Lg propagates efficiently with predominant periods of about 1 to 1 1/2 s across the stable regions of Asia: the Indian shield, the Eurasian platform, the Tarim basin, and (less efficiently) eastern China. This suggests that the structure is relatively homogeneous and Q must be high in the crust in these regions. For paths across the Tien Shan, where the crust is abnormally thick, Lg is observed with somewhat less sharp arrivals, but nevertheless, frequencies are high, and amplitudes large. For paths along the Tien Shan, signals are weaker, and arrivals are still less sharp. Lg was not observed for paths crossing the Tibetan Plateau. We infer that the change in crustal structure on the margins of the Tibetan Plateau and at the Tien Shan may disrupt the wave guide for Lg and may scatter energy out of the relatively homogeneous wave guide of the more stable crust. Alternatively or in addition, Lg may not propagate across the Tibetan Plateau because of an unusual velocity structure or because of high attenuation in the crust.

Relocations of earthquakes, recorded by a local network of stations in Afghanistan and Tadjikistan in 1966 and 1967, indicate a narrow seismic zone (width ≲30 km) dipping steeply into the mantle to a depth of 300 km beneath the Pamir and Hindu Kush ranges. Very low seismicity was observed at depths less than about 70 km, the approximate depth of the Moho. Clear gaps in activity exist also within the zone of intermediate depth seismicity. One gap, about 50 km wide near 37°N and at depths greater than 100 km, separates a steeply northward dipping zone to the southwest from a steeply southeastward dipping zone to the northeast. This gap probably marks either a tear in the downgoing slab or a gap between two oppositely dipping slabs. Fault plane solutions, determined by Soboleva for events between 1960 and 1967, generally show steeply plunging T axes approximately within the planar seismic zone. They therefore are grossly similar to those at island arcs where no deep earthquakes occur and presumably result from gravitational body forces acting on a relatively dense slab of lithosphere. At the same time there is a very large variation in the fault plane solutions, much larger than is common at island arcs. Appendix is available with entire article on microfiche. Order from the American Geophysical Union, 2000 Florida Ave., N.W., Washington, DC 20009. Document J80‐003; $1.00. Payment must accompany order.

No abstract

No abstract

The 214 m high Toktogul Dam will impound a 19.5‐km3 reservoir on the Naryn River in central Kirgizia. The reservoir lies along the Talas‐Fergana fault, a major tectonic feature of central Asia. The nature of Cenozoic movements along the fault is in debate; however, there is considerable evidence of present‐day right‐lateral movement. The region is one of moderate to high seismicity. A magnitude 7.6 earthquake occurred in 1946, approximately 65 km from the dam, near the intersection of the Talas‐Fergana fault and Chatkal ranges. Construction of the dam was completed in 1975 and by late 1979 the water level in the reservoir reached three fourths its final planned depth. Low‐magnitude, induced earthquakes began as the water level first exceeded 100 m in late 1977. A microearthquake survey using portable instruments was carried out in July‐August 1978 and a seven‐station telemetered network was installed in October 1978. The largest of the induced earthquakes has been of magnitude 2.5; the activity is in the immediate vicinity of the dam at very shallow depths (2–5 km) and is related to water level. Preliminary analysis of data gathered during 1978–1979 shows a zone of activity, extending to 5 km depth, supporting geological evidence for an active reverse fault beneath the dam and parallel to the canyon of the Naryn River. Where the reservoir crosses the Talas‐Fergana fault, 15 km upstream from the dam, the water level has not yet reached 100 m and seismicity along the Talas‐Fergana has not shown any significant change related to filling of the reservoir.