We investigate the distribution of focal depths for earthquakes that do not appear to be associated with zones of recent subduction, using both new results from analyses of individual events recorded at teleseismic distances and published data for both microearthquakes and larger events. The deepest events in oceanic regions occur in old lithosphere (≥100 Ma), and excluding earthquakes in active mountain belts, the deepest crustal events occur in old cratons (tectonic age ≥800 Ma). Therefore, the temperature at the source region is likely to be an important factor determining whether deformation occurs seismically or not. From estimates of the temperatures at depths of the deepest events, we conclude that those limiting temperatures are about 250°–450°C and 600°–800°C for crustal and mantle materials, respectively. In several regions of recent continental convergence, in addition to shallow crustal seismicity, there is seismic activity in the uppermost mantle. The lower crust, however, is essentially aseismic. We infer that both the upper crustal and the mantle seismic regions correspond to zones of relatively high strength and that they are separated by a zone of lower strength in the lower crust where aseismic, ductile deformation predominates. This simple interpretation is qualitatively in agreement with extrapolated values of brittle and ductile strengths of geologic materials studied under appropriate pressure and temperature conditions in the laboratory. A low‐strength zone in the lower crust might allow detachment of crystalline nappes from the underlying mantle (and lower crustal) lithosphere. The apparently greater strength of mantle materials than crustal materials at the same temperature implies that oceanic lithosphere is much stronger than continental lithosphere, and this difference may account for why plate tectonics works well in oceanic regions but not in continents.

We combined precise focal depths and fault plane solutions of more than 40 events from the 20 September 1999 Chi-Chi earthquake sequence with a synthesis of subsurface geology to show that the dominant structure for generating earthquakes in central Taiwan is a moderately dipping (20 degrees to 30 degrees ) thrust fault away from the deformation front. A second, subparallel seismic zone lies about 15 kilometers below the main thrust. These seismic zones differ from previous models, indicating that both the basal decollement and relic normal faults are aseismic.

We determined the fault plane solutions and focal depths of 17 earthquakes beneath the Shillong Plateau and the northern Indoburman ranges by combining results from the inversion of long‐period P and SH waveforms and amplitudes, from polarities of first motions, and from the identification of pP and sP phases on short‐period seismograms. Fault plane solutions of 15 earthquakes show mixtures of thrust and strike‐slip faulting, but the P axes for these events are nearly horizontal and consistently oriented north‐northeast–south‐southwest. All of these earthquakes occurred at depths greater than 29 km. Beneath the Shillong Plateau, one event occurred at a depth of 52 km. The relatively large depths for earthquakes in an intraplate setting suggest that the Indian lithosphere in this area is especially cold. Earthquakes beneath the northern Indoburman ranges define a gently east‐southeast dipping zone from 30 to 45 km beneath the Bengal basin to 40 to 90 km beneath the ranges. This zone seems to steepen and connect with the zone of intermediate depth seismicity that dips eastward beneath Burma. These earthquakes cannot have occurred along the interface between a subducting Indian plate and the overriding Indoburman lithosphere, because the P axes, not the nodal planes, are parallel to the north‐south trending seismic zone. Although a couple of the earthquakes might have occurred within the Indoburman lithosphere, most of this seismicity seems to have occurred within the Indian plate, recently and currently being subducted eastward beneath the Indoburman ranges. The consistent north‐northeast trend of the P axes implies that the orientation of maximum compressional strain in the Indian plate throughout its northeastern pan is nearly perpendicular to that responsible for roughly north‐south trending folds of the Indoburman ranges. Thus, either recently in geologic time (since 1 Ma) the orientation of maximum compression changed dramatically, or, more likely, the deformation in the Indoburman ranges is decoupled from that in the underlying Indian plate. Meanwhile, the seemingly identical northward displacement of India and the Indoburman ranges with respect to south China must be accommodated farther east, along the Sagaing and other faults.

Seismic moments for 12 major

We compare synthetic and recorded P wave forms to place constraints on the focal depths and fault plane solutions of 16 crustal earthquakes beneath the highest parts (>4000 m) of the Tibetan plateau. Fault plane solutions for all 16 events show combinations of normal and strike slip faulting with T axes oriented approximately east–west. None of these solutions show thrust faulting. Thus the data corroborate previous inferences that the active tectonics are dominated by east–west extension. Focal depths for all 16 events are less than 15 km and appear to be between 5 and 10 km. This style of deformation and these depths of faulting are similar to those in the Basin and Range province of the western United States. Two intermediate depth events below the crust of southern Tibet also show primarily normal faulting with east–west T axes. The solution for one, discussed by Chen et al. (1981), is unambiguous. The solution for the other, the event of August 1, 1973 (27.59°N, 89.17°E, 85±10 km, mb = 4.9) is less certain. Both apparently occurred in the mantle beneath a thick, aseismic lower crust, and their occurrence suggests that brittle deformation occurs there in response to a stress field similar to that operating at shallow depths beneath Tibet.