Kenneth M. Cruikshank
1992 Landers Earthquake

Kenneth M. Cruikshank
1992 Landers Earthquake

Fleming, R.W., Messerich, J.A., and Cruikshank, K.M., 1998. Fractures along a portion of the Emerson fault zone related to the 1992 Landers, California, earthquake: Evidence for the rotation of the Galway-Lake-Road block. Geological Society of America Map and Chart Series. MCH082.

FRACTURES ALONG A PORTION OF THE EMERSON FAULT ZONE RELATED TO THE 1992 LANDERS, CALIFORNIA, EARTHQUAKE: EVIDENCE FOR ROTATION OF THE GALWAY-LAKE-ROAD BLOCK

Robert W. Fleming
Geologic Hazards Team
U.S.Geological Survey, Denver, CO 80225

James A. Messerich
Geologic Photogrammetry Laboratory
U.S. Geological Survey, Denver, CO 80225

Kenneth M. Cruikshank
Department of Geology
Portland State University, Portland, OR 97201-0751

Table of Contents

Abstract

Part of the surface rupturing associated with the 1992-Landers, California, earthquake that occurred along the Emerson fault zone was mapped from stereo pairs of aerial photographs. The principal structure on this part of the fault zone is the Galway-Lake-Road rotated block. The Galway-Lake-Road rotated block is a 3 by 1 km oblate structure that is bounded on all sides by right-lateral, strike-slip fault motion; the circumscribed block rotated counterclockwise. Strike-slip displacement across the throughgoing part of the fault on the southwest side of the block was 5-6 m. Displacement on the northeast side of the block was about 1 m. Throughgoing displacement on the Emerson fault zone is therefore 4 to 5 m, and is consistent with reported displacement of 3 to 4 m farther northwest and 3.7 to 4.5 m farther southeast. The amounnt of rotation of the circumscribed block was equivalent to the displacement on the northeast side and was about 1 m.

Fault-related structures are a product of different levels of fracturing. Individual faults and tension cracks are the basic elements of shear zones. Interactions within and between shear zones produce structures of different scales. For the Landers-earthquake surface rupture, the largest regional-level structure is the Landers/Big Bear rotated block. The rupture belt is a nearly perfectly circular segment, about 80-km long, that activated along an arc of 60o with the radius centered near the San Andreas fault at San Bernardino, CA. Local-level structures are individual fault segments, stepovers between fault segments, rotated blocks, uplifted tectonic ridges, duplexes, and various distinctive fracture patterns associated with releasing and constraining bends. Study of each of these structural types illustrates the necessity of recognizing a structure along a fault before measuring offsets along faults that make up the structure. Detailed mapping is required to understand fracture kinematics and a larger, more general map view is required to understand boundary conditions for smaller structures.

MCH082-Cover.gif (154978 bytes) Oblique aerial view looking southeast showing the southerly part of the Galway-Lake-Road rotated block. The main rupture belt of the Emerson Fault zone extends from the lower left corner to the small white spot (playa) in the upper right of the view. The low hill in the middle of the view is at the southeast end of the block and was uplifted about 2 m. Part of Galway Lake Road is visible on the left extending to the dry Galway Lake bed in the upper left. (Photo courtesy of I. K. Curtis Services, Inc., Burbank, CA).

Introduction

Ground Rupture

The Landers earthquake, Ms 7.5, occurred on June 28, 1992, in a sparsely populated area between Palm Springs and Barstow in southern California. This earthquake was accompanied by more than 80 km of spectacular surface rupture that broke to the surface on parts of four previously known faults and on short stretches of several other previously known and unknown faults as well. This text accompanies a photogeologic map of part of the Emerson fault zone in the vicinity of the site where the Galway-Lake-Road crosses the Emerson fault zone (fig. 1 and inset map on photogeologic map); large strike-slip displacement has been reported at this location. The maximum right-lateral surface offset was reported to be in the range of about 5.5 to 7 m. At other locations, the typical offset was 3 to 4 m across any major rupture belt. The four principal faults that ruptured during the Landers earthquake are, from the southeast end, the Johnson Valley, the Homestead Valley, the Emerson, and the Camp Rock faults (see fig. 1 or index map to photogeologic map). The faults were all connected by stepover structures to produce a near continuous belt of surface rupture that is much longer than the mapped traces of any of the individual faults. The traces of the portions of the faults that did rupture to the ground surface are shown with a heavy line, and traces of faults that did not produce surface rupture are shown with a light line on figure 1.

mch082-fig-1.gif (8640 bytes)

Figure 1. Simplified map of known faults (light lines) and faults that produced surface rupture (dark lines) during the Landers earthquake. Also shown on photogeologic map as index to map area.Click on image for larger version.

Our previous research at Landers (Johnson and others, 1997) has illustrated that surface rupture and deformation, while complicated in initial appearance, are understandable in the context of patterns of fractures along a shear zone (Fleming and Johnson, 1989). Virtually all the surface rupture produced during the Landers earthquake was in the form of belts of fractures. Some fractures were part of a broad shear zone ranging from about 10 to 500 m wide along a principal fault zone. These broad shear zones also contained zones of concentrated shearing in more narrow bands of fractures. Other fractures were in belts peripheral to the principal fault zone.

In our previous research we recognized the importance of context of fracturing, but only in a limited sense. We recognized that the understanding of surface rupture begins with study of individual fractures, and that the orientation, position, and kinematic function of individual fractures are the building blocks of surface rupture patterns. Orientation and position place a fracture in an appropriate context for understanding the deformation that created the fracture.

Anomalous Displacement Measurement

There are two purposes in making the map. One purpose is to investigate the causes of the local, high offset measured along the Galway-Lake-Road stretch of the Emerson fault zone. The investigation consisted of a special photogeologic study of low- and moderate-altitude aerial photographs of this part of the Emerson fault zone. A photogeologic study of fracture patterns in the rupture belt could provide a larger spatial context to understand the surprisingly large offset. The map revealed the Galway-Lake-Road rotated block, which is about 3 km long and 1 km wide. Once the rotated block was recognized, the reason for the large reported offset could be understood.

Whereas measurements of strike-slip offset for a single rupture zone along the Johnson Valley, Homestead Valley and Emerson fault zones indicate generally 3 to 4 m of right-lateral shift (Sowers and others, 1994; Unruh and others, 1994; Johnson and others, 1997), measurements across the Emerson Fault zone at Galway-Lake-Road range from nearly [see note 1] 5 to 7 m of right-lateral shift (Sieh and others, 1993; Hart and others, 1993; Irvine and Hill, 1993; Hamilton, oral comm., 1996). The measurements at the road crossing are anomalously large relative to measurements elsewhere on the Emerson Fault zone. Displacement at the power-line road crossing of the Emerson fault zone, nearly 6 km to the northwest, was about 2.9 m. Displacement at the Tortoise Hill tectonic ridge along the Emerson fault zone (Fleming and Johnson, 1997), about 4 km to the northwest, was about 3.1 m. Displacements south of Galway-Lake-Road were reported to be 3.7 to 4.5 m (Hart and others, 1993).

Thus Galway Lake Road is in a region where overall displacement declines in the direction of fault propagation from perhaps 4 m in the south to 3 m in the north, and there is no obvious reason for the displacement to jump to 6 m locally.

Context of Structures Along a Fault

The second purpose in making the map is to further illustrate coactive, fault-related structures that formed hierarchically at least five different scales in a variety of settings along earthquake ruptures at Landers (e.g. Johnson and others, 1997).

At a regional level, the largest structure is the Landers-Big Bear rotated block (fig. 2). Block rotation is the result of strike-slip shift along a curved, not straight, rupture belt. The broad, arcuate Landers rupture, concave toward the west, followed parts of four previously mapped faults that were connected by right-stepping stepover structures to produce a nearly-continuous zone of right-lateral surface rupture. In the southern part of the main rupture belt, right-shift was noted on the Burnt Hill fault (fig. 2). In addition, coactive slip was noted on short segments of at least seven other previously recognized faults (Hart and others, 1993). Individual faults did not rupture along their entire lengths (fig. 1). The Johnson Valley and Emerson fault zones ruptured along about half of their mapped lengths. Two-thirds of the Camp Rock fault zone and nearly all of the Homestead Valley fault zone ruptured. Combined, the circular belt of surface rupture has a center of rotation along the San Andreas fault about 80-90 km to the west, near the city of San Bernardino. The Big Bear earthquake (fig. 2), which followed the Landers main shock by three hours, was located within the rotated block. The focal mechanism for the Big Bear earthquake was left-lateral and apparently conjugate to the right-lateral Landers main shock.

mch082-fig-2.gif (8330 bytes)

Figure 2. Rupture belt (thick lines) of the Landers earthquake of 28 June 1992. The four principal faults that ruptured form a broad arc of 60o that is centered on the San Andreas fault zone near the city of San Bernardino. The circular arc of surface rupture and the radius of the circular arc are each about 80 km in length. This is apparently the largest-scale structure activated during the earthquake, although coseismic surface rupture was reported on several other fault zones (Hart and others, 1993). The 1 x 3-km Galway-Lake-Road rotated block structure is within the north end of the rupture belt in an east-facing concave section of the Emerson fault zone.Click on image for larger version.

The second level of structures are the active parts of various fault zones that make up the Landers-Big Bear rupture zone, including the Emerson, Homestead Valley and Johnson Valley fault zones (fig. 1). These active parts of fault zones are composed of smaller, third-level structures that include releasing duplex structures and pull-apart basins that are present in stepovers between fault zones. They also include en echelon fault elements, tectonic ridges, twisted flower structures, and rotated blocks that occur within the fault zones (Johnson and others, 1997). Still-smaller structures at the fourth level include rotated blocks, belts of shear zones, releasing duplex structures and small basins in releasing bends, and thrusts and domes at restraining bends (Johnson and others, 1993, 1994, 1997; Fleming and Johnson, 1997). At the smallest scale, the fifth level, we mapped and described en echelon tension cracks and faults, complex and compound fractures, and shear zones (Johnson and others, 1993).

The mechanical behavior of rock should produce surface rupture that follows pre-existing faults as long as the orientation of the faults is compatible with the tectonic deformation that is relaxed by the earthquake. Instead, the Landers rupture zone utilized only those parts of the pre-existing faults that were connected by stepovers to form the broad curve of the Landers rupture. The stepovers contain distinctive internal patterns of surface rupture. The stepover between the Johnson Valley and Homestead Valley fault zones was a releasing duplex that perhaps contained a component of rotation, as described later. There, relatively long right-lateral, strike-slip faults were connected by shorter right-lateral faults oriented 30° to 45° clockwise to the longer faults (Johnson and others, 1997). The stepover between the Homestead Valley and Emerson fault zones was apparently a duplex structure (Zachariasen and Sieh, 1995). The stepover between the Emerson and Camp Rock fault zones was at least partly accomplished through a series of open tension cracks oriented about 60° clockwise to the trends of the two fault zones; the structure is a pull-apart basin (Fleming and others, 1997).

In the process of analyzing fault-related structures at Landers, we have learned that the full understanding of fault-related structures not only requires study of individual fault-related structures at the next lower level, but must also include study of structures at the next higher level. Detailed mapping is required to understand fracture kinematics and a larger, more general map view is required to understand boundary conditions. This is true whatever the level of an investigation.

Principal Conclusions

There are two principal conclusions resulting from this investigation. The first is the recognition of a structure, the Galway-Lake-Road rotated block, that is surrounded by fracture zones with right-lateral shift (fig. 3A). The circumscribed block rotated counterclockwise during slip on the Emerson fault zone. The second conclusion is that the anomalously large displacement that was measured where Galway Lake Road crosses the main rupture belt of the Emerson fault zone is a direct result of the rotation (fig. 3B). The displacement at the point of measurement at Galway Lake Road is made up of two parts: One part is the amount of throughgoing displacement on the main rupture belt of the Emerson fault zone. The second part is displacement due to rotation of the block. We will show that the rotational displacement is at least one meter and is part of the observed slip at the point of measurement of the anomalously large offset. Thus, throughgoing displacement on the Emerson fault zone at Galway Lake Road is the total slip minus the component due to rotation.

In following pages we will describe the fracture systems along the Emerson fault zone in the vicinity of the Galway Lake Road, and document the two principal conclusions.

MCH082-Fig-3.gif (5499 bytes)

Figure 3. Cartoon of deformation in the area of the Emerson fault zone and Galway Lake Road. A. Fault segments represent the mapped shear zones shown on the photogeologic map. B. Connected fault segments circumscribe the outline of the Galway-Lake-Road rotated block. (Adapted from a sketch by Arvid M. Johnson).Click on image for larger version.

How the map was made

Photography

Shortly after the Landers earthquake, two sets of aerial photographs were taken of the area where Galway Lake Road crosses the Emerson fault zone. Nearly the entire 80 km length of the rupture belt was photographed in black-and-white at an approximate scale of 1:6,000, by I. K. Curtis Services, Inc. [see Note 2] The width of coverage was about 1.5 km. In addition, I. K. Curtis Services, Inc., took a 4-km-long strip of color aerial photographs at a scale of 1:2,400, centered at the crossing of the Emerson fault zone by Galway Lake Road. The width of this coverage was about 0.6 km. Both sets of photographs were obtained in diapositive form for photogrammetric analysis. Several oblique aerial photographs, taken with view directions both to the north and south, provide spectacular images of the surface ruptures produced along the belt (fig. 4).

MCH082-Fig-4.gif (230342 bytes) Figure 4. View of main rupture belt of Emerson fault zone looking southeast. Galway Lake Road crosses the belt near an abrupt bend of 10o that restrains slip in the fault zone. Stanford Hill is the low hill about 1.5 km southeast of the place where Galway Lake Road crosses the main rupture belt. Beyond the hill, the fault zone continues to the small white playa at the base of the dark hills in the upper right part of the photograph. Peripheral fractures on the northeast side of the rotated block are not visible at this scale. (Photo courtesy of I. K. Curtis Services, Inc., Burbank, CA). Click on image for larger version.

Photogrammetry

The map was made entirely from aerial photographs using a computer-assisted Kern stereoplotter. Ground control for the photographs was established by us in the field. Control points were selected that were visible in photographs and could be located within a few centimeters on the ground; these were surveyed with a total station. These points provided optimum control of tilt, scale, and orientation of the photos and were located along the stretch of the 20 photos (19 stereo models). Photo control between the surveyed points was bridged by obtaining x-y-z coordinates for photo-identifiable points that were visible on the survey-controlled stereo models. The bridging of control was adjusted every few stereo pairs by several field-surveyed points. In this way the entire strip of photographs was adjusted to the set of surveyed ground control points. A UTM grid was established later using Global Positioning System measurements for several of the points; the UTM grid conforms to that shown on the Iron Ridge 7½-minute quadrangle.

The horizontal positions of surveyed control points are accurate to about 10 cm. Because of refraction errors in measuring vertical angles, the vertical control points are probably good only to about 30 cm. Relative positions of closely-spaced points on the photographs are much better known and accurate to a few centimeters, both horizontal and vertical.

Fractures were mapped from the photos, and data were stored digitally as CADMAP files. The map contains more than 12,000 individual fractures. A fracture was mapped as a line unless it had distinct characteristics of a scarp or a thrust fault. We strove to map only features that were definitely fractures and not trails, rills, or tracks. We decided to err on the conservative side and did not map visible linear features on the photos that were not definitely fractures.

Fractures interpreted from both the 1:2,400 and 1:6,000-scale photography are shown on the map. Topographic contour lines were mapped from 1:2,400-scale photographs; the area of detail essentially consists of a strip about 500-m wide that is centered along the main rupture belt. Fractures in areas without topographic contours were mapped from the 1:6,000-scale photographs. The part of the map made from 1:6,000-scale photos was used to extend the mapped area to the northeast. Because the 1:6,000-scale photography also covered the area of the 1:2,400-scale photography, a direct comparison of fracture visibility as a function of scale could be made. Five to ten times more fractures were visible on the larger-scale photography than could be seen in the same area on the 1:6,000-scale photography. [see note 3]

The number of fractures visible, even on the 1:2,400-scale photographs, is only a subset of those that could be directly seen in the field. For comparison, detailed maps of the southeast part of this map area were previously made in the field by Aydin and Du (1995) and Cruikshank and others (unpubl. map cited in Arrowsmith and Rhodes, 1994). They established control points with a total-station surveying instrument and fractures were mapped by point-reoccupation, in the same way as with plane-table methods. Aydin and Du (1995) and Arrowsmith and Rhodes (1994) show that the field mapping can not only see more fractures but also can determine the kinematics of each individual fracture.

Our photogeologic map does lack, in several key locations, detailed information on the fractures, and air photo coverage of an even larger area would help in determining the context or structural framework of the area of the map. Nevertheless, this map provides such a framework for more detailed mapping by others and has illustrated that the dominant structure along this stretch of the Emerson fault zone is the Galway-Lake-Road rotated block.

The fractures along the Emerson Fault zone

The kinematics of the fractures along the Emerson fault zone are interpreted from the sense of stepping of the fractures across narrow zones. Left-stepping fractures generally indicate right-lateral shear; right-stepping fractures generally indicate left-lateral shear. A fracture that contained a bulge or buckle parallel to its trace was mapped as a thrust fault. A fracture with significant vertical displacement, and which was not clearly compressional, was mapped as a scarp. Individual fractures may contain more than one sense of slip, and a scarp might be the product of either normal or reverse movement. Thus, the kinematics of individual fractures that have been mapped photogrammetrically cannot be read without some ambiguity. The methods of interpretation of deformation and kinematics from maps of fractures are described in Fleming and Johnson (1989) for strike-slip faults that bound landslide elements and by Johnson and others (1997) and Martosudarmo and others (1997) for surface rupture produced by the Landers and Loma Prieta earthquakes, respectively.

Main Belt of Fractures

The map presents part of the Emerson fault zone (fig. 1 and the 1:2,500-scale map) showing fractures in, and peripheral [see note 4] to a main rupture belt in a zone that is about 4 km long and 1.5 km wide. There is a main, throughgoing belt of shear zones that extends across the entire map and about 5 km farther to the southeast and 10 km farther to the northwest. Most of the displacement is carried in this main belt of shear zones. The trend of the main rupture belt contains curves, but is oriented, overall, at about N30oW. For descriptive purposes, three places are identified as principal points of reference on the main belt of fractures for the Galway-Lake-Road rotated block. Near the southern end of the rupture belt is Stanford Hill, a low hill (fig. 4) between 40 and 50 m high, that has been described and informally named by Aydin and Du (1995). About 1.5 km to the northwest of Stanford Hill, the main rupture belt crosses Galway Lake Road. Another 1.6 km to the northwest, the main rupture belt crosses Bessemer Mine Road (fig. 5).

MCH082-Fig-5.gif (170190 bytes)

Figure 5. View looking northerly along Emerson fault zone. The area seen in this view is about the same as shown on the photogeologic map of fractures. Stanford Hill, with its west-facing scarp, is in the lower right. Galway Lake Road crosses the rupture belt and trends across the photo in a line that is a little above the middle part of the view. Farther northeast is the Bessemer Mine Road, which crosses the main rupture belt a little below and to the right of the low hills in the upper left part of the view. The rupture belt continues to the northwest out of the view on the right side of these low hills. The Galway-Lake-Road rotated-block structure is on the right side of the main rupture belt. (Photo courtesy of I. K. Curtis Services, Inc., Burbank, CA). Click on image for larger version.

The main rupture belt is intensely broken by a large number of vertical fractures, the traces of which are mostly parallel or oriented 10-20° clockwise with respect to the trend of the belt (fig. 6). The intervening ground between visible fractures is also highly broken, and the rupture belt has a bulged or "mole-track" appearance that is evidently a result of dilation of material in the shear zone. Our experience in detailed mapping in the main belt of fractures (Johnson and others, 1993, 1997) is that all kinematic types of individual fractures may occur within the zone. Most common features are tension cracks and short, right-lateral fault segments that are oriented from a few degrees to nearly 45o clockwise to the trend of the fracture belt. However, there are also thrusts, normal faults, oblique-slip faults, and even left-lateral faults within the right-lateral fault zone.

MCH082-Fig-6.gif (106129 bytes)

Figure 6. Northwest-looking view of main rupture belt where it crosses the Galway Lake Road. Fault offset at the Galway Lake Road was the largest measured; values ranged from about 5 to 7 m. Main rupture belt is about 15 m wide and splinter fractures extend away from and make an acute angle with the main rupture belt. The kinematics of individual fractures within the belt vary widely, but overall, the belt is a right-lateral, strike-slip fault zone. (Photo by W. A. Bryant, used with permission, California Division of Mines and Geology).

The main rupture belt varies from a single fault, or very narrow shear zone, about 450 m northwest of Galway Lake Road, to a broad zone perhaps 25 m wide at Galway Lake Road. Typically, the belt of intense fracturing ranges from 10 to 15 m width. Associated with the main rupture belt are a few groups of fractures that appear to extend away from the main belt at a clockwise angle of about 30o. This type of fracture, which also carries right-lateral offset, was termed "splintering" by H. F. Reid (1910) and was noted by several different investigators in surface rupture produced in the 1906 San Francisco earthquake. Zones of splintering fractures are, for example, on the northeast side of the main rupture belt to the northwest of Galway Lake Road (3822910N and 540080E, UTM coordinates, see photogeologic map) and on the southwest side of the main rupture belt to the southeast of Galway Lake Road (3822425N and 540240E). Within the map area, splintering fractures appear to be associated with a small change in orientation of the main rupture belt.

Peripheral Fractures and Fracture Zones

Peripheral fractures and fracture zones are the indicators or keys to the geologic structures. In the absence of peripheral fractures, the main rupture belt would simply contain right-lateral, strike-slip movement. The peripheral fractures that are outside the main rupture belt indicate that more deformation occurred than simple strike-slip motion. These fractures, which typically occur in long narrow zones, are generally tension cracks or right or left-lateral fault zones but can have virtually any kinematic function. [see note 5]

The way that we interpret geologic structure from the fracture patterns is as follows: The expectable pattern for pure strike-slip faulting is a belt of fractures from a few meters to perhaps 500 m wide that is interpreted to be the main rupture belt (Johnson and others, 1994). Fractures in the ground surface that are outside of (peripheral to) this simple belt also indicate deformation. The deformation on the peripheral fractures produces recognizable geologic structures that are adjacent to but not part of the main rupture belt. Thus, the interpretation of a structure is deduced from interpretation of the kinematic function of the zones of peripheral fractures in the overall context of the strike-slip deformation. The function of the zones is a result of differential displacement first, as deduced from individual fractures, and then from groups of other fractures accomplishing the same deformation.

The vertical movement within and along the main rupture belt can be inferred from the shape of the contour lines where they cross the zone. Contour lines that cross the main rupture belt are inferred to have been smooth curves before the earthquake. Any disruption of a smooth curve after the earthquake is interpreted to be a result of differential vertical movement during the earthquake. For example, the 885 m contour that crosses the rupture belt about 900 m northwest of Galway Lake Road is apparently bulged over a width of nearly 200 m. This same contour line, about 300 m southeast of Galway Lake Road, is bulged in the same way over a width of perhaps 140 m. The maximum vertical change is less than one 5-m-contour interval and is probably in the range of 0.5-1 m. As the main rupture belt approaches Bessemer Mine Road from the southeast, the vertical movement changes from bulging to a shallow depression. [see note 6]

The Structures along the Emerson Fault zone

Two significant geologic structures on this part of the Emerson fault zone can be identified on the map. The first, simpler structure was produced at a small restraining bend. The second is the rotated-block structure.

Fractures at a Restraining Bend

There is a restraining bend in the main rupture belt about 125 m northwest of the intersection with Galway Lake Road. The rupture belt makes an easterly-concave bend of 10°. Peripheral fractures occur on both sides of the main rupture belt (fig. 7). On the east (or concave) side, there is a complex zone of fractures with two distinct orientations. One set of right-lateral fractures is subparallel to the main rupture belt. The other set makes an acute angle of about 30o with the right-lateral fractures and is roughly parallel to a group of splinter fractures that join the main belt another 100 m to the northwest. The sense of stepping identifies these fractures as a left-lateral, strike-slip fault zone. The two sets of peripheral fractures are a conjugate pair and signify north-south compression on the concave side of the restraining bend (fig. 7).

MCH082-Fig-7.gif (17444 bytes)

Figure 7. Map of fractures produced at a restraining bend in the main rupture belt at Galway Lake Road. The map was excerpted from the photogeologic map at the 10o restraining bend in the Emerson fault zone. The upper (northeast) side of the main rupture belt is concave, and peripheral fractures are conjugate right- and left-lateral shear zones. Here, the right-lateral shears are subparallel to the main rupture belt; the left-lateral shears make a clockwise angle of about 30o with the right-lateral shears. On the southwest (convex) side of the rupture belt there is a fan-shaped group of fractures of mostly right-lateral fault zones and a few, smaller left-lateral fault zones. Approximate directions of maximum compression are shown by opposing arrows. On opposite sides of the main rupture belt, these directions are different by 80° indicating the the fractures are a product of localized deformation. Click on image for larger version.

Interestingly, the left-lateral fractures in the conjugate pair and the previously described, right-lateral splinter fractures in the main rupture belt (Reid, 1910), separated by only 100 m on the concave side of the main rupture belt, have a different sense of slip but have approximately parallel orientations.

On the convex side of the restraining bend are three long zones and another two or three short zones of opening/right-lateral shear fractures (fig. 7) that extend to the edge of the map, 400 m west of the main rupture belt. These right-lateral fractures trend approximately normal to the main rupture belt. There are a few, weakly developed left-lateral fault zones that make an acute angle of about 50o with the right-lateral fracture zones. The left and right-lateral ruptures on this side of the main rupture belt also represent a conjugate set. The fracture pattern indicates that the direction of maximum compression is about east-west. Overall, the structure produced in the 10o restraining bend extends for more than 600 m away from the main rupture belt; lack of low-altitude photographs limits the detection of the extent of fractures on the northeast side and lack of any coverage limits the detection on the southwest side. Minimum extent of the area of peripheral fractures that are associated with the small, 10o restraining bend is 300 by 600 m. Maximum compression directions on opposite sides of the main rupture belt, as indicated by the conjugate fault zones, are oriented about 80° apart. Direction of maximum compression on the concave side of the restraining bend is about 20° clockwise (N10oW) from parallel to the main rupture belt (N30oW). On the convex side of the bend, the maximum compression direction (about E-W) is oriented about 25-30° clockwise from the normal (N60oE) to the main rupture belt. The large-angular difference in orientation of maximum compression on opposite sides of the main rupture belt is strong evidence that the fracturing is localized and a result of the small restraining bend in the main rupture belt. There is no indication here that the restraining bend caused any kind of pushup or ridge-like structure. Indeed, the structure formed is in a topographic low of the main rupture belt.

The two most significant observations about the fractures at this restraining bend are the lack of vertical uplift and the presence of fracturing on both sides of the main rupture belt. There is no question that localized compression is developed in an area surrounding a restraining bend along a strike-slip fault. However, the formation of tectonic ridges and other styles of vertical deformation in a restraining bend is not supported by the deformation at the bend near the Galway Lake Road.

The Galway-Lake-Road Rotated Block

In this section, we describe the fractures associated with the Galway-Lake-Road rotated block. We begin with the peripheral fracture zones that bound the northeast side of the block, continue with a description of fractures in the rotated block, and conclude with a brief description of fractures at the ends of the block near Bessemer Mine Road and Stanford Hill. More detailed descriptions of the structures at the ends of the block are in the Appendix.

Peripheral Fracture Zone along the Galway-Lake-Road Rotated Block

The peripheral fractures in the Galway-Lake-Road rotated block are contained in five or six narrow belts of right-lateral shear that are arranged in a curved, en echelon pattern (fig. 3). They are up to 1 km northeast of the main rupture belt. Each belt consists of individual left-stepping fractures. The belts of fractures also step left. Thus, the right-lateral sense of shift is evident in relations between individual fractures as well as between belts of fractures.

The right-lateral fracture zones that define the peripheral fractures curve into the main rupture belt at both ends. On the southeast end, the curving peripheral, right-lateral rupture ends near the base of Stanford Hill. The peripheral zone, beginning at the base of Stanford Hill, steps left to a 1.3-km-long, curved shear zone that offsets Galway Lake Road in a right-lateral sense by 1.05 m. [see note 7] The long, curved right-lateral zone contains some complex stepping fractures, and another 500-m-long, right-lateral fracture zone at the place where it is about parallel to the main rupture belt.

On the northwest end, the long, curved zone steps left to two more curved, right-lateral zones that trend back toward the main rupture belt. The projected intersection with the main rupture belt is about 500 m southeast of a complex knot of fractures at Bessemer Mine Road. These long, curved fractures terminate about 200 m from the main rupture belt.

Except for fractures associated with structures at the ends of the block near Bessemer Mine Road and at Stanford Hill, none of the other peripheral fractures shown on the map are part of the Galway-Lake-Road rotated block. To describe the fracturing in the rotated block, we first must identify and mentally discard peripheral fracture zones that are not part of the rotated block structure. [see note 8] We are left with the simplified belts of fractures shown in figure 3A.

The Fractures at the Ends of the Galway-Lake-Road Rotated Block

Fractures at the ends of the rotated block are complex, and probably a result of geometric incompatibility between the shapes of rigid blocks and the imposed deformation (Gabrielov and others, 1996). The northwest end of the block structure is a result of that incompatibility and apparently includes fractures along the main rupture belt from Bessemer Mine Road to the projected intersection point with the belt of peripheral fractures about 500 m southeast.

At the intersection of Bessemer Mine Road with the main rupture belt and continuing southeast, there are two superimposed deformation patterns. First, is the throughgoing, right-lateral deformation on the main rupture belt of the Emerson fault zone. The other pattern is one of opening or stretching in a direction of about N80oE. Stretching is indicated both by the vertical offset along the main rupture belt and by the swarms of tension cracks just east of the main rupture belt (fig. 8). Because the opening is apparently confined to the northeast side of the main rupture belt, the opening or stretching deformation moves material in the direction N80oE, away from the main rupture belt. Note that these fractures are all northwest of the projected intersection between the main rupture belt and the curving peripheral fracture zones at 3823540N and 539775E. Thus, they are on the outside of the rotated-block structure.

On the other, southeastern end of the rotated block at Stanford Hill, the fractures are within the bounding peripheral fractures. Thus, Stanford Hill is part of the rotated-block structure, and the hill should should be displaced toward the northeast and away from the main rupture belt. The observed deformation in the hill is vertical uplift of 2 m or more and complex groups of fractures that have ambiguous kinematics. The fracturing may be due to the presence of a ramp structure (double curve) in the main rupture belt at the hill. If so, fractures would be expected on both sides of the main rupture belt, and the only fractures on the southwest side are a localized response to the small restraining bend. The groups of fractures at both ends of the rotated-block structure are described more fully in the appendix.

The Galway-Lake-Road Rotated Block

The fractures outlining the Galway-Lake-Road rotated block define an oblate or football-shaped structure. The structure extends from a few hundred meters southeast of Bessemer Mine Road to the southeast end of Stanford Hill (fig. 8). The main rupture belt bounds the southwest side, and several belts of right-lateral, strike-slip (peripheral) faults bound the northeast side. Rotation is a required part of the deformation because all the bounding fractures contain the same right-lateral sense of strike-slip motion (fig. 3B).

Differential displacements associated with the Galway-Lake-Road rotated block

Measurement of relative displacement [see note 9] across the main rupture belt of the Emerson fault zone was made by several investigators (Sieh and others, 1993; Hart and others, 1993; Irvine and Hill, 1993; and Hamilton, oral comm., 1996) as well as photogrammetric measurements by us (fig. 8). We report the horizontal component of displacement in two ways (table 1), as a fault-parallel offset and as a fault-normal offset [see note 10]. A complete description of the displacement vector would show positive or negative dilation as well as vertical offset. The fault-parallel measurement is sensitive to the orientation of the offset feature where it crosses the rupture belt. If the offset feature trends normal to the rupture belt, the values are the same, but as the offset feature becomes more inclined to the rupture belt, the apparent horizontal, fault-parallel offset increases. Some investigators correct for the obliquity of the reference line to achieve a fault-normal offset, but dilation is customarily ignored.

MCH082-fig-8.gif (4795 bytes)

Figure 8. Map of the Galway-Lake-Road rotated block structure in which fractures that are not believed to be related to the structure have been removed. Remaining features that are shown are index contours (25 m), a few roads and trails, and the belts of fractures. Also shown with letters are locations of a displacement measurement, with fault-parallel offset shown in meters; data are in Table 1. The Galway-Lake-Road rotated block extends from a little southeast of Bessemer Mine Road to the southeast end of Stanford Hill; rotation of 0.11 degrees induced at least 1.05 m of counterclockwise displacement on the structure.

Table 1 - Relative displacement of selected points

Point Location(from Galway Rd) Normal Displacement (m) In Direction of Fault
A 1600 m NW 4.75 5.0
B 940 m NW 3.5 4.75
C 800 m NW 4.0 5.0
D At Galway Lk Rd 3.51 5.2
E 325 m SE 5-6.5 5.5-7.0
F 1080 m east 1.0 1.05
G 1100 m SE 4.7
H 1030 m SE 4.0
I 1000 m SE 4.85
J 1230 m SE
K 1320 m SE

Components of relative displacement at selected points are shown on figure 8 and in table 1. Points begin at the northwest end at Bessemer Mine Road, point A, and extend to the mid-point of Stanford Hill, point K. At Bessemer Mine Road, the component of fault-parallel displacement across a zone about 100 m wide was 5.0 m. The component of fault-parallel displacement was about the same along the main rupture belt all the way to Galway Lake Road, where we measured 5.2 m across a 75 m wide zone. Point E, 325 m south of Galway Lake Road, displayed 5.5 to 7.0 m of right-lateral shift on a motorcycle track across the rupture belt. The tracks on opposite sides of the rupture belt were not parallel and the minimum value of 5.5 m is probably closer to the actual fault-parallel component of displacement.

Points G, H, and I were taken from the map—compiled by Cruikshank and others (unpublished data, reported in Arrowsmith and Rhodes, 1994)—covering part of the rupture belt at Stanford Hill. The map lists only fault-parallel components of displacement and vertical offset. Fault-parallel components of displacement vary from 4.0 to 4.85 m. All the measured points at Stanford Hill, points G through K, contain significant vertical offset, with the hill uplifted relative to the southwest side of the rupture belt. Maximum differential vertical uplift was at point G, where 2.0 m was measured across one scarp and a total of 2.3 m on two scarps. Other vertical offsets obtained by adding values from across several smaller, overlapping scarps were typically 1.1 to 1.5 m.

In summary, the main rupture belt along the southwest side of the Galway-Lake-Road rotated block, between Bessemer Mine Road and Galway Lake Road, underwent about 5 m of right-lateral, fault-parallel, horizontal displacement. From Galway Lake Road to at least 325 m farther to the southeast along the same rupture belt, there was 5.2 to at least 5.5 m of right-lateral, fault-parallel shift. Still farther to the southeast, at the northwest end of Stanford Hill, the component of fault-parallel horizontal displacement is in the range of 4 to 5 m.

The maximum vertical shift reported by Cruikshank and others (unpubl. data reported in Arrowsmith and Rhodes, 1994) is 2 m, northeast-side-upthrown, on a single scarp at Stanford Hill. At the other end of the structure, near Bessemer Mine Road, the main rupture belt is down-dropped, a few tens of centimeters to perhaps a half meter in a graben. The faults that bound the graben are down-dropped along fractures within and on both sides of the belt.

Offset across the northeast side of the rotated block could be measured where Galway Lake Road crosses the peripheral fracture zones. The location, 1.08 km southeast of the intersection of the main rupture belt with Galway Lake Road (point F, fig. 8; table 1), is offset 1.05 m right-lateral in the direction of the fracture zone (1.0 m in the normal to the road).

The pattern of faults on the map defines the boundaries of the rotated block. The deformation at the ends of the block at Stanford Hill and near Bessemer Mine Road is complicated and at least partly a result of the oblate shape of the block (Gabrielov and others, 1996). At the area near Bessemer Mine Road, the structure contains opening or stretching toward the east. This direction, opening toward the east, is the expected direction and style of displacement on the outside of the block on the northwest end. If the area of the graben were inside the bounding peripheral faults and, hence, part of the rotated block, shortening features would be expected in this direction. Stanford Hill is on the inside of the peripheral belt of fractures and therefore, is part of the rotated block structure. In this case, rotation causes displacement of Stanford Hill toward the north and away from the main rupture belt. The remainder of the fault zones that bound the rotated block are the main rupture belt and the peripheral, right-lateral fault zones. Thus, the block is entirely bounded by features compatible with right-lateral deformation, which requires rotation within the block relative to materials outside the block.

The amount of rotation in the block can be estimated from the displacement data. The main rupture belt carries about 5.5 m of total displacement. Part of this total is throughgoing displacement along the Emerson fault zone; another part is due to the rotation of the circumscribed block. On the other (east) side of the block, displacement at point F was 1.05 m. This 1.05-m displacement is the amount of relative displacement of the block across the peripheral fault zone. If we assume that the same value of rotation applies to the side (southwest) of the block bounded by the main rupture belt, we conclude that the non-rotational part of the differential displacement is about 4.5 m, i.e. 5.5 m minus 1 m. The displacement due to rotation is 1 m [see note 11]. The amount of rotation can be expressed in terms of angular deformation. There was 1.05 m of rotational displacement on the perimeter of the 1 by 3 km block assuming a 500 m radius. The result is a maximum of about 0.11 degrees of counterclockwise rotation.

The value of 4.5 m for throughgoing displacement on the Emerson fault zone is perhaps a little larger, but in the same range as the values of differential horizontal offset measured along strike-slip faults throughout the Landers area. At a power-line crossing and at Tortoise Hill Ridge, 6 to 4 km to the northwest of the map area, the differential horizontal displacements were 2.9 to 3.1 m. Differential displacements on the Emerson fault zone south of Galway Lake Road are apparently 3.7 to 4.5 m (Hart and others, 1993).

Right-lateral offset of 4.5 m across the entire Galway-Lake-Road rotated block is smaller than the values of 5 to 7 m reported for the main rupture belt of the Emerson fault zone by Sieh and others (1993), Hart and others (1993), Irvine and Hill (1993), Hamilton, oral comm. (1996), and also measured by us. The key to the correct measurement of differential displacement is recognition of the rotated block structure. In order to obtain the net displacement across the Emerson fault zone, the displacement on the east side of the block must be subtracted from that on the west side. The counterclockwise rotation of the circumscribed block increases the displacement along the main rupture belt.

Discussion

Recognition of fault-related structures, such as rotated blocks in rupture belts produced by earthquakes, offers new information about how to measure differential displacement across faults. In studies of other areas of surface rupture (Johnson and Fleming, 1993; Johnson and others, 1993, 1994), belts of shear zones were recognized to distribute differential displacements across a zone from 10 to 500 m wide. Measurement and summation of shift across individual fractures cannot be expected to accurately summarize the deformation. In this paper, we have shown that measurements even across zones of fractures cannot be confidently summed without a knowledge of the kind of structure containing the fracture zones. In the case of the Galway-Lake-Road rotated block, the right-lateral deformation in the peripheral fault zones must be subtracted from the right-lateral deformation in the main rupture belt to obtain an approximation of net translational shift across a rupture zone. If displacement on the main rupture belt was added to that on the peripheral rupture belt that crosses Galway Lake Road 1.08 km to the southeast, on the opposite side of the rotated block, the estimated right-lateral shift of 6 to 6.5 m would be too large by 2 m. By making an appropriate interpretation of structure, we have seen that we obtain a reasonable value of about 4.5 m. If we were to sum the differential displacements on all the peripheral faults that cross Galway Lake Road, there would be five right-lateral faults and two left-lateral faults in addition to the main rupture belt. Differential displacement obtained by adding the offsets could reach perhaps 8 to 8.5 m. This is a ridiculous total according to measurements at other places along the Landers rupture.

Hidden within the overall complex pattern of rupture produced by the Landers earthquake are several other candidates to be recognized as rotated block structures. Another rotated block that is similar to the Galway Lake rotated block may exist about 25 km southeast of the map area. The location contains overlapping traces of the right-lateral Johnson Valley and Homestead Valley fault zones (fig. 1), which are subparallel and about 1.5 km apart. These faults strike northwesterly and were connected together by a group of stepover faults in a broad (1.5 km) zone called the Kickapoo stepover (Sowers and others, 1994). The more northerly fault of the stepover is called the Kickapoo fault zone. The southerly side of the stepover contains a poorly developed, right-lateral fault zone. Displacement on the fault zones is in the range of 2.5 to 3.0 m on the Johnson Valley, 1.5 to 2.0 m on the Homestead Valley, and 1.5 to 2.0 m on the Kickapoo fault. On the southerly bounding zone of the Kickapoo stepover, a discontinuous group of faults have up to 20 cm of offset. These faults are all right-lateral, strike slip and all are connected so that they bound an area nearly 3-km-long north-south by about 1.5-km-wide east-west. Thus, the same situation exists here as at the Galway Lake Road where a block is completely surrounded by right-lateral, strike-slip faulting. To carry the analogy a step further, there appears to be up to about 20 cm of displacement attributable to counterclockwise rotation in the Kickapoo block.

There are other similarities between the Kickapoo and Galway-Lake-Road rotated blocks. Near the junction between the Kickapoo fault and the Homestead Valley fault at the north end of the structure, Goatsucker Hill (Johnson and others, 1997) contains many of the same characteristics as Stanford Hill. The southeast end of the hill was uplifted about 0.3 m along the Homestead Valley fault. The Plio-Pleistocene beds, which here rest on crystalline igneous and metamorphic rocks, dip 15 to 25° away from the ridge much like the beds at Stanford Hill.

Another rotational structure is the Landers-Big Bear rotated block (fig. 2). Here, there might be a rupture belt along the San Andreas fault that is concave toward the east, with the center of rotation of the open, curved fault in the vicinity of San Bernardino (Jennings, 1975 and 1994). In this analogy, the arc of the Landers rupture belt corresponds to the zones of peripheral fractures of the Galway-Lake-Road rotated block. Rotation of the Landers-Big Bear block was a maximum of 5 m in 80 km or about 3.6 x 10-3 degrees. The difference here is that only the northwest side of the Landers-Big Bear block was displaced at the ground surface during the Landers earthquake.

Our observations suggest that the formation of a rotated block is due to the geometry of the main rupture. The Emerson fault zone contains an open, double curve along the southwest side of the rotated block, resulting in a distinctively shaped, main rupture belt. The shape of the main rupture belt is similar to a cross-section through a dinner plate or saucer. The cross-section of the saucer contains a straight section, a concave toward the east bend, another relatively straight section, another, more complicated concave bend toward the east, and another relatively straight section. From the northwest end of the map to about 3823500N and 539800E, the rupture belt is straight. From there to the southwest end at Stanford Hill, the rupture belt is concave toward the northeast. Fault-parallel displacement of the block along the main rupture belt would cause rotation of the material on the concave side of the saucer-shaped rupture.

Conclusions

We prepared this map 1) to learn how much detail on fracturing could be obtained with 1:2,400-scale aerial photographs, as compared to 1:6,000-scale photographs and field mapping and 2) to see whether we could understand the anomalously large displacements reported in the literature for the Emerson fault zone at the crossing of Galway Lake Road. Through serendipity we recognized the rotated block structure.

With respect to scales of photographs, we learned that 5 to10 times more fractures are visible on the 1:2,400-scale than on the 1:6,000-scale aerial photography. The number of fractures that are visible increases, approximately, in proportion to the ratio of the areas of maps at each of these scales covering the same general area of ground.

The photogrammetric mapping is not a substitute for detailed field mapping [see note 12], because the kinematic information obtained from individual fractures is important to understanding the rupture patterns. On the other hand, with the photogrammetric mapping at 1:6,000, we could recognize geologic structures in the earthquake ruptures that were not recognizable in detailed field maps because of limitations in the size of the mapped areas (Johnson and others, 1994). Clearly, in future earthquakes, it would be smart to combine photogeologic mapping with field mapping (Lajoie, oral communication, 1994).

Study of the setting of the intersection of Galway Lake Road with the main rupture of the Emerson fault zone indicates that the unusually large displacement along the main rupture belt was exaggerated because of a rotated block structure. The relative translational displacement for that part of the Emerson fault zone was augmented by a displacement of at least 1.05 m induced by rotation of the Galway-Lake-Road rotated block. Furthermore, recognition of the rotated-block structure reveals that amounts of surface displacement cannot be estimated from simple measurement of shift across shear zones and summing the individual amounts to get a total. In the case of the Galway-Lake-Road rotated block, the right-lateral shift on the peripheral zone is subtracted from the right-lateral shift on the other side of the block (the main rupture belt) to obtain a net, throughgoing displacement amount.

Perhaps the most important outcome of this research is the recognition of the significance of context of a fault-related feature being studied. Individual fractures are the guides to shear zones. The kinematics of individual fractures within a broad shear zone must be compatible with the overall shear in the zone. In this way, the shear zones provide a context for individual fractures within the zones. Groups of shear zones interact to produce structures such as duplexes, stepovers, tectonic ridges, and rifts. These structures also need to be placed in their proper context. For example, we noted that rifts formed in areas where two strike-slip faults, with the same sense of shear but different orientations, intersect. Likewise, a rotated block forms where there is a saucer-shaped excursion from a straight trace in the trend of the main rupture belt.

At Landers, we cannot continue to develop the hierarchy of structures that begins with individual fractures, extends to groups of fractures, and extends still further to structures of different sizes. Unfortunately, there is a lack of aerial photographic coverage and of detailed field mapping to provide the necessary structural context for structures larger than the Galway-Lake-Road rotated block. The next level of information that we have is at the regional-level scale of the Landers-Big Bear surface rupture, which is also a rotated block. Who knows how many other fault-related structures occur at intermediate sizes between that of the Galway-Lake-Road rotated block (on the order of 1 to 3 km in dimension) and the Landers-Big Bear rotated block (on the order of 30 to 50 km in dimension)? There seems to be no alternative to detailed and complete study of a deformed zone.

Acknowledgments—Support for Fleming was from the Nuclear Regulatory Commission, and support for Cruikshank was from the National Science Foundation. The mapping was done in the U.S. Geological Survey Laboratory for Geologic Photogrammetry, M.S. 913, Denver, CO 80225. This laboratory offers a capability for traditional analog mapping on stereo plotters as well as modern digital and analytical photogrammetry. This project would not have been possible without full support of the laboratory. Critical reviews were provided by Arvid Johnson, Purdue University, Mark Hudson, and Karl Kellogg, both of the U.S. Geological Survey. Their comments greatly improved this presentation.

References Cited

Antonellini, M., Arrowsmith, J. R., Aydin, A., Christiansen, P., Cooke, M., Cruikshank, K., Du, Y., and Wu, H., 1992, Complex surface rupture associated with the North Emerson Lake fault zone, caused by the 1992 Landers, CA earthquake: results of detailed mapping: EOS, v. 73, p. 362.

Arrowsmith, J. R., and Rhodes, D. D., 1994, Original forms and initial modifications of the Galway Lake Road scarp formed along the Emerson fault: Bulletin of the Seismological Society of America, v. 84, p. 511-527.

Aydin, A. and Du, Y., 1995, Surface rupture at a fault bend: the 28 June 1992 Landers, California, earthquake: Bulletin of the Seismological Society of America, v. 85, no. 1, p. 111-128.

Fleming, R. W., and Johnson, A. M., 1989, Structures associated with strike-slip faults that bound landslide elements: Engineering Geology, v. 27, p. 39-114.

Fleming, R. W., and Johnson, A. M., 1997, Growth of a tectonic ridge: Geology, v. 25, p. 323-326.

Fleming, R. W., Johnson, A. M., and Messerich, J. A., 1997, Growth of a tectonic ridge: U. S. Geological Survey Open-file Report 97-153, 94 p, 6 Plates.

Gabrielov, A., Keilis-Borok, V., and Jackson, D.D., 1996. Geometric incompatibility in a fault system: Proceedings of the National Academy of Science USA, v. 93, p. 3838-3842.

Hart, E. W., Bryant, W. A., and Treiman, 1993, Surface faulting associated with the June 1992 Landers earthquake, California: California Geology, v. 46, no. 1, p. 10-16.

Irvine, P. J. and Hill, R. L., 1993, Surface rupture along a portion of the Emerson fault: California Geology, v. 46, p. 23-26.

Johnson, A. M., and Fleming, R. W., 1993, Formation of left-lateral faults at Loma Prieta within Summit ridge shear zone: Journal of Geophysical Research, v. 98, p. 21,823-21,837.

Johnson, A. M., Fleming, R. W., and Cruikshank, K. M., 1994, Shear zones formed along long, straight traces of fault zones during the 28 June 1992 Landers, California, earthquake: Bulletin of the Seismological Society of America, v. 84, p. 499-510.

Johnson, A. M., Fleming, R. W., Cruikshank, K. M., Martosudarmo. S. Y., Johnson, N. A., Johnson, K. M., and Wei, W., 1997, Analecta of structures formed during the 28 June 1992 Landers-Big Bear, California earthquake sequence: U. S. Geological Survey Open-file Report 97-94, 58 p, 7 plates.

Martosudarmo, S., Johnson, A. M., and Fleming, R. W., 1997, Ground fracturing at the southern end of Summit Ridge caused by October 17, 1989 Loma Prieta, California, earthquake sequence: U. S. Geological Survey Open-file Report 97-129, 24 p., 5 Plates.

Reid, H. F., 1910, Report of the State Earthquake Investigation Commission, II—The mechanics of the earthquake: Carnegie Institution of Washington, Washington, D. C., 192 p.

Sieh, K. E., and 19 others, 1993, Near-field investigations of the Landers earthquake sequence, April to July, 1992: Science, v. 260, p. 171-176.

Sowers, J. M., Unruh, J. R., Lettis, W. R., and Rubin, T. D., 1994, Relationship of the Kickapoo fault to the Johnson Valley and Homestead Valley faults, San Bernardino County, California: Bulletin of the Seismological Society of America, v. 84, p. 528-536.

Terres, R. R., and Sylvester, A. G., 1981, Kinematic analysis of rotated fractures and blocks in simple shear: Bulletin of the Seismological Society of America, v. 71, p. 1593-1605.

Unruh, J.R., Lettis, W.R., and Sowers, J.M., 1994, Kinematic interpretation of the 1992 Landers earthquake. Bulletin of the Seismological Society of America, v. 84, p. 537–546.

Zachariasen, J., and Sieh, K., 1995, The transfer of slip between two en-echelon strike-slip faults—A case study from the 1992 Landers earthquake, southern California: Journal of Geophysical Research, v. 100, p. 15,281-15,301.

APPENDIX
Descriptions of Fractures at the Ends of the Rotated Block

There are distinctive fracture patterns, representing different kinds of deformation, at each end of the rotated block. The mapping from the aerial photographs does not provide enough detail to fully understand the structures at either end, but the observations do constrain how they formed. In the two sections in this appendix, we describe the deformation patterns at the ends of the rotated block.

Bessemer Mine Road End of the Galway-Lake-Road Rotated Block

The northwest end of the rotated block contains a zone of complex fracturing along and east of the main rupture belt, extending southeast from about Bessemer Mine Road for about 500 m. There is no bend here in the trace of the main rupture belt, but the belt widens to about 100 m from the more typical width of 10-15 m. What may appear to be a restraining step in the main rupture belt at Bessemer Mine Road is only a widening of the belt. The widening and unusual fracture pattern may be related to the junction to the southeast with the arcuate peripheral fractures of the rotated block.

Within the knot of fractures at Bessemer Mine Road, right-lateral faults are generally parallel to the trend of the main rupture belt, and large open tension cracks are oriented about 35-40o clockwise to the right-lateral faults. The structure is, in part, a graben in which the main rupture belt is down-dropped up to 0.5 m between bounding faults. The structure in the road at Bessemer Mine Road is 300 to 500 m northwest of the projected intersection of the belt of peripheral fractures of the rotated block with the main rupture belt. Thus, the structure is outside the rotated block and represents opening that is oriented at about N80oE (fig. 8).

The right-lateral, strike-slip peripheral fractures at the north end of the rotated block appear to end at about 3823600N and 539975E. Two additional short, narrow zones of right-lateral faults that occur about 300 m north of this end of the more or less continuous fracturing are 10 to 20o clockwise (N50oW) from the peripheral fractures in the north end of the block. These two fault zones, centered on 3823810N and 539865E and 3823890N and 539840E, apparently define the width of peripheral fractures in the north end of the rotated block. If so, the deformed zone at the end of the block is about 200 m wide.

Between the two belts of right-lateral fractures are five belts of tension fractures. Within this 200-m wide zone and near the main rupture belt, parallel fractures that are 100 to 200 m long are oriented about N10oW (i.e. 20° clockwise from the main rupture belt). The belts of fractures do not show any consistent pattern of stepping that would indicate strike-slip deformation; rather, they appear to be tension cracks that subsequently shifted to produce a down-to-the-west offset. They make an acute angle of 40-50° with the peripheral, curving right-lateral fault zones of the rotated block described above. The area between these five narrow zones of fractures and the main rupture is down-dropped. The down-dropping can be inferred from the vertical offset across the belt in three of the zones, and from the deflection of contour lines at the other two zones. The amount of down-dropping measured across a few of the fracture zones varied from a few centimeters to about a half meter.

The spacing between these five zones of tension/down to the west fractures increases with increasing distance from the main rupture belt. Proceeding outward from the main rupture belt, the spacing increases from 25 m between the first two zones to about 35 m between the next two, and about 75 m between the outer three zones. The easternmost zone of tension cracks is oriented about 10o counterclockwise with respect to the westernmost band of tension cracks.

There is another swarm of tension fractures on the other (southwest) side of the main rupture belt (centered about at 3823625N and 539650E) that has the same orientation as the ones on the northeast side. These fractures could be a coeval extension or continuation of the tension cracks on the northeast side of the main rupture belt, although they lack any evidence of vertical movement, and are not separated into distinct zones, as are their potential counterparts across the main rupture belt.

This zone of tension/down to the west fractures accommodates opening in a N70-80oE direction. Insofar as the direction of opening is toward the east, the outside of the rotated block at the northwest end moved away from the main rupture belt. This position and sense of movement indicated by the tension cracks is the expected deformation in that location. The directions of shift on the outside of the rotated block is, from the northwest end, extension at N70-80oE. Proceeding further around the perimeter of the rotated block, the strike-slip faults change orientation. Materials on the outside the block are displaced by strike-slip faults oriented S60oE curving to S30oE. The eastern side of the block at about mid-width contains right-lateral faults oriented between S20oE and S5oE. The part that trends toward Stanford Hill is S5oW.

The northwest end of the main belt of rupture at Bessemer Mine Road, where it widens to about 100 m, contains bounding faults that are right-lateral with a significant normal component. The easternmost fault is down to the west about 0.4 m at 3824055N and 539538E, and the westernmost fault is down to the east about 0.3 m at 3824000N and 539460E. Within the zone, fracturing is a complex mixture of left-lateral and right-lateral faults and tension cracks. The fractures that are generally parallel to the trace of the main rupture are tension cracks, and those oriented clockwise from the trace of the rupture belt are right-lateral, strike-slip faults. The left-lateral, strike-slip faults are oriented counterclockwise from the trend of the main rupture belt. Both the deflection of the topographic contour lines and the vertical shift on the fractures indicate that the belt was down-dropped during the fault movement; due principally to opening in a N70-80oE direction.

All these kinematic features are expected in a general way by the models of intersecting faults described by Gabrielov and others (1996). They show that, where fault blocks are rigid and are bounded by faults with the same sense of slip (all right-lateral or all left-lateral), a geometric incompatibility is necessary. In this case, the geometric incompatibility leads to a divergence of the block corners. Perhaps the knot of fractures at Bessemer Mine Road and the depression near the main break are results of the incompatibility.

Stanford Hill End of Galway-Lake-Road Rotated Block

At Stanford Hill, on the southeast end of the Galway-Lake-Road rotated block, there is a south to southwest-facing scarp as much as 2-m-high along the entire uplifted hill (fig. 4 and jacket cover for map, see note 13). Plio-Pleistocene lake beds dip about 30° to the north in the north-northeast end of the hill and 64° to the east in the southeast (Aydin and Du, 1995). The increase of dip to the southeast indicates progressively increased tilting as the block was translated along the main rupture belt in the Landers and in prior fault-movement episodes. As material was translated to the southeast from the southeast end of the map during earlier faulting events, it was moved beyond the rotated block structure to a relatively straight part of the main rupture belt. Beyond the map area to the southeast, there are deeply dissected, tilted beds that perhaps represent the displaced parts of the rotated block structure that formed during earlier episodes of fault movement.

The fractures on the other, southwest side of the main rupture belt appear to be the result of a restraining bend at Stanford Hill (bend at 3821620N and 540885E). These fractures are similar to ones that were described at the convex side of the restraining bend on the main rupture belt just northwest of its intersection with Galway Lake Road.

With photogrammetric mapping we can identify two additional groups of fractures in the Stanford Hill. The first is a swarm of tension cracks that strike approximately north on the northwest end of the hill (3821820N and 540860E). The tension cracks trend directly toward a group of thrust faults that strike easterly along the base of Stanford Hill. We have seen and described similar structures at other locations, most notably the Happy Trail shear zone on the Johnson Valley fault (Johnson and others, 1993, 1994) and Summit Ridge in the San Andreas fault zone (Johnson and Fleming, 1993). The fractures form as tension cracks oriented at a clockwise angle to the main rupture trend. The cracks divide the ground into prismatic blocks that are rotated by further right-lateral shearing. The rotation produces a small component of left-lateral shift on the tension cracks and shortening parallel to their lengths (Terres and Sylvester, 1981). Some of these are shown in Arrowsmith and Rhodes (their fig. 5, 1994) and Aydin and Du (their fig. 6b, 1994).

The other distinctive fracture group in Stanford Hill consists of two north-side-up, curving, bedding-plane fracture zones. One fracture group was mapped as a thrust fault and the other as a scarp. The more northerly of the two is about 75 m north of the label "Stanford Hill". On the west side, the fracture zone begins (3821250N, 540900E) near the north-trending tension cracks described above. Here, the zone looks like it contains a component of right-lateral shift. About 75 m from the west end, a thrust-like feature occurs parallel to the fracture zone, and the north side is up relative to the south side. Overall this fracture zone is nearly 300 m long. The second fracture zone is parallel to the first and occurs about 75 m south of the label "Stanford Hill". The zone was mapped largely as a scarp with the north-side up; however, this scarp may also be a reverse fault similar to the other one that is 150 m to the north. On the west side of the zone, the scarp is replaced by a group of short, curving fractures and a few short, thrust-like features. On the east side of the zone, the scarp occurs together with a sinuous thrust-like structure that curves toward the south and ends. Overall, this group of fractures is about 200 m long. The kinematics of these two fracture zones is ambiguous. Both the thrust fault of the first fracture zone and the scarp of the second zone could represent reverse, bedding-plane faulting. The other fractures associated with the two zones locally indicate both left and right-lateral shift.

The fracturing that bounds or intersects these two fracture zones likewise does not reveal the deformation within Stanford Hill. A group of fractures on the west side of the two fracture zones is a 100-m-wide group of discontinuous fractures oriented about N20°E that are predominantly left-lateral faults. On the east side of the two fracture zones are generally north-south ruptures consisting of thrust faults and short strike-slip fault zones, some showing left-lateral and some showing right-lateral offset. In general, the left-lateral faults strike more easterly than the right-lateral faults, but there are places where both left- and right-lateral faults have the same orientation. The main rupture belt trends about N40°W to the northwest of Stanford Hill; at the hill, the main rupture belt makes a 20° bend to a strike of N60°W. The main rupture belt maintains this orientation for about 300 m, and reverts to a trend of N40°W near the southeast edge of the map. The curving bedding-plane fracture zones are sub-parallel to the ramp-like section of the main rupture belt [see note 14].

If the fracturing in Stanford Hill were solely the result of material being translated across a ramp, there should be similar fracturing and tilting of beds on the southwest side of the main rupture belt. Because there are not similar fractures on the southwest side, these fractures in Stanford Hill are probably due to movement of the rotated block.

Endnotes

Use "back" button to return to text

Note 1. The range from 5 to 7 m offset at the crossing of the Emerson fault zone by Galway Lake Road is primarily a result of ambiguity caused by the irregularities of reference lines along Galway Lake Road and by curvature of the Emerson fault zone near Galway Lake Road. Differences in estimates of the angle between the trace of the rupture belt and the direction of the road could account for the entire range of reported offset.

Note 2. Use of trade or company names does not imply endorsement by the U.S. Geological Survey.

Note 3. For maps that represent 1 square km of ground, for example, the areas of the map sheets are 0.174 m2 at 1:2,400 and 0.028 m2 at 1:6,000, or a ratio of about 6.

Note 4. The terms main rupture belt and peripheral fractures are used in a special way here. In general, the fractures in the main belt are all contained within the broad shear zone that carries the bulk of the displacement along the Emerson fault zone. All other fractures are peripheral to the main belt. The main belt of fractures is basically a right-lateral, strike-slip shear zone. The peripheral fractures that are associated with the main rupture belt indicate a departure from a simple throughgoing fault zone and show such features as stepovers between faults, results of curves or bends in the main rupture belt and, as in the case described here, a rotated block.

Note 5. Curiously, where the peripheral fractures approach the main belt of fractures, they cannot be traced to intersect with it. This aspect of non-intersection of the fracture traces was noted throughout the Landers earthquake rupture belt, by us as well as others (Aydin and Du, 1995). Most likely, the segmented aspect of these fractures results from the fact that the propagation direction of the rupture was upward and not parallel to the rupture traces. During the Landers earthquake, the fault rupture propagated from southeast to northwest. At seismogenic depths, continuous, throughgoing fractures would be expected in the direction of fault rupture propagation. As the fault rupture approached the ground surface, the velocity was retarded and propagation direction of the rupture contained a vertical component.

Note 6. A reviewer has correctly pointed out that the interpretation of deformed contour lines assumes that landscape processes (erosion-deposition) alter topography on a short time scale relative to rupture frequency. In support of our assumption, we note that the vertical changes indicated by the deflected contours are consistent with the appearance of the main rupture belt. The belt along Galway Lake Road is elevated relative to ground on both sides and has the appearance of a wide "mole track" (fig. 6). Furthermore, thrust faults occur on one or both sides of the main rupture belt as if the material was squeezed out. Near Bessemer Mine Road, low scarps on both sides of the main rupture belt indicate that the ground was downdropped relative to ground outside the belt. For about 300 m southeast of Bessemer Mine Road, the contours show a small depression in the area where down-dropping occurred. Thus, observations about the styles of deformation in the main rupture belt supports the suggestion that the changes in shape of the contour lines is a result of vertical deformation.

Note 7. About 150 m farther east along the Galway Lake Road is a left-lateral zone that appears to be conjugate to the right-lateral zone. The trace of the road was too irregular here to measure the displacement on the left-lateral fault, but it was less than 0.5 m.

Note 8. The groups of peripheral fractures that are not part of the rotated block structure are as follows: On the extreme northwest end of the map on the northeast side of the main rupture belt is a group of north-south-trending, right-lateral fault zones that extend to the edge of the map, and probably beyond. These fractures are apparently part of a structure that is well-developed farther northwest. They are part of a pull-apart basin between the Camp Rock fault zone and the Emerson fault zone (fig. 1). Such basin-connecting stepover fractures occur for another 5 km to the northwest adjacent to the main rupture belt of the Emerson fault zone.

There is another, poorly-organized group of fractures on the southwest side of the main rupture belt just across from the stepover fractures (between 3824210N to 3824450N and 539060E to 539300E, UTM coordinates, see map). These fractures occur in exposed crystalline bedrock, and their significance is unknown.

Another poorly-developed zone of fractures extends from Bessemer Mine Road nearly to Galway Lake Road on the southwest side of the main rupture belt. These fractures are subparallel to the main rupture belt and are 400 to 500 m southwest of it. These fractures are not part of the main rupture belt, nor are they part of the Galway-Lake-Road rotated block. The fractures curve toward the main rupture belt on both ends, and all appear to be either tension cracks or right-lateral fault zones.

Finally, the fractures at a small restraining bend on Stanford Hill (Aydin and Du, 1994) are not thought to be part of the rotated block. Most, if not all, of the fractures on the southwest side of the main rupture belt at Stanford Hill are due to the bend (3821625N and 540800E). These fractures on the southwest side of the main rupture belt are tension cracks/right lateral faults. Similar to the faults on the convex side of the bend at Galway Lake Road, these fractures appear to be the product of fault-normal (about east-west) compression. The deformation on the northeast side of the rupture belt at Stanford Hill is very complex; there are several kinds of deformation superimposed on each other. To understand the deformation, the fractures caused by the presence of the restraining bend need to be isolated and subtracted from the fractures due to other causes.

It is tempting to ascribe much of the fracturing on Stanford Hill to the ramp (double bend) in the bounding main rupture belt. However, if the ramp structure caused the deformation in Stanford Hill, the same kind of deformation would be expected on the other (southwest) side of the ramp. Comparable fractures are not there however, and the only fractures on the southwest side of the main rupture belt at Stanford Hill appear to be related to the small restraining bend of the main rupture belt. Thus, by a process of elimination, it appears that much of the fracturing in Stanford Hill is associated with its position at the southeast end of the rotated block.

Note 9. Relative displacement is a site measurement made across a fault, fault zone or rupture belt that shows orientation and magnitude of separation of one side relative to the other.

Note 10. There are three components to a displacement vector that are needed to specify the complete relative displacement across a fault or fault zone. One component is the vertical offset; difference measurements can be made directly from one side of the zone to another. The horizontal components of the displacement vector are more difficult to measure. Typically, displacement is measured across the offset of a linear feature, such as a road, that crosses an entire zone of rupture. This horizontal component is the normal distance between two offset line segments that were co-linear prior to the earthquake. An unknown component that would complete the definition of the net differential displacement is at right angles to the offset measured normal to the line segments. The unknown component would be positive or negative dilation in the rupture belt.

We found a few places along the rupture belt where a narrow fault zone severed a bush, and the root ball of the bush was broken into two parts. For that situation, the complete displacement vector including relative vertical change could be measured, and we typically found positive dilation normal to the rupture belt. Unfortunately, we did not find places where dilation could be measured along the Emerson fault zone. Thus, we have followed custom and report offset in the direction of the rupture belt. Reporting, as we do here, a fault-parallel offset, we misrepresent the actual net differential displacement but, unfortunately, we are unable to measure the dilation in the rupture belt photogrammetrically.

Note 11. The net translational dispacement along the Emerson fault zone may be even less than 4.5 m. In a reconnaissance of the rupture zones in March, 1997, we noted seven peripheral, strike-slip fracture zones that cross Galway Lake Road. The seven zones occurred in two groups, the first group of which contains three fault zones shown on our map. The first zone is at a distance of 1.08 km along the road from the main rupture belt; this zone carried right-lateral shift of 1.05 m. There was another right-lateral zone 90 m farther along the road at 1.17 km. This zone appears to be a continuation of the one mapped farther to the south on the aerial photographs (ends on map at 3822120N and 541300E) but not visible in the photographs at the road crossing. The stepping sense of the fractures revealed right-lateral shear, but the amount of displacement could not be determined in the field. The next fracture zone was a left-lateral fault and is shown on the photogeologic map (crosses Galway Lake Road at 3822290N and 541370E). Therefore, we know that a second right-lateral fault zone crossed Galway Lake Road only 90 m beyond the zone having 1.05 m displacement. The total right shift on this side of the rotated block cannot be measured on the photographs, but the 1.05 m appears to be a minimum value.

The second group of fractures begins almost 400 m farther along Galway Lake Road from the left-lateral fault at 1.24 km (i.e. at the end of the first group of fractures). This group contains four individual fault zones--a right lateral at 1.62 km, a left lateral at 1.67 km, a right lateral at 1.78 km, and finally, another right lateral at 1.96 km. The right-lateral fault zones all trend about N10-15oW, and the left-lateral fault zone that appears to be conjugate to the right-lateral zones strikes N25-30oE. These four zones were recognized in the field, but were outside our limit of coverage of aerial photographs. The four zones are shown on the reconnaissance map of Irvine and Hill (1993) and apparently connect with the Emerson fault zone about 1 km southeast of Stanford Hill. Thus, their relationship to the Galway-Lake-Road rotated block is ambiguous. The presence of these fault zones, noted by Irvine and Hill (1993) and also by us in the field during 1997, could indicate that the rotated block is much larger than shown on our map. At a minimum, the presence of additional fault zones over a still larger area than we could map illustrates the necessity of having a complete map. The map should be at an appropriate level of detail of an area larger than the area of the structure in question. Without the map of the larger area, our interpretation of the rotated block structure lacks context in the same way a map of the main rupture belt at Bessemer Mine Road would make no sense without the map of the larger area.

Note 12. Mapping scales range from about 1:100 to 1:200 for detailed maps within a belt of shear zones to about 1:500 where we are interested in relations between individual fractures and shear zones.

Note 13. Detailed maps of some of the fractures on the northeast side of the main rupture belt at Stanford Hill were made by Aydin and Du (1994), Arrowsmith and Rhodes (1994), and Antonneli and others (1992).

Note 14. Ramp structures are a characteristic feature of thrust faults, where the fault surfaces are subhorizontal. In strike-slip faults, a ramp structure is indicated by a double bend in the fault zone. In this case, the fault surfaces are approximately vertical.

Geology Department
http://www.pdx.edu/geology

Copyright © 1994-2020 · K.M. Cruikshank ·
http://geomechanics.research.pdx.edu

PSU Logo