are relatively high, in the range of 4800-5200 m/s, and in some cases constant velocities could be applied in the whole range of depths, 4778 ± 188 and 4753 ± 326 m/s, respectively, for the Permian and Devonian. Silurian rocks are
SEISMIC WAVE VELOCITIES IN THE SEDIMENTARY COVER OF POLAND
995
characterized by relatively low velocities; at a depth of 2000 m they are by about 500 m/s lower than the average value. Finally, for the Ordovician and Cambrian, the velocity-depth dependence is close to the average line. The amount of data for Ordovician and Cambrian is small in comparison to younger periods, and in practice the use of the average dependence could be satisfactory in those cases.
In the studied sedimentary cover, a distinct increase of velocity with depth and large velocity differentiation is observed. The lowest values are observed in the uppermost TQ deposits (about 1800 m/s), while the highest values are observed for the Permian and Devonian (about 5000 m/s). The ve-locity contrast at the Triassic–Permian boundary is of the order of 1000 m/s, forming the most distinct seismic discontinuity in the sedimentary cover (about 1200 m/s at 1000 m depth, and about 800 m/s at 2000 m depth).
5. V(z) ANALYSIS FOR GEOLOGICAL PERIODS AND UNITS
A large amount of the VSP data spread across Poland allows for a more pre-cise analysis, at least for some geological periods and units. In the Carpathi-ans, the entire sedimentary cover reached by drillings is basically composed of flysch deposits (TQ, Cretaceous and Upper Jurassic). Such an analysis for the Carpathians was presented earlier in Fig. 5, and here will not be repeated. So, a combination of 9 periods and 11 geological units should give 99 for-mulas. In fact, this number is lower because of the lack of some deposits in the sedimentary cover within some geological units (e.g., Cretaceous is miss-ing in the Upper Silesian block), or because of an insufficiently deep pene-tration by wells (e.g., Older Palaeozoic strata beneath the Pomorze–Kujawy anticlinorium). Also, the amount of available data for the periods and units is not uniform and changes for all the studied units and periods.
Examples of selected cases are shown in Fig. 7. For the Cretaceous, in area A the data down to 1200 m depth are well approximated by a linear function. A similar situation is found for the Silurian in area A, the Triassic in area B1, the Cambrian in area B1, and the Cretaceous in area Ca. In all these cases, we can apply a linear approximation in the shown depth range, which means, we have a sufficient number of data points for depths. For the Permian in area B4, the amount of available data is relatively big, with a sig-nificant scattering. However, average values are close to constant through the whole depth range. For the Devonian in area B2, and the Cambrian in area Cf, the data coverage is very limited, and in both cases the approxima-tion by a constant value seems to be the best solution. The problem in all the above-mentioned examples appears when we need a value of velocity for the depth outside of this range. A constant or linear interpolation could lead to
M. GRAD and M. POLKOWSKI
996
Fig. 7. Examples of velocity-depth distribution for chosen areas and geological peri-ods. Thick deep green lines are an average velocity-depth dependence for the whole area shifted by value x to attain the best fit of data. Other explanations as in Fig. 3. Colour version of this figure available in electronic edition only.
underestimated or overestimated values. In such a case, we suggest extrapo-lation using an average relation shifted by value x
This formula better expresses the velocity gradient in the shallow and deep part of the diagram. For example, for the Cambrian deposits in area B1 the use of linear formula determined for the depth interval 2601-3880 m could lead to an overestimation of velocities both for shallower and deeper depths (Fig. 7). The value of shift x in the above-mentioned examples changes from
SEISMIC WAVE VELOCITIES IN THE SEDIMENTARY COVER OF POLAND
997
Fig. 8. Velocity-depth distribution for all units and geological periods. The numbers of used boreholes are given in brackets. Explanations as in Fig. 7. Values of coeffi-cients with standard deviations for interpolated polynomials are given in Table 1. Colour version of this figure available in electronic edition only.
–428.0 m for Triassic in area B1, to 766.3 m for Cambrian in area Cf . It is a good illustration of velocity range between different units and periods, which reaches about 1200 m/s for average values at the same depth.
An example of a good approximation through the whole depth range can be observed for the Triassic in area B2 (Fig. 7). In this case, the averaged
M. GRAD and M. POLKOWSKI
998
Tabl
e co
ntin
ues o
n ne
xt p
age
Tab
le 1
Se
ism
ic P
-wav
e ve
loci
ties o
f the
sedi
men
tary
cov
er in
Pol
and
SEISMIC WAVE VELOCITIES IN THE SEDIMENTARY COVER OF POLAND
999
Exp
lana
tion:
A, B
1, ...
, Cf –
geol
ogic
al u
nits
of
Pola
nd s
impl
ified
fro
m S
okoł
owsk
i (19
68),
see
Figu
re 1
; a, b
, c –
coe
ffic
ient
sof
pol
ynom
ial V
p =
a +
bz
+ cz
2 w
ith s
tand
ard
devi
atio
ns; Δ
– d
epth
inte
rval
for t
he V
SP d
ata
(in m
); Σ
– to
tal l
engt
h of
VSP
dat
a(in
m);
x –
shift
of g
loba
l cur
ve d
escr
ibed
by
Eq. (
2) w
ith st
anda
rd d
evia
tion.
M. GRAD and M. POLKOWSKI
1000
curve (red line), approximation curve (black line), average extrapolation (thin light green line) and shifted curve (thick dark green line) are in fact identical, with a difference of just a few tens of meters (the value of shift is x = 20.8 m).
The collection of diagrams for all geological periods and units is pre-sented in Fig. 8, and corresponding values of coefficients for interpolated polynomials with standard deviations, depth ranges, and values of shift x are listed in Table 1. Additionally, in this table the depth range for the VSP data is given by Δ, and Σ gives total length of VSP data (both in meters). Figure 8 is valuable because in a compact form it gives visual information about the amount of data (numbers in brackets correspond to the number of boreholes used), scattering of data, quality of fit, and where the data are not available. Together with Table 1 it gives a full description of velocity changes with depth for all geological periods and units.
6. SUMMARY AND CONCLUSIONS
The results of the velocity analysis presented in this paper are derived from the most comprehensive study of the area of Poland published so far. In the complex area of junction of major tectonic units: the Precambrian East European Craton, the Palaeozoic Platform of Central and Western Europe, and the Carpathian thrust belt, seismic P-wave velocities in the sedimentary cover were analyzed using vertical seismic profiling data from 1188 bore-holes. The dependence of velocity versus depth has been approximated by polynomials down to a 6000 m depth (a) for the whole area, (b) for regional geological units, (c) for geological periods from the Tertiary and Quaternary to the Cambrian and, in greater detail, (d) for the periods and regional geo-logical units.
General formula. The most general formula for the territory of Poland describes the velocity versus depth relation by a third-order polynomial (Fig. 3 and Fig. 9a) which is given by formula (2) with shift x = 0. It is the first formula for the whole country based on such a huge amount of VSP data. Taking into account the complicated structure of the sedimentary cover, appurtenance to so different geological units and periods and very significant scattering of the data, this approach could be used for making very general comparisons. Similar compilations for the area of the Swabian Molasse Basin (John 1956), the Pannonian Basin − Great Hungarian Plain (20 wells in eastern Hungary; Mészáros and Zilahi-Sebess 2001) and the Norwegian shelf (linear velocity-depth trend in sedimentary rocks from 60 wells; Storvoll et al. 2005) are compared with our results in Fig. 9b. The lowest average velocities are observed for the Norwegian shelf and the
SEISMIC WAVE VELOCITIES IN THE SEDIMENTARY COVER OF POLAND
Fig. 9: (a) Summary of data of P-wave seismic velocities of the sedimentary cover measured in boreholes in Poland. The data were sampled with 1 m depth and the map shows the number of points in the depth interval of 50 m and the velocity inter-val of 50 m/s. The total number of points is about 2.3 million, which corresponds to about 2300 km of total length of vertical seismic profiling in boreholes. Colour scale shows a density of data. (b) Our result is compared to other regions: Pannonian Ba-sin (Mészáros and Zilahi-Sebess 2001), Swabian Molasse Basin (John 1956) and Norwegian shelf (Storvoll et al. 2005). Colour version of this figure available in electronic edition only.
largest average velocity on the territory of Poland. The youngest deposits of the Pannonian Basin and the Swabian Molasse Basin have intermediate val-ues. Higher values of velocities in Poland, particularly at the depth range 1000-4000 m, could result from the occurrence of high-velocity Permian and Devonian deposits within relatively large areas (e.g., B2, B3). The general formula could be used in more regional studies, e.g., in global seismic tomography.
Formulas for specific units. Formulas for velocity-depth trends for geo-logical units give better space distribution, and take into account regional differentiation of average velocity for 14 geological units simplified from the map of Sokołowski (1968): A, B1, ..., Dc (see Fig. 1). For each individual geological unit, an approximated polynomial is given (Fig. 3). Average velocities are stable, and second-order polynomials describe well the increase of velocity with depth. We recommend to use these formulas in the depth range of formula application (marked in Fig. 3 by horizontal dashed lines), and use constant velocities below.
Formulas for periods. Another approach of velocity is given for geolog-ical periods. For the Tertiary–Quaternary, a common formula could be applied at the whole Lowland (Fig. 4) and for the Carpathians (Fig. 5). For
M. GRAD and M. POLKOWSKI
1002
other geological periods, 9 formulas are given (Cretaceous to Cambrian; Fig. 6). The TQ velocities are the lowest at the Lowland (about 1800 m/s), by about 500 m/s lower than those in the Carpathians. Velocities for periods are stable, being the highest for Permian, Devonian (about 5000 m/s) and the oldest Palaeozoic. The velocity contrast at the Triassic-Upper Permian (Zechstein) boundary is of the order of 1000 m/s, creating a reflecting “screen” for seismic waves, particularly effective in near-vertical seismic reflection surveys. For this reason, the detailed seismic structure of the sub-Permian strata is poorly known (e.g., within geological units B2, B3, B4). Because the velocity increase in the Permian and Devonian rocks is not sig-nificant, constant velocities could in some cases be applied in the whole range of depths, 4778 ± 188 and 4753 ± 326 m/s, respectively. The known stratigraphy formulas for the periods give a simple and easy way for deriving the distribution of velocity in sedimentary complex.
Detailed formulas for periods and regions. A large amount of VSP data permitted an even more detailed analysis of seismic velocities carried out for particular periods and geological units. Figure 7 and Table 1 give a collec-tion of 71 formulas both for individual geological units A, B1, ..., Dc (Fig. 1) and geological periods from the Cretaceous to the Cambrian. They allowed for construction of averaged velocity distributions when stratigraphy and geological units are known. Data of this kind could be useful for regional studies relying on seismic velocities: identification and location local seismic events (e.g., Wiejacz and Rudziński 2010), deep seismic soundings, receiver function analysis (e.g., Trojanowski and Wilde-Piórko 2012), local and regional seismic tomography. Better understanding of seismic velocities for sedimentary cover in the area of Poland could also be useful, for example, for gravity and thermal modelling. Averaged velocities, however, would not be satisfactory enough for near-vertical reflection seismic and very local geophysical investigations – in such a case, data from individual boreholes nearby the seismic profile or in the investigated area should be used for calibration.
***
Results of VSP data analysis presented in this paper complement the previous results (e.g., Śliwiński 1965, Grad 1987, 1991), particularly for the Carpathians, and provide the velocity- depth relationship for all periods and geological units in a systematic way. Dense coverage by boreholes and about 2300 km of VSP measurements in boreholes give a reliable distribution of Vp velocity with depth and for rocks of different age. This is important, because the seismic P-wave velocity is a basic characteristic in investiga-tions of the structure and composition of the sedimentary basins and crystal-
SEISMIC WAVE VELOCITIES IN THE SEDIMENTARY COVER OF POLAND
1003
line crust (e.g., Christensen and Mooney 1995). Seismic velocities are often interpreted together with other physical parameters: density in gravimetric investigations (e.g., Grabowska and Raczyńska 1991, Królikowski and Petecki 1997, Krysiński et al. 2000, Grabowska et al. 2011) and thermal in heat flow modelling (e.g., Majorowicz et al. 2003).
Acknowledgments . The VSP data collected in the 1980s are from the Archives of the PBG Geophysical Exploration; more recent data are from the Central Geological Archives.
R e f e r e n c e s
Berthelsen, A. (1992), From Precambrian to Variscan Europe. In: D.J. Blundell, R. Freeman, and S. Mueller (eds.), A Continent Revealed – The European Geotraverse, 153-164, Cambridge Univ. Press, New York.
Berthelsen, A. (1998), The Tornquist Zone northwest of the Carpathians: An intra-plate-pseudosuture, GFF 120, 223-230.
Christensen, N.I., and W.D. Mooney (1995), Seismic velocity structure and compo-sition of the continental crust: A global view, J. Geophys. Res. 100, B6, 9761-9788, DOI:10.1029/95JB00259.
Grabowska, T., and M. Raczyńska (1991), Structure of the Earth’s crust on the Pol-ish lowland in the light of gravimetric modelling, Publs. Inst. Geophys. Pol. Acad. Sc. A19, 230, 85-110.
Grabowska, T., G. Bojdys, M. Bielik, and K. Csicsay (2011), Density and magnetic models of the lithosphere along CELEBRATION 2000 profile CEL01, Acta Geophys. 59, 3, 526-560, DOI: 10.2478/s11600-011-0007-3.
Grad, M. (1987), Velocities of seismic waves in sedimentary cover of the East European platform, Kwart. Geol. 31, 97-114 (in Polish with English sum-mary).
Grad, M. (1991), Seismic wave velocities in the sedimentary cover of the Palaeozoic platform in Poland, Bull. Pol. Acad., Earth Sci. 39, 1, 13-22.
Grad, M., T. Trung Doan, and W. Klimkowski (1991), Seismic models of sedimen-tary cover of the Precambrian and Palaeozoic platforms in Poland, Publs. Inst. Geophys. Pol. Acad. Sc. A19, 236, 125-145.
Grad, M., T. Janik, J. Yliniemi, A. Guterch, U. Luosto, T. Tiira, K. Komminaho, P. Środa, K. Höing, J. Makris, and C.-E. Lund (1999), Crustal structure of the mid-Polish trough beneath the Teisseyre-Tornquist Zone seismic pro-file, Tectonophysics 314, 145-160.
Grad, M., S.L. Jensen, G.R. Keller, A. Guterch, H. Thybo, T. Janik, T. Tiira, J. Yliniemi, U. Luosto, G. Motuza, V. Nasedkin, W. Czuba, E. Gaczyński,
M. GRAD and M. POLKOWSKI
1004
P. Środa, K.C. Miller, M. Wilde-Piórko, K. Komminaho, J. Jacyna, and L. Korabliova (2003), Crustal structure of the Trans-European suture zone region along POLONAISE’97 seismic profile P4, J. Geophys. Res. 108, B11, 2541, DOI: 10.1029/2003JB002426.
Grad, M., A. Guterch, G.R. Keller, T. Janik, E. Hegedűs, J. Vozár, A. Ślączka, T. Tiira, and J. Yliniemi (2006), Lithospheric structure beneath trans-Carpathian transect from Precambrian platform to Pannonian basin: CELEBRATION2000 seismic profile CEL05, J. Geophys. Res. 111, B03301, DOI: 10.1029/2005JB003647.
Grad, M., A. Guterch, S. Mazur, G.R. Keller, A. Špičák, P. Hrubcová, and W.H. Geissler (2008), Lithospheric structure of the Bohemian Massif and adjacent Variscan belt in central Europe based on profile S01 from the SUDETES 2003 experiment, J. Geophys. Res. 113, B10304, DOI: 10.1029/ 2007JB005497.
Guterch, A., M. Grad, T. Janik, R. Materzok, U. Luosto, J. Yliniemi, E. Lueck, A. Schultze, and K. Foerste (1994), Crustal structure of the transition zone between Precambrian and Variscan Europe from new seismic data along LT-7 profile (NW Poland and eastern Germany), C.R. Acad. Sci. Ser. II, 319, 1489-1496.
Janik, T., J. Yliniemi, M. Grad, H. Thybo, T. Tiira, and POLONAISE P2 Working Group (2002), Crustal differentiation across TESZ along POLONAISE’97 seismic profile P2 in NW Poland, Tectonophysics 360, 129-152.
Janik, T., M. Grad, A. Guterch, J. Vozár, M. Bielik, A. Vozárova, E. Hegedűs, C.S. Kovács, I. Kovács, G.R. Keller, and CELEBRATION 2000 Working Group (2011), Crustal structure of the Western Carpathians and Pannonian Basin: seismic models from CELEBRATION 2000 data and geological im-plications, J. Geod. 52, 97-113, DOI: 10.1016/j.jog.2010.12.002.
John, H. (1956), Das Geschwindigkeitsproblem im Bereich der Schäb Vorland-smolasse, Erdöl und Kohle 9, 5, 290-299.
Kamińska, E., and J. Zagórski (1978), Analysis of seismic velocities in Warsaw Synclinorium (Analiza prędkości sejsmicznych w synklinorium warszawskim), Nafta 33, 293-300 (in Polish).
Królikowski, C., and Z. Petecki (1997), Crustal structure at the Trans-European suture zone in northwest Poland based on gravity data, Geol. Mag. 134, 5, 661-667.
Krysiński, L., M. Grad, and POLONAISE Working Group (2000), POLONAISE’97 – Seismic and gravimetric modelling of the crustal structure in the Polish basin, Phys. Chem. Earth A, 25, 4, 355-363, DOI: 10.1016/S1464-1895(00) 00057-0.
Majorowicz, J.A., V. Čermak, J. Šafanda, P. Krzywiec, M. Wróblewska, A. Guterch, and M. Grad (2003), Heat flow models across the Trans-European Suture Zone in the area of the POLONAISE’97 seismic experiment, Phys. Chem. Earth 28, 9-11, 375-391, DOI: 10.1016/S1474-7065(03)00059-7.
SEISMIC WAVE VELOCITIES IN THE SEDIMENTARY COVER OF POLAND
1005
Mészáros, F., and L. Zilahi-Sebess (2001), Compaction of sediments with great thickness in the Pannonian Basin, Geophys. Trans. 44, 1, 21-48.
Młynarski, S. (1984), The structure of deep bedrock in Poland on the basis of refrac-tion results, Publs. Inst. Geophys. Pol. Acad. Sc. A13, 160, 87-100.
Pożaryski, W., W. Brochwicz-Lewiński, and H. Tomczyk (1982), Sur la caractére hétérochronique de la Ligne Teisseyre-Tornquist, entre Europe centrale et orientale, Tectonique, C.R. Acad. Sci. Paris 295 Sér. II, 691-696.
Ptak, A. (1966), Preliminary regional elaboration of seismic velocities in Warsaw Synclinorium (Wstępne regionalne opracowanie rozkładu prędkości w obrębie synklinorium warszawskiego), Geofiz. Geol. Naft. 5-6, 133-136 (in Polish).
Skorupa, J. (1974), Seismic velocity map of Poland 1:500 000, Wyd. Geol., Warszawa.
Sokołowski, J. (1968), Geology and structure of the regional units of Poland from the point of view oil exploration (Charakterystyka geologiczna i struk-turalna jednostek regionalnych Polski pod kątem poszukiwań bitumitów), Surowce Mineralne 1, 7-58 (in Polish).
Stefaniuk, M., and W. Klityński (2007), Magnetotelluric map of Carpathians, Table III.1. In: Geophysical Atlas of Carpathians, CAG PGI, Warszawa.
Storvoll, V., K. Bjørlykke, and N.H. Mondol (2005), Velocity-depth trends in Meso-zoic and Cenozoic sediments from the Norwegian Shelf, AAPG Bulletin 89, 3, 359-381, DOI: 10.1306/10150404033.
Śliwiński, Z. (1965), Velocity model of the western fore-Sudetic area (Model pręd-kościowy zachodniej części obszaru przedsudeckiego), Przegl. Geol. 13, 313-316 (in Polish).
Środa, P., W. Czuba, M. Grad, A. Guterch, A.K. Tokarski, T. Janik, M. Rauch, G.R. Keller, E. Hegedűs, J. Vozár, and CELEBRATION 2000 Working Group (2006), Crustal and upper mantle structure of the Western Carpathi-ans from CELEBRATION 2000 profiles CEL01 and CEL04: seismic mod-els and geological implications, Geophys. J. Int. 167, 737-760, DOI: 10.1111/j.1365-246X.2006.03310.x.
Świtek, J. (1983), Seismic and acoustic characteristic of the Palaeozoic strata in Pomerania-Warsaw Synclinorium from borehole data (Charakterystyka sejsmiczna i akustyczna utworów paleozoicznych synklinorium pomorsko-warszawskiego na podstawie pomiarów w odwiertach), Nafta 39, 41-46 (in Polish).
Toporkiewicz, S. (1984), The structure of the consolidated basement of the Palaeo-zoic Platform in south-west Poland, Publs. Inst. Geophys. Pol. Acad. Sc. A13, 160, 101-117.
Trojanowski, J., and M. Wilde-Piórko (2012), Velocity structure beneath the Bohe-mian Massif from Monte Carlo inversion of seismic receiver function, Acta Geophys. 60, 1, 76-91, DOI: 10.2478/s11600-011-0047-8.
M. GRAD and M. POLKOWSKI
1006
Wiejacz, P., and Ł. Rudziński (2010), Seismic event of January 22, 2010 near Bełchatów, Poland, Acta Geophys. 58, 6, 988-994, DOI: 10.2478/s11600-010-0030-9.
Znosko, J. (1975), Tectonic units of Poland against the background of the tectonics of Europe, Geol. Inst. Anniversary Bull. 252, 61-75.
Znosko, J. (1979), Teisseyre-Tornquist tectonic zone: some interpretative implica-tions of recent geological and geophysical investigations, Acta Geol. Pol. 29, 365-382.
Żaruk, J. (1971), Analysis of vertical seismic profiling in boreholes of Pomerania Synclinorium (Analiza wyników profilowania prędkości w otworach synk-linorium pomorskiego), Przegl. Geol. 19, 138-145 (in Polish).
