The_impact_of_earthquakes_on_mining_oper.pdf

The Impact of Earthquakes on Mining Operations W.A. Lenhardt It is well known, that mining activity can result in seismic events. On the other hand, tectonic – or natural – earthquakes are able to cause distress to mine openings and open cast mines which may lead to instable situations. It will be shown, that natural earthquakes have a limited impact on underground mining operations but should not be neglected at surface operations and tunnel openings. Der Einfluss von Erdbeben auf den Bergbau.  Bergbau kann zu seismischen Ereignissen führen. Andererseits beeinträchtigen natürliche Erdbeben auch bergbauliche Aktivitäten. Diese tektonischen Erdbeben wirken sich unterschiedlich auf Tag- unter Untertagebauten aus. Solche äußeren Einflüsse haben aber meist eine sehr begrenzte Auswirkung auf Untertagebauten, während Auswirkungen in Tagbauten und an Tunnelportalen nicht zu unterschätzen sind.

1. Introduction External influences on mining operations are considered a serious threat because they occur unexpected, seldom and at erratic long time intervals. Despite drastic and sudden fluctuations in global market values of mineral resources, natural hazards such as torrential rains, floods, fires or earthquakes can cause interruptions of the mining operations. This article deals with the latter aspect – earthquakes. These effects from tectonic movements are rare in Austria, but happen. Therefore, they are usually not considered during planning and the following operation of mine workings. The following paragraphs deal with this possible impact of earthquakes on surface and underground mining operations within the frame of the seismo-tectonic potential in the Alps.

2. Surface To estimate the impact an earthquake can have on open cast mines or cut and cover developments, this task is treated similar to a landslide problem. Common to both situations are the steep slopes which result from the mining layout. Besides, an open cast mine consists of several levels or terraces which are separated by few tens of meters in contrast to cut-and-cover operations. The steeper a slope, the higher the likelihood of rockfall and slides, especially when not properly drained or when exposed to heavy rain. Epicentral distance and the magnitude of an event govern mainly its severity and determine how widespread these effects actually can be observed. Earthquakes in the Alps have the potential of reaching seismic Univ.-Doz. Dr. W. A. Lenhardt, Department of Geophysics, Zentralanstalt für Meteorologie und Geodynamik (ZAMG), Hohe Warte 38, 1190 Wien / Österreich.

BHM, 154. Jg. (2009), Heft 6

Fig. 1: Distance of landslides versus magnitude, based on an Arias-intensity of 0.11 m/s (Harp and Wilson2)

magnitudes above 6 in e.g. Friuli 19761 causing possibly landslides up to distances of 50 km from the epicentre (see Fig. 12) thus effecting Austrian territory. Whether or not a slope can resist the transient ground motions depends mainly on the current geomechanical properties of the ground, the slope angle and on its water saturation3. To judge the potential, where slides triggered by earthquakes, are likely to affect open cast mines in Austria, empirical estimates of ground shakings and few theoretical concepts are employed. Since the bench face angle in open cast mines is always higher than the overall slope angle, slope instabilities between ramps of an open cast mine are more likely than in a natural environment as dictated by nature. In 1970 Arturo Arias proposed a way to determine objectively the intensity of ground shaking by measuring

© Springer-Verlag

Lenhardt

249

empirical estimates of ground shakings and few theoretical concepts are employed. Since the bench face angle in open cast mines is always higher than the overall slope angle, slope instabilities between ramps of an open cast mine are more likely than in a natural environment as dictated by nature. In 1970 Arturo Arias proposed a way to determine objectively the intensity of ground shaking by the acceleration of transient seismic waves. The integral Ilocal = 1.5 log10 (IA) + 9.15 – 0.45 log10 (z) – 1.3 α (R-z) (5) measuring the acceleration of transient seismic waves. The integral of the square of the ground of the square of the ground acceleration4 over time 4 acceleration over time Equation (5) demonstrates, that deeper earthquakes (z = 10 km) lead only to slightly smaller macroseismic  (1) intensities (~0.5°, which is also the inherent inaccuracy of macroseismic intensities) when compared with shal(1) low ones (e.g. z = 1 km), while their Arias-intensity is with with kept constant. This difference depends also very slightly g … acceleration due to gravity (approx. 9.81 m/s² at of 47.5 g … acceleration due to gravity (approx. 9.81 m/s² at a geographical latitude on°)the hypocentral distance which can be ignored in this Td ... durationaofgeographical signal above threshold, forof practical reasons, theoretically the integral should be infinite latitude 47.5 °) context because the focal depth “z” is usually < 10 km. became known as “Arias intensity” represents square root ofThe the energy Td ... duration of signal aboveand threshold, forthe practical reasonper formass this thus obvious independence of the focal having units of “m/s”.theoretically It does not represent a ground velocity, its units might indicate that. reasons; the integral should be although depth can be traced back to the data set used. Only infinite landslides (z = 0) and their Arias-intensities were studied, and hence, the term of the focal depth in equation became known as “Arias intensity” and represents the (5) is a residual of equation (3) and (4), which deal with square root of the energy per mass thus having units of real earthquakes of various focal depths. In general one “m/s”. It does not represent a ground velocity, although can state, that the Arias-intensity serves indeed as a its unit might indicate that. good indicator for the otherwise subjectively determined This “intensity” must not be confused with the macromacroseismic intensity. Hence, an Arias-shaking intenseismic intensity scale5, which describes the subjective sity value of 0.11 m/s results in an integer value of the intensity of shaking as reported by people and associth macroseismic intensity of 7. This intensity corresponds ated building damage on surface (see also Table 2, 4 roughly to ground accelerations in excess of 1 m/s². column – EMS-98). The 10 % probability of exceedance in 50 years of such Table 1: Some Arias-intensities and their meaning2 accelerations applies to zone 4 of the Austrian Building Code for Earthquake Resistant Design based on IA-Minimum Category Description Lenhardt9. Because “P” in equation (2) was chosen “0”, value equation (5) represents an average conversion from the macroseismic intensity to the Arias intensity – or in other 0.11 m/s I Falls, disrupted slides, avalanches words, the macroseismic intensity of 7 or the associated 0.32 m/s II Slumps, block slides, earth flows ground acceleration of 1 m/s² can be exceeded in 50 0.54 m/s III Lateral spreads and flows % of the cases. Accordingly, areas with ground accelerations in excess of 1 m/s² in Fig. 2 are exposed to an Since Arias-intensity values have been found to be earthquake related landslide hazard of 5 % in 50 years, typical for certain effects in nature (Table 1), they lend given that local geological and mining layout conditions themselves to be converted into local macroseismic permit mass movements in the first place. For macintensity degrees, to delineate regions in Austria where roseismic intensities smaller than degree 7, empirical this potential is existing. This can be done by substitutrelations which may differ slightly from equation (2) and ing the magnitude in Harp and Wilson’s formula derived consequently from equation (5) should be used (e.g. from larger earthquakes Travasarou et al.10). log IA = Mw – 2 log10 (R) – 4.1 – 0.5 P (2) Regions in Austria, which can be exposed to relative with high ground accelerations – by Austrian standards – are Mw … moment magnitude Namlos and Innsbruck in Tyrol, the Mürz Valley in Styria P ... deviation from mean value in units of standard and the southern part of the Vienna Basin. The region of deviations Katschberg is also thought of being an additional area R … hypocentral distance (km) of high shaking potential, for the epicentre of the 1201 earthquake has been relocated from Murau in Styria to by the magnitude derived from the scaling law of Shebathe region of the Katschberg11 in Carinthia/Salzburg. In 6 lin , which links the macroseismic epicentral intensity “I0” addition, the southern part of Carinthia also qualifies to the seismic magnitude “M” as an area where slope instabilities can be potentially M = 2/3 I0 + 2.3 log10 (z) – 2 (3) and substitute the term of the epicentral intensity by the macroseismic local intensity based on the Sponheuer formula7 I0 = Ilocal + 3 log10 (R/z) + 1.3 α (R-z)

(4)

in which “z” represents the focal depth in km. The attenuation coefficient “α” is usually ranging from 0.001 to 0.004 with a typical value at the lower bound of 0.001/ km. In the original work, Shebalin6 derived his magnitude from surface waves. However, experience has shown, that the relation also holds for events of small magnitudes in the Eastern Alps8. Therefore, the quoted equation (3) is used in Austria for more than 15 years to estimate intensities from measured local magnitudes. In the following, it is set Mw = M and P = 0, and it results in

250

Lenhardt

Fig. 2: Effective ground accelerations in Austria with 10% probability of exceedance in 50 years (Lenhardt9)

© Springer-Verlag

BHM, 154. Jg. (2009), Heft 6

triggered by earthquakes from the northern part of Italy – Friuli1. It has been shown empirically, that seismic events of M6.5 have the potential of triggering landslides up to 100  km from the epicentre12,2, which is equivalent to a macroseismic intensity of „VI“, if the focal depth lies within the upper crust – meaning less than 10 km. Grünthal5 even states an intensity of „V“ as the lower limit for possibly triggering landslides. Besides, high peak accelerations of short duration and high frequency – as usual in Austria – have not been found to trigger large landslide masses with volumes exceeding 100 000 m³. Slope Instabilities without Earthquake Triggering Whether earthquakes are capable of destabilising slopes depends on many factors, despite the obvious distance dependent attenuation of earthquake ground motions and the magnitude of the generating earthquake. Voight and Pariseau13 mention two earthquakes in 1969 in Hope in Canada of magnitude M3.0, which were accompanied by landslides. Much earlier, the same region was affected by a much stronger earthquake of magnitude M7.5 in 1872. There was no evidence of landslides in the latter case, however, possibly due to lack of information. This phenomenon seems to be characteristic for slope instabilities, because most tend to occur without an earthquake at all. A slope needs to be at the brink of failure already, if an earthquake should be able to destabilise it – or in other words – rock/ground conditions must have dramatically deteriorated or fractured locally. Records from local ground motions near landslides are usually sparse and hinge on the deployment of seismic instrumentation (e.g. Harp and Wilson2, Glawe and Moser14). Besides, the analysis of recorded seismic signatures of slope instabilities turns out to be a difficult one for seismologists. If they are fortunate and one of their stations was near enough to the rock mass movement and recorded the event, it will give them the time and the record of ground motions at the station and possible at others, too. This helps to ascertain the type of tremor: tectonic earthquake, explosion, blast, landslide, collapse in karst regions, ultrasonic boom from an airplane, meteorite – just to name a few. By the way, about 50 % of all tremors analysed by the Austrian Seismological Survey at ZAMG – which are about 600 per year in Austria only – are related to blasts (opencast or tunnelling), despite of up to 4000 earthquakes abroad. However, seismologists know that a „magnitude“ of a slope failure would not adhere to the definition of a tectonic earthquake magnitude, which implies a certain stress drop, signal duration, frequency content and a predefined mechanism, and therefore refrain from stating a magnitude. Nevertheless, seismic records are very valuable when documenting rock mass movements of any kind, provided enough seismic stations covering the surrounding are available to gain a good picture of the seismic pattern in terms

of amplitude, frequency content and duration of the process. This can help to identify, how many rockslides occurred with which power, in which sequence and whether an earthquake triggered them.

3. Underground Tectonic earthquakes can affect underground operations too, although to a much lesser extent as will be shown. Despite of poor support, the main reason for such damage and its human and financial consequences is mainly the proximity and magnitude of sudden tectonic movements. Experience has shown, that underground structures suffer damage only under specific circumstances. According to Dowding and Rozen15 damage to underground structures become relevant once certain levels of accelerations are exceeded (Table 2). Most tectonic earthquakes pass unnoticed underground, however. Little damage is experienced in few cases. Examples in Austria are the earthquakes near Judenburg in Styria in 1916, during which mining carriages were shaken in a nearby mine coal underground in Fohnsdorf and on the occasion of the earthquake at Obdach in 1936 (Hammerl and Lenhardt17). During the latter earthquake, only sudden dust was observed in the underground working, whereas on surface already wide spread damage to buildings was observed. Therefore, the documentation of Dowding and Rozen15 becomes even more valuable. What it points out is that the macroseismic intensity experienced on surface is much higher when compared with those underground. The reason for this discrepancy is mainly the effect of the free surface, which theoretically alone is capable to double ground movements on surface on one hand, and on the other, the lack of resonance of structures underground when compared with those on surface. Underground openings are confined by the rock mass and support, whereas surface structures are free to respond horizontally. The influence on underground structures depends on geomechanical properties of the surrounding rock mass, the overburden – which alters the effect of the free surface from which measurements are usually gathered – and the azimuth of the earthquake, its stress drop, rupture direction and seismic magnitude. In the case of large earthquakes the rupture velocity and major aftershocks within the time span of dynamic loading should be considered as well, as it was the case for the Sumatra earthquake on December 26 in 2004. Concentrating on Austria, however, possible local earthquakes may be considered. Macroseismic intensities of degree 8 are seldom in Austria, and it is hard to estimate their recurrence period, but about ten appear to have happened during the past 1000 years. And those occurred in the Vienna Basin, the Mürz and Inn Valley. No larger events – that is intensity

Table 2. Damage to underground structures after Dowding and Rozen15 in Berger16 Estimated peak ground Underground Damage on surface to buildings acceleration on surface (m/s²) damage

Macroseismic intensity (EMS-98)

< 1.77 1.77–2.45 2.45–4.9 > 4.9

7 8 8–9 10

none few minor little larger

BHM, 154. Jg. (2009), Heft 6

frequent heavy damage heavy damage to many buildings massive major

© Springer-Verlag

Lenhardt

251

degree 9 or above – are known today in these areas. In addition, effects from earthquakes in Friuli 1976 have caused havoc in Carinthia but did not exceed intensity 7 anywhere on Austrian territory during the past. Estimates of possible maximum earthquakes on Austrian territory have been carried out and lead to the conclusion, that earthquakes of magnitude 6 and above cannot be substantiated18. Magnitudes based on fault lengths alone according to Wells and Coppersmith19 might appear applicable, one should be careful when applying it to the local low stress-strain regime, however. Deformation rates are pretty small in the Alps when compared with those of the Southern Alps, where most of the strain is already consumed. Hence, the strain rate in the Alps on Austrian territory is much less than those in the southern counterpart in Friuli, as the high seismicity during the past centuries in this region demonstrates. 3.1. Dynamically Imposed Strains The impact of tectonic earthquakes on underground structures can be outlined as follows: Firstly, the relevant dynamic strains and the parameters are observed, which drive them. Secondly, these parameters are estimated. To estimate the impact of seismic waves impinging at an angle of 45° – a guesstimate, when assuming earthquakes sources may originate anywhere – on underground structures such as tunnels20, it may be written ε

lmax

= [vmax/(2 vs)] + [0.7 r amax / (vs²)]

(6a)

ε

lmin

= [vmax/(2 vs)] - [0.7 r amax / (vs²)]

(6b)

with vmax... underground peak particle velocity (m/s) amax... underground peak acceleration (m/s²) vs... shear wave velocity (m/s) r... radius of tunnel (m) The second term in equation (6a) and (6b) refers to the induced curvature of the tunnel. As will be shown in the following paragraph, this term is rather small due to the relatively low seismically imposed acceleration at moderate seismic magnitudes and realistic source distances. The maximum circumferential deformation ‘εθR’ due to racking of the concrete lining in the tunnel as imposed by a shear wave for compressive and tensile strains at the hanging and sidewall can be determined by εθR = (2 vmax / vs) [(d / r) + (3 Em r / (16 El d))]

3.2 Seismic Loads To determine these seismic loads, the relevant peak particle velocities “vmax” and accelerations “amax” acting underground are needed to calculate the relevant strains. Since mainly empirical correlations from observations on surface are reported in literature, one is forced to introduce corrections for induced underground movements, based on scarce experience. Tamura et al.21 have already shown, that ground movements observed underground at a depth of 67 m were approximately half of those which were measured on surface. These data were based on earthquakes with focal depths between 10 and 80 km, and the recorded earthquakes occurred 50 – 250 km from the underground structure and earthquake magnitudes ranged from 3.5 to 7.5. After the earthquake in 1976 in Friuli in Italy, Berardi et al.22 confirmed this observation. The following equations permit the calculation of the peak ground velocity (“PGV”) in m/s log10(PGV) = M – 1.66 log10(R) – 5.3

(8)

and of the peak ground acceleration (“PGA”, after Smit23) in m/s² log10(PGA) = 0.868 M - log10(R) – 0.001509 R – 3.77 (9) at the surface above the underground structure, whereby “M” equals the seismic magnitude and the hypocentral distance “R” is given in km. These equations take into account the petering out of ground motions with increasing distance due to geometrical spreading, scatter and attenuation as observed in the Alps. Taking half of these values approximate relevant velocities “vmax” (= PGV/2) and accelerations “amax” (= PGA/2) due to impinging seismic waves underground and allows to calculate the needed longitudinal and circumferential strains brought about during earthquake loads. These loads are minute when compared with dynamic loads reported from rockbursts (Wagner24). Already from Table 2 it became apparent that only relatively high ground movements in excess of 1.77 m/ s² on surface may lead to damages underground. Taking half of it, which accounts mainly for the effect of the free surface, 0.9 m/s² are left. In Fig. 3 corresponding surface accelerations for hypocentral distances due to several magnitudes are given. At a focal depth and thus hypocentral distance of 6 km below the epicentre and shallower, only an earthquake of magnitude M>5.5 is

(7)

whereby d... thickness of tunnel lining (m) Em... modulus of elasticity of surrounding rock (N/m²) El... modulus of elasticity of tunnel lining (N/m²) These two formulas show that underground strains are a result of the impinging seismic shear waves – in particular the maximum of the 1st (= vmax) and 2nd (= amax) derivative of their time series of ground displacements as observed underground in the ultimate vicinity of a tremor. At larger distances from the source the rock mass behaves in a much more elastic way and high differences in circumferential strain (εθR) due to the long wave length in respect to the size of the underground structure should not occur. Therefore, this is an extremely conservative approach.

252

Lenhardt

Fig: 3: Peak ground accelerations as function of the hypo­ central distance according to Smit23 on surface

© Springer-Verlag

BHM, 154. Jg. (2009), Heft 6

Fig: 4: Peak velocities depending on hypocentral distance according to the Seismological Service of Austria (ZAMG) on surface

capable to cause damage underground. In Fig. 4 corresponding velocities on surface are given for the same scenario. Again, it can be noticed that ground velocities become irrelevant above a similar hypocentral distance. This calculation demonstrates, that relatively high seismically induced stresses tend to occur only within few wave lengths from the underground cavity – a distance which is mainly consumed already by the focal depth. In addition, only within a distance corresponding to the focal depth circumferential strain differences between the sidewall and the hanging or footwall are relevant. Axial strains in the tunnel remain much smaller, almost by two orders of strain units, thus posing no threat to underground structures (Fig. 5). Besides, field stresses must be accounted for. They are capable to counteract dynamically induced stresses – considering the tensile nature of dynamical ground movement – to such an extent, that no harm is taking place in underground excavations at all, especially when properly supported. This fact explains the rare observance of underground damage due to an earthquake – in particular in deeper mine workings, where the overall stress regime is very high and external dynamical stresses due to earthquakes contribute only little to the prevailing stress regime. This limit is defined by the capability of the rock mass to sustain extensional stresses. Should an earthquake occur within a few wave lengths – that is within few kilometres from its source (or hypocentre) – more detailed calculations and other approaches should be considered, however. The hypocentral distance should be given special attention, for even in the epicentre (which is above the hypocentre), the distance between a potential hypocentre from an underground opening can be relatively large, considering the usual focal depth of 5–10 km of earthquakes in the Alps8. This was also noted in Figures 3–4. As an example: The recent earthquake in L’Aquila on April 6, 2009 had a magnitude M6.3, having an epicentral intensity of degree 9 and maximum ground acceleration of about 5 m/s², left 297 dead and more than 40.000 homeless25. Under these circumstances, an underground mine would have suffered little damage below the epicentre despite its closer hypocentral distance but due to the reduced ground motions underground and the lack of resonance of underground openings.

BHM, 154. Jg. (2009), Heft 6

Fig: 5: Resulting strains from a magnitude 5.5 earthquake, as expected underground

4. Conclusion Seismic waves from earthquakes cause less dynamical loads underground when compared with those on surface. The reason for this discrepancy can be explained mainly by surface effects. In addition, the amplitudes of impinging waves need to be superimposed on the current field stresses to judge their actual impact. Since prevailing tectonic stresses are usually much higher than those induced from transient shear waves from tectonic earthquakes, it is not surprising, that the observed underground effects are much less than those experienced on surface. Effects on mining operations from earthquake loads need to be differentiated. Surface operations are much more likely to be affected by earthquakes than underground workings, however. This principle applies worldwide. The situation in Austria is less serious, though, as seismic magnitudes of earthquakes are rather small to cause damage and thus pose a danger underground. However, the situation appears slightly elevated in open cast mines and at tunnel portals where even local earthquakes can do some damage. Therefore, areas in or next to the Vienna Basin, the Mürz Valley and near Innsbruck should be monitored close by to ascertain the continuation of services in terms of traffic and to meet safety expectations. This can be achieved most effectively by monitoring of tunnel access points of important transport routes and by seismic stations at intervals of at least 10 km, once a tunnel is longer than that. Such monitoring would ascertain a continuous control of underground and surface stability. The previous considerations restrict the necessary effort to prove the earthquake resistance of underground structures mainly to their portals during and after mining operations have been conducted, unless geological structures are intersected during the development. This is when induced seismicity may become a major factor affecting not only the advance of underground operations but also the safety of mining personnel.

© Springer-Verlag

Lenhardt

253

References Lenhardt, W.A.: Earthquake triggered landslides in Austria – Dobratsch revisited. Jb. Geol. B.–A., Vol. 147, Vienna, (2007), 193–199. – 2 Harp, E.L. and R.C. Wilson: Shaking intensity thresholds for rock falls and slides: Evidence from the Whittier Narrows and Superstition Hills earthquake strong motion records. Bull.Seism.Soc.Am., 85, (1995), 1739–1757. – 3 Sassa, K., H. Fukuoka, J.H. Lee, and D.X. Zhang: Measurement of the apparent friction angle during rapid loading by the high–speed high–stress ring shear apparatus – Interpretation of the relationship between landslide volume and the apparent friction during motion. Landslides (Bell, ed.), Balkema, Vol.1, (1991), 545–552. – 4 Arias, A.: A measure of earthquake intensity. In: Seismic Design for Nuclear Power Plants. The M.I.T. Press. Hansen, R.J. (ed.), (1970), 438–483. – 5 Grünthal, G. (ed.): European Macroseismic Scale 1998 (EMS–98). Conseil de l’Europe, Cahiers du Centre Européen de Géodynamique et de Séismologie, Luxembourg, Volume 15, (1998), 99 pages. – 6 Shebalin, N.: Correlation between earthquake magnitude and intensity. Studia geophys. et geod. 2, (1958), 86–87. – 7 Sponheuer, W.: Methoden zur Herdtiefenbestimmung in der Makroseismik. Freiberger Forschungsheft C88, Akademie–Verlag Berlin, (1960), 117 pages. – 8 Lenhardt, W.A., Ch. Freudenthaler, R. Lippitsch, and E. Fiegweil: Focal–depth distributions in the Eastern Alps based on macroseismic data. Austrian Journal of Earth Sciences, Vol. 100, (2007), 66–79. – 9 Lenhardt, W.A.: Regional earthquake hazard in Austria. Proc. of ‘10th European Conference on Earthquake Engineering’, Duma, G. (ed.), Balkema, (1995), 63–68. – 10 Travasarou, T., J.D. Bray, and N. Abrahamson: Empirical attenuation relationship for Arias Intensity. Earthquake Engineering Struct. Dyn., Vol. 32, (2003), 1133–1155. – 11 Hammerl, Ch.: Das Erdbeben vom 4.Mai 1201. Mitteilungen des Inst. f. Österr. Geschichtsforschung, Bd.103/3– 4, (1995), 350–368. – 12 Masannat, Y. M.: Geotectonics and earthquake risks in Jordan. Publ. in „Ground Motion and Engineering Seismology“, Developments in Geotechnical Engineering 44, A.S. Cakmak (ed.), Elsevier, (1987), 3–18. – 13 Voight, B. 1

254

Lenhardt

and W.G. Pariseau: Rockslides and avalanches: An introduction. Publ. in „Rockslides and avalanches“, (1978), 1–63. – 14 Glawe, U. and M. Moser: Messtechnische und theoretische Bearbeitung von Bergzerreißungen und Blockbewegungen. Felsbau 11, Nr.5, (1993), 235–250. – 15 Dowding. C.H. and A. Rozen: Damage to rock tunnels from earthquake shaking. Journal of the Geotechnical Engineering Division, ASCE, Vol. 104, No. GT2, (1978). – 16 Berger, E.: Konzeptionelle Überlegungen zum Verhalten von Untertagebauten während Erdbeben. Nagra informiert, 3/87, (1987), 15–20. – 17 Hammerl, Ch. and W.A. Lenhardt: Erdbeben in Österreich. Leykam Verlag, Graz, (1997), 191 pages. – 18 Lenhardt, W.A.: Erdbebenkennwerte. In Erdbebenberechnung von Talsperren, Band 2. Bundesministerium für Land– und Forstwirtschaft, Österreichische Staubeckenkommission, (1996). – 19 Wells, D.L. and K.J. Coppersmith: New empirical relationships among magnitude, rupture length, rupture width, rupture area and surface displacement. Bull.Seism. Soc.Am., Vol.84, No.4, (1994), 974–1002. – 20 Smirnow, T.P. , M.A. Elioff, J.E. Monsees, and J.L. Merritt: Design and behaviour of the Los Angeles metro during seismic events. Proc. Mechanics of Jointed and Faulted Rock II, Rossmanith, H.–P. (ed.), Balkema, (1995), 905–911. – 21 Tamura, T., C. Mizukoshi, and T. Ono. Characteristics of earthquake motion at the Rocky Ground. Proc. of 4th World Conf. on Earthquake Engineering, Santiago, Chile, (1969). – 22 Berardi, R., F. Capozza, and L. Zonetti: Analysis of rock motion accelerograms recorded at surface and underground during the 1986 Friuli seismic record. Proc. OECD Specialist Meeting on the 1976 Friuli Earthquake and the Antiseismic Design of Nuclear Installations, Rome, Italy (1977). – 23 Smit, P.: Strong motion attenuation model for central Europe. In: Proceedings of 11th European Conference on Earthquake Engineering (1998). – 24 Wagner, H.: Support requirements for rockburst conditions. Proc. 1st Int.Symp. on Rockburst and Seismicity in Mines (Gay and Wainwright, eds.). Johannesburg, SAIMM, 209–218. – 25 Cocco, M.: Report on the L’Aquila earthquake at the General Assembly of the European Geoscience Union (EGU) in Vienna, 23.4.2009 (2009).

© Springer-Verlag

BHM, 154. Jg. (2009), Heft 6