Transactions, SMiRT 19, Toronto, August 2007 Paper # K02/3

FREQUENCY CONTENT AND ATTENUATION LAWS FOR SIERRAS PAMPEANAS EARTHQUAKES AND THEIR DIRECT IMPLICATIONS ON THE SEISMIC ASSESSMENT FOR LIFE EXTENSION OF EMBALSE NPP.

Roberto D. Martino1,3, Aldo A. Bonalumi1,3, and Ricardo J. Rocca2,3

1Departamento de Geología Básica, 2Departamento de Construcciones Civiles. 3Facultad de Ciencias Exactas, Físicas y Naturales. Universidad Nacional de Córdoba. Av. Vélez Sarsfield 1611. X 5016 GCA - Córdoba. República Argentina. E-mails: [email protected], [email protected], [email protected]

ABSTRACT

The Embalse NPP (ENPP), located in Córdoba, Argentina (lat: -32.23º, long -64.44º), is currently being reassessed in order to pursue a renewal of the operating license, due in 2010. ENPP site is located at a valley within the Sierras Pampeanas (Pampean Ranges) of Córdoba, which constitute the most eastern group of the Pampean Ranges Province. They are broken Andean foreland blocks, built as a consequence of the flat- lying orientation of the Nazca subducting plate. In the decade of 1970 it was determined a design earthquake based on little instrumental data. It was used a floating earthquake as MCE with a magnitude M = 6.5 and a maximum horizontal acceleration of 0.26 g. In the following decades several earthquakes of medium size magnitude (M = 5.5 to 6.4) within the Pampean Ranges have been registered. Data is not enough for defining a regional attenuation law, but data points adjust to the proposed relations for Central and Eastern US (CEUS). With new tectonic and seismological data, a deterministic analysis indicates that the magnitude size of the controlling earthquake remains reasonable, but changes in attenuation laws and design spectra will have to be considered for the renewal of the operating license The analysis of elastic response spectra shows intermediate shape forms compared with those used for stable continental zones like CEUS and the unstable continental zones like West US. This shape has been determined using response spectra from different sites of the Sierras Pampeanas.

GEOLOGICAL SETTING

Regional Setting The Pampean Ranges (PR, 28º-34º SL), at central Argentina, constitute a lozenge shape core of basement blocks composed of igneous and metamorphic rocks complexes. In between of these blocks there are basins filled with tertiary and cretaceos sediments. Geotectonically the PR are the present foreland of the Andean cordillera. This vast region have been divided into a series of accreted terranes with different provenance and age, produced during distinct Early Palaeozoic accretional events to the pacific margin of Gondwana. These terranes are limited by means of sutures, whose geometry on a map and in a section are still in discussion discussed and it is subjected to several interpretations due to the scarce geological and geophysical information (Fig. 1a) The PR have been formed during the Andean Orogeny, when the Nazca plate arrives at the western side of the Southamerican plate producing cretaceous basin inversion [1], uplift and clockwise (seeing North) tilting of basement blocks, developing the nowadays landscape. It should be noted that some attention that the Nazca plate between 29º y 33º SL deepens to 100 kilometers and then it is horizontalized up to 5º. It remains by several hundreds of kilometers in this last angle, ending near the limit between the Pampean Ranges at Córdoba and the Chacoparanaense Basin (near 64º00´WL). Here, the Nazca plate became vertical and plunge into the mantle [2,3]. This general behavior of the Nazca plate is spatially correlated with the main block Transactions, SMiRT 19, Toronto, August 2007 Paper # K02/3 uplift in the upper plate (Southamerican Plate) of the entire Pampean Ranges. The uplift began ca. 10 Ma ago [4]. The geometry describe above is related to the subduction of the Juan Fernández aseismic ridge whose velocities are in the order of 6.3 mm/y and with an azimuth of 79.5º. Nowadays seismicity along the Pampean Ranges indicates an active tectonics in the entire region. Pampean Ranges near Córdoba are the most eastern group of north-south basement uplifts (sierras) emergent from Quaternary sediments, separated by intermontane basins filled with Mesozoic and Cenozoic sediments. Here the blocks are composed of Late Precambrian-Early Palaeozoic metamorfic rocks intruded by Late Palaeozoic granitoids. In the western central part of the Figure 1b Neogene trachyandesites and pyroclastic deposits form prominent outcrops on top of the clockwise tilted blocks. The most eastern block is named the Sierra Chica and at the bottom of it the ENPP is built. In the following, the more important fault near – and far-field, related to the plant, is described (Figura 1c).

The Sierra Chica Fault (SCF) The Sierra Chica Fault is one of the most important faults of the Pampean ranges, extending its escarpment of fault by more than 200 kilometers. ENPP site is very close to the SCF (less than 1 kilometer). The analysis of smaller faults allowed determining two events of deformation in the SCF: 2 My (Event B) and 0.8 My (Event A) [5]. The FSC do not have geochronological data that would permit to know the movement ages.

The La Calera – Salsipuedes Fault ( = Pampean uplift, PU) This reverse fault gives an small chain, roughly parallel to FSC, integrated by small isolated hills. It is believe that is newer than the FSC but there is no geochronology on it. It is the more eastern thrust recognized nowaday and it is located at 50 kilometer of the ENPP.

Soconcho Lineamient (SL) It is an extensive fragile deformation shear zone that cut the entire Sierra Chica and it is superimposed to an old ductile shear zone. On its NE limit a fracture is nucleated. It presents some features of seismogenic reactivation, marked by the presence of pseudotachylites. The last movement is unknown but it cut the SCF.

Transactions, SMiRT 19, Toronto, August 2007 Paper # K02/3

FIGURA 1. 1a. Map showing the Pampean Ranges Province from Central Argentina with terrane limits and relationships with other geological provinces (based and modified from [6,7]). 1.b. Map from Pampean Ranges of Córdoba. ENPP is the Embalse nuclear power plant. Asterisk mark Sierra Chica at this latitude. 1.c. Structural sketch of the main faults sorrundings ENPP (see text for more details) at Sierra Chica, considered in this paper. Transactions, SMiRT 19, Toronto, August 2007 Paper # K02/3

SEISMOLOGICAL REGIONS

The suture between the Pampia and Cuyania terrains (Figure 1) is located to the west of Valle Fertil fault, and shows a milonitic zone, that presents a structure of important seismic reflection. It constitutes the limit between West Sierras Pampeanas (WSP) and East Pampeanas Orientales (ESP).

FIGURA 2. Differences in seismic activities between Western (WSP) and Eastern (ESP) Pampean Ranges. The Valle Fértil Lineament is the limit between both WSP and ESP. The dot point are epicenters registered in a ten year period. Modified from [7].

The Sierras Pampeanas constitute an active region seismic in the transarc Andean zone, although a noticeable difference between the two differentiated regions exists. The Western Pampean Mountain ranges are most active, with inverse focal mechanisms that reach up to 25 km. The Eastern Pampean Mountain ranges present inverse and strike type focal mechanisms with smaller depths of 10 km. In the last 10 years, an average of 12 earthquakes with M > 4,0 have been generated in WSP, whereas single 3 have epicenter in ESP.

PREVIOUS DESIGN LEVEL

In the decade of 1970 it was determined a verification earthquake based on little instrumental data. It was used a floating earthquake as MCE. It was averaged two PGA values. The first was obtained from an event with a magnitude M = 6.5 at 20 km depth and the second came from an earthquake that produced at the site an intensity MMI = VIII. The maximum horizontal acceleration adopted was 0.26 g. In the following decades, several earthquakes of medium size magnitude (M = 5.5 to 6.4) within the Pampean Mountain ranges have been registered. Data is not enough for defining a regional attenuation law, but data points adjust to the proposed relations by [8] for Central and Eastern US (CEUS). In March 2003, there was an small earthquake (M = 4.3) near ENPP, that confirmed this law.

DETERMINISTIC ANALYSIS OF MCE

Transactions, SMiRT 19, Toronto, August 2007 Paper # K02/3

Current IAEA Deterministic Procedures [9]call for analysis of main faults surrounding NPP. When the magnitudes are analyzed relating them to the size of the faults, empirical regressions were used [10]. Some authors are postulating the hypothesis of earthquakes in the Pampean Mountain ranges, with magnitudes M > 7,0 [11]. The main argument is based on the measurements of superficial offset observed in trenches and other manifestations of seismic activity like paleo-avalanches and paleo-liquefation. Those indicators are less robust than the rupture length, for the determination of the seismic capacity. Superficial outcrop displacements generated by an earthquake vary throughout the fault extension, and the paleo-liquefaction and the slides have a geotechnical complexity. Table 1 shows the computations of Magnitude of the principal fault surrounding the ENPP (Figure 1 c).

TABLE 1. Maximum Magnitude of fault based on the length of superficial rupture

Fault Style Total Length Rupture Magnitude Type Lenth Sierra Chica (FSC) Reverse 42 km 21 km 6.6 Pampean Uplift (PU) Reverse 60 km 30 km 6.8 Soconcho lineament (SL) Strike 25 km 12 km 6.3 El Molino fault Reverse 51 km 25 km 6.7

Peak Ground Accelerations In order to evaluate the PGA at the site, different attenuation laws have been used. Table 2 summarizes the values corresponding to the near and distant earthquakes. The Sierra Chica Fault FSC has not exposed outcrops with holocene displacement in the section close to the CNE. There are evidences in Santa Rosa (15 km to then north), in a section of the fault that would be disconnect of the portion that is adjacent to the ENPP. The offset characteristic of the fault in Santa Rosa is pleistocenic but following regulations of IAEA, it must be considered in the analysis. The Pampean Uplift fault EP is one of most recent of the region and for that reason an earthquake generated in her has been postulated . There are not detail studies that allow its characterization and therefore conservative values have been taken. The Molinos fault is a satellite of Comechingones Fault system (51 km toward the west). Movements in the last 800 to 1300 years have been dated and therefore a source is considered holocenic (< 10000 years). The Soconcho fault is a left lateral fault that affects all the Pampean system and therefore it is younger than the FSC. For that reason also it enters the consideration like active, although measured data of their movement do not exist. The closest of the distant sources, is the Valle Fertil fault that is the limit between the West and East Sierras Pampeanas.

TABLA 2. Peak ground accelerations

FAULT MAGNITUDE DISTANCE CAMPBELL TORO IDRISS (km) [12] [8] [13] FSC 6.6 15 0.25 g 0.37 g 0.30 g PU 6.8 25 0.14 g 0.25 g 0.20 g SL 6.3 15 0.15 g 0.27 g 0.21 g El Molino 6.7 55 0.06 g 0.09 g 0.08 g Valle Fertil 7.8 240 0.024 g 0.024 g 0.04 g

Predominant Periods For a local earthquakes at a distance of 15 km, the results of the more recent empirical model [14], for an earthquake of Ms = 6.6 gives a value of the Predominant Period Tp of approximately 0.23 sec. According to Idriss (2006) the Predominant Period is approximately equal to 0.20 sec. For the case of the distant earthquake, to the distance of 240 km, a Predominant Period Tp approximately equals to 0.45 sec.

Spectral Shapes Transactions, SMiRT 19, Toronto, August 2007 Paper # K02/3

A comparison of the spectral shapes of the March 2003 earthquake has been made, in relation to the shapes of the active crustal zones of the west of the United States (WUS) and the shapes of stable seismic zones of the center of the United States (CEUS), propose by [15]. The horizontal components have spectral shapes that have intermediate characteristics between the above WUS and CEUS. A similar situation has been observed in earthquakes measured in the Pampean Mountain ranges of Tucumán (540 km north of ENPP). On the basis of these evidences, an intermediate spectral form have been defined for the Sierras Pampeanas as shown in Figure 3.

3,0

2,5

2,0

1,5

1,0 PAA/PGA

0,5

0,0 0,01 0,1 1 10 PERIOD (sec)

WUS CEUS -1C CEUS - 2C PAMPEAN RANGES

FIGURE 3. Comparison of elastic response spectra shapes of Pampean Ranges and the proposed for WUS and CEUS by Silva et al (2001)

CONCLUSIONS

By the time of the construction of ENPP, en the late 1970, it was used a floating earthquake as MCE with a magnitude M = 6.5. With the new data, the magnitude size remains reasonable, but changes in attenuation laws and design spectra will have to be considered for the renewal of the operating license. The advance of the tectonic knowledge in the last 25 years has allowed us to have a better characterization of the sismotectonic region of the site where the CNE is located. On the other hand, the measurements of accelerations produced by earthquakes help us to define the seismic parameters. The 200 km length FSC can be divided it in several sections, with different mechanical behavior due to its geometry. The main inference is the presence of geometric barriers which limit the propagation of the rupture, and therefore the size and magnitude of the earthquakes, as it is reflected in Table 1. The parameters that arise from the analysis of the hypotheses that are expressed in Table 2, indicate that the MCE can reach a magnitude Mw = 6.6 with maximum peak accelerations between 0.25 and 0.37g, and a predominant period between 0.20 and 0.23 seg. The analysis of elastic response spectra of different Sierras Pampeanas earthquakes shows intermediate shape forms compared with those used in CE US and West US. This shape has direct implications in the new ENPP seismic evaluation .

REFERENCES

[1] Schmidt, C.J., Astini, R.A., Costa, C.H., Gardini, C.E. y Kraemer, P.E., “Petroleum basins of South America”. Cretaceous rifting, Alluvial Fan Sedimentation, and Neogene Inversion, Southern Sierras Pampeanas, Argentina”. En: A.J. Tankard, R. Suárez S., and H.J. Welsink, AAPG Memoir 62, 1995, pp. 341-358. Transactions, SMiRT 19, Toronto, August 2007 Paper # K02/3

[2] Booker, J.R., Favetto, A. and Pomposiello, M.C., “Low electrical resistivity associated with plunging of the Nazca flat slab beneath Argentina”. Nature, 2565, doi: 10.1038, pp. 1-4. [3] López de Luchi, M., Favetto, A., Pomposiello, M.C. and Booker, J.R., “Magnetotelluric evidences for the suture between the Ríao de la Plata and Pampean cratons at 31º40´S, Córdoba Province, Argentina”. 6th International Symposium on Andean Geodynamics (ISAG 2005, Barcelona), Extended Abstracts, pp. 446-449. [4] Jordan, T. y Allmendinger, R., “The Sierras Pampeanas of Argentina. A modern analogue of Rocky Mountain foreland deformation”. American Journal of Sciences, 1986., Vol. 286, pp. 737-764. [5] Martino R.D, Kraemer, P.E. Escayola, M. P., Giambastiani, M. and Arnosio, M. “Transecta de las Sierras Pampeanas de Córdoba a los 32º LS”. Revista de la Asociación Geológica Argentina, Vol 50, 1995, pp. 60-77. [6] Alvarado P., Beck S., Zandt, G., Araujo, M. and Triep, E., “Crustal deformation in the south-central Andes backarc terranes as viewed form regional broad-band seismic waveform modeling”. Geophysics Journal International, Vol. 163, 2005, pp. 580-598. [7] Alvarado P., Machuca B.G. and Beck S. “Comparative seismic and petrographic crustal study between the Western and Eastern Sierras Pampeanas region (31ºS)”. Revista de la Asociación Geológica Argentina, Vol. 60, 2005, pp. 787-796. [8] Toro, G.R., Abrahamson, N.A. and Schneider, J.F., “Model of strong ground motions from earthquakes in Central and Eastern North America: best estimates and uncertainties”. Seismological Research Letters, Vol. 68, 1997, pp. 58-73. [9] IAEA (2002) Evaluation of Seismic Hazards for Nuclear Power Plants. Safety Guide No NS-G-3.3. [10] Wells D.L. and Coppersmith K.J., “New empirical relationships among magnitude, rupture length, rupture witdth, rupture area and surface displacement”. Bulletin of Seismological Society of America. Vol 84, pp 974-1002. [11] Costa C.H., Murillo M.V., Sagripanti G.L. and Gardini C.E., “Quaternary intraplate deformation in the southeastern Sierras Pampeanas, Argentina”. Journal of Seismology, Vol 5, 2001, pp. 399-409. [12] Campbell, K.W. and Bozorgnia, Y., “Empirical Ground Motion Model for the Average Horizontal Component of PGA, PGV and SA at Selected Spectral Periods Ranging from 0.01-10.0 Seconds”. Interim Report for USGS Review, 2006 . [13] Idriss, I.M. ”Empirical model for estimating the average horizontal values of pseudo-absolute spectral accelerations generated by crustal earthquakes”. Interim Report Issued for USGS Review, Vol.1., 2007. [14] Rathje E.M., Abrahamson N.A. and Bray J.D. “Simplified frequency content estimates of Earthquake Ground Motions”. Journal of Geotechnical and Geoenvironmental Engineering ASCE, Vol. 124, No 2, 1998, pp. 150-159. [15] Silva W.J., Youngs, R.R. and Idriss, I.M., “Development of Design Response Spectral Shapes for Central and Earstern U.S. (CEUS) and Western U.S. (WUS) Rock Site Conditions”. Proceedings of the OECE-NEA Workshop on Engineering Characterization of Seismic Input. Nov. 15-17, 1999. NEA/CSNI/R (2000) 2, Vol.1, 2001, pp. 185-268.