SEISMICITY CHANGES IN POLYPHYTO DAM AREA (NW GREECE) K. Pavlou 1 , G. Kaviris 1 , K. Chousianitis 2 , G. Drakatos 2 , V. Kouskouna 1 , K. Makropoulos 1 1 - University of Athens, Geophysics-Geothermics Department, Greece 2 - National Observatory of Athens, Geodynamic Institute, Greece kpavlou@geol.uoa.gr; gkaviris@geol.uoa.gr; chousianitis@noa.gr; g.drakat@noa.gr; vkouskouna@geol.uoa.gr; kmacrop@geol.uoa.gr Introduction The three reservoirs of Aliakmonas river (NW Greece), i.e. Polyphyto, which is the second greatest res- ervoir in Greece, Sfikia and Asomata, are located downstream of each other. The basin of the study area belongs to the Pelagonian zone which consists of metamorphic rocks (crystalline substratum) covered by carbonate rocks and flysch. Data The epicenters and the source parameters are obtained from the catalogue of Makropoulos, Kaviris & Kouskouna (2012) which is an updated earthquake catalogue for Greece and adjacent areas. The Makropoulos & Burton (1981) and Makropoulos, Drakopoulos & Latoussakis (1989) catalogues, cover- ing the time span 1900-1985 were updated for the period 1986-2009, using instrumental parameters from the bulletin of the ISC, apart from magnitude. For Ms, the same procedure as in the previous ef- forts was applied, i.e. using the ISC body wave magnitude (m b ) and a regression equation. In the up- date used in the present study, Mw is also included for the whole period 1900-2009, calculated ac- cording to a regression equation. The parameters of an earthquake catalogue, as well as its completeness, play a key role in the credibil- ity of any attempt to assess the seismic regime of a region and, therefore, on how realistic the seismic hazard assessment is. In order to assess the degree of completeness of the instrumental catalogue of Makropoulos et al. (2012), that is to find the magnitude above which the catalogue can be considered as reasonably complete, or alternatively to assign time intervals in which a certain magnitude range is likely to be completely reported, the method introduced by Stepp (1971) was applied. The analysis is based on determining the time period required to restore a stable release rate of a specific range of magnitudes. Table 1 summarizes the results of the application of the test of completeness on the Makropoulos et al. (2012) catalogue. Events with magnitude above 4 are completely reported for the last 34 years of the catalogue (1976-2009). The earthquakes of half a magnitude unit larger are com- pletely reported for the last ϲϬ years of the Đatalogue aŶd eǀeŶts ǁith Ms≥ϱ.Ϭ or Ms≥ϱ.ϱ after ϭ9ϰϬ and 1911, respectively. No earthquake with magnitude 6 or greater seems to have been omitted for the whole period (i.e. 1900-2009). Table 1: Results of the completeness test of the Makropoulos et al. (2012) instrumental earthquake catalogue. In order to investigate the seismicity of the study area during the instrumental period (i.e. 1900-2009) the catalogue was divided into three time windows: the first up to 1973, the second up to 1994 and the third 1900-2009. The selection criteria were the date (1974) of impoundment of the Polyphyto reservoir and the strong Kozani earthquake (Ms=6.5) which occurred in 1995. The data used are lim- ited to distances up to 70Km around the three artificial lakes. The spatial distribution of the seismic ac- tivity for distances up to 50Km is illustrated in Fig. 2. Fig. 2: Spatial distribution of seismicity (M s ≥ϰ) for the period ϭ9ϬϬ-2009. Seismicity The temporal distribution of events by magnitude and number ( Fig.3, Fig.4, respectively) was carried out and compared for the three selected periods. Significant increase of seismicity was observed since 1974, possibly ini- tiated by the impoundment of Polyphyto reservoir (1974) and by the two downstream reservoirs (i.e. Sfikia in 1984 and Asomata in 1985). The occurrence of the Kozani (1995) earthquake dominates in the 1900-2009 period. Fig.3: Temporal distribution by number: Left: for period 1900-1994 - Right: for period 1900-2009 Fig.4: Temporal distribution by size: Left: period 1900-1994 - Right: period 1900-2009 Seismic Hazard The most suitable statistical approach for the available sample is the theory of Extreme Values founded by Gum- bel in 1939 which has extensively been applied to seismic hazard problems (Epstein and Lomnitz, 1966; Shakal and Willis, 1972; Karnik and Schenkova, 1977; Burton, 1978; Makropoulos, 1978; Makropoulos and Burton, 1985). Extreme Values distribution follows three different types of asymptote distributions which do not depend on the parent population distribution. The estimation of maximum expected ground motion parameters, includ- ing PGA, requires the application of the first asymptote, i.e. of the equation: Φ(x) (1) = exp [-exp (-a(x-u))], a>0 (1) where Φ(x) represents the probability of occurrence of an earthquake with magnitude less than or equal to x. The values of the peak ground acceleration (PGA) with a 90% probability of no exceedance in 50 years, in the vi- cinity of Polyphyto Dam, are calculated using the extreme values method and the HAZAN software (Makropoulos and Burton, 1986). Three time periods were selected, in order to examine the effect of the Dam to the seismic hazard of the region. The first covers the time period before the impoundment of the Dam (1900-1973), the sec- ond lasts until the occurrence of the Kozani-Grevena Earthquake (1900-1994) and the third refers to the whole in- strumental period. PGA values for the first period vary between 50 cm/sec 2 and 175 cm/sec 2 , for the second be- tween 75 cm/sec 2 and 200 cm/sec 2 , while for the whole instrumental era between 100 cm/sec 2 and 225 cm/sec 2 . The obtained values indicate a clear PGA increase after the impoundment in the whole region, as it can be ob- served in the map, where the differences of PGA values between the period before the impoundment (1900- 1973) and the one before the Kozani earthquake (1900-1994) are presented. It is worth noting that the PGA in- crease is evident at the epicentral region of the Kozani earthquake, before its occurrence. The largest increase (90 cm/sec 2 ) is observed SE of the Dam, where the highest PGA values are obtained for both the second and the third period. Within distances smaller than 40km from the Dam, the only area where PGA values reduced is the one lo- cated close to its NE edge. The south-west part is covered by ophio- lites (Jurassic-Upper Cretaceous) and mo- lasse sediments which belong to the Mesohellenic trench, while the northern part of the broader area (up to 50Km around the artificial lakes) is covered by carbonate rocks that belong to Axios zone and ophiolites (Fig.1). The main faults that dominate the broader Polyphyto reservoir area are: a) a 70km long fault zone which consists of several subparallel faults with an ENE-WSW to NE- SW direction, parallel to the Polyphyto ar- tificial lake, b) the Chromio-Vari, Pontini- Pilori and Feli faults located SW of the edge of the Polyphyto lake, striking in a E- W direction and c) the Vegoritis-Ptolemais fault system with a NE-SW direction. Fig.1: Map of the main geotectonic zones of the broader study area (50Km around the three artificial lakes). Surface Wave Magni- tude Time required for stable recurrence rate (yr) Period of completely reported events M s ≥ ϰ.Ϭ 5 1976-2009 M s ≥ ϰ.ϱ 20 1950-2009 M s ≥ ϱ.Ϭ 25 1940-2009 M s ≥ ϱ.ϱ 45 1911-2009 M s ≥ ϲ.Ϭ 60 1900-2009 References Burton, P.W., 1978. The application of extreme value statistics to seismic hazard assessment in the European area, Proc, Symp., Anal. Seismicity and on Seismic Risk, Liblice, 17-22 October 1977. Academia, Prague 1978. Epstein, B., and Lomnitz, C., 1966. A model for the occurrence of large earthquakes, Nature, 211, p. 954–956. Karnik, V., and Schenkova, Z., 1977. The third asympotic distribution in earthquake statistics, Proc. Sympr. Anal. Seismicity and on Seismic Risk, Liblice, 17-22 October 1977, Academia Prague 1978. Makropoulos, K.C., 1978. The statistics of large earthquake magnitude and an evaluation of Greek seismicity, Ph.D. thesis, Univ. of Edinburgh. Makropoulos, K.C. and Burton, P.W., 1981. A catalogue of seismicity in Greece and the Adjant Areas, Geophys. J.R. Astr. Soc., and Microfiche GJ65. Makropoulos, K.C. and Burton, P.W., 1985. Seismic Hazard in Greece: I Magnitude recurrence. Tectonophysics, 117, 205-257. Makropoulos, K.C. and Burton, P.W., 1986. "HAZAN": A Fortan program to evaluate seismic-hazard parameters using Gumbel's theory of extreme value statistics. Computer and Geosciences, Vol.12/1, 29-46. Makropoulos, K.C., Drakopoulos, J., and Latoussakis, J.B., 1989. A revised and extended earthquake catalogue in Greece since 1900. Geophys. J. Int. 98. Makropoulos K., Kaviris G. and Kouskouna V., 2012. An updated and extended earthquake catalogue for Greece and adjacent areas since 1900. Nat. Hazards Earth Syst. Sci., 12, 1425-1430. Shakal, A.F., and Willis, D.E., 1972. Estimated earthquake probability in the North Circum - Pacific Area. Bull. Seism. Soc. Am., v. 62, 6, p. 1397-1410. Stepp, J.C., 1971. An investigation of Earthquake Risk in the Puget Sound Area by use of the type I Distribution of Largest Extremes, Ph.D Thesis, Pensylvania State Univ. Conclusions  The number of earthquakes that occurred in the study region increased after the impoundment of the Poly- phyto Dam (1974).  The results of the seismic hazard analysis indicated an increase of the PGA values, with a 90% probability of no exceedance in 50 years, after the Polyphyto impoundment in the whole region, with the exception of the vicin- ity of the NE edge of the Dam.  The largest increase is observed SE of the Dam, where the highest PGA values are obtained for both the second (1900-1994) and the third (1900-2009) period (200 and 225cm/sec 2 , respectively).  The calculated increase of both the number of earthquakes and of PGA values may be attributed to the im- poundment and the water level fluctuations of the Polyphyto Dam. Acknowledgements The present study was funded through the 7 th FP projeĐt EPO“: EuropeaŶ Plate OďserǀiŶg “ysteŵ ǁhiĐh is the iŶtegrated solid Earth “ĐieŶĐes researĐh iŶfrastruĐture approved by the European Strategy Forum on Research Infrastructures (ESFRI) and is included in the ESFRI Roadmap. National and Kapodistrian University of Athens