ישובי הגליל סבלו נזקים רציניים כתוצאה משתי רעידות האדמה בשנת 1837 ובשנת 1921. גלישת הקרקע ושימוש בחומרים לא-יציבים לבנייה היו בין הגורמים לנזק העיקרי. מסקנות החוקרים היו שיש לאמוד את הערכות לרעידת אדמה באזורים השונים על מנת להתכונן לאירועים הצפויים.
Two major earthquakes occurred in northern Israel during the last two centuries. The first earthquake, the "Safed earthquake", occurred on 1/1/1837 and the second, which was less pronounced, occurred on 11/7/1927. Safed was completely destroyed in the 1837 earthquake and Arab villages in its vicinity were severely damaged. The earthquake of 1927 was less pronounced in the Galilee but caused heavy damage in Nablus.
Estimates of the intensity, magnitude and epicenter location of these earthquakes were made by Arieh (1967), Ben-Menahem et al. (1976) and Vered and Striem (1977). The epicenter of the Safed earthquake, according to Vered and Striem (1977) was probably in the Jordan Rift Valley, east of Safed and the magnitude of the earthquake was estimated to be 6.8 and 6.25 to 6.5 by Arieh (1967) and Vered and Striem (1977) respectively. These estimates of magnitudes and epicenter locations were based on macroseismic data.
Arieh (1967) gives the magnitude of the 1927 earthquake as 6.25, and Vered and Striem (1977) locate its epicenter near the Damiye Bridge in the Jordan Valley.
THE EFFECTS OF THE 1837 EARTHQUAKE
Vered and Striem (1976a) present a compilation of detailed descriptions of the damage suffered by Jewish and Arab settlements in the Galilee region as a result of the 1837 earthquake. In two of the sites which suffered heavy damage (intensity 8 or more) landslides or poor soil conditions appear to have been responsible for much of the damage.
In Safed (Fig. 1), which was totally destroyed, an eyewitness account of the damage, as quoted by the above authors and reproduced here, appears to indicate that a landslide was the direct cause of the damage: ". . . The whole surface of the ground, on the top and round the sides of the hill, bore signs of the frightful convulsions; - here great masses of rock rent and torn; there, huge fissures in the earth, half filled with loose clay and stones from the shivered sides." "We saw (near Safed) rents and cracks in the rocks and earth . . ." At Gush Halav (Djish, El-Jish),". . . A large rent in the ground . . . when seen nearly three weeks afterwards, was about a foot wide and fifty feet long. It was described as being much larger." "We saw (near Safed) rents and cracks in the rocks and earth, though not as large a fissure as at Jish which I examined that morning . . ."
Vered and Striem (1976a) also quote descriptions of the damage suffered by Ein Zeitun, Tiberias, Lubya, Sejera and Reina (Fig. 1). Heavy damage was also reported from numerous places in Lebanon.
In contrast to the above mentioned places which suffered heavy damage, some nearby sites either completely escaped damage or were only slightly damaged. Kafr Kanna, for example, which is situated 4 km from Reina was undamaged, as was quoted by the above authors. "In the apparent capricious course the earthquake struck the neighbouring villages but spared Kafr Kanna from destruction." Nazareth, which is situated 3 km from Reina, suffered only moderate damage (Amiran, 1951). According to Amiran, Sasa Fig. 1) was not affected by the earthquake, whereas at El-Jish, which Is 4.5 km away, the earthquake was of destructive force.
The sharp contrast in the effects of the 1837 earthquake at adjacent localities in the Galilee region suggests that local ground conditions played an important part in the amount of damage suffered at any particular locality
INTERPRETATION OF AERIAL PHOTOGRAPHS AND OF EYEWITNESS DESCRIPTIONS
Ancient Safed, Gush-Halav (Jish), Ein Zeitun and Reina are all situated on the slopes of steep hills, as was a large part of Lubya Sejera was situated on a gentle slope and partially on the floor of a valley (wadi). The presence of old landslides at Gush-Halav, Reina and at the site of Ein Zeitun was observed on aerial photographs. The outlines of these landslides are marked on Figs. 2, 3, and 4. Reina was also heavily damaged in the 1927 earthquake. Aerial photographs reveal here a landslide and the local sheik confirmed that the buildings which were damaged during the 1927 earthquake were situated within the landslide zone. He also claimed that at the time of the earthquake cracks and tears were formed In the ground. It is assumed that the damage of 1837 also occurred as a result of a landslide which was triggered by the earthquake. Aerial photographs didn't reveal the presence of landslides at the site of ancient Safed, probably because of extensive, post-earthquake reconstruction.
At Gush-Halav aerial photographs revealed a few landslides other than the old landslide. According to the village church guard, one of them moved in the winter of 1969 (at the same time the well-known landslide occurred at Moghar (Wiseman et al., 1970)). Part of the village of Lubya was founded on a steep slope which, under the appropriate conditions of saturated soil triggered by an earthquake, would be prone to movement. Indirect evidence for past occurrence of landslides comes from the following quotation taken from Vered and Striem (1976a): "I recognized no ruins there with the exception of those houses thrown down by the earthquake." This suggests that the damaging agent was of an extremely selective nature, and affected only particular areas, as in the cases of a landslide.
THE CAUSE AND MECHANISM OF LANDSLIDES
In all the places where landslides took place in 1837, the foundation materials were either clay, soft marl and chalk, or soil. In contrast to this, Sasa and Kafr Kanna, which are founded on hard limestone and dolomite, suffered little or no damage during the earthquake.
Safed, Ein Zeitun, Jish, and Reina which suffered extensive damage were founded on dip slopes consisting of soft bedrock and soil. The strata are inclined by at least 20¡. This combination of soft bedrock and steep dip slope is most likely to produce landslides.
It is suggested that the damaged section of Lubya was situated on the hill slope and founded on soil, whereas the undamaged section had a solid rock foundation. Where the foundation material is also saturated (as at the places mentioned above, where the presence of numerous springs indicates a shallow water table) the likelihood of landslides is increased.
The greater damage resulting from the 1837 earthquake when compared to that of 1927 may be due in part to the fact that the former took place during winter, when the ground was wet and the water table was high, whereas the earthquake of 1927 took place in summer when the water table is generally lower and the soil is dry.
Wiseman et al. (1970) show that saturation of unconsolidated material may increase the likelihood of landslides, such as occurred in Moghar where a landslide followed the heavy rains of January, 1969. Another landslide occurred at the same locality 40 years earlier. This was also preceded a period of heavy rainfall, as indicated by peak levels in the Sea of Galilee, which drains the East Galilee region. Arieh et al. (1977) show that even minor earthquakes (e.g. the September 2, 1973 event) may initiate landslides.
Low frequency vibrations, which predominate at a distance from the epicenter, are most likely to trigger landslides. The occurrence of landslides is, therefore, not necessarily indicative of strong shaking or of proximity to the epicenter.
DAMAGE RESULTING FROM POOR FOUNDATION MATERIAL
Tiberias and Sejera, which are situated on level ground, were both damaged in 1837 (Amiran, 1950; Vered and Striem, 1976a). Tiberias suffered damage for a second time in 1927. It is clear that landslides cannot account for the damage in either of the above cases. It is possible, however, that poor ground conditions combined with proximity to the epicenter could have been responsible for the damage. Nablus, for example, was only partially destroyed in 1837 and 1927. In 1837 "One quarter of the town was ruined, one quarter severely damaged,the rest slightly damaged" (quoted from Vered and Striem , 1976b). In 1927 "About 300 houses collapsed in the southern part of the city on the lower slopes of Mt. Gerizim and on the alluvial round between Mt. Gerizim and Mt. Ebal, . . ." (Amiran, 1951). An inspection of the site of Sejera showed that it had been founded on a thick (over 8 m at one location) layer of alluvium. A seismic refraction survey carried out during the course of this study near the ruins of the old walls of Tiberias indicated unconsolidated material (pwave velocity 1800 ft/sec.) which was saturated from 5 m down. This unconsolidated material reaches a total thickness of 15 m. Unconsolidated sedimentary deposits are known to intensify the vibration caused by an earthquake (Lajoie and Helley, 1975 and Borcherdt et al., 1975). It can therefore be concluded that the damage suffered by Tiberias and Sejera was, at least in part, due to the fact that they were founded on unconsolidated alluvium. Arieh et al. (1977) suggest that unconsolidated ground produced an exaggerated response during the earthquake of September 2, 1973.
DISCUSSION AND CONCLUSIONS
It has been shown that much of the severe damage suffered by certain settlements in the Galilee region as a result of the 1837 and 1927 earthquakes was directly caused by landslides and poor foundation material. In contrast to this, some settlements (in the same area) which were founded on solid rock were either spared or suffered minor damage. It is concluded that the degree of damage is not only a function of the intensity of the earthquake and the distance to the epicenter, but is also closely related to local geology and topography - i.e., type of foundation material, dip of strata, surface slope, degree of ground saturation and the interaction of these various factors.
Vered and Striem (1977) located the epicenter of the 1837 earthquake and estimated its magnitude on the basis of the degree of damage reported from various settlements. They did not take into account the local ground conditions. Close analysis of the type of damage to each locality as well as correlation between local conditions and the resulting damage may indicate that the epicenter was not necessarily where indicated by Vered and Striem Based on the results of this study it is believed that the intensity of the earthquake was probably less than previous estimates.
It is concluded that since the Galilee area is one of known earthquake risk, and since much of the damage caused may be the result of the poor siting of settlements, the compilation of seismic risk evaluation maps for this area would have very high priority Such a map of the Safed region is at present being prepared by the GSI and it is hoped that future damage can be prevented, or at least alleviated by taking into account in landuse planning the seismic risk and other aspects of the geology of any particular area
ACKNOWLEDGEMENTS We wish to thank Avi Olshina for editing the manuscript and Ms. Jill Wolf for typing it
REFERENCES Amiran, D.H.K, 1961 A revised earthquake catalogue of Palestine, Israel Exploration Jour., v. 1 (4), p. 223-246.
Arieh, E. 1967. Seismicity of Israel and adjacent areas. Geol. Survey of Israel Bull. no. 43.
Arieh, E., Peled, U., Kafri, U., and Shaal, B. 1977. The Jordan Valley earthquake of September 2, 1973. Israel Jour. of Earth Sci. v. 26, p. 112-118.
Borcherdt, R D., Joyner, W.B., Warrick, R.E., and Gibbs, J.F., 1975. Response of local geologic units to ground shaking, p. 52-67; in R.D. Borcherdt (ed.), Studies for Seismic Zonation of the San Francisco Bay Region, U.S. Geol. Survey Prof. Paper 941-A.
Lajoie, K. R. and Helley, E.J., 1975. Differentiation of sedimentary deposits for purposes of seismic zonation, p. 39-51, in R.D. Borcherdt (ed.), Studies for Seismic Zonation of the San Francisco Bay Region, U.SO Geol. Survey Prof. Paper 941-A.
Vered, M and Striem, H:L., 1976a. The Safed earthquake of 1/1/1837 and its implication on seismic risk evaluation in Israel. Israel Atomic Energy Commission, Licensing Div. IA-LD-1-105.
Vered, M. and Striem, H.L., 1976b. A macroseismic study of the July 11, 1927 earthquake. Israel Atomic Energy Commission Licensing Div. IA-LD1-107
Vered, M. and Striem, H. L., 1977. A macroseismic study and the implication of structural damage of two recent major earthquakes in the Jordan Rift, Bull. of the Seismological Soc. of America, v. G7, no. 6, p. 1607-1613.
Wiseman, G., Hayati, G., Frydman, S., Aisenstein, B., David, D., Flexer, A., 1970. A study of a landslide in the Galilee, Israel. Technion Israel Inst. of Technology, Faculty of Civil Engineering, Pub. no. 146.
MINISTRY OF ENERGY AND INFRASTRUCTURE Report HYDRO/5/78 - Jerusalem - June 1978
Damage Caused By Landslides During the Earthquakes of 1837 and 1927 in the Galilee Region By D. Wachs and D. Levitte