Version classiqueVersion mobile
OpenEdition Books

Risque, nature et société

Lucien Faugères
Christiane Villain-Gandossi

II. Étude de cas. La prévision des séismes (Grèce, Albanie)

Earthquake hazard assessment in Albania

Eduard Sulstarova


L'Albanie est un pays dont l'activité sismique atteint des niveaux proches de ceux de la Grèce et de la Turquie. Avec la croissance de la population et l'urbanisation, les enjeux économiques de la prévention des séismes deviennent de plus en plus importants. Une politique efficace de prévention et de protection passe d'abord par une bonne connaissance des aspects historiques et géographiques de la sismicité (importance de la cartographie pour établir une régionalisation sismique), et par l'étude des facteurs géologiques. Cette étude vise à la mise en relation des données structurales et tectoniques. Elle combine la cartographie des foyers, la cartographie des failles, les levers de cartes néotectoniques, de cartes historiques des épicentres, des intensités, des magnitudes. Toutes ces méthodes permettent d'aboutir à un zonage sismique régional, à l'échelle du pays, indispensable pour répartir les efforts à accomplir en matière de prévention et de protection.

Texte intégral


1Albania, as a country of high seismic activity in Europe, is among the most frequently hit by damaging earthquakes, next to Turkey and Greece. Only during this century, earthquakes in Albania and nearby have generated 7.2% of the total energy released by all the shallow earthquakes occurring in Europe during this period. At nowadays' level of development, with the increasing density of population in urban areas and the rising investment requirements the damage will be greater each time if the seismological, seismological-engineering and earthquake-engineering studies of the probable events are not taken into consideration. In our country, measures are taken to reduce the human losses and the material damage that an earthquake may cause.

2Earthquake hazard is a complex function defining the probability of occurrence, at any given site and during any given period of time, of earthquake ground motion of various intensities. Earthquake hazard is determined entirely by natural processes going on deep within the Earth's crust and, at the present time, at least, is not susceptible to human influence.


3Earthquake hazard assessment is connected with the exact time and place of occurrence, as well as the intensity of the future event. Earthquake hazard cannot be changed, but it is nevertheless essential to be accurately known, in order that the most appropriate technical-economic measures be taken to reduce the damage that the earthquakes may cause to a point where the overall risk is brought down to an acceptable level for society; the definition of this level is a task of considerable difficulty.

4For earthquake hazard assessment we have acted in the following way:


5As a result of many years' endeavour, we have gathered and checked ample material from a period of about 2000 years. All the data collected are computed so we might compile a "Catalogue of Earthquakes in Albania" till 1970 (Sulstarova 1975) and a catalogue for 1971-1985 (Sulstarova 1986). At the same time, we have compiled an album of isoseismal maps, which contains the isoseismals of 160 earthquakes occurring during 1800-1985. Nineteen destructive earthquakes occurred in Albania during the last 90 years, i.e. 1901-1989. These earthquakes caused more than 400 human losses and destroyed or heavily damaged about 35,000 buildings.


6The observed surface effects of the earthquakes are expressed in degrees (I), velocity (cm/sec), or acceleration (cm/sec2). Recently we have set up a strong motion network which at present contains 29 accelerographs SMA-1 and 45 seismoscops WM-II. As we have not got any information from this network yet, the surface effect on the earthquake is evaluated according to the MSK-64 scale. For the macroseismic field, the following relations are found:

7Io=1.9M-3.061og h-0.61; h<10km, 4.0<M<7.5 (1)

8Io=1.75M-4.551og h-3.45; 10<h<40, 4.0<M<7.5 (2)

9As an average for all Albania: Io=2.1-41og h + 0.44; 4.0<M<7.5 (3)

10On the basis of these relations, we come to the conclusion that in Albania and nearby tremors of 6 degrees are caused by earthquakes of M>4.9; of 7 degrees by earthquakes of M>5.5; of 8 degrees by earthquakes of M>6.0; while the ones of 9 degrees by earthquakes of M>6.6. Considering the isoseismal maps of the strongest earthquakes, the maximum intensities observed in all inhabited centres, and Albanian earthquake foci maps (taking into account the above relations of the macroseismic field), the map of the maximum observed intensities for the period 1800-1985 was compiled (fig. 1). This map shows the maximum surface effects in our country.


11The most appropriate mapping of earthquake parameters is based on the epicentre maps, which represent a fundamental information on the seismic activity, the zones where the earthquakes are generated, and hence they can be used for seismic activity studies. For the study of the spatial distribution of seismic activity, a series of maps of earthquake epicentres have been compiled. In this paper are presented: (1) the map of epicentres of historical earthquakes for the period from II-III BC up to 1900 (fig.2), (2) the map of the earthquake epicentres for the period 1901-1987 representing two ranges of magnitude: M>5.0 for 1901-1970, and M>4.0 for 1971-1987 (fig. 3), and (3) the map of earthquake epicentres for the period 1977-1987 of M>3.0 (fig. 4).

12From these maps it may be seen clearly that earthquake foci are concentrated in the Ionian-Adriatic zone, along the Ionian and Adriatic coast, mainly in the external area of the Albanids, in the vicinity of where the "promontory" of the Adriatic plate comes into collision with the Albanids orogene (Albanids microplate): in Korca-Ohri-Peshkopi (Diber) along the Drini fault zone at the Eastern part of Albania (the internal area). These two seismic zones of NW-SE and N-S direction are intersected by the Fieri-Elbasani-Dibra transversal belt, which is very active and has a NE-SW direction. The Ionian-Adriatic zone and the Fier-Elbasan-Diber transversal belt extend beyond the Albanian territory over a large part of the Balkan region. Thus, the Ionian-Adriatic seismogenetic zone extends along all the Adriatic and Ionian coast line and joins the Aegean arc southwards. The Fieri-Elbasani-Diber transversal belt continues towards Skopje up to Bulgaria. Considering the epicentre map for the earthquakes of M>3.0 (for the period 1977-1987, fig. 4) it is obvious that in the zones where the strong earthquakes have occurred (in the two zones and the belt mentioned above), also weak earthquakes, of greater number and frequency, occur continuously. This map also determines some transversal seismogenetic lines as the Shkodera-Peja, the Vlore-Tepelene-Korce-Lake of Prespa ones etc.

Fig. 1. Maximum intensities (1800-1988)

Fig. 1. Maximum intensities (1800-1988)

Fig. 2. Historical earthquakes

Fig. 2. Historical earthquakes

Fig. 3. Epicenters (1901-1987)

Fig. 3. Epicenters (1901-1987)

Fig. 4. Epicenters (1977-1987)

Fig. 4. Epicenters (1977-1987)

13In the seismogenetic zones and lines of Albania and in the surrounding area, as in the Balkans in general, the epicentres of earthquakes occur in narrow belts which follow the zones of active tectonic faults, and generally appear distinctly in the morphology of the continent or at sea floor. The concentration of the epicentres in narrow belts leads to the conclusion that the earthquake foci in Albania and nearby are mainly of linear type.

14The foci of Albanian earthquakes are generally located in the granitic layer, therefore they have shallow foci of 10-20 km; few earthquakes have depths exceeding 50-70km.

15The final objectives of seismic hazard studies are to define the focus, the magnitude and the time of the future event. As, up to the present time, it has been impossible to account for all the processes that take place in an earthquake focus, our efforts aim at determining the average seismic hazard combining various empiric relations and statistical models or defining some criteria of earthquake distribution in time and space. Such analysis requires homogeneity of seismological data in space and time within the energy range, which, for our country, can be carried out considering all the events of Io>6 degrees for the period 1901-1985.

16Frequency distribution of earthquakes, as a function of their magnitude, has first-hand importance to the study of the seismicity of a region. The law of earthquake recurrence is already known (Gutemberg-Richter, 1954):

17log N(M) = a-bM (4)

18Making use of the least-square method and the data of 1901-1985, in our country, the earthquake recurrence law for one year was found to be:

19log N = 5 74-1.09 M; 4.6<M<7.5 (5)

20whereas using the method of maximum likelihood (Utsu 1965, Aki 1965) the following relation was obtained:

21log N = 5.74-1.10 M;4.6<M<7.5 (6)

22As can be seen, both methods give more or less the same result. Coefficient b = 1.10 approximates the world value for shallow earthquakes (Utsu 1971) and also the value obtained for the shallow earthquakes of Europe, b = 1.01 (Karnik 1979).

23The most popular probabilistic approach is the application of the Gumbel theory of largest value. We have used the first Gumbel asymptotic distribution and later his third one.

24In the first Gumbel distribution the probability model (Epstein et al. 1966) has been used. For this model to be applied, the supposition that earthquakes are events independent of each other and constitute a Poissonian process must be made. From this model we have obtained the


log N = 3.96-0.80M 4.6<M<7.5



RD (M) = 1-exp (-aD-BM)


25a and b are coefficients found with the method of the least squares and have the respective values:

26a = 9119.025, b = 1.8437, in formula (7) a = In a/ln(10), b = b/ln(10).

27Using equation (8), we can see that 7-degree earthquakes (M>5.5) should be expected in our country every 5 years, earthquakes of 8 degrees (M>6.0) every 10 years, while 9-degree ones (M>6.6) every 30 years (probability 80-97%).

28Recently the third Gumbel asymptotic distribution has been used:

29where K>0, x <w, u <w, w-is the upper magnitude limit, K = 1/1 is the shape parameter, u - is the characteristic largest value.

30The three parameters were calculated by the least-square method making use of the data of 1901-1985 (Sulstarova 1986); annual intervals were used, too. The curve of the third asymptotic distribution defined by w = 7.46, l = 0.26 k = 3.83 and u = 4.99 fits the observed data quite well.

31Using the third Gumbel distribution, the value Ms = 7.46 can be considered as the magnitude of the largest possible event in the Albanian area. Based on this, and making use of the data of 1800-1983, the map of the maximum expected magnitudes has been compiled (fig. 5) (Kociaj 1986). According to this map, the maximum expected magnitude is Mmax = 7.1 - 7.5.


32The period of historical seismological data and instrumental recordings is relatively short, consequently it is difficult to determine the seismic activity of a country based on seismological data only, and especially on the strongest probable earthquakes, the recurrence period of which is relatively long.

33Earthquakes are phenomena accompanying the present geological situation and manifestations of the deep tectonic processes. There is a close connection between earthquakes and the neotectonic structure which crops up on the surface. Earthquakes lead to new changes in the geological structure, therefore it is indispensable to carry out special geological geophysical and geodetic studies to evaluate the seismic hazard.

34Intending to study the seismicity of Albania, special seismotectonic studies are carried out, widely using the geological and geophysical studies performed by other specialists on the geological and tectonic construction of our territory.

35The geodetic studies on today's Earth crust movements in our country are at their beginning. In some zones of Vlora, Durres, Shkodra and Tirana special geodetic networks have been set up.

36The seismotectonic, geophysical and earthquake focal mechanism studies (Aliaj 1980, 1983, 1986; Sulstarova 1980, 1983, 1986) point out that our territory and the surroundings are divided by some blocks ormicroplates", by deep faults of extension (fig. 6). The seams bordering these "microplates" (longitudinal and transversal deep fault zones) are seismogenetic structural elements. The main cause of the seismic activity in these borders is the collision between the two great plates: the Euroasiatic and the African ones. In this process an important role is played by the "Adriatic promontory" of the African plate which, due to its Northern motions in and near our territory, exerts stresses and is responsible for the seismic activity of our country.

37Confronting the fault zones with the earthquakes generated by them, the seismoactive deep fault zones in Albania, their expected seismic energy and their respective mechanism are determined (fig. 7) (Sulstarova 1986). Based on the geological and tectonic structure of our country, on the dimension of the microplates as well as the seismological data, the maximum magnitude of earthquakes in Albania should not exceed M max 7.2 (Sulstarova 1975, 1983, 1986).

Fig. 5 Map of the maximum magnitudes (M max). The third distribution of the extreme values according to the data of 1980-1983 for M sup./égal à 5.1

Fig. 5 Map of the maximum magnitudes (M max). The third distribution of the extreme values according to the data of 1980-1983 for M sup./égal à 5.1


38According to the seismotectonic and seismological data we have compiled the map of zones of occurrence probability in Albania (fig. 8) (Sulstarova et al. 1980, 1982).


39From the map of zones of earthquake foci (fig. 8) (prepared according to the average radius of the attenuation of intensity from the focal zone), and according to the principle that an earthquake may occur at every point of the focus, we have compiled the map of the seismic regionalization of the PSR of Albania on a scale of 1:500 000 (fig. 9).

40The map of seismic regionalization represents the maximum expected intensity for mean soil conditions expressed in degrees according to the MSK-1964 scale.

41For mean soils we have accepted those of the most inhabited centres of our country (most ordinary soils), where the basic intensity is equal to the maximum intensity observed: thick loose quaternary deposits of deep-level underground waters, below 5m from the surface of the earth (sands, fine sands, clay, fine clay, gravel). These soils are situated at the fields of the country. In this category we have also included the hard and semi-hard rocks at a relief mean differentiated.

42On the seismic map three categories of zones are distinguished: zones of I = VIII, VII and VI degrees. At some strong focal zones, due to bad soil conditions, the seismic intensity may attain 9 degrees. The seismic map represents the expected seismic hazard in Albania for 100 future years. This map was approved by our government in December 1979, and is included in our aseismic design technical regulations.

Fig. 6. Neotectonic map.

Fig. 6. Neotectonic map.

Fig. 7. Fault zones in Albania and today tectonic field according to the focal mechanism and macroseismic field.

Fig. 7. Fault zones in Albania and today tectonic field according to the focal mechanism and macroseismic field.

Fig. 8. Zones of earthquakes foci of Albania on the basis of seismological and geological data.

Fig. 8. Zones of earthquakes foci of Albania on the basis of seismological and geological data.

Fig. 9. Map of the seismic regionalization of Albania.

Fig. 9. Map of the seismic regionalization of Albania.



1. K. Aki. (1965), "Maximum likelihood estimate of b in formula log N = a - bM and its confidence limits”, Bull. Earth. Res. Inst. Tokyo Univ., 43, pp 237-239.

2. Sh. Aliaj. (1979), Seismotectonic and geological criteria of the seismicity of Albania, Thesis. Seismological Center, Tirana (in Albanian).

3. Sh. Aliaj. (1983), "The main seismotectonic characteristics of Albania, The earthquake of April 15, 1979", "8 November" Pb. House, pp 358-384.

4. Sh. Aliaj. (1987), "Seismoactive faults in Albania", Buletini i Shkencave Gjeologjike, Nr.2, pp 123-135 (in Albanian).

5. B. Epstein, C. Lomnitz. (1966), "A model for occurrence of large earthquakes", Nature, 211, 5052, pp. 954-9546.

6. E.J. Gumbe. (1962), Statistics of extremes, New York, Columbia University Press.

7. V. Karnik. (1979), "Large European Earthquakes of the 20th Century", Tectonophysics, 53, pp. 159-169.

8. S. Kociaj. (1986), The seismic hazard of the Earth crust in Albania and its precision in building site, Thesis, Seismological Center, Tirana.

9. A. Roca et al. (1984), "Application of the Gumbel III law to seismic data from Southern Spain", Engineering Geology, Vol. 20, Nr. 1/2.

10. C.H. Scholz (1986), "The frequency-magnitude relation of microfracturing in rock and its relation to earthquake", Bull. Seism. Soc. Am., 58, pp. 319-415. Report of WG A, UNDP/UNESCO Project RER/79/014, Paris.

11. E. Sulstarova, S. Kociaj. (1986), "Catalogue of earthquakes of Albania for the years 1971-1985", Seismological Center, Tirana.

12. E. Sulstarova. (1986), The focal mechanism of earthquakes and tectonic stress field in Albania, Thesis, Seismological Center, Tirana (in Albanian).

13. T. Utsu. (1965), "A method for determining the value of b in formula log n = a - bM showing the magnitude frequency relation for earthquakes", Geophys. Bull. Hokkaido Univ., 13, pp. 89-103.

14. T. Utsu. (1971), "Aftershocks and earthquake statistics (III)", J. Fac. Sci. Hokkaido Univ. Geoophys., 3, pp. 379-441.

15. E. Sulstarova (1974), The Albanian Seismicity, Thesis, Seismological Center, Tirana (in Albanian).

16. E. Sulstarova, E., Kociaj, S. (1975), "Catalogue of earthquakes of Albania", Seismological Center, Tirana (in Albanian).

17. E. Sulstarova, S. Kociaj, Sh. Alia. (1980), "Seismic Regionalization of the PSR of Albania", "8 Nentori" Pb. House, Tirana.

18. E. Sulstarova, S. Kociaj, Sh. Aliaj. (1982), Map of earthquake foci in Albania, Final

Table des illustrations

Titre Fig. 1. Maximum intensities (1800-1988)
Fichier image/jpeg, 224k
Titre Fig. 2. Historical earthquakes
Fichier image/jpeg, 217k
Titre Fig. 3. Epicenters (1901-1987)
Fichier image/jpeg, 318k
Titre Fig. 4. Epicenters (1977-1987)
Fichier image/jpeg, 312k
Fichier image/jpeg, 18k
Titre Fig. 5 Map of the maximum magnitudes (M max). The third distribution of the extreme values according to the data of 1980-1983 for M sup./égal à 5.1
Fichier image/jpeg, 124k
Titre Fig. 6. Neotectonic map.
Fichier image/jpeg, 213k
Titre Fig. 7. Fault zones in Albania and today tectonic field according to the focal mechanism and macroseismic field.
Fichier image/jpeg, 369k
Titre Fig. 8. Zones of earthquakes foci of Albania on the basis of seismological and geological data.
Fichier image/jpeg, 376k
Titre Fig. 9. Map of the seismic regionalization of Albania.
Fichier image/jpeg, 194k


Séismological Center, Tirana, Albanie.

© Éditions de la Sorbonne, 1996

Conditions d’utilisation :