Showing papers on "Fault (geology) published in 1983"

Continental fault structure and the shallow earthquake source

Abstract: Plate boundaries in continental crust are generally less sharply defined than in the oceans, with seismicity spread over broad areas. Interplate displacements appear to be largely accommodated by networks of major fault zones. A simple 2-level model for these important structures accounts for the depth distribution of most continental earthquakes, and for the observed range of faulting styles and associated rock deformation textures. The model consists of a seismogenic frictional slip regime overlying quasi-plastic mylonite belts wherein shearing is largely accommodated aseismically, due mainly to the changing response of quartz to deformation with increasing temperature. Shear resistance increases with depth to a peak value in the vicinity of the frictiona1/quasi-plastic transition and then decreases rapidly. The depth to which microseismic activity extends appears inversely related to regional heat flow and can be satisfactorily modelled as the frictional/quasi-plastic transition for different geotherms using laboratory determined flow laws for quartz-bearing rocks. Larger earthquake ruptures ( M > 5.5) tend to nucleate near the base of the seismogenic regime in the region inferred to have the highest shear resistance and concentration of distortional strain energy. Consideration is also given to the depression of isotherms and seismic activity in regions of thrusting, and to the question of the downward continuation of major fault zones through the lithosphere. Decoupling of the upper crust on flat-lying shear zones may accompany higher-level dip-slip (and perhaps in some circumstances, strike-slip) faulting, being favoured by above average continental heat flow and a high quartz content in the middle or deep crust. The average level of deviatoric stress within the seismogenic regime remains an outstanding problem.

Nucleation and growth of strike slip faults in granite

Abstract: Fractures within granodiorite of the central Sierra Nevada, California, were studied to elucidate the mechanics of faulting in crystalline rocks, with emphasis on the nucleation of new fault surfaces and their subsequent propagation and growth. Within the study area the fractures form a single, subparallel array which strikes N50°–70°E and dips steeply to the south. Some of these fractures are identified as joints because displacements across the fracture surfaces exhibit dilation but no slip. The joints are filled with undeformed minerals, including epidote and chlorite. Other fractures are identified as small faults because they display left-lateral strike slip separations of up to 2 m. Slickensides, developed on fault surfaces, plunge 0°–20° to the east. The faults occur parallel to, and in the same outcrop with, the joints. The faults are filled with epidote, chlorite, and quartz, which exhibit textural evidence of shear deformation. These observations indicate that the strike slip faults nucleated on earlier formed, mineral-filled joints. Secondary, dilational fractures propagated from near the ends of some small faults contemporaneously with the left-lateral slip on the faults. These fractures trend 25°±10° from the fault planes, parallel to the direction of inferred local maximum compressive stress. The faults did not propagate into intact rock in their own planes as shear fractures. Rather, adjacent faults were linked together by secondary, dilational fractures. Extensive secondary fracturing between faults produced larger fault zones that accommodate 10–100 m of left-lateral slip. As deformation progressed, faulting evolved from relatively short, closely spaced faults to longer, more widely spaced fault zones.

The Sistan suture zone of eastern Iran

Abstract: A deformed accretionary prism and a flanking forearc basin extending from Birjand southeast to Zahedan, Iran, record the destruction of an arm of the Neo-Tethys during Senonian-Paleocene time and consequent collision of the Afghan and Lut eratonic blocks. The accretionary prism at 32 °N is subdivided into two northwest-trending en echelon belts termed the “Ratuk” and “Neh” complexes, respectively. On the east, the Ratuk complex is characterized by ophiolitic block-against-block or serpentinite-matrix melange and large fault slivers of epidote blueschist tectonite. The Ratuk complex was built prior to Maastrichtian time. The Neh complex to the southwest is Senonian to Eocene in age and includes, in addition to ophiolitic melange, weakly metamorphosed marine sedimentary rock exposed in extensive belts bounded by steeply dipping faults. The Sefidabeh forearc basin deposits onlap both the Neh and Ratuk complexes and the southwest margin of the Afghan block. They make up as much as 8 km of Cenomanian to Eocene terrigenous elastics and carbonates that display a complex but coherent stratigraphy. Facies relations demonstrate the uplift and subaerial exposure of the Ratuk structural high, followed by its subsidence contemporaneous with construction of the Neh complex and calc-alkalic volcanism on the northeast (inner) side of the basin. The accretionary prism-forearc basin polarity, the structural vergence and general younging of the accretionary prism to the southwest, as well as the position of the (relatively) high P T metamorphic rock on the inner side of the prism are consistent with northeast-dipping subduction. Widespread emergence of the entire belt and the initiation of folding of the Sefidabeh basin deposits during middle Eocene are interpreted to be consequences of the entry of the Lut block into the subduction zone. Continued convergence of the continental blocks is expressed by a regional system of folds and transcurrent faults corresponding to east-northeast compression. These structures are buried by mildly deformed Miocene volcanic rocks. Extensive post-Miocene right-slip faulting is inferred to be an effect of Miocene “terminal” collision of Arabia and Eurasia.

Nonlinear strain buildup and the earthquake cycle on the San Andreas Fault

Abstract: Two contrasting models of the earthquake deformation cycle on strike slip faults predict significant temporal declines in shear strain rate near the fault, accompanied by a progressive broadening of the zone of deformation adjacent to it. In the thin lithosphere model, transient deformation results from flow in the asthenosphere due to stress relaxation following faulting through most or all of the lithosphere. For an earth model with a thick elastic lithosphere (plate thickness » depth of seismic slip), transient motions are due to postearthquake aseisrnic slip below the coseismic fault plane. Data from the San Andreas fault indicate a long-term temporal decrease in strain rate that persists for at least 30 years and may extend through the entire earthquake cycle. Observations support a cycle-long rate decrease and a temporal spreading of the deformation profile only if movement cycles on the northern and southern locked sections of the fault are basically similar. If so, the usually lower strain rates and broader deformation zone currently observed on the southern San Andreas represent a later evolutionary stage of the northern locked section, where a great earthquake is a more recent occurrence. Although the data allow some extreme models to be discarded, no sufficiently strong constraints exist to decide between the thin and thick lithosphere models. Regardless of the appropriate model the geodetic observations themselves indicate that strain buildup is sufficiently nonlinear to cause significant departures from recurrence estimates based on linear strain accumulation and the time-predictable model.

Structural analysis and interpretation of the surface deformations of the El Asnam earthquake of October 10, 1980

Abstract: The El Asnam earthquake of October 10, 1980, represents the major seismotectonic event of these last decades in the West Mediterranean area. It has induced a large amount of surface breaks. This earthquake reveals the importance of compressional phenomena that characterize nowadays the tectonics of North Africa. The tectonic analysis of surface rupture shows the complexity of the deformation mechanism. The principal mechanism consists of the activation of a NW-SE trending thrust fault, accompanied with a left-lateral motion and with an intense deformation of the northwestern overthrusting block. An accurate description of these deformations shows that, for a large part, the breaks geometry depends closely on the substratum structure. Finally, the analysis of the dislocations on irrigation ducts just above the fault traces allowed a quantitative study of the deformation.

The evolution of Middle America and the Gulf of Mexico–Caribbean Sea region during Mesozoic time

Abstract: A plate-tectonic model for the evolution of Middle America and the Gulf of Mexico-Caribbean Sea region is presented. The model, which is based upon the existence of the Mojave-Sonora megashear, incorporates into the Triassic Pangea reconstruction three microplates between North and South America, thus avoiding the overlap of the Bullard fit. These plates are the Yaqui, bounded on the north by the Mojave-Sonora megashear; the east and west Maya plates, bounded on the north by the Mexican volcanic zone and on the south by a predecessor of the Motagua fault zone; and the Chortis plate (parts of Guatemala and Honduras). During Late Jurassic time, as North America split away from Europe, Africa, and South America, shear, with left-lateral sense of displacement, occurred along the transform faults that bounded the micro-plates. If ∼800 km of left-lateral displacement along the Mojave-Sonora megashear, ∼300 km along the Mexican volcanic belt, and ∼1,300 km along a proto-Motagua megashear are restored, and if Yucatan and Cuba are rotated to fit against northern South America, then (1) a curvilinear belt of late Paleozoic rocks that show lithologic as well as paleontologic similarities extends across the reconstruction and links outcrops in Texas, eastern Mexico, nuclear Central America, and Colombia; (2) a Mediterranean-like sea is delineated that was a precursor of most of the present Gulf of Mexico; (3) correlation is implied between the distinctive quartzose San Cayetano Formation of Cuba and the Caracas and Juan Griego Groups of Venezuela. Geometric constraints suggest that probably shear initially occurred along the Mexican volcanic zone near the end of the Middle Jurassic. Subsequently, probably about 160 m.y. ago, displacements that total ∼800 km began along the Mojave-Sonora megashear. Contemporaneously, Yucatan and fragments of pre-Cretaceous rocks that compose parts of central and western Cuba migrated northward toward their present positions. Rotation of Yucatan was facilitated by considerable displacement along the proto-Motagua zone and along a zone that is probably coincident with the modern Salina Cruz fault. Accumulation of widespread major salt units of Late Jurassic (Callovian to early Oxfordian) age in the Gulf Basin probably occurred contemporaneously with the arrival of these blocks at their present positions. Clastic units that interfinger with some of the youngest salt units and rim the Gulf of Mexico have not recorded major recognized translations since their accumulation. Clockwise rotation of South America and the Chortis plate occurred during Early Cretaceous time. This movement, which was manifested by subduction of Jurassic ocean floor against the previously rifted precursor of the island of Cuba and under parts of Hispaniola and Puerto Rico, is recorded by circum-Caribbean orogeny. Abrupt changes in the relative motions between North and South America during Late Cretaceous time may have resulted in extension and outpourings of basalt upon the Jurassic rocks of the ocean floor of the Venezuelan Basin. West of Beata Ridge, sea-floor spreading formed the Colombian Basin. Related subduction occurred as the Chortis plate (including part of Central America, the Nicaraguan Rise, and southeastern Cuba) was sutured against the Maya East plate along the present Motagua fault and Cayman Trench. Our model is constrained by published geologic data, the relative positions of North and South America from Atlantic sea-floor magnetic anomalies, and the requirement that the major transform faults be compatible with the poles of rotation for the appropriate relative motions between North and South America. Paleomagnetic data from Middle America are sparse but do not conflict with the predicted motions of some of the microplates, especially Chortis.

The 1979 Homestead Valley Earthquake Sequence, California: Control of Aftershocks and Postseismic Deformation

Abstract: The coseismic slip and geometry of the March 15, 1979, Homestead Valley, California, earthquake sequence are well constrained by precise horizontal and vertical geodetic observations and by data from a dense local seismic network. These observations indicate 0.52±0.10 m of right-lateral slip and 0.17±0.04 m of reverse slip on a buried vertical 6-km-long and 5-km-deep fault and yield a mean static stress drop of 7.2±1.3 MPa. The largest shock had MS = 5.6. Observations of the ground rupture revealed up to 0.1 m of right-lateral slip on two mapped faults that are subparallel to the modeled seismic slip plane. In the 1.9 years since the earthquakes, geodetic network displacements indicate that an additional 60±10 mm of postseismic creep took place. The rate of postseismic shear strain (0.53±0.13 μrad/yr) measured within a 30×30-km network centered on the principal events was anomalously high compared to its preearthquake value and the postseismic rate in the adjacent network. This transient cannot be explained by postseismic slip on the seismic fault but rather indicates that broadscale release of strain followed the earthquake sequence. We have calculated the postearthquake stress field caused by the modeled coseismic slip. We assume that failure is promoted when the sum of the shear stress plus 0.75 times the faultopening stress increases. Most aftershocks concentrate at points where the stresses are enhanced by 0.3 MPa (3 bars) or more; aftershocks are nearly absent where postearthquake stresses decrease by 0.3–0.5 MPa. Isolated off-fault clusters of aftershocks that locate at one fault length from the rupture plane are explainable by this hypothesis. We find that ground rupture and postseismic creep take place where near-surface stresses are calculated to increase within the preexisting fault zones. Two patches that extend 4 km from both ends of the seismic fault exhibited neither aftershocks nor measurable postseismic creep. The sensitivity of aftershocks and ground rupture to changes in stress that are less than 5% of the earthquake stress drop demonstrates that the region around the earthquakes was within a few percent of its failure threshold before the main shocks. The preearthquake stress field and the stress required for failure must also have been nearly uniform.

Earthquake frequency distribution and the mechanics of faulting

Abstract: The level of intraplate seismicity in Japan generally shows a positive correlation with the density of Quaternary faulting. In southwest Japan, where intraplate seismicity is concentrated on land, rates of seismic moment release ( M˙0) are similar when calculated from either the 400-year historical record of seismicity or geologically determined slip rates of Quaternary faults. A data set of 18 earthquakes with seismic moments (M0) ranging from ∼0.01 to 3×1027 dyn cm shows a relationship between rupture lengthl and M0(log M0 = 23.5+1.94 · log l). When seismic moment on each Quaternary fault is assumed to occur in discrete events every T = M0/ M˙0g years (where M0 is estimated for a rupture extended over the entire fault length, and M˙0g is proportional to the slip rate of each Quaternary fault), the moment frequency distribution of earthquakes (log N =A − B · log M0) predicted from the geologic record is virtually identical to that seen with the 400-year record of seismicity. In contrast, if it is assumed that earthquakes on each fault occur according to the Gutenberg-Richter relation, we obtain poor agreement with the observed seismicity. Thus, while regional seismicity satisfies the relation log N =A − B · log M0 (or equivalently, log N = a − b · log M, where M is magnitude), it appears that seismicity on individual faults does not. This further implies that the primary factor that leads to the magnitude frequency distribution in regional seismicity studies is the relative distribution of the slip rates and lengths of preexisting faults.

Average seismic source-parameter relations for mid-plate earthquakes

Abstract: Mid-plate earthquakes, which are defined as having their epicenters at least 500 km from plate margins, represent a small portion of the total global seismicity. However, because those occurring within continental interiors are characterized by anomalously large damage areas, they are of particular interest. Empirical mb , Ms , and M o data are used to develop scaling relations for mid-plate earthquakes. From them, it is found that the corner period, T o2, of their far-field displacement spectra varies as the one-fourth power of the seismic moment, which indicates that average stress drop increases as seismic moment increases. By use of Savage's (1972) rectangular, bilateral dislocation model, the derived spectra of mid-plate earthquakes yield estimates of fault rupture length and width, average fault displacement, and source rise time as a function of seismic moment. These relations show that large mid-plate earthquakes, such as the 1886 South Carolina and 1811 and 1812 New Madrid events, do not require large fault rupture length.

Melt origin of fault-generated pseudotachylytes demonstrated by textures

Abstract: A petrographic and electron-microprobe study of fault-generated pseudotachylytes from the Outer Hebrides Thrust Zone, Scotland, demonstrates that the textures have resulted from the primary crystallization of a clast-laden melt rather than the devitrification of a glass, or by crushing and cataclasis. The observations are consistent with these pseudotachylytes having formed by frictional fusion followed by rapid quenching—implying faulting at shallow crustal depths.

Collision, rotation, and the initiation of subduction in the evolution of Sulawesi, Indonesia

Abstract: The island of Sulawesi, Indonesia, has been shaped and deformed as a result of collision with the Sula platform, a sliver of continental material from the northern margin of Australia-New Guinea. The collision has resulted in rotation of the north volcanic arm of Sulawesi and the development of the accretionary wedge of the North Sulawesi trench. The North Sulawesi trench changes laterally from a zone of no active deformation in the eastern part to a wide accretionary wedge in the west. Early stages of thrusting produce a steep frontal slope (8°–16°), indicative of relatively high basal shear stress, whereas the more advanced (western) zone of thrusting produces a gentle (2°) slope, consistent with low basal shear stress. Reported paleomagnetic data suggest post late Eocene counter-clockwise rotation of the North Arm, and the offshore geophysics are explained by a pivot of the North Arm with respect to the Celebes basin about the eastern end of the arc. Convergence between the north Banda basin and Southeast Sulawesi is documented by the presence of the Tolo thrust. Its outcrop is strongly arcuate and its accretionary wedge varies in width from a minimum of a few kilometers at each end to a maximum of 30–40 km in the central part. The northern end transforms to the leftlateral Matano fault, with a reported offset of 20 km. The southern end of the thrust projects toward the deformed rocks of Buton, but the structural relations there are not clear. The Matano fault zone appears to connect westward with the Palu fault, which forms the western transform of the North Sulawesi trench. The Palu-Matano fault system acts as a trench-trench transform between the North Sulawesi trench and the Tolo thrust, and this system is described by the same rotation pole as that for the Sulawesi North Arm.

Interpretation of seismic reflection profiling data for the structure of the San Andreas fault zone

Abstract: A seismic reflection profile crossing the San Andreas fault zone in central California was conducted in 1978. Results are complicated by the extreme lateral heterogeneity and low velocities in the fault zone. Other evidence for severe lateral velocity change across the fault zone lies in hypocenter bias and nodal plane distortion for earthquakes on the fault. Conventional interpretation and processing methods for reflection data are hard-pressed in this situation. Using the inverse ray method of May and Covey (1981), with an initial model derived from a variety of data and the impedance contrasts inferred from the preserved amplitude stacked section, an iterative inversion process yields a velocity model which, while clearly nonunique, is consistent with the various lines of evidence on the fault zone structure .

COCORP profiling across the Southern Oklahoma aulacogen: Overthrusting of the Wichita Mountains and compression within the Anadarko Basin

Abstract: COCORP (Consortium for Continental Reflection Profiling) deep reflection profiles recorded across the Wichita Mountains and Anadarko Basin suggest that significant crustal shortening occurred in the final stages of the evolution of the Southern Oklahoma aulacogen. The crystalline rocks of the Wichita Mountains were thrust in Pennsylvanian time northeastward over sedimentary rocks of the Anadarko Basin along a series of faults with moderate (average 30° to 40°) and southwesterly dips. These faults can be traced possibly as deep as 20 to 24 km. Listric thrust faults and hanging-wall anticlines developed in the sedimentary rocks of the basin. These features contrast with conventional interpretations of Pennsylvanian structures as the result of predominantly vertical movements along high-angle faults, and they suggest that Pennsylvanian downwarping of the Anadarko Basin was at least partially due to thrust loading. Truncations of reflections from Cambrian-Ordovician rocks in the deepest part of the basin suggest normal faulting, which would support ideas of an early extensional stage in the aulacogen cycle. The distinctive Precambrian layering seen on earlier COCORP data recorded south of the Wichita Mountains cannot be recognized under the Anadarko Basin, and the Proterozoic basin containing that layering may have been bounded on its north side by a Precambrian fault. This inferred fault was probably twice reactivated during formation of the Southern Oklahoma aulacogen—once during late Precambrian(?)-Early Cambrian extension, and again during Pennsylvanian compression. The popular view that aulacogens originated from radial rifting of updomed, homogeneous continental crust is probably too simplified, and a more important constraint on their location and development may be the nature of pre-existing lines of weakness.

Seismicity and rates of relative motion on the plate boundaries of Western North America

Abstract: Summary. The consistency of earthquake data and plate tectonic models and other estimates of fault slip, may be tested by estimating rates of motion from the earthquakes using the concept of seismic moment. The contribution of individual events may simply be summed, but a generally better estimate of long-term average slip rate is obtained by integration over magnitudefrequency of occurrence relations. Estimates of fault motion rates associated with earthquakes are possible within about a factor of 2 using this approach if the major sources of uncertainty are given careful consideration; e.g. incompleteness and inaccuracies in the earthquake data; empirical moment-magnitude relations and the effect of their stochasticity; fault ‘widths’ or depth extents; the recurrence relations; maximum magnitudes and the form of truncation at the maximum magnitude. A formulation for the recommended magnitude density truncation is developed. Application of the latter method to the earthquake data of the offshore transform faults of the Juan de Fuca ridge system, the Queen Charlotte fault zone and the northern Vancouver Island area in each case gives good agreement with rates from plate tectonic models. For the southern San Andreas fault and Gulf of California area there is also good agreement with previous estimates obtained by summing the contributions of individual events. However, the displacement rate in the margin convergence zone of southern British Columbia, Washington and Oregon computed from the seismicity is at least a factor of 10 lower than from plate models and from other convergence estimates, and primarily aseismic slip is suggested. The fault motions as a function of time computed from the seismicity records have also been plotted and compared to the longterm average rates. The plots permit estimates of the minimum present accumulated elastic strain, and show if there are any temporal relations among earthquake displacements on different fault zones.

Earthquakes in the Orozco Transform Zone: Seismicity, source mechanisms, and tectonics

Abstract: As part of the Rivera Ocean Seismic Experiment, a network of ocean bottom seismometers and hydrophones was deployed in order to determine the seismic characteristics of the Orozco transform fault in the central eastern Pacific. We present hypocentral locations and source mechanisms for 70 earthquakes recorded by this network. All epicenters are within the transform region of the Orozco Fracture Zone and clearly delineate the active plate boundary. About half of the epicenters define a narrow line of activity parallel to the spreading direction and situated along a deep topographic trough that forms the northern boundary of the transform zone (region 1). Most focal depths for these events are very shallow, within 4 km of the seafloor; several well-determined focal depths, however, are as great as 7 km. No shallowing of seismic activity is observed as the rise-transform intersection is approached; to the contrary, the deepest events are within 10 km of the intersection. First motion polarities for most of the earthquakes in region 1 are compatible with right-lateral strike slip faulting along a nearly vertical plane, striking parallel to the spreading direction. Another zone of activity is observed in the central part of the transform (region 2). The apparent horizontal and vertical distribution of activity in this region is more scattered than in the first, and the first motion radiation patterns of these events do not appear to be compatible with any known fault mechanism. Pronounced lateral variations in crustal velocity structure are indicated for the transform region from refraction data and measurements of wave propagation directions. The effect of this lateral heterogeneity on hypocenters and fault plane solutions is evaluated by tracing rays through a three-dimensional velocity grid. While findings for events in region 1 are not significantly affected, in region 2, epicentral mislocations of up to 10 km and azimuthal deflections of up to 45° may result from assuming a laterally homogeneous velocity structure. When corrected for the effects of lateral heterogeneity, the epicenters and fault plane solutions for earthquakes in region 2 are compatible with predominantly normal faulting along a topographic trough trending NW–SE; the focal depths, however, are poorly constrained. These results suggest an en echelon spreading center or leaky transform regime in the central transform region.

Source times and scaling relations of large earthquakes

Abstract: Source time is a kinematic fault parameter corresponding to the duration of seismic source time functions and is accurately determined from phase spectra of long-period surface waves. Source times are determined for 36 large earthquakes during the last three decades. Scaling relations among source times, seismic moments, and fault dimensions are derived. Seismic moment is proportional to the cube of source time, and fault dimension is proportional to source time. Source times for low-angle thrust earthquakes along deep-sea trenches are found to be longer than those of other types, such as intraplate shocks and deep shocks. They are also significantly longer than the rupture times expected from a Haskell model, suggesting that, generally, there exists an introductory stage of faulting that precedes the main stage of the rupture propagation.

An aftershock study of the El Asnam (Algeria) earthquake of 1980 October 10

Abstract: Summary. An array of 28 portable seismic stations was operated in the region of El Asnam following the magnitude 7.3 (Ms) thrust earthquake of 1980 October 10. Locations of 494 events are presented in this paper and provide an indication of the overall form of the aftershock distribution. Tests to establish location accuracy (particularly depth) reduce this set to 277 events which, it is argued, are well constrained. P-waves alone are used in this study as a consequence of a debate about the reliability of reading S-phases. From the reduced set of 277 events, 81 events provide well-constrained focal mechanisms. The locations are presented in the form of maps and cross-sections, and discussed in relation to information already derived from field mapping of surface breaks and teleseismic studies of the waveforms of the main event. The zone of surface faulting (including secondary normal faulting) extended for 35 km but the aftershock distribution extends for twice this distance. Along the part of the fault which experienced substantial displacement in the main shock, the fault plane itself appears to be devoid of aftershocks, although many lie in the footwall beneath the fault. At junctions between segments of thrust faulting, strike-slip motion occurs. This is apparent in the aftershock focal mechanisms, and in the surface ruptures in one place. The large number of aftershocks in the north-east area appears to be due to the reactivation of a fan-like system of smaller reverse faults associated with surface folding. Activity at the south-west end is considerably less than that in the north-east, and is not obviously associated with recognizable geological or morphological features.

Complex distribution of large thrust and normal fault earthquakes in the Chilean subduction zone

Abstract: Summary. We present the results of a systematic study of events with M, > 6 in northern Chile (20-33"S), for the period between 1963 and 1971. Medium to large earthquakes near the coast of this region are of three types: (1) Interplate events at the interface between the downgoing slab and the overriding South American plate. These events can be very large reaching magnitudes greater than 8. (2) Intra-plate earthquakes 20-30 km inside the downgoing slab. They have fault mechanisms indicating extension along the dip of the slab and may have magnitudes up to 7.5. (3) Less frequent,M, - 6 events that occur near the top of the downgoing slab and have thrust mechanisms with an almost horizontal E-W compressional axis. This type of mechanism is very different from that of the events of type 1 which are due to shallow dipping reverse faulting. There is a rotation of about 30" of the compressional axis in the vertical plane between events of types (1) and (3). Three groups of events near 32.5", 25.5" and 21"s were studied in detail. Depth and mechanisms were redetermined by P- wave modelling and relative locations were obtained by a master event technique. Near 32.5"S, only events of types 1 and 2 were found in the time period of this study. At the two other sites, the three types of events were identified. This shows clearly that there are compressive stresses at the top of the slab and extension at the centre, a situation which is usually found in the areas where a double Benioff-zone has been identified in the seismicity.

Fault Type Predictions from Stress Distributions on Planetary Surfaces: Importance of Fault Initiation Depth

Abstract: The prediction of fault type on planetary surfaces from model stresses calculated at depth is discussed. These fault-type predictions yield different faults than those predicted using the surface criteria commonly employed in geophysical models. For elastic-plate flexure models of mascon loading on the moon, stresses calculated at the surface predict the occurrence of strike-slip faulting at the radial distance where grabens are found. Normal faults bounding lunar grabens and thrust faults responsible for wrinkle ridges are analyzed. It is found that the former initiate at the mechanical discontinuity that separates the breccia of the megaregolith from in situ fractured rock and that the latter initiate at the mechanical discontinuity between basalt layers and the underlying basin floor. The difference between elastic constants for the outer few kilometers of brecciated megaregolith and the underlying lunar lithosphere are evaluated. Superposing nonisotropic stresses resulting from the weight of overburden to the depth of the relevant mechanical discontinuity yield stresses that predict wrinkle ridges in the basin centers and grabens outside the basin margin, and eliminate the predicted zone of strike-slip faults.