Originally published as:

Orlecka-Sikora, B., Papadimitriou, E. E., Kwiatek, G. (2009): A study of the interaction

among mining-induced seismic events in the Legnica-Głogów Copper District, Poland. - Acta
Geophysica, 57, 2, 413-434

DOI: 10.2478/s11600-008-0085-z
 A study of the interaction among mining induced seismic events in the
Legnica–Glogow Copper District, Poland

B. Orlecka–Sikora1, E.E. Papadimitriou2, G. Kwiatek3

1
Faculty of Geology Geophysics and Environmental Protection, AGH University of Science and Technology, al.
Mickiewicza 30, 30–059 Kraków, Poland, e–mail: orlecka@geol.agh.edu.pl – corresponding author
2
Department of Geophysics, University of Thessaloniki, GR54124 Thessaloniki, Greece, e–mail:
ritsa@geo.auth.gr
3
GeoForschungsZentrum Potsdam, Dept. 3.2 Deformation and Rheology, Telegrafenberg D427, 14473 Potsdam,
Germany, e–mail: kwiatek@gfz–potsdam.de

Abstract

Seismic hazard assessment in mining areas is of paramount importance for the nearby built
environment since local events, although of small or moderate magnitude, because their locations
are just beneath and very superficial, they have caused serious damage and often loss of life, and
on top of that they exhibit a very high occurrence rate. Based on the fact that this activity also
exhibits time dependence, as has been shown by several authors previously, and that small stress
perturbation can enhance or prohibit future occurrences, we have applied the Coulomb stress
transfer technique to investigate interactions among seismic events induced by mining works in
the Rudna Mine in the Legnica–Głogów Copper District in south–west Poland. The coseismic
stress changes due to tremor occurrences are only a small component of the stress field in mining
areas and are not capable of generating a future seismic event. But if the rock mass at the location
of the next event is already close enough to failure they can move it into the failure regime.
Therefore, we have examined the influence of the cumulative static stress changes due to
previous events on the generation of subsequent ones. For this purpose we considered events with
energy greater than 105J with a known focal mechanism that occurred in the LGCD area during
the time period 1993–1999. We then calculated Coulomb stress changes (∆CFF) after the
occurrence of each event inverting each time the derived stress field according to the faulting
type of the next event in the sequence of seismic events. We assumed the state of stress before
the first seismic event in the analyzed sequence to be zero and the time changes of stress were
referred to this initial value. At each stage of our calculations we attempted to correlate the
location of the seismic event with the calculated values of ∆CFF.
The results of this study indicate that in many cases strong mining tremors produce changes
in the state of stress of a sufficient magnitude to influence subsequent events. The location of

1
over 60% of events is consistent with stress–enhanced areas where the values of positive ∆CFF
were above 0.01 MPa. For most of the events located inside areas of a calculated negative ∆CFF,
their modelled rupture zone was partially located inside stress enhanced, providing thus
additional evidence for possible triggering at the nucleation point.

Key words: LGCD, Rudna Mine, induced seismicity, Coulomb stress changes

1. INTRODUCTION

The mining history in the Lower Silesia (LS) area in south–west Poland goes back to the second
half of the 1930s. Currently, copper ore is exploited in the Legnica–Głogów Copper District (LGCD),
located in LS, in three mines: namely Lubin, Polkowice–Sieroszowice and Rudna. The long lasting
mining activity has caused perturbation of the stress field in the rock mass, resulting in intense seismic
activity. The mine seismological networks record every year several thousands events of local
magnitudes ranging from 0.4 to 4.5. Within a large set of induced seismic events, rockbursts are the
most serious phenomena in mining operations. Although rockbursts in the Polish mines are only a
small subset of seismic events, they cause much damage with severe casualties. In the period 1985–
2006 (22 years), on average three rockbursts were triggered annually by tremors, which resulted in
over 10 accidents and two fatalities (Kłeczek, 2007). In the LGCD, the seismic activity also occurs
directly beneath urbanized and industrialized areas, thus causing very strong ground motions affecting
the natural and built environment. The strongest events have caused peak ground accelerations of
almost 2.0 m/s2 (Lasocki, 2005).
McGarr and Simpson (1997) classified seismicity accompanying mining exploitation as “induced”
seismicity, which is expected to occur in response to stress changes that are comparable to typical
earthquake stress drops. The generation process of mining induced seismic events is complicated by
various time–variable factors of both natural and anthropogenic origin. In some regions of Poland,
induced seismicity is also affected by the local geological and tectonic conditions. Strong tremors are
considered as a result of interaction between the mining, lithostatic and tectonic stresses (e.g. Marcak,
1985; Gibowicz, 1990). Studies of seismicity in Polish mines have confirmed that at least two types of
seismic events are observed, those directly connected with mining works and those associated with the
movement on major geological discontinuities (e.g. Kijko et al., 1985, 1987; Idziak et al., 1991;
Gibowicz and Kijko, 1994; Gibowicz and Lasocki, 2001). These two types of seismicity result in
complex and multimodal magnitude distribution (e.g. Lasocki, 2001).

Lasocki (1992a, b) proved that the series of seismic events from individual mining stopes are not
Poissonian, but show a time–variability in the long term. Several authors have also provided the
evidence for interrelations of mining seismic events in the form of the spatio–temporal clustering of
seismicity (e.g. Trifu et al., 1993; Gibowicz, 1997; Orlecka–Sikora and Lasocki, 2002). Kijko (1997)

2
in studying the interevent time of seismicity from one of the deep gold mines in South Africa observed
a trigger effect indicating internal correlations in the earthquake occurrence process. Similar
conclusions have been derived by Węglarczyk and Lasocki (2008) who applied the Hurst rescaled
analysis to the interevent times of seismicity from LGCD in Poland. Studies of the spatio–temporal
distribution of seismicity recorded in the Creighton Mine in Canada, carried out by Marsan et al.
(1999), confirmed the existence of a stress diffusion mechanism and its influence on the stronger
events occurrence. Gibowicz (2006) found that the occurrence of mining–induced seismic events is
capable of increasing the probability of a second event. The author analyzed the seismicity from the
Wujek and Ziemowit coal mines and two copper mines in LGCD, namely the Polkowice and Rudna,
and observed seismic doublets and multiplets in magnitude range of 0.7–3.5. They also noticed the
correlation between the dominant direction of the doublets' spatial distribution and the azimuths of the
nodal planes of some events forming doublets.

In recent years many studies of temporal and spatial patterns of earthquake occurrence have
concentrated on examining possible correlations between earthquakes and the role of stress transfer in
this process. Chinnery (1963) observed that changes in the stress field produced by a shear failure
appeared in an area larger than one fault length. Later studies have presented the stress changes
influencing the space–time patterns of aftershocks (e.g. Das and Scholz, 1981; Stein and Lisowski,
1983). Nowadays, fault interaction is considered as an integral part of the seismic hazard assessment,
and is often investigated by static stress transfer models at different spatial and time scales (e.g. Harris,
1998 and the references therein; Steacy et al., 2005 and the references therein).

Static stress changes are calculated considering a seismic source as a dislocation imbedded in a
half–space and are expressed in terms of the Coulomb failure function (CFF) (e.g. King et al., 1994;
King and Cocco, 2001). The advantage of using this approach is that besides that the absolute value of
stress is not known, stress changes can be calculated on the basis of details of geometry and the slip
direction of earthquake rupture. The stress changes affecting subsequent events are typically of the
order of 0.01 MPa, which are a small fraction of the corresponding stress drop (e.g. Harris, 1998 and
the references therein). Coulomb stress changes can adequately explain the occurrence patterns of both
small and strong events, and have been used as a powerful tool for the assessment of future seismic
hazard in certain areas. Relevant studies have shown that this method is more effective if in addition to
the accumulated stress changes due to the coseismic slips only, the long term tectonic loading is taken
into account thus determining the evolution of the stress field in time (e.g. Deng and Sykes, 1997;
Papadimitriou and Sykes, 2001).

The obtained results in natural seismicity encouraged us to apply the Coulomb stress transfer
technique to investigate interactions among seismic events induced by mining works. The coseismic
stress change associated with a given tremor is only one small component of the stress field in mining
areas and is not enough to generate another seismic event, as is also the case in strong earthquake

3
occurrence. Nevertheless, if the rock mass at the location of the next event is already close enough to
failure the first tremor can trigger the second one by introducing a positive Coulomb stress change to
move it into the failure regime. Therefore, calculation of the static stress changes and their association
with the spatial pattern of future occurrences will greatly contribute to the assessment of a future
seismic hazard in LGCD.

Aiming to examine the influence of the cumulative static stress changes due to previous events on
the generation of the next ones, the spatio–temporal earthquake occurrence in the Rudna mine area in
LGCD in Poland is investigated and at each stage of the calculations the future occurrences are
correlated with the resulted stress field, each time inverted according to the faulting type of the next
event whose triggering is inspected.

2. METHOD

The first step in the analysis of faults interaction by modeling the static stress transfer is the
calculation of the stress field associated with a particular fault. Static displacements, strains and
stresses are calculated by solving the elastostatic equation for a dislocation on an extended fault in an
elastic, isotropic and homogeneous half–space (e.g. King and Cocco, 2001). The displacement field
produced by a dislocation uj across a surface  in a uniform elastic half–space is given by Voltera’s
formula (Steketee, 1958a, b):

 u l  u j u k 
ui   u j  jk  i    i  i  k dS , (1)
 l   
   k  j 

where  jk is the Kronecker delta,  and  are the Lamé’s constants, uij is the ith component of

displacement at (x1, x2, x3) due to a point force of unit magnitude (x1, x2, x3) in the j–direction and  k

is the outward normal vector to . Fault geometry, the slip distribution and the strain nuclei are
necessary for the computation of the static displacement. Chinnery (1961, 1963) and Okada (1985,
1992) have derived analytical expressions for the static displacement and strain fields caused by a
finite rectangular fault at the Earth’s surface and at any depth. The stress tensor components are
calculated from the strain eij according to Hooke’s law for an isotropic medium:

2
sij   ij ekk  2eij , (2)
1  2
where  is the Poisson ratio and  ij is the Kronecker delta.

The second step is the choice of criterion characterizing failure in rocks. Among various criteria
the more widely used is the Coulomb failure criterion (Jaeger and Cook, 1979; Scholz, 1990), which
quantifies the closeness of a fault to failure. This criterion has been used by a number of authors to

4
study the distribution of aftershocks after a large earthquake and to investigate static stress changes on
other faults in its vicinity (e.g. Das and Scholz, 1981; Stein and Lisowski, 1983; Reasenberg and
Simpson, 1992; Stein et al., 1992; Hudnut et al., 1994; King et al., 1994; Parsons et al., 1999).
According to the Coulomb criterion, failure occurs on a plane when the Coulomb stress f exceeds a
specific value:

f =  + (n + p), (3)

where  is the shear stress on the failure plane, n is the normal stress, p is the pore fluid pressure and
 is the coefficient of friction. Both n and  are calculated for a fault plane at the observation point
from the stress tensor defined by equation (2). The difference in the sign of  indicates whether the
potential for a slip on the plane is right– or left–lateral. The coefficient of friction and the fluid pore
pressure can be combined together as the apparent coefficient of friction, ΄. Then the Coulomb
failure function (CFF) (3) can be rewritten as:

f =  + ΄n, (4)

The change in shear stress is positive for increasing shear stress in the direction of a relative slip on the
observed fault, the normal stress is positive for increasing tensional normal stress. A positive value of
CFF moves a fault toward failure, a negative value of CFF moves it away from failure.
In the next step, the cumulative changes in stress due to the consecutive seismic events in the
analyzed time series are calculated. We assume the state of stress before the first seismic event in the
analyzed sequence to be zero and the time changes of stress are referred to this arbitrary zero baseline.
In our work we do not take into account the mining and lithostatic stress changes caused by mining
works, assuming that the stress changes were caused only by a coseismic slip. We put certain
coseismic displacements on ruptured planes and add the changes in the components of the stress tensor
as they occurred in time. It should be noted that stress is a tensional quantity and thus lobes with
positive and negative stress changes must be viewed in the context of a specific type of faulting.

3. DATA

The seismic network at the Rudna Mine is composed of 32 vertical Wilmore MK2 and MK3
seismometers recording ground velocity. The seismometers are located at the level of copper ore
deposits (except for a few ones located in shafts), at depths ranging from 300 m down to 1000 m. The
signals are transmitted in analogue form with FM modulation by standard cables used in mining up to
a central recorder located on the ground surface. Thereupon, signals are digitized with a sampling
frequency equal to 500 Hz and 14–bit resolution. The whole system works in a triggering mode, the
frequency band is from 0.5 to 150Hz and the system dynamics is approximately equal to 70dB
(Domański et al., 2002). The seismic network records annually several thousands of mining–induced

5
seismic events with local magnitudes, ML, ranging from 0.4 up to 4.5, with the completeness cutoff
being about 1.2. These events are considered to be directly related to copper ore mining.

For the following analysis, we selected the subset of 217 events with ML≥2.0 (seismic energy
equal to 1.0×105 J) that were recorded between June 6, 1993 and August 27, 1999. Figure 1 depicts the
study area, along with the seismic stations and the spatial distribution of the seismic events
incorporated in our stress changes calculations. Table 1 gives information on the annual number of
events above a certain magnitude (energy class).

[FIGURE 1 SOMEWHERE HERE]

[TABLE 1 SOMEWERE HERE]

The events, located by the mine’s staff, exhibit uncertainties in the epicentral coordinates of the
order of ±50 m, which are much smaller than the uncertainty in depth determination (Leśniak &
Pszczoła, 2008). This is because of the location of seismometers mainly at the level of copper ore
deposits. There, one can expect seismic P waves propagating almost horizontally along the geological
layers. It causes the problems with a proper detection of the first arrivals by the primarily vertical
sensors. As a consequence, the spatial distribution of seismic stations makes the whole seismic
network nearly planar (except for stations located in shafts), what results in a worse quality of the
vertical component of location (Kijko, 1982). However, most of events in LGCD area occur very close
to the exploitation level, directly above it or in an anhydrite layer c.a. 60 m above (Wiejacz, 1991)
what may be a constrain on sources’ depth. The estimated depth of analyzed events ranged from 0.3
km down to 1.1 km and, because accurate depth determination is not feasible, we performed the
following calculations with the depth finally fixed to 0.9 km for all events, which is representative for
the study area.

Two independent procedures were applied to the analyzed subset of seismic events: the moment
tensor inversion in time domain for the determination of the focal parameters and a spectral analysis
for the calculation of source parameters (Domański and Gibowicz, 2008). The detailed focal
mechanism solutions were provided by the Institute of Geophysics of the Polish Academy of Sciences.
The moment tensor inversion was based on the work of Fitch et al. (1980) and the special software
adjusted to the geological situation within mines. The input parameters were the amplitude and
polarity information on the first P–wave displacement pulses Fortunately, the waveform field in the
LGCD area displays the domination of refracted waves in the first P–wave arrivals over the direct P–
wave onsets, which are seen only at the closest distances (<1000 m). We observed significant head
waves refracted from the crystallic baseground located below the exploitation level and composed of
igneous rocks (incidence angle ~60º) and from anhydrite layer from above (incidence angle ~115º).
Both types of waves could be detected and used for improving the coverage of the focal sphere and the

6
quality of fault plane solutions. Haskell’s source model was assumed (Haskell, 1953) which requires
the rupture time to be calculated from the average first P–wave pulse durations. According to Wiejacz
(1991), this form of seismic source is expected to be a good approximation to the real mining–induced
data in the LGCD area. The deviatoric, pure shear and full moment tensor were calculated using the
L1 norm as a measure of the misfit and the method of the Lagrange multipliers (Wiejacz, 1991) was
used in seismic moment tensor inversion. The uncertainties of strike, dip and rake do not exceed 15º.
The analysis of unconstrained and decomposed moment tensors revealed that the type of process
responsible for the mine tremor source nucleation is not uniform. The generalized source mechanism
of the analyzed mining–induced seismic events contained about 15 per cent of the isotropic component
ISO, about 15 per cent of the uniaxial compressional or extensional component CLVD and about 70
per cent of the shear component DC in the total moment tensor solution. Fig. 2 presents the histograms
of the particular components contributed to the total seismic moment tensor solution of the analyzed
seismic events. The participation of the shear components in the focal mechanism is the most
significant, and the higher values of CLVD and/or ISO component may be predominantly attributed to
the inaccuracy of vertical component of location and, partially, to the insufficient focal sphere
coverage, as was pointed out by Wiejacz (1991), among others. Table 2 gives information on the fault
plane solutions of the strongest events.

[TABLE 2 SOMEWERE HERE]

[FIGURE 2 SOMEWERE HERE]

Spectral analysis was performed for a source–receiver distance greater than 3 km to suppress the
influence of the near field effects and the complexity of the wave field. The amplitude of waveforms
was corrected due to the vertical component used. A numerical integration filter was applied to the
ground velocity records to obtain the displacement waveforms. The selected parts of P– and S–phases
were tapered using a 10% von Hann’s window, then the FFT was applied to both the ground velocity
and displacement waveforms. The correction for instrument response has not been applied due to the
higher frequency content of the recorded signals. The spectra were corrected for attenuation using
QP=400 and QS=200, which represent the typical values of the quality factor used in LGCD. Spectral
parameters were estimated from P and S components of vertical seismometers assuming the -2 of
Brune’s model (Brune 1970, 1971) and the methodology developed by Andrews (1986). The J and K
integrals provided by Snoke (1987) were used to estimate seismic moment and corner frequency and
were corrected for limited frequency band. For the seismic moment, we applied the correction for
radiation pattern. Free surface, as well as site corrections was not used due to the downhole location of
seismic sensors.

7
 It is very well known that the estimates of the seismic source size are heavily model–dependent.
However, the quasidynamic model of circular fault of Madariaga (1976) with KP=2.01 and KS=1.32
for P and S waves, respectively, provides reasonable results in adequate agreement with independent
observations in mines (e.g. Gibowicz and Kijko, 1994 and the references therein). The calculated
circular fault radius and the seismic moment for each of the analyzed events are also given in Table 2.
Typically we obtained at least 20 estimates of source parameters from both P and S phases for each
event. The spectra that generated the outliers were reconsidered/removed and finally the average
values of source parameters were calculated. The error of magnitude and source radius estimation (i.e.
standard deviation), required for stress changes calculations, equaled to ±0.1 and ±50 m, respectively.

4. STRESS CHANGES CALCULATIONS AND EARTHQUAKE OCCURRENCE

Stress changes calculations were performed by the use of the software Coulomb 3.0 (Toda et al.,
2005; Lin and Stein, 2004). As it has been mentioned above, the stress field is inverted each time
according to the faulting type of the next event. There is, however, no adequate information for the
selection of the actual fault plane between the two determined. Although some generally accepted
criteria were taken into account (a smaller dip angle in the case of dip slip events, similarity in the
orientation of nearby events, etc), this selection is not unambiguous. Although it would result in
intrinsic uncertainty when correlating stress changes and events locations, if this correlation were to be
proved significant then the expectedly stronger correlation for the set of actual nodal planes would be
significant as well.

The geomechanical parameters of rocks in the LGCD vary in a wide range depending on the type
of rocks (e.g. Piestrzyński, 1996). We performed calculations with the shear modulus and the Poisson
ratio fixed as 2.2·104 MPa and 0.25, respectively, and the apparent coefficient of friction was taken to
be equal to 0.8 throughout our calculations.

The circular fault radius and seismic moment were used to calculate two additional parameters
necessary for the stress transfer model application, namely the fault dimension and average
displacement. The values of the circular fault radius were recalculated to the length and width of the
corresponding rectangular fault. The coseismic displacement along the ruptured plane was estimated
from the seismic moment, by taking the shear modulus equal to 2.2·104 MPa and considering the fault
area (e.g. Aki and Richards, 1980). Information on the rupture models of the events included in our
calculations is given in Table 3.

[TABLE 3 SOMEWERE HERE]

8
 Fig. 3 presents histograms of the frequency of seismic tremor occurrence as a function of CFF
changes (∆CFF) due to previous occurrences. The first histogram presents the frequency of all the
analyzed seismic events, while the second one only the events with a seismic moment larger or equal
to 1013 [N·m]. Negative changes in Coulomb stress denote a decreased likelihood of fault rupture,
while positive CFF changes denote an increased likelihood of failure. About 61 per cent of the total
considered seismic tremors occurred at locations of positive changes in stress created by previous
events. Most of these events, about 50 per cent, were located inside regions of ∆CFF values larger than
0.01 MPa. In the cases of stronger tremors, those with Mo≥1013 [N·m], the percentage of events
occurred at locations with positive changes is almost the same and more particularly, equal to 63% of
the total number of events. The remaining events occurred in areas of calculated negative ∆CFF but
for most of them the modeled rectangular fault is located partially inside stress enhanced areas,
although the hypocentre is located inside stress shadows (Fig. 4d, f). These observations provide
evidence for their possible triggering considering the nucleation point being inside positive stress
areas.

[FIGURE 3 SOMEWERE HERE]

Fig. 4 shows snapshots of the cumulative stress changes in the study area at a depth of 0.9 km in
the G–1/7 field of the Rudna Mine. The mine field is the unit of mine division, section of mine, the
region with the part of copper ore deposit intended for the extraction by the determined exploitation
method. The ΔCFF was calculated by taking into account all the events in our dataset, while in Figure
4 only the events that are marked to have occurred in G–1/7, are depicted. The white in Fig. 4
indicates no significant changes in Coulomb stress, blue regions denote negative changes and yellow
to red regions represent positive CFF changes. The mining works in the G–1/7 field were carried out
in the right bottom quadrant of pictures in a NW–SE direction.

[FIGURE 4 SOMEWERE HERE]

Fig. 4a shows the cumulative Coulomb stress changes due to the first 76 tremors in the studied
dataset, calculated according to the fault plane solution of the next event that occurred on 23.04.1996
in the G–1/7 field of the Rudna Mine. These events created in the G–1/7 field a shadow zone in a SW–
NE direction and two bright zones in the north–western and south–eastern parts (Fig. 4a). The
23.04.1996 event occurred at the border between bright and shadow zones, making triggering by
∆CFF doubtful. Then 8 other tremors with magnitude from 2 to 2.6 occurred in the vicinity of G–1/7,
5 in bright zones and 3 in shadows.

9
 Fig. 4b presents the state of the accumulated stress changes due to the first 85 seismic events in our
data sample, calculated according to the faulting type of the next event. These events caused a large
shadow zone in G–1/7 except for two smaller bright zones in the northern part of the area. The next
event with local magnitude of 2.9 occurred on 07.11.1996 in one of these two stress–enhanced zones.

Fig. 4c presents the state of stress just before the occurrence of the 88th event occurring on
08.11.1996, inverted according to its faulting type. After the occurrence of the previous event, on
07.11.1996, a bright zone was created at the site of the 88th event thus providing evidence of its
probable triggering; evidence that is more substantiated by the small difference in the occurrence time
of the two events (just one day).

The cumulative ∆CFF were calculated for faulting in agreement with the focal plane solution of
the next tremor, which occurred on 19.12.1996 (Fig. 4d). The epicenter of this event is located in a
patch of the shadow zone but its rupture zone is partially located inside the bright zone. The partial
location of the fault inside a stress enhanced area provides evidence of possible triggering.
After the 19.12.1996 event, 51 more tremors with ML>2.0 occurred in the Rudna mine, 21 of them
in the vicinity of G–1/7. Fig. 4e shows the state of stress just before the 09.09.1998 event of ML2.8.
Positive ∆CFF occupied a large zone in the western part of the area, continuing to the north–east and
two smaller zones in the south–east. The epicenter of the 09.09.1998 tremor was located inside a
bright zone, and more particularly at a site where the positive ∆CFF had a value equal to 0.26 MPa.

Fig. 4f presents the cumulative Coulomb stress changes associated with the sequence of events
considered earlier and additionally, the next 20 occurred in the Rudna mine before 13.10.1998. The
stress shadow is now limited to 3 zones, the first of them starting at the centre of the G–1/7 field and
continues to the north, the second located to the east part of the study area, and the third at the
southern part. The next event occurred on October the 13th 1998, with its epicenter being inside a
shadow zone, while about 30% of its rupture area is located inside a stress–enhanced zone. One can
observe that the large bright zones in the south–western and south–eastern part of G-1/7 still did not
experience events. The explanation is that the seismic events are induced by mining works and events
are expected to occur in places where exploitation is carried out.
The snapshot of Fig. 4g indicates the cumulative stress changes until just before the 04.07.1999
event. It includes the stress changes caused by 29 additional events which occurred during 9 months
from the last 13.10.1998 event in the Rudna mine and 10 of them in surrounding fields of the G–1/7
area. The epicenter of the 04.07.1999 event is located in a small patch of negative ΔCFF, although the
largest part of its rupture area is inside a stress enhanced area. Fig. 4h depicts the state of stress just
after this event, calculated for the fault plane solution of the next 05.07.1999 event, (ML=2.9), which is
located at a place of positive ∆CFF equal to 0.03 MPa. The stress drop caused by this event is one
order larger than the corresponding ∆CFF and equal 0.67MPa.

10
 The addition of coseismic stress changes associated with the occurrence of the next inspected 5
events in the Rudna mine is shown in Fig. 4i. These events caused enlargement of the shadow zone in
the north–eastern part of the presented area. The next event occurred on 12.08.1999 in a stress–
enhanced zone, where the ∆CFF was as before, equal to 0.03 MPa.

The occurrence of the 12.08.1999 event caused the appearance of an additional stress shadow zone
on the south–west (Fig. 4j). In the area where the exploitation was carried out the next event happened
on 13.08.1999. The epicenter of this event is located at the border between a bright zone and a stress
shadow.

5. CONCLUSIONS

We investigated possible interactions through perturbations of the stress field due to the coseismic
slip of the mining induced seismic events that occurred in the Rudna mine in the Legnica–Głogów
Copper District. For this purpose we modeled the static stress changes, represented by the Coulomb
failure function, due to the sequence of 217 seismic events of local magnitudes ranging from 2.0 to
4.2, which occurred in this mine between June 6, 1993 and August 27, 1999. The coseismic
displacements used in this study were simplified to be vectors across a rectangular fault in elastic half–
space. We calculated Coulomb stress changes after the occurrence of each event according to the
faulting type of the next event in the dataset. At each stage of calculations we examined the possible
triggering effect by correlating the particular event location and the stress–enhanced zones. In the
calculations we chose one of the nodal planes from the catalog at random, assuming that if a
correlation between event locations and CFF changes is significant the expectedly stronger correlation
for the set of actual nodal planes would be significant as well.

The results of our preliminary study indicate that strong mining tremors are capable of producing
changes in the state of stress of a sufficient magnitude to move a specified area into the failure regime.
We found that the location of more than 60% of the analyzed events is consistent with the stress–
enhanced areas. Most of the events were located inside regions of positive ∆CFF, larger than 0.01
MPa. The remaining events were located inside areas of calculated negative ∆CFF but for most of
them the modeled rupture zone was located partially inside areas of positive ∆CFF. Furthermore, more
than 15 per cent of these events had a doubtful focal mechanism.

We presented the model of stress evolution in the G–1/7 section of the Rudna mine. Among the 10
events that occurred in the G–1/7 field between 23.04.1996 and 12.08.1999, 5 were located in stress–
enhanced zones, 2 events were located at the border between the positive and negative stress changes.
The epicenters of 3 events were located inside shadow zones, while the largest part of the rupture areas
of 2 of them were well inside a bright zone. These results provide evidence of possible triggering due
to stress transfer by previous events occurrence in the G–1/7 field. The modeling of stress changes due

11
to seismic events along with the progress of mining works can provide additional information for
determining more probable zones for future tremor location.

There are several assumptions and uncertainties in our approach. The errors of event location in
the LGCD are about ±50 m for epicentral determination and much higher for depth. This last
uncertainty led us to assume a nucleation depth equal to 900 m. The next group of parameters that
have an impact on the results of analysis and are characterized by a wide range of possible values are
the elastic moduli of rocks in the Rudna mine. The selected values reflect the averaged geomechanical
conditions in LGCD, where the seismogenic zone is composed of limestones, dolomites, anhydrites
and sandstones. The apparent coefficient of friction that was taken equal to be 0.8 throughout our
calculations, is much higher than for Coulomb stress transfer modeling performed for natural
earthquakes because rocks in the mining area are characterized by a higher cohesion than is the case in
seismic fault zones. The proper determination of rupture plane (strike, dip and rake) from nodal planes
provided by the fault plane solution plays a key role in Coulomb stress transfer studies. In our
calculations, the choice of a particular fault plane depended on its similarity to the focal mechanisms
of previous events that had occurred in the vicinity of the studied earthquake. In some sections of the
mine the fault plane orientation parameters for seismic events are similar, indicating a common origin.
If this information was not accessible, the fault plane was chosen randomly. We considered only
events with rather reliably determined fault plane solutions. Therefore we had to assume that the state
of stress before 06.1993 was zero.

Although a number of uncertainties are involved in the calculations, the results of the preliminary
stress changes modeling in the Rudna mine in LGCD are promising, since it has been shown that most
of the events occurred in areas where stress was enhanced due to the occurrence of previous events.
For this reason, this approach can be incorporated in seismic hazard assessment studies. On the other
hand, much work and detailed investigation should be carried out to prove the role of Coulomb stress
transfer in the generation process of the mining induced seismicity.

A c k n o w l e d g e m e n t s . This work was prepared within the framework of the research project No.
PBS–Grecja/10/2007, financed by the Ministry of Education and Science of Poland during the period
2007 to 2009. We thank S.J. Gibowicz, B.M. Domański and P. Wiejacz for sending us the unpublished
catalog of focal mechanism solution and spectral analysis parameters for strong seismic events from
Rudna Mine.

References

Aki, K. and P. Richards (1980), Quantitative Seismology: Theory and Methods, W.H. Freeman, San
Francisco.

12
Andrews, D. (1986), Objective determination of source parameters and similarity of earthquakes of
different size, In: S. Das, Boatwright, J. and C. Scholz (ed.), Earthquakes source mechanics,
37, 259–267.

Brune, J. N. (1970), Tectonic stress and spectra of seismic shear waves from earthquakes, J. Geophys.
Res. 78, 4997–5009.

Brune, J. N. (1971), Correction, J. Geophys. Res. 76, 5002.

Chinnery, M.A. (1961), The deformation of the ground around a surface fault, Bull. Seism. Soc. Am.
51, 355–372.

Chinnery, M. A. (1963), The state of stress changes that accompany strike-slip faulting, Bull. Seism.
Soc. Am. 53, 921–932.
Das, S. and C.H. Scholz (1981), Theory of time–dependent rupture in the earth, J. Geophys. Res. 86B,
6039–6051.

Deng, J. and L. R. Sykes (1997) Evolution of the stress field in southern California and triggering of
moderate-size earthquakes: a 200-year perspective, J. Geophys. Res. 102, 9859–9886.
Domański, B. and S.J. Gibowicz (2008), Comparison of source parameters estimated in the frequency
and time domains for seismic events at Rudna copper mine, Poland, Acta Geophys. 56, 324 –
343. DOI: 10.2478/s11600-008-0014-1.

Domański, B., Gibowicz, S. J., Wiejacz, P. (2002), Source time function of seismic events at Rudna
copper mine, Poland, Pure Appl. Geophys. 159, 131–144.
Fitch, T. J., McCowan, D. W., Shields, M. W. (1980), Estimation of seismic moment tensor from
teleseismic body wave data with application to intraplate and mantle earthquakes, J. Geophys.
Res. 85, 3817–3828.
Gibowicz, S.L. (1990), The mechanism of seismic events induced by mining. In: Rockburst and
seismicity in mines, Rotterdam, Balkema, 3–27.

Gibowicz, S.J. (1997), An anatomy of a seismic sequence in a deep gold mine, Pure Appl. Geophys.
150, 393–414.

Gibowicz, S.J. (2006), Seismic doublets and multiplets at the Polish coal and copper mines, Acta
Geophys. 54, 142-157. DOI 10.2478/s11600-006-0014-y.
Gibowicz, S. J. and A. Kijko (1994) An Introduction to Mining Seismology, Academic Press, San
Diego.

Gibowicz, S.J. and S. Lasocki (2001), Seismicity induced by mining: Ten years later, Advances in
Geophysics, 44, 39–181.

Harris, R. A. (1998) Introduction to special session: stress triggers, stress shadows, and implications
for seismic hazard, J. Geophys. Res. 103, 24347–24358.

13
Haskell, N. A. (1953), The dispersion of surface waves in multilayered media, Bull. Seismol. Soc. Am.,
43, 17–34.

Harris, R. A. and R.W. Simpson (1996), In the shadow of 1857: the effect of the great Ft. Tejon
earthquakes in southern California. Geophys. Res. Lett. 23, 229–232.
Hudnut, K. W., Y. Bock, M. Cline, P. Fang, Y. Feng, J. Freymueller, X. Ge, W. K. Gross, D. Jackson,
M. Kim, N. E. King, J. Langbein, S. C. Larsen, M. Lisowski, Z. K. Shen, J. Svarc, Zhang, J.
(1994), Co-seismic displacements of the 1992 Landers earthquake, Bull. Seismol. Soc. Am.,
84, 625–645.

Idziak, A., Sagan, G, Zuberek, W.M. (1991), The analysis of energy distribution of seismic events
from the Upper Silesian Coal Basin, Publ. Inst. Geophys. Pol. Acad. Sci. M–15 (235), 163–
182, (in polish).

Jaeger, J.C. and N.G.W. Cook (1979), Fundamentals of rock mechanics (3rd ed.), London, Chapman
and Hall.
Kijko, A. (1997), Keynote lecture: seismic hazard assessment in mines. In: S.J. Gibowicz and S.
Lasocki (ed.), Rockbursts and Seismicity in Mines, Balkema, Rotterdam, 247–256.
King G.C.P. and M. Cocco (2001), Fault Interaction By Elastic Stress Changes: New Clues From
Earthquake Sequences. Advance in Geophysics, 44, 1–38.

Kijko, A., Drzęźla B., Mendecki, A. (1985), Why the extremal seismic events distribution have the
bimodal character? Acta Montana, 71, 225-244, (polish).
Kijko, A., Drzeźla, B., Stankiwicz, T. (1987), Bimodal character of extreme seismic events in Polish
mines, Acta Geophys. Pol. 35, 157–166.

Kijko, A., Dessokey, M. M., Głowacka, E., Kazimierczyk, M. (1982), Periodicity of strong mining
tremors in the Lubin copper mine, Acta Geophys. Pol. 30, 221–230.
King, G, C. P., Stein, R. S., Lin, J. (1994), Static stress changes and the triggering of earthquakes,
Bull. Seismol. Soc. Am., 84, 935–953.
Kłeczek, Z. (2007), Sterowanie wstrząsami górotworu LGOM (Control of rock-mass bursts in Polish
Copper Mines LGCD). In: Warsztaty Górnicze 2007 „Zagrożenia naturalne w górnictwie”,
Ślesin k. Konina, 4–6 czerwca 2007 r. Bezp. Pr. Ochr. Śr. Gór. 2007 nr 6, 25–27.

Lasocki, S. (1992a), Non–Poissonian structure of mining induced seismicity, Acta Montana, 84, 51–
58.

Lasocki, S. (1992b), Weibull distribution for time intervals between mining tremors, Publs. Inst.
Geophys. Pol. Acad. Sci. M–16 (245), 241–260.

Lasocki, S. (2001), Quantitative Evidences of Complexity of Magnitude Distribution in Mining–

induced Seismicity: Implications for Hazard Evaluation. In: G. van Aswegen, R.J. Durrheim,
W.D. Ortlepp (ed.), 5th Int. Symp. Rockbursts and Seismicity in Mines "Dynamic rock mass

14
 response to mining", Magalisberg, 17–20 September 2001 SAIMM S27, Johannesburg, South
Africa, 543–550.

Lasocki, S. (2005) Probabilistic Analysis of Seismic Hazard Posed by Mining Induced Events, In: Y.
Potvin, M. Hudyma (ed.), The Sixth International Symposium on Rockbursts and Seismicity in
Mines "Controlling Seismic Risk" Proceedings, ACG, Perth, 151–156.

Leśniak, A., Pszczoła, G. (2008), Combined mine tremors source location and error evaluation in the
Lubin Copper Mine (Poland), Tectonophysics 456, 16-27.

Lin, J. and R.S. Stein (2004), Stress triggering in thrust and subduction earthquakes, and stress
interaction between the southern San Andreas and nearby thrust and strike–slip faults, J.
Geophys. Res. 109, B02303, doi:10.1029/2 003JB002607.
Madariaga, R. (1976), Dynamics of an expanding circular fault, Bull. Seismol. Soc. Am. 66, 639–666.

Marcak, H. (1985), Geophysical models of development of destruction process in the rock mass prior
to rockburst. Publ. Inst. Geophys. Pol. Sc. Acad. M–6 (176), 149–174, (in polish).

Marsan, D., Bean, Ch.J., Steacy, S., McCloskey, J. (1999), Spatio-temporal analysis of stress diffusion
in mining–induced seismicity system, Geophys. Res. Lett. 26, 3697–3700.
McGarr, A. and D. W. Simpson (1997), Keynote lecture: a broad look at induced seismicity, In: S. J.
Gibowicz and S. Lasocki (ed.), Rockbursts and Seismicity in Mines, Balkema, Rotterdam,
385–396.
Okada,Y. (1985), Surface deformation due to shear and tensile faults in a half-space. Bull. Seism. Soc.
Am. 75, 1135–1154.
Okada,Y. (1992), Internal deformation due to shear and tensile faults in a half–space. Bull. Seism. Soc.
Am. 82, 1018–1040.
Orlecka–Sikora, B. and S. Lasocki (2002), Clustered structure of seismicity from the Legnica–Głogów
copper district, Publs. Inst. Geophys. Pol. Acad. Sc., M–24 (340), 105–119, (in polish).

Papadimitriou, E. E., and L. R. Sykes (2001), Evolution of the stress field in the northern Aegean Sea
(Greece), Geophys. J. Int. 146, 747–759.

Parsons, T., Stein, R. S., Simpson, R. W., Reasenberg, P. A. (1999), Stress sensitivity of fault
seismicity: a comparison between limited–offset oblique and major strike–slip faults, J.
Geophys. Res. 104, 20183–20202.
Piestrzyński, A. (1996), Monografia KGHM Polska MIedz SA. Centrum Badawczo-Projektowe Miedzi
"Cuprum". Wroclaw.

Reasenberg, P. A. and R. W. Simpson (1992), Response of regional seismicity to the static stress
change produced by the Loma Prieta earthquake, Science, 255, 1687-1690.

15
Scholz, C.H. (1990), The mechanics of earthquakes and faulting, Cambridge, U.K., Cambridge
University Press, 439.

Snoke, J. A. (1987), Stable determination of (Brune) stress drops, Bull. Seismol. Soc. Am., 77, 530-
538.
Steacy S., Gomberg J., Cocco, M. (2005), Introduction to special section: stress transfer, earthquake
triggering and time–dependent seismic hazard, J. Geophys. Res. 110,
doi:10.1029/2005JB003692.

Stein, R. S. and M. Lisowski (1983), The 1979 Homestead Valley earthquake sequence, California:
Control of aftershocks and postseismic deformation, J. Geophys. Res. 88, 6477–6490.

Stein, R.S., King, G.L.P., Lin, J. (1992), Change in failure stress on the southern San Andreas fault
system caused by the 1992 magnitude=7.4 Landers earthquake, Science, 258, 1328–1332.

Steketee, J.A. (1958a), On Volterra’s dislocations in a semi-infinite elastic medium. Can. J. Phys. 36,
193–205.

Steketee, J.A. (1958b), Some geophysical applications of the elasticity theory of dislocations. Can. J.
Phys. 36, 1168–1198.
Toda, S., Stein, R. S., Richards–Dinger, K., Bozkurt, S. (2005), Forecasting the evolution of seismicity
in southern California: Animations built on earthquake stress transfer, J. Geophys. Res.
B05S16, doi:10.1029/2004JB003415.
Trifu, C.-I., Urbancic, T.I., Young, R.P. (1993), Non-similar frequency-magnitude distribution for
M<1 seismicity, Geophys. Res. Lett. 20, 427–430.
Węglarczyk, S. and S. Lasocki (2008), Studies of long and short memory in mining–induced seismic
processes, Acta Geophysica (submitted).
Wiejacz, P. (1991), Investigation of focal mechanisms of mine tremors by the moment tensor
inversion, Ph. D. Thesis, Inst. Geophys. Pol. Ac. Sci., Warsaw, Poland.

16
Table 1. Annual number of events per magnitude / energy class in the Legnica–Głogów Copper
District in the years 06.1993–08.1999. ML denotes the local magnitude, while energy (E) is expressed
in joules.

Table 2. Source parameters of ML ≥ 3.0 mining induced seismic events that occurred between 08.1994
– 08.1999 in the Rudna mine area in the LGCD. The first three columns provide information on the
occurrence date and time, X, Y, Z denote location in the local Cartesian coordinate system, ML
denotes the local magnitude, E energy, M0 seismic moment, r is the circular fault radius. The last six
columns provide information of the fault plane solution (strike, dip and rake for both nodal planes).

Table 3. Rupture models for mining induced seismic events with ML>3.0 that occurred between
08.1994 – 08.1999 in the Rudna mine area in the Legnica-Głogów Copper District, included in the
Coulomb stress transfer calculations. L and W denote the length and width of the rectangular fault,
respectively, expressed in meters, u is the mean displacement, ss, along strike and ds, along the dip
direction, expressed in centimeters.

17
Table 1
Number of seismic events

ML E [J]
Year 105 106 107 108 109

2 2.5 3 3.5 4

1993 0 1 0 0 0
1994 11 3 9 1 0
1995 34 11 4 2 0
1996 8 4 0 0 0
1997 9 9 10 0 0
1998 43 21 9 1 0
1999 17 6 2 1 1
Total 122 55 34 5 1

18
Table 2
Origin Hypocentre Source size Mechanism plane A [deg] Mechanism plane B [deg]
Year Data Time X [m] Y [m] Z [m] M E [J] M0[N·m] r[m] Strike Dip Rake Strike Dip Rake
1994 31-Aug 20:33 34850 5633 -890 3.4 7.3E+07 9.4E+13 267 213 55 78 54 36 107
1994 1-Sep 21:27 34386 5532 -870 3.0 1.2E+07 3.7E+13 228 240 66 116 9 34 45
1994 2-Sep 21:19 33033 8497 -910 3.1 1.5E+07 3.4E+13 231 139 54 99 304 37 78
1994 14-Oct 18:17 32904 4480 -750 3.0 1.1E+07 1.0E+13 171 188 89 108 282 18 4
1994 18-Oct 11:12 33432 8817 -940 3.3 4.2E+07 5.2E+13 185 196 77 77 61 19 134
1994 22-Nov 13:29 28512 7028 -740 3.4 7.3E+07 4.6E+13 277 114 85 -86 255 6 -129
1994 5-Dec 23:45 28354 7053 -740 3.1 1.4E+07 3.3E+13 265 138 88 -91 351 2 -57
1994 10-Dec 14:35 32687 4778 -760 3.3 3.9E+07 1.0E+14 306 33 82 73 278 18 154
1994 21-Dec 13:26 27203 9390 -690 3.6 2.0E+08 2.9E+14 502 348 62 -99 187 28 -73
1994 22-Dec 9:23 27419 9023 -680 3.0 1.1E+07 1.4E+13 178 12 61 -80 171 31 -108
1995 31-Mar 2:57 27680 7622 -550 3.1 9.5E+07 4.4E+13 202 140 87 -87 272 8 -138
1995 26-May 4:57 33500 8600 -1050 3.8 2.9E+08 5.6E+14 416 140 41 -92 323 49 -89
1995 24-Aug 16:46 34986 5601 -800 3.3 3.6E+07 9.1E+13 280 318 74 -119 202 33 -31
1995 14-Sep 19:39 32699 4699 -470 3.2 3.1E+07 3.3E+13 167 17 87 140 109 50 3
1995 1-Oct 13:06 32818 4298 -900 3.0 8.8E+06 1.7E+13 186 63 57 86 251 34 96
1995 10-Oct 18:17 33352 8678 -920 3.0 9.6E+06 1.9E+13 230 309 89 -95 209 5 -11
1995 23-Nov 6:03 28452 6952 -740 3.6 1.4E+08 7.6E+13 280 123 54 -82 290 37 -100
1995 2-Dec 10:13 34493 5812 -650 3.0 1.1E+07 2.5E+13 301 179 46 -88 356 44 -92
1997 11-Mar 20:13 31569 5947 -700 3.1 2.0E+07 1.6E+14 467 130 47 97 299 43 82
1997 17-Apr 12:35 31656 10539 -775 3.4 6.0E+07 1.6E+14 445 356 69 -48 107 46 -151
1997 26-Apr 10:52 31640 9885 -1025 3.0 1.2E+07 5.5E+13 305 0 88 -64 94 26 -175
1997 26-Apr 14:23 31233 6155 -975 3.2 2.9E+07 7.3E+13 348 133 72 -80 283 20 -119
1997 29-Apr 17:11 30826 6060 -900 3.2 2.8E+07 3.7E+13 275 233 47 89 54 43 91
1997 13-Jun 15:22 31624 6007 -1050 3.1 2.1E+07 9.6E+13 374 101 64 36 353 58 149
1997 28-Jun 21:50 31229 6201 -725 3.4 6.9E+07 1.1E+14 387 360 52 76 202 41 107
1997 3-Jul 22:00 31618 10542 -950 3.1 1.5E+07 1.1E+14 428 335 79 -83 124 13 -120
1997 23-Jul 23:49 31674 7112 -800 3.1 1.4E+07 4.8E+13 245 103 88 174 193 84 2
1997 30-Aug 16:43 30432 6067 -825 3.0 1.1E+07 4.7E+13 252 271 64 -121 145 40 -44
1998 31-Jul 16:22 31223 5928 -750 3.0 8.8E+06 4.1E+13 284 97 62 84 288 29 100
1998 31-Jul 18:51 30449 6147 -750 3.3 4.5E+07 6.1E+13 279 78 87 75 336 15 167
1998 28-Aug 14:59 28792 6805 -720 3.0 1.1E+07 5.6E+12 202 223 84 111 327 22 15
1998 31-Aug 5:40 31252 9910 -900 3.2 3.0E+07 9.1E+13 378 210 77 98 357 15 58
1998 10-Sep 5:03 32473 8882 -930 3.4 5.5E+07 6.0E+13 325 99 67 94 270 23 81
1998 22-Sep 21:56 27591 8608 -750 3.3 3.5E+07 2.6E+13 331 44 47 85 231 43 95
1998 3-Oct 15:49 32977 8808 -860 3.2 2.7E+07 2.9E+13 215 110 85 -92 312 5 -68
1998 20-Oct 2:07 27472 8510 -750 3.1 1.9E+07 3.9E+13 346 198 68 93 10 23 83
1998 5-Nov 10:23 32686 4663 -600 3.0 8.6E+06 2.3E+13 265 120 76 83 325 16 114
1998 6-Nov 13:13 30112 6167 -770 3.5 1.1E+08 6.9E+13 337 91 66 100 249 26 70
1998 28-Nov 21:18 32375 8892 -920 3.3 5.0E+07 5.5E+13 314 330 68 44 220 50 150
1998 10-Dec 0:20 28209 6173 -690 3.2 3.0E+07 4.5E+13 318 88 57 56 319 46 131
1999 18-Jul 5:45 31165 9966 -810 4.2 2.0E+09 5.1E+14 582 192 88 89 39 3 117
1999 27-Jul 16:30 29088 7448 -900 3.2 2.8E+07 5.1E+13 279 170 66 99 328 26 70
1999 12-Aug 1:37 31000 6109 -750 3.1 1.9E+07 3.7E+13 241 359 88 93 125 4 36
1999 23-Aug 3:20 27428 8375 -700 3.6 1.5E+08 7.5E+13 360 30 77 -51 135 40 -160

19
Table 3
Origin Source size Slip [cm] Origin Source size Slip [cm]
Year Data Time r[m] L[m] W[m] u ss sd Year Data Time r[m] L[m] W[m] u ss sd
1994 31-Aug 20:33 267 580 386 1.91 -0.40 1.87 1997 29-Apr 17:11 275 598 398 0.70 -0.02 0.70
1994 1-Sep 21:27 228 495 330 1.03 0.45 0.93 1997 13-Jun 15:22 374 812 541 1.00 -0.81 0.59
1994 2-Sep 21:19 231 501 334 0.92 0.14 0.91 1997 28-Jun 21:50 387 841 560 1.06 -0.26 1.03
1994 14-Oct 18:17 171 371 247 0.50 0.16 0.47 1997 3-Jul 22:00 428 928 619 0.90 -0.11 -0.90
1994 18-Oct 11:12 185 402 268 2.20 -0.51 2.14 1997 23-Jul 23:49 245 532 354 1.16 1.15 0.12
1994 22-Nov 13:29 277 601 401 0.87 -0.06 -0.87 1997 30-Aug 16:43 252 546 364 1.07 0.55 -0.91
1994 5-Dec 23:45 265 575 384 0.68 0.01 -0.68 1998 31-Jul 16:22 284 616 410 0.73 -0.07 0.73
1994 10-Dec 14:35 306 664 443 1.55 1.39 0.67 1998 31-Jul 18:51 279 605 403 1.14 -0.29 1.10
1994 21-Dec 13:26 502 1090 727 1.67 0.27 -1.64 1998 28-Aug 14:59 202 437 292 0.20 0.07 0.19
1994 22-Dec 9:23 178 386 258 0.64 0.19 -0.61 1998 31-Aug 5:40 378 821 547 0.92 0.13 0.91
1995 31-Mar 2:57 202 439 292 1.56 -0.10 -1.55 1998 10-Sep 5:03 325 706 470 0.82 0.05 0.82
1995 26-May 4:57 416 903 602 4.69 0.14 -4.69 1998 22-Sep 21:56 331 719 479 0.34 -0.03 0.34
1995 24-Aug 16:46 280 608 405 1.68 0.82 -1.47 1998 3-Oct 15:49 215 467 311 0.90 0.03 -0.90
1995 14-Sep 19:39 167 363 242 1.71 1.32 1.09 1998 20-Oct 2:07 346 750 500 0.48 0.03 0.47
1995 1-Oct 13:06 186 404 269 0.71 -0.05 0.71 1998 5-Nov 10:23 265 575 383 0.48 -0.06 0.47
1995 10-Oct 18:17 230 499 333 0.52 0.04 -0.52 1998 6-Nov 13:13 337 732 488 0.87 0.15 0.86
1995 23-Nov 6:03 280 608 405 1.40 -0.19 -1.39 1998 28-Nov 21:18 314 681 454 0.81 -0.58 0.57
1995 2-Dec 10:13 301 653 436 0.40 -0.01 -0.40 1998 10-Dec 0:20 318 690 460 0.65 -0.36 0.54
1997 11-Mar 20:13 467 1013 675 1.07 0.14 1.06 1999 18-Jul 5:45 582 1263 842 2.18 -0.04 2.18
1997 17-Apr 12:35 445 966 644 1.19 -0.80 -0.88 1999 27-Jul 16:30 279 606 404 0.95 0.16 0.94
1997 26-Apr 10:52 305 662 441 0.85 -0.37 -0.76 1999 12-Aug 1:37 241 523 349 0.92 0.05 0.92
1997 26-Apr 14:23 348 756 504 0.88 -0.16 -0.86 1999 23-Aug 3:20 360 781 521 0.84 -0.52 -0.65

20
Figure 1. Spatial distribution of the analyzed events with 2.0<ML<2.9 (gray small circles) and ML>3.0
(black circles) along with the seismic stations (triangles). The dashed rectangle outlines the area of
study shown in Figure 4.

Figure 2. Histograms of a. the isotropic component ISO, b. the linear vector dipole component CLVD
and c. the double couple component DC in the total moment tensor solutions of the analyzed seismic
events of the Rudna Coal Mine.

Figure 3. Histogram of the frequency of the occurrence of seismic events of a. magnitude ML≥ 2.0 and
b. seismic moment M0≥1013 [N·m], that occurred during 06.1993 – 08.1999 in the Rudna Mine area in
the Legnica–Głogów Copper District as a function of ∆CFF.

Figure 4. Accumulated stress changes in the G-1/7 field of the Rudna Mine in LGCD, the west part of
Poland from 06.06.1993 to 13.08.1999. Coulomb stress is calculated for faults at a depth of 0.9 km.
The stress pattern is calculated for the faulting type of the next mine event in the data sample. Changes
are denoted by a color scale shown in the right part of each figure (in bars). The Coulomb failure
function (CFF) is taken to be equal to the state of stress after the first 76 tremors that occurred between
06.06.1993 and 23.04.1996. The green line denotes the fault trace projected up-dip at the Earth’s
surface, the black one is the intersection of target depth with fault plane, the grey rectangle is the
frame of the fault projected to the surface. a. coseismic Coulomb stress changes before 23.04.1996, the
stress field is inverting according to the fault plane solution of the 23.04.1996 event; b. stress
evolution until just before the 07.11.1996 event; c. the state of stress just before the occurrence of the
08.11.1996 tremor; d. cumulative stress changes until just before the 19.12.1996 event; e. the state of
CFF before the event on 09.09.1998; f. cumulative stress changes before the event occurred on
13.10.1998; g. the stress pattern just before the 04.07.1999 tremor; h. the state of stress before the
05.07.1999 event; i. stress evolution just up to the event occurred on the 12.08.1999; j. the stress filed
before the occurrence of the 13.08.1999 event.

21
Figure 1.

22
a. b. c.
80 100 70
90
number of observations

number of observations

number of observations
70 60
80
60
70 50
50 60
40
40 50
30
30 40
30 20
20
20
10 10
10
0 0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60 70 80 90 0 10 20 30 40 50 60 70 80 90 100
%ISO %CLVD %DC

Figure 2.

23
a. b.
90 26
24
80
22
number of observations

number of observations
70 20
60 18
16
50 14
40 12
39% 61% 10 37% 63%
30
8
20 6
4
10
2
0 0
-1.8 -1.4 -1.0 -0.6 -0.2 0.2 0.6 1.0 1.4 1.8 -1.8 -1.4 -1.0 -0.6 -0.2 0.2 0.6 1.0 1.4 1.8
CFF [MPa] CFF [MPa]

Figure 3.

24
a. b.

c. d.

e. f.

25
g. h.

i. j.

Figure 4.

26