AN ABSTRACT OF THE THESIS OF

Craig L. Schneider for the degree of Master of Science in Geology presentedon March 8. 1994. Title: Pre-Pliocene Structural Geology and Structural Evolution of the Northern Los Angeles Basin. Southern California

Redacted for Privacy Abstract approved: Robert S/Yeats

Detailed subsurface structure contour maps and cross sections have shown the northern Los Angeles basin to be underlain by a south facing monocline that is complicated by secondary faults and folds. The monocline forms a structural shelf that marks the northern boundary of the Los Angeles central trough. The monocline and associated structures are called the Northern Los Angeles shelf. Isopach maps show that during the Miocene, the predominant structural style was extension. Thick accumulations of volcanic and volcaniclastic rocks, controlled by normal faults, hada very different depositional pattern than during the Pliocene. At approximately the beginning of the Pliocene extension changed to compression resulting in the reactivation of the Miocene normal faults in a reverse sense and the beginning of the formation of the monocline and secondary structures. Thick growth sequences were deposited to the south of the growing monocline toward the present day Los Angeles central trough. Fault-bend and fault-propagation fold models are inadmissible solutions to explain the growth of the monocline. A basement-involved shear model may explain some of the details of the secondary structures. Analysis of the Pliocene growth strata shows that the monocline and secondary structures, the South Salt Lake, the East Beverly Hills, and the Las Cienegas anticlines, all began to form near the beginning of the Pliocene. All of the secondary structures became inactive prior to the Upper Pico during the Late Pliocene. Thick accumulations of Upper Pico growth strata attest to continued monoclinal folding after the secondary structures became inactive. The growth strata record both the structural growth and the shortening associated with growth and therefore allow the dip of the monocline causing fault or shear zone (the Monocline fault) to be calculated. In the East Beverly Hills area, the growth strata yield a dip of 61°. At Las Cienegas the dip of the Monocline fault is 62°. These dips are maximum values based on the assumption the growth strata record all shortening. The fault slip rates for the Monocline fault are similar in both areas, 1.1-1.2 mm/yr in the East Beverly Hills and 1.3-1.5 mm/yr. in Las Cienegas. The resulting horizontal convergence rates are also similar, .5-.6 mm/yr and .6-.7 mm/yr respectively. The Quaternary marine gravels have been deformed into a broad east-west trending fold, the Wilshire arch. Elastic and non-elastic methods of modeling the blind fault (Wilshire fault), over which the deformation occurred, yield much greater shortening rates than for the Pliocene. The non-elastic method involves modeling the arch as a fault-bend fold. This model predicts a 15° north-dipping thrust with a slip rate of 1.5-1.9 mm/yr and a horizontal shortening rate of 1.4-1.8 mm/yr. The elastic method involves matching the observed deformation to that produced on the free surface by slip on a fault in an elastic half-space. The elastic dislocation model predicts a right-lateral reverse slip solution with an oblique-slip rate of 2.6-3.3 mm/yr. This solution yields a horizontal shortening rate of 1.4-1.8 mm/yr. These higher shortening rates suggest that there was a marked change in tectonic style at the end of the Pliocene from high-angle faulting and tectonic subsidence to shallow faulting and uplift. Pre-Pliocene Structural Geology and Structural Evolution of the Northern Los Angeles Basin, Southern California

by

Craig L. Schneider

A THESIS

submitted to

Oregon State University

in partial fullfillment of the requirements for the degree of

Master of Science

Completed March 8, 1994

Commencement June, 1994 APPROVED: Redacted for Privacy

Professor of Geology in charge of m4or Redacted for Privacy department ofeosciences

Redacted for Privacy

Dean of Graduate Sch

Date thesis is presented March 8. 1994

Typed by Craig L. Schneider for Craig L. Schneider ACKNOWLEDGMENTS

Many people have been instrumental to the completion of this work. Chevron U.S.A. Inc. and Unocal Corporation contributed both the primary data set, over 400 oil well files, and the financial support for reproduction of the data. I would like to thank Linda Thurn, Bill Bart ling, and Wayne Tobiasz at Chevron and Larry Greene and Charles Roberts at Unocal for their assistance in obtaining oil well data and for gathering assorted bits of data whenever it was crucial. I would also like to thank Gregg Blake of Unocal for his AAPG summary of the biostratigraphy of the Los Angeles basin and for assisting in the release of paleontologic data which proved invaluable to the study. Financial support for the project was provided by both, the National Earthquake Hazard Reduction Program (NEHRP), Contract No. 14-08-0001-G1967, administered by the United States Geological Survey and the Southern California Earthquake Center (National Science Foundation). I thank my coworkers. All aspects of the project have been collaborative between myself, Cheryl Hummon, Bob Yeats my advisor, Gary Huftile my advisor and principal resource while Bob was on sabbatical, and Hiro Tsutsumi. Ross Stein, of the United States Geological Survey, enthusiastically instructed me on the method of elastic dislocation modeling as well as provided insightful discussion on the structure of the earth's crust and its response to stress. My understanding of elastic modeling was also aided by discussions with Geoff King. Wayne Narr, at Chevron, introduced me to an exciting new hypothesis describing fault and fold interaction within crystalline rocks. Tom Wright, whose lucid description of the structure of the Los Angeles basin has proved a valued reference, has been a part of the discussion since the beginning of the project. I would also like to thank my friends and fellow students, Rene La Berge, Steve Moothart (Moot), Nickoli Greenstone, Stor Nelson, In-Chang Ryu, Daniel la Assail, Chris Goldfinger, Jeff Templeton, Dave Maher, Ken Bevis, Erwin Tome, Brian Desmarais, Christian Braudrick, Jennifer Crum, Kerry Mammone, Chuck Payne, and Peter Powers who have expanded my education in innumerable ways. Lastly and most important - To my family who have supported me through many years of school, I thank you. TABLE OF CONTENTS

INTRODUCTION 1 Structural Setting 4 Newport-Inglewood fault zone 4 Santa Monica fault system 5 North Los Angeles shelf 6 Stratigraphy 6 Lithostratigraphy 6 Biostratigraphy 9 Growth Strata 12

MIOCENE EXTENSION 15 The San Vicente fault 15 The Las Cienegas fault 16

PLIOCENE COMPRESSION 18 The San Vicente fault 18 The South Salt Lake anticline 20 The East Beverly Hills anticline 20 The Las Cienegas fault 22 Summary 23

STRUCTURAL KINEMATICS 24 Key Structural Observations 24 Problems with Classical Methods 24 Fault-Bend Fold Model 25 Fault-Propagation Fold Model 27 Basement Shear Deformation 27

GROWTH STRATA ANALYSIS 32 Introduction 32 East Beverly Hills area 34 Structural Observations 34 Growth Calculations 36 Shortening 37 Fault Dip 39 Las Cienegas area 42 Structural Observations 42 Growth Calculations 42 Shortening Calculations 44 Fault Dip Calculations 44 Fault Slip and Shortening Rates 44 Summary 45

DISCUSSION AND CONCLUSIONS 48

BIBLIOGRAPHY 50

APPENDIX 56 LIST OF FIGURES

Figure

1. Location map of the study area showing major tectonic provinces 2 and faults.

2. Location map of geographic features within the study area. 3

3. Lithostratigraphic column of the units within the study area. 7

4. Biostratigraphic column for units within the northern Los Angeles basin. 11

5. Effect of sedimentation rate vs. growth rate on resultant growth strata 13 geometry.

6. Parameters for calculating the vertical component of folding (Growth) 14 from growth sediments.

7. Schematic unbalanced cross-section illustrating the evolution of the 21 South Salt Lake anticline by reactivation of the San Vicente fault.

8. Comparison of growth strata geometry. 26

9. Four types of folding that could result from a fault-fault-axial surface 28 (a-a') triple junction.

10. Kinematic development of a footwall-shear, fault-bend fold anticline 29 (Narr, 1990).

11. Deformational geometry of the basement surface. 31

12. Calculation of shortening and fault dip. 33

13. Cross section through the East Beverly Hills area, showing the position 35 of the pin lines used in the shortening calculations.

14. Cumulative shortening vs. cumulative relative subsidence for the East 41 Beverly Hills and Las Cienegas areas.

15. Cross section through the Las Cienegas area, showing the position 43 of the pin lines used in the shortening calculations.

16. Relative timing of structures in the northern Los Angeles basin. 47

A.1. Structure contours and dip data showing deformation of the base of 57 Quaternary marine gravels (Hummon, 1994).

A.2. Fault-bend fold model of the Wilshire arch. 59 A.3. Dislocation parameters used in the elastic dislocation models. 61

A.4. Effect of dislocation burial depth on free surface deformation. 62

A.5. Effect of dislocation width on free surface deformation. 62

A.6. Minimum and maximum two dimensional elastic dislocation models 63 of the Wilshire fault.

A.7. Three dimensional elastic dislocation model for the same fault geometry65 as the 2D maximum slip solution shown in Figure A.6.

A.8. Three dimensional elastic dislocation model of the Wilshire arch using 67 a right-lateral strike slip component.

A.9. Best fit three dimensional elastic dislocation model of the Wilshire arch.68

A.10.Comparison of elastic and non-elastic solutions. 70

A.11.Location and focal mechanism of the possible earthquake on the 72 Wilshire fault. LIST OF PLATES

Plate 1. Location of cross sections (Hummon and Schneider).

Plate 2. Cross section A-A', East Beverly Hills, Saturn CH (Hummon and Schneider).

Plate 3. Cross section B-B', East Beverly Hills, Packard drillsite (Hummon and Schneider).

Plate 4. Cross section C-C', St. Elmo CH-1 ( Hummon and Schneider).

Plate 5. Cross section D-D', Union Pacific Electric drillsite, Las Cienegas area (Hummon and Schneider).

Plate 6. Cross section E-E', Fourth Avenue drillsite, Las Cienegas area (Hummon and Schneider).

Plate 7. Cross section F-F', Murphy drillsite, Las Cienegas area (Hummon and Schneider).

Plate 8. Structure contour map of the Pico/Repetto contact (Hummon).

Plate 9. Structure contour map of the base of the Repetto (Schneider).

Plate 10. Structure contour map of the Delmontian Stage-Mohnian Stage contact (Schneider).

Plate 11. Structure contour map of the top of the nodular shale (Lower Mohnian Stage, Lower Member of the Puente Formation (Schneider).

Plate 12. Structure contour map of the Las Cienegas, San Vicente, and North Salt Lake faults (Schneider).

Plate 13. Isopach map of Pico strata, Upper Fernando Formation (Hummon). Plate 14 Isopach map of Repetto strata, Lower Fernando Formation (Hummon). Plate 15 Isopach map of Delmontian Stage strata (Upper Member of the Puente Formation) (Schneider).

Plate 16 Isopach map of Mohnian Stage sandstones (Schneider). PREFACE

This work has been a cooperative effort, using the same data set, between myself and Cheryl Hummon. Cheryl Hummon was responsible for mapping the Pliocene and Pleistocene subsurface structure. I was responsible for mapping the pre-Pliocene subsurface geology and the structural evolution for the study area as a whole. This division of work has made both thesis interdependent on the work of the other. For example, in determining the timing of various structures and the growth history of the northern Los Angeles shelf it was essential that I examine the geometry of the Pliocene growth strata. Also, because in many places the Pliocene strataare only controlled by widely spaced drillsites, it was necessary that Hummon utilize the deeper structure, which I was responsible for mapping, to make shallow interpretations. Beyond the sharing of data and structural interpretations, my effort and understanding has been greatly aided by continuous geologic discussions with Cheryl Hummon.. The thesis is divided into two parts. The first part outlines the pre-Pliocene structural geology and the structural evolution of the northern Los Angeles basin. The second part "Locating the Wilshire fault: Elastic and Non-elastic Approaches", was undertaken as part of a study of deformed Quaternary marine gravels within the study area. This work, as part of the study of the Quaternary marine gravels, is accepted for publication in Geology and is presented here as an appendix. PRE-PLIOCENE STRUCTURAL GEOLOGY AND STRUCTURAL EVOLUTION OF THE NORTHERN LOS ANGELES BASIN, SOUTHERN CALIFORNIA

INTRODUCTION

The purpose of this study was to determine the location, geometry, and evolution of subsurface structures in the northern Los Angeles basin (Figure 1). The principal database is electric logs from over 400 directionally-drilled oil wells from the San Vicente, Salt Lake, South Salt Lake, East Beverly Hills and Las Cienegas oil fields,as well as numerous exploratory wildcat oil wells (Figure 2, Plate 1). Structuralcross sections, structure contour maps, isopach maps, and analysis of growth packages document the structural evolution of the northern Los Angeles basin. The geometric and kinematic evolution of the structures is investigated by various models of fault and fold interaction. The Los Angeles basin is at the boundary between the California Continental Borderland, including the Peninsular Ranges, and the Transverse Ranges in southern California (Figure 1). This late Cenozoic basin formed within the evolving transform margin along the western edge of the North American plate. The California Continental Borderland and Peninsular Ranges are characterized by northwest- trending right-lateral strike-slip faults, each of which takes up a part of the overall right-lateral motion between the North American and Pacific plates. The Transverse Ranges are characterized by west-trending reverse faults that have formed in response to the formation of the restricting "big bend" of the San Andreas fault system. At approximately 30 Ma, the Pacific-Farallon spreading center began subducting beneath the leading edge of the North American plate at the approximate future position of the Los Angeles basin, initiating the formation of the transform margin (Atwater, 1970). During the Oligocene to late Miocene, during development of the transform margin, the Los Angeles region accumulated sediments due to transtension.Abundant volcanism (Yerkes et al. 1965, Yeats, 1968, Campbell and Yerkes, 1976, and Weigand, 1982), strike-slip crustal block rotation (Kamerling and Luyendyk, 1979; Luyendyk et al., 1980), and normal faulting occurred at this time. Beginning in the early Pliocene, the Los Angeles basin began to take its present shape (Yeats and Beall, 1991) as extension changed to compression (Campbell and Yerkes, 1976), reactivating pre-existing normal faults as reverse faults and producing a thick sedimentary 119'00' 118°00' 34°30' it set, Andreas Transverse Ranges! San Gabriel fellit Santa Mountains

Ser? %II 11 1 1 1 11 111 1 II 1 1 1 1 11 11 \ Santa Susana Fault Gabriel,fault 11\0 San Fernando Valley /minM\\'illi Sierra 11111111111111111111111111111111111\1"W" fault Santa Monica Mountains

34'00' 6 ider faultPeninsular Malibu Coast- Los Angeles ,,046 Ranges Santa Monica fault +19 41114%,,, IP z Figure 2 .0 oz. 0, -WI -es awe "/, .S 46 Palos Verdes 67 6)11- (76.41 fault 4), -i,Santa Ana Mountains Continental Borderland I

Figure 1.Location map of the study area showing major tectonic provinces and faults. The Transverse Ranges are characterized by west trending,left-lateral reverse oblique faults.The Continental Borderland and the Peninsular Ranges are characterized by northwest-trending right-lateral, strike-slip faults. 1 2 34° 7' 30.. kilometers \\ \ II I \I/ SantaMonica \ WIOUntanS \ \A \\0- '4" \ \ WM* Sherman oilfield Elysian San Vicente Hills oilfield Jade-Bunram drillsite La Brea Tar Pits A.; .> ....,,,,"^.,,...... C-2) / Wilshire Boulevard acs / South Sat Lake , ..I., . oil field 31 Ass '''' /./ Pacific Electric f 01 '$,:a.West Pico &Melte ,g.,..., Fourth Avenue Cheviot ', \113, Dr) 8 te Packard drillsIte McMICappal* , East Beverly Hills Hills -- s oil field Drillsite Santa -- Freewa Santa = Monica Las Cienegas oil field urphy drillsite \\ VArt ,,\ Baldwin Hills ; 34° 00' 118° 30' (Beverly Hills 7.5' quad) 118° 22' 30" (Hollywood 7.5' quad) 118° 15' Figure 2. Location map of geographic features within the study area. Stippled lines indicate topographicfault scarps (Dolan and Sieh (1992). Oil fields use in this study, are shaded. The East BeverlyHills oil field consists of both the Packard and Electric, Fourth Avenue, and Murphy Drillsites of the West Pico drillsites. Only the Pacific Las Cienegas oil field were mapped in thisstudy. The location of the Newport- Inglewood fault zone in the Baldwin Hills is from Dibblee(1991a, 1991b). 4

sequence. The recent 1994 Northridge and 1987 Whittier Narrows earthquakes attest to continuing compression and highlight the hazard associated with blind faults inthe greater Los Angeles metropolitan area. This study focuses on the northern Los Angeles basin in thearea bounded by the Santa Monica Mountains to the north, the Elysian Hills and Cheviot Hillsto the east, and the Baldwin Hills and Los Angeles central trough to the south. The studyarea is a south-sloping plain of alluvial fans that extend from the Santa Monica Mountains, giving little topographic evidence for the complex geologicstructures in the subsurface.

Structural Setting

Located at the boundary between the Continental Borderland and the Transverse Ranges tectonic provinces, the northern Los Angeles basin containsstructures common to both (Figure 2).

Newport-Inglewood fault zone The right-lateral Newport-Inglewood fault bounds the studyarea on the west and reflects the tectonic style of the California Continental Borderland (Figure 2). Wright (1991) restricts the fault to south of the northern side of the Baldwin Hills, but recognizes that the topographic lineament on the east side of the Cheviot Hillsmay reflect the northwestern continuation of the Newport-Inglewood faultas described by Soper (1943). This topographic feature has been termed the West Beverly Hills lineament (Dolan and Sieh, 1992; Figure 2). West of the Newport Inglewood fault the basement is greenschist- and blueschist-facies metamorphic rock of the Catalina Schist Formation (Schoellhamer and Woodford, 1951; Yeats, 1973). South of the Los Angeles basin, in the San Joaquin Hills, the middle Miocene San Onofre Breccia (Catalina Schist debris) overlies Peninsular Ranges plutonic basement, suggesting that the Newport-Inglewood fault at least at this location, acted as a boundary between basement blocks and was active as early as Miocene (Yeats, 1973). This boundary is obscured by a thick cover of Miocene and Pliocene strata within the Los Angeles basin, and Yeats (1973) suggests that the boundary may lie east of the Newport- Inglewood fault zone in the central and northern Los Angeles basin. At the Baldwin Hills, in the Inglewood oil field, Mohnian (?) sandstones thin onto the anticlinally 5

folded lower Mohnian Nodular Shale, suggesting structural growth at this time (Wright, 1991). The major time of movement on the Newport-Inglewood faultwas late Pliocene and Pleistocene, however, as evidenced by substantial thinning of Pico sandstones onto the Baldwin Hills (Wright, 1987). Isopachs of these intervals reveal approximately 4000 ft (1200 m) of late Pliocene and Pleistocene right-lateral offseton the Newport-Inglewood fault (Wright, 1987). The 1920 Inglewood and 1933 Long Beach earthquakes document continuing movement on the Newport-Inglewood fault.

Santa Monica fault system The left-lateral oblique Hollywood fault bounds the studyarea to the north (Figure 2), and is part of the larger Malibu Coast-Santa Monica-Hollywood-Raymond Hill fault system. This system makes a left step at the West Beverly Hills lineament. The Santa Monica fault system has been considered the southern boundary of the Transverse Ranges tectonic province. This study, and studies by Davis et al. (1989) and Hauksson (1990), suggest that structures in the subsurface south of the Santa Monica fault system are also responding by reverse faulting, indicating that the southern boundary of the Transverse Ranges province extends south of the Santa Monica Mountains into the northern Los Angeles basin. Correlation of lower Miocene shoreline facies and Cretaceous through Paleocene sedimentary and crystalline rocks between the western Santa Monica Mountains and the Santa Ana mountains documents 90 km of left-lateral displacement (Yeats, 1976). Lamar (1961) based on electric log correlations, correlates the Mohnian Stage Tarzana submarine fan facies described by Sullwold (1960) between the northern Santa Monica Mountains and the Los Angeles basin, and points out that it has been offset left-laterally 8 km. Redin (1991) also noted the correlation of this facies with the Mohnian stratigraphy within the Los Angeles basin and suggests approximately 10-15 km of post-Mohnian left-lateral offset along the Santa Monica fault system. In the Beverly Hills oil field, the Santa Monica fault shows 3600 m (12,000 ft) of north-side- up vertical separation of the basement surface. Reverse separation on the Santa Monica fault system began at about the end of the Miocene (Campbell and Yerkes, 1976). This period of compressive structural growth marks the initiation of the Los Angeles basin in its present shape. Scarps associated with the Santa Monica fault west of the Newport Inglewood fault and with the Hollywood fault attest to continuing movement on the fault system (Dolan and Sieh, 1992). 6

North Los Angeles shelf South of the Santa Monica fault system is a series of subsurface anticlines that form a structurally high shelf and define the northern margin of the Los Angeles trough. Davis et al.(1989)interpret this zone of structures to be the frontal folds of the larger Santa Monica Mountains anticlinorium. As illuminated by seismicity this trend of structures was termed the Elysian Park fold and thrust belt by Hauksson (1990).Wright(1991)uses the term "Northern Shelf" to define this trend. Although the structures are probably related to the same fault system that uplifted the Santa Monica Mountains, I prefer to use the descriptive term "shelf". A structure contour map of the base of the Repetto locates the major structural features of the study area (Plate9).The predominant structural feature of the shelf is a south-vergent, basement-cored monocline that is complicated by smaller secondary faults and folds. From east to west, these secondary foldsare: the Las Cienegas anticline, the South Salt Lake anticline and the East Beverly Hills anticline. These smaller folds have acted as structural traps for hydrocarbon accumulation. Thus, oil exploration has revealed much of the subsurface geology of the north Los Angeles basin. Structural cross sections and contour maps, combined with isopach maps, reveal two stages of tectonic development of the northern Los Angeles shelf. During the Miocene, normal faults controlled sedimentary depocenters. The shelf and folds that are explored for oil today began forming near the beginning of the Pliocene.

Stratigraphy

Lithostratigraphy The oldest sedimentary units within the Los Angeles basin are located on the western slopes of the Santa Ana Mountains (Figure 1) and are Late Cretaceous in age. However, nowhere within the study area are sedimentary rocks older than Middle Miocene encountered, except for the possible occurrence of Paleocene redbed and granite-wash sandstone at the base of the Morgan-Brown U-6 #1 well (Plate 1; Yeats, 1973).The stratigraphic divisions and nomenclature used in this study are shown graphically in figure3. Within the study area,middle Miocene volcanic and volcaniclastic marine sedimentary rocks nonconformably overlie metamorphic rocks that are considered 7

General Graphic Formation Lithologic Thickness Lfthologlo Composite Thickness Depositional Epoch Staisi /Division Description (1000's ft) Column Electric Log Range Envronmenta Marine bailie. Marine gravels and sandstones. 0.450 ft Gravels (0-137 nil Innernerftic Pleistocene uvemeeriar 1 Massive micaceous shales and 0.6300 ft Inner nerttic to Cr- ienturka" sitstones interbedded with sand. (0-1900 in) upper bathyal I ... u c i a) - Fine to coif -grained sandstone - m 0-4600 ft it'.. Interbedded with safely micaceous Lower bathyal Repettian sttm.,... and shais. (0.1400 m) orI t a.

A kyr taddadcliaternalyousailtstone and shale with a regionally extensive 750.1750 ft Upper to patof bentonites at the base. (210-530 m) middle bathyal

Int erbeddedeandstone, microarrinated sittstone and shale. SPtstone and shale with two interbeddecteMonites. _a ... a "Tarzana Fan" i5 3v _ 1000-4500 ft Upper to ct. x Thick amalgamated masshe (305-1370 rn) mode bathyal s sandstone.

.§ : u I .4' North of the Las Cienegas and South Salt Lake areas te nodular shale is I wedeln by thick amalgamated sandstones continuous with t hose of ru the Mbddle Member of the Puente 100-2000 ft Upper to formation. -4-----..,.. (304000 ml -..-II-.1- middle bathyal Phosphatic, glauconftic shale 'Nodular StaleShale', ", Interbedded with -i--i--,i- sedately. erezirceice 4 Interbeddedtuffaceoussiltstoneand 11zoom 0-'3000 ft Inner neritic to - Luisi a n F sandstone. Rare basalt flows . MI (0 >900m) middle bathyal Onsirseesse Las Cienegas area: Gray-green, ,... Ieeieefteee chloriteschist spidote, alkate-bearing chlorite schist. pift~er, 0f ise~ftel Santamalice Hollywood fault zone: Maw-black ePerrereffee0 marcasite-bearing Mate. efoftenePerel Slate piosieweeeeo

1. Pleistocene and Pliocene: Natland, 1952; Miocene: Kleinpell, 1938, 1980 2. Wissler,1943, 1958 3. Blake, 1991

Figure 3.Lithostratigraphic column of the units within the study area. Thickness scale is generalized because many of the units show great thickness variation. South of the San Vicente fault, (Plate 12), the top of the Lower Mohnian coincides with the top of the glauconitic phosphatic shale. North of the San Vicente fault, the top of the Lower Mohnian is within massive sandstone. The composite electric log shows the general character of the units and does not intend to show true thicknesses. 8

basement. Basement rocks are reached by oil wells in two locations, the Las Cienegas anticline (Plate 6) and along the southern margin of the Hollywood faultzone, where basement is penetrated at the Sherman oil field (Figure 2) and Laurel CH (Plates 1 and 2). In the Sherman oil field the basement is described by industry mudlogas blue-black, marcasite-bearing slate (Arden P.E. #2) suggesting that it could correlate with the Santa Monica Formation, which is present just to the northacross the Hollywood fault and is Jurassic in age. This correlation wouldsuggest a small left- lateral offset across the Hollywood fault at this location. At the Las Cienegas oil field in contrast, the metamorphic rocks are predominantlygray-green, epidote and albite- bearing chlorite schist as seen in the Unocal Fourth Avenue and Murphy drillsitesas well as the Union-Signal-Texam U-19-1 exploratory well (Plate 1). A similar basement lithology is found at the Inglewood oil field to the south. Here, east of the Newport-Inglewood fault zone, Yeats (1973) describes the basementas albite and chlorite-muscovite schist. These metamorphic rocks have been, tentatively, interpreted as Jurassic in age (Sorensen, 1984, Yeats, 1973). Basement is unconformably overlain by the middle Miocene Topanga Formation. The Topanga Formation is composed of interbedded marine sandstone, siltstone, tuffaceous siltstone, tuffs, and rare basalt flows or intrusions (Gilmore #5; Plate 3). Two wells, the Morgan-Brown U-19 (Plate 2) and the Jade-Buttram Gilmore #5 (Plate 3), penetrate over 900 m (2953 ft.) true stratigraphic thickness of Topanga Formation without reaching the base. Thus, within the study area, the maximum thickness of the Topanga Formation is only constrained to be greater than 900 m. In the Las Cienegas area, the Topanga is only 0-200 m (0-656 ft.) thick. Overlying the Topanga Formation is the middle to late Miocene Puente Formation (Eldridge and Arnold, 1907). The Puente Formation is divided into Lower, Middle, and Upper Members which thicken dramatically to the north. The Lower Member consists of a basal glauconitic, phosphatic nodular shale interbedded with sandstones and overlain by thick, massive sandstones. The thickness of the overlying sandstones increases to the north from < 50 m in the Las Cienegas and East Beverly Hills anticlines to 305 m (1000 ft) in the Union Paramount CH and the San Vicente area (Plates 2 and 7). The fine-grained, glauconitic nature of this member suggests a condensed section or very slow sedimentation rates. This member is equivalent in lithology and time to the carbonate marlstone member of the Monterey Formation in the Santa Barbara-Ventura basins (Isaacs, 1980) and the Lower Member of the 9

Monterey Formation in the Santa Maria basin (Pisciotto, 1978; Woodring & Bramlette, 1950). The Middle Member of the Puente Formation is characterized by thick amalgamated sandstones, which are continuous with those of theupper part of the Lower Member, with minor interbedded siltstone and shale This thick sandsequence is correlated by Redin (1991) to the submarine Tarzana Fan described by Sullwold (1960). The thickness of the Middle Member varies from >610m (>2000 ft) at the north end of the field area to 20 m (65 ft) farther south at the Las Cienegas anticline. Overlying the thick amalgamated sandstone are interbedded sandstones and microlaminated, cherry siltstones which form the top of the Middle Member. The Upper Member of the Puente Formation is predominantly composed of diatomaceous siltstones and shales. In the Las Cienegas area and the East Beverly Hills syncline, however, the top of the Upper Member has thick (10-100 m) sandstone intervals. Near the base of the Upper Member are two bentonites (only 10 m apart), easily recognized in electric log, that are used to identify and map the Upper Member. The siliceous lithology of the Upper Member and the chert-rich microlaminated nature of the Middle Member suggests that these strataare equivalent to the upper siliceous facies of the Monterey Formation (Pisciotto and Garrison, 1981). The Middle and Lower Members of the Puente Formation thicken towards the Los Angeles basin margins and have a regional depositional pattern similar to the Topanga Group (Blake, 1991) Overlying the Puente Formation is the Fernando Formation of early Pliocene early Pleistocene age (Eldridge and Arnold, 1907; Blake, 1991). The Fernando Formation marks a change in depositional style in that it shows dramatic thickening to the south, toward the Los Angeles central trough, rather than to the north as in the Puente Formation. Based on common usage by oil company geologists, the Fernando Formation is informally divided into two members, the Repetto member and the overlying Pico member. The Repetto consists of interbedded fine- to coarse-grained sandstone, siltstone and mudstone. The Pico is finer grained than the Repetto and is predominantly massive siltstone and mudstone with minor interbedded silty sandstone.

Biostratigraphy Biostratigraphic correlation uses the biochronology presented by Blake (1991), which is based on the benthic foraminifera zonation of Kleinpell (1938, 1980), Wissler 10

(1943, 1958), and Nat land (1952). Correlation of the benthic foraminifera biostratigraphy with planktonic biostratigraphies and the radiometric time scale, based on fission-track (Obradovich and Naeser, 1981) and K-Ar (Turner, 1970) dating of ashes, has allowed the age calibration of the benthic foraminifera zones (Figure 4; Blake, 1991). Benthic foraminifera occurrence is strongly controlled by water depth, thus the zonations can be time transgressive. However, within the study area, the base of the Delmontian Stage is consistently overlain by a characteristic bentonite. This close correlation suggests that within thestudy area, stage boundaries approximate time lines. To the east, the zonal correlations cross bentonite horizons, suggesting greater correlation errors (Wright, 1991). Four stages comprise the middle and late Miocene: from oldest to youngest these are, the Relizian, Luisian, Mohnian, and Delmontian Stages (Kleinpell, 1938). Wissler (1943, 1958) assigned six divisions (A- F) to correspond to foraminifera zones of the Delmontian, Mohnian, and Luisian Stages (Figure 4).Nat land (1952) divided the Pliocene and Pleistocene stratigraphic sections into the Repettian, Venturian, Wheelerian, and Hallian Stages. The Venturian through Ha llian Stages have been shown to be locally time-transgressive (Ponti et al, 1993). However, these divisions have been widely used by industry paleontologists and have been adopted, with local revision, for this study. The oldest Miocene sedimentary rocks penetrated in the study area are Luisian Stage volcaniclastic rocks and basalts which are considered within Division "F". The division `B" /"F" boundary coincides with the boundary between the Puente and Topanga Formations and the Luisian and Mohnian Stages. This boundary is correlated to an age of 13.9±0.1 Ma (Blake, 1991). The top of division "E" corresponds with the top of the Lower Member of the Puente formation and is marked by the first downhole occurrence of Bulimina uvigerinaformis and Baggina californica (Wissler, 1943). This boundary is correlated to an age of 8.75±0.15 Ma (Blake, 1991). Division "D" is defined by the occurrence of the crushed foram "Renulina" (Cassidulinella Renulinaformis ?). Because this is the only form restricted to division "D", differentiation between "C" and "D" is often difficult. The top of this division is correlated at 7.4±0.4 Ma (Blake, 1991). Division "C" is marked by the first occurrence of the Bolivina hughesi and Bolivina decurtata assemblage (Wissler, 1958), and the top is assigned an age of approximately 6.5 Ma (Blake, 1991). The Delmontian Stage is composed of the "A" and "B" divisions and corresponds to the Upper Member of the Puente Formation. The top of the 11

Time (Ma) Epoch Depositional Formation Stage /Division2Environment3 Marine Ha Man Gravels Innerneritic 1 Pleistocene ---. tiv,iheelerian

C)m Inner neritic to 2 a v entUrian upper bathyal -8 1 3 2 3 5 u 4u co .:..1m o LL w 4 = a. Repettian Lower bathyal 0_ '' ov _ cc 1 ca Lu

c's A 41 e upper to 6 i.e B middle bathyal c) 7 v C 2 Upper to 8 J"13_i -0 middle bathyal eu -±- D 4.,c 9 I c (a 10 'E c4' -g 4.) U 11 o _ -t- cu Upper to 0 E middle bathyal 12 _1 N "T3 13 t"0 14 qeLuisian F Inner neritic to middle bathyal 15 ,c09

p.. chloriteschist or w140? SantaMonica 1 Slate

1. Pleistocene and Pliocene: Natland, 1952; Miocene: Kleinpell, 1938, 1980 2. Wissler,1943, 1958 3. Blake, 1991

Figure 4. Biostratigraphic column for units within the northern Los Angeles basin. Vertical axis is time in millions of years. The depositional environments based on benthic foraminiferal assemblages (Blake, 1991)are shown for reference. The Late Pliocene and Pleistocene benthic foraminifera Stages can be locally time-transgressive (Ponti et al., 1993; Hummon, 1994). 12

Delmontian Stage is marked by the first downhole occurrence of Rotalia garveyensis. The top of the Delmontian Stage is assigned an age of 4.95±0.15 Ma (Blake, 1991). The top of the Repettian Stage is estimated at 2.5 Ma (Blake, 1991) and thetop of the Pico member is 0.9±0.1 Ma (Blake, 1991). Both the Repetto and Pico Membersare divided into upper, middle, and lower intervals.

Growth Strata

The principal data set that we use to evaluate the timing of the growth of structures is the geometry and thickness variation of the sediments that are deposited during structural growth. This reasoning was used above as evidence for the timing of motion on the Newport-Inglewood fault. The following discussions use thickness variations of syndeformational sediments to define timing. Thus, a description of the method is presented here. As folds grow, they can change the topography or bathymetry of the overlying surface either by uplift or subsidence and can thus affect sediment depositional patterns. Sediments affected by tectonic activity are termed growth strata (Suppe et al., 1992). During relative uplift or subsidence of an active fold, syndeformational sediments will be ponded around the active structure. If the rate of sedimentation or deposition exceeds the deformation rate, then no topographic relief will form, but strata will be thinned on the top of the structure (Figure 5a). If the sedimentation rate equals the deformation rate then sediment will be deposited adjacent the growing structure but not on the crest of the structure. This structurally controlled non- deposition or slow deposition will allow the accumulation of glauconite on the crest of the structure (Figure 5b). If the sedimentation rate is less than the deformation rate, then no sediment will be deposited over the top of the structure which may result in the formation of an unconformity (Figure 5c). If sedimentation rate exceeds growth rate, then the thickness of strata adjacent to a structure minus the thickness of growth strata over the crest of the structure equals the vertical component of structural growth (Figure 6). Conversely, if growth rate exceeds sedimentation rate and erosion has truncated the crest of the structure, the structural growth equals the thickness of growth strata adjacent to the structure plus the thickness of strata, growth or pregrowth, that has been eroded from the crest. 13

a) Sedimentation Rate > Growth Rate

r0 ri h Strata Thin deposition on crest

Pregrowth Strata b) Sedimentation Rate = Growth Rate Non-deposition on crest; glauconite accumulation

c)Sedimentation Rate < Growth Rate ...,;,/%1IPV\OWp Erosion of crest and formation of unconformity

Figure 5. Effect of sedimentation rate vs. growth rateon resultant growth strata geometry (modified from Suppe et al., 1992). 14

Gr owth Strata Thickness Thickness over structure 101.1/*. *11161 11.111.,,,, in basin tGrowth Pre-Growth Strata

Figure 6: Parameters for calculating the verticalcomponent of folding (Growth) from growth sediments. Thickness in basin- Thickness over structure = Growth (modified from Suppe et al., 1992). 15

MIOCENE EXTENSION

San Vicente fault

The San Vicente fault, first described by Jacobson and Lindblom (1987), isan approximately east-west striking, blind reverse fault thatpasses through the southern margin of the San Vicente oil field, through the core of the South Salt Lake anticline, and north of the Las Cienegas oil field (Plate 12). The San Vicente fault marksa boundary between relatively thin (610 m, 2000 ft) Mohnian Stagestrata in the footwall south of the fault and thick (1524 m, >5000 ft) Mohnian Stagestrata in the hangingwall on the north (Plate 2). It is therefore interpretedas a north-dipping, Mohnian Stage normal fault. This is analogous to the thickening of the Topanga Formation in the hangingwall of the Whittier fault (Yeats, 1987). Although interpreted as a Mohnian Stage normal fault, the San Vicente fault showsreverse separation of the Mohnian and Luisian strata. This separation indicates that the San Vicente fault was reactivated in a reverse sense at sometime after the Mohnian. The compressive reactivation of the San Vicente fault is discussed in the next section. Only three oil wells (Chevron San Vicente #7, #17, and #39A) penetrate the fault (Plates 2 and 12). However, an isopach map of Mohnian Stage sandstones shows the position of the dramatic change in thickness (Plate 16). This thickness change coincides with the position of the South Salt Lake anticline axis east of where the fault location is constrained by wells (Plates 5, 6, and, 7). Therefore, the anticlinal axis is used to map the location of the San Vicente fault east of the San Vicente oil field. The San Vicente fault is not found in the Elysian Hills to the east of the studyarea. The deepest rocks encountered within the Miocene "graben", bounded on the south by the San Vicente fault, are Topanga Formation (Luisian Stage) volcaniclastic sedimentary rocks interbedded with basalt flows. South of the fault, in the Union 4th Ave #7A well at Las Cienegas oil field Good Shepard area, the Luisian is absent, and lower Mohnian Stage strata directly overlie metamorphic basement (Plate 6). North of the fault, however, the Gilmore #5 well in the Jade-Buttram drillsite, reaches total depth after penetrating 914 m (3000 ft) of Luisian strata. Also, to the northeast, the Ambassador Corehole (approximately 4000 ft east of Wilton Corehole #2,[Plate 1]) reaches total depth 610 m (2000 ft) within Luisian Stage strata. Therefore, at least north of the Las Cienegas area, the San Vicente fault marked an abrupt change in thickness as early as the Luisian Stage. 16

At the East Beverly Hills anticline, the complete section of the Topanga Formation is not penetrated by wells. The Packard P-50 wellpenetrates approximately 229 m (750 ft) of Topanga Formation. At the Inglewood oil fieldto the south, a complete section, 884 m (2900 ft) thick, of Topanga Formation is encountered (Wright, 1991). Thus, at East Beverly Hills anticline, there isno evidence for or against Luisian Stage activity on the San Vicente fault. As stated above, the Mohnian Stage sandstone isopachsare controlled by the San Vicente fault (Plate 16). Isopachs are measured normal to the base of Delmontian Stage strata (Upper Member, Puente Formation) to the top of the nodular shale (Lower Member, Puente Formation). Therefore, this map includes sandstones of the Tarzana submarine fan described by Sullwold (1960), which were deposited during the C, D, and upper E divisions of the Mohnian Stage (Figure 3). Plate 16 (along cross-section F-F') shows that at the location of the San Vicente fault, the thickness of Mohnian Stage sandstones dramatically thins to the south from >1173m (>3850 ft) at Hobart CH to 480 m (1575 ft) at Texam U-19-1(Plate 7). This thickness variation requiresa depositional slope of 13°. Typical slopes of modern submarine fansvary from <1° to 7°, but original depositional slopes rarely exceed 1° (Normark et al., 1993). Therefore, the deposition of Mohnian Stage sandstones is interpreted to have been structurally controlled by the San Vicente fault (Plate 7). This sandstone package thins to less than 100 m thick in the Inglewood oil field where it is termed the City of Inglewood zone (Wright, 1987). Similar thickness changes are present to the east at the San Vicente oil field, suggesting that by Mohnian time, the San Vicente faultwas active along its entire length within the study area.

The Las Cienegas fault

The Las Cienegas fault is a north-dipping, blind reverse fault. It is located beneath and cuts the southern limb of the Las Cienegas anticline (Plate 6). Barbat (1958) first described the Las Cienegas fault as a west-northwest trending fault that extends from the Santa Monica fault to south of downtown Los Angeles. He mapped this fault based on the disparity between thick basin fill in the Los Angeles central trough and relatively thin fill in the Los Angeles downtown oil field. The discovery and development of the oil-producing intervals of the Las Cienegas anticline during the 1960's allowed a more accurate delineation of the orientation of the fault (Mefferd, 17

1970). Gardett (1971) suggested that the Las Cienegas fault splaysto the west of the Las Cienegas field to underlie the East Beverly Hills and South Salt Lake folds.This study suggests that the Las Cienegas fault does not splay to underly the East Beverly Hills anticline. However, it is interpreted that the Las Cienegas and San Vicente faults are splays above a blind, south-vergent fault at depth and that each shallow splay has a different yet kinematically related evolution. Mefferd (1970; his Plate IV) shows that the deepest well at the Murphy Area, Murphy #1, penetrates the Las Cienegas fault and 610 m (2000 ft) of the Lower Member of the Puente Formation in the footwall block. Although Mefferd showsno dip data, Davis et al. (1989), citing this cross section, interpret the Las Cienegas fault as a south-dipping, normal fault bounding a late Miocene and early Pliocene graben. More recent industry paleontologic examination of the Murphy #1 well indicates that the well, after passing through the fault, encounters lower Delmontian Stage strata rather than lower Mohnian strata before bottoming 183 m (600 ft) below the top of the middle Member of the Puente Formation. At the Fourth Avenue drillsite, both the Fourth Avenue #16 and #17 wells penetrate the Las Cienegas fault (Plate 6). Both wells pass through the fault into lower Delmontian Stage strata and well #16 penetrates 274 m (900 ft) of the Middle Member and 168 m (550 ft) of the Lower Member of the Puente Formation within the footwall. There is a similar thickness of middle Puente Formation 274 m (900 ft) in the hangingwall, suggesting that at least during the late Mohnian Stage (Divisions "C" and "D") there was little, if any, normal movement on the Las Cienegas fault. There are no data to determine the full thickness of the Lower Member of the Puente Formation in the footwall of the Las Cienegas fault. Thus, it is possible that the Las Cienegas fault was active as a normal fault prior to deposition of the Middle Member of the Puente Formation. Although there is no Topanga Formation on the crest of the Las Cienegas anticline, there is Topanga Formation within the fault zone at the base of the Fourth Avenue #17 well and just above the fault in #16 (Plate 6). Presumably, the fault is "sampling" lithologies from the footwall and carrying it with the hangingwall. Because there is Topanga Formation in the footwall and not in the hangingwall, sometime prior to the Upper Mohnian, the Las Cienegas fault uplifted the hangingwall such that Topanga Formation was either not deposited or was eroded. This evidence, and the interpretation that the San Vicente fault was extensional during deposition of the Topanga Formation, suggests that the Las Cienegas fault was also a normal fault. 18

PLIOCENE COMPRESSION

The Pliocene Fernando Formation thickens dramaticallyaway from the structural shelf toward the Los Angeles trough. This depositional pattern suggests that the northern Los Angeles shelf and associated secondary foldswere growing throughout the Pliocene and controlled sediment depositional patterns. The first evidence of a change to compressive deformationnear the beginning of the Pliocene is the stratigraphic thinning of Upper Delmontian sandstonesonto the crest of the East Beverly Hills (Plate 2) and South Salt Lake anticlines (Plates 3 and 15). Also, north of the Las Cienegas area, the Upper Member of the Puente Formation (Delmontian Stage) shows slight thickening south of and adjacentto the fold (the eastern extension of the South Salt Lake anticline) associated with the San Vicente fault (Plate 7). This suggests that the strata were deposited during compressional folding, after normal movement on this fault had ended. Similar stratigraphic thinning occurs within the Delmontian Stage at the Las Cienegas anticline as well (Plates 6, 7, and 15). Thus, subsequent to the Upper Mohnian Stage, the San Vicente and Las Cienegas faults were reactivated as reverse faults.

San Vicente fault

As stated above, the San Vicente fault is interpreted as a reactivated Miocene normal fault. This interpretation is supported by the observation that apparentreverse separation of Mohnian Stage strata increases with stratigraphic position. Notice in plate 3 that the Lower Mohnian Stage strata show a slight normal separation. Moving up stratigraphic section, the top of Division "D" shows approximately 335 m (1100 ft) of reverse separation. This apparent increase is because during normal movement, the top of the Lower Mohnian Stage strata was downdropped more than the top of the Division "D" strata. Thus during reactivation the Division "D" strata show more reverse separation than the Lower Mohnian Stage strata (Williams et al., 1989). The top of the Mohnian Stage (top of Division "C") shows approximately 300 m (1000 ft) of separation or about the same as the top of Division "D" indicating that normal growth was minor during Division "C" time. North of the Las Cienegas area, the Delmontian Stage strata show no thickening north of the San Vicente fault and are thus considered to be deposited after normal 19

motion on the San Vicente fault had ceased. Therefore, offset and folding of the Delmontian Stage strata can be used to describe the variation inreverse motion along the length of the fault. North of the Las Cienegas area, the Delmontian Stage strata are folded but not cut by the San Vicente fault (Plates 7 and 10). Structural relief on the top of the Delmontian Stage strata across the San Vicente fault increases westward from 152 m (500 ft) north of the Murphy drillsite (Plate 7) to approximately 518m (1700 ft) north of the Union Pacific Electric drillsite (Plate 5). Between Union CH 20 (Plate 4) and the South Salt Lake field (Plate 3), the San Vicente fault cuts the base of the Delmontian Stage strata. This is the approximate position of the northwestern end of the Las Cienegas fault. Offset increases to the west where, at the southern margin of the San Vicente oil field (Plate 2), most of the Delmontian Stage section is truncated by the San Vicente fault except for the very upper Delmontian Stage strata which overlie the fault with angular unconformity. This overlap relationship is defined by correlation of only two oil well electric logs. If the overlap relationship is correct then the reverse motion at this location is pre-Repettian age. However, in the South Salt Lake anticline (Plate 3) the unconformity is middle Repettian inage. Thus, there is some uncertainty to when the reverse motion on the San Vicente fault ended in the vicinity of the San Vicente oil field. In the hanging wall, of the fault, the Delmontian Stage strata and upper Division "C" strata have been eroded and the Pliocene overlies Division "C" strata with angular unconformity (Plates 2 and 3). The observation that slip on the San Vicente fault decreases at the position where the Las Cienegas fault begins suggests that in the Las Cienegas area, the Las Cienegas fault takes up most of the shortening. However, to the west past the tip of the Las Cienegas fault, shortening is taken up by the San Vicente fault. Between the Los Angeles trough and the top of the structural shelf, the base of the Pliocene has approximately 3200 m (10,500 ft) of structural relief. The San Vicente fault, however, only vertically separates the base of the Delmontian Stage a maximum of approximately 700 m (2350 ft). Therefore, reverse motion on the San Vicente fault only accounts for about 20% of the structural relief represented by the monoclinal shelf. 20

South Salt Lake anticline

The South Salt Lake anticline is located at the anticlinal hinge of the monocline (Plate 3). The anticline is interpreted to be caused by reverse motionon the San Vicente fault. The crest of the anticline is eroded and overlapped by Middle Repetto strata, suggesting that major growth of the fold had ended by this time. This is later than at the San Vicente oil field where the San Vicente fault is overlapped byupper Delmontian sandstones. Because the fault cuts the Delmontian Stage strata witha cut- off angle of 130°, rather than 70° which would be expected fora reactivated north- dipping normal fault, the Delmontian Stage must have been folded prior to beingcut by the fault (Figure 7). Reactivation of the normal fault wouldcause a fold to form over the fault tip as modeled by Mitra (1993). Continued fault propagation could result in a break-through of the forelimb. The small north-dipping limb of the South Salt Lake anticline (Plate 3) suggests that the folding of the Delmontian Stage strata was, at least in part, due to fault-propagation or fault-bend folding. After the end of reverse motion on the San Vicente fault (middle Repettian Stage), the monocline continued to grow and influence deposition of thick Pliocene strata in the Los Angeles trough. Therefore, some of the folding at the South Salt Lake anticline was due to subsequent monoclinal folding.

East Beverly Hills anticline

As stated above, stratigraphic thinning of Delmontian Stage sandstones onto the crest of the East Beverly Hills anticline (Plate 2), suggests that the fold began to grow at this time. The East Beverly Hills anticline is interpreted as a shear fold within the synclinal axis of the monocline. Three lines of evidence support this interpretation: 1) No wells in the south limb of the fold cut a fault. 2) The fold can be constructed to depth without requirement of a fault between wells in the south limb of the fold and the Adamson and Genesee CH within the Los Angeles trough. 3) The non-parallel, simple fold geometry and thickening of the Upper and Middle Members of the Puente Formation in the core of the East Beverly Hills anticline (Plate 3) suggests bedding- parallel shear (Ramsay, 1974) is an important factor in the formation of the fold. 21

a) End of Upper Mohnian

Delmontian Mohnian eeeeeeeeeeeeeee e eN.%%%%N.N.N.N.%%% eeeeeeeee /Jo' 70° %%%%%%%% %%%%%%%%%/////111/ b) Beginning of Pliocene

eee / e e I eeeee

c) Middle Repetto Delmontian

Mohnian /%%%%%%%%%%%%%% /11//010/1///1 /%%%%%%%%%%%%%%% ///1/////////11/

Figure 7. Schematic unbalanced cross-section illustrating the evolution of the South Salt Lake anticline by reactivation of the San Vicente fault. a) End of Upper Mohnian. Delmontian Stage strata depositedon normal fault-controlled Mohnian Stage strata. The cutoff angle of the normal fault is approximately 70°. b) Early Pliocene. Delmontian Stagestrata are folded by reactivation of the San Vicente fault. Propagation of the fault into the overlying Mohnian and Delmontian Stage strata results ina fault-propagation fold. The Mohnian Stage strata must bearea balanced. c) Middle Repetto.The base of the Delmontian Stage strata in the forelimb of the fault-propagation fold is cut by the San Vicente fault with a cutoff angle of approximately 130°. (Modified from Mitra, 1993) 22

If the anticlinal axis of the monocline is fixed, then a space problem develops within the synclinal axis as the fold grows. This space problem can be accommodated by detachment or bedding-parallel shear folding or faulting (Ramsay, 1974). This style of folding is common in Laramide structures in Wyoming and Colorado and has been termed out-of-the-syncline or "rabbit-ear" folding (Brown, 1988). The East Beverly Hills anticline is interpreted as a rabbit-ear fold and is thus kinematically tied to and synchronous with the formation of the monocline.

Las Cienegas fault

As described above, during the Luisian and Mohnian Stages, the Las Cienegas fault was a south-dipping normal fault. In its present geometry, it is a north-dipping reverse fault. Therefore, reactivation of the Las Cienegas fault required that the fault be rotated into a north-dipping position during formation of the Las Cienegas anticline. It is suggested here that this rotation could occur by monoclinal folding above a deeper fault or shear zone within the metamorphic basement. As soon as this deeper fault encountered the rotated Las Cienegas normal fault, reactivation could occur. The Delmontian Stage strata show depositional thickening away from the Las Cienegas anticline and the folds associated with the San Vicente fault (Plate 6 and 7). This thickening is interpreted as resulting from deposition during growth of these structures and is analogous to thickening in the East Beverly Hills syncline (Plates 2 and 3). Thus, the Las Cienegas anticline is interpreted to have started forming during the Delmontian Stage. The Las Cienegas anticline is located at the southern edge of the structurally high, metamorphic, Las Cienegas basement block (Plate 6). The Las Cienegas anticline is in the hangingwall of the Las Cienegas fault and is cored by metamorphic rocks. There is a marked structural change between the Pacific Electric and Fourth Avenue drillsites. In the Murphy and Fourth Avenue drillsites (Plate 6), the Las Cienegas fault cuts the base of the Delmontian Stage rocks in the forelimb of the anticline. However, to the west at the Pacific Electric drillsite, the fault does not cut the forelimb but is interpreted as propagating into the bedding parallel contact between schistose basement and the base of the Puente Formation (Plate 5). Structure contours (Plate 23

12) on the Las Cienegas fault show a steepening of the faultplane at this location. A kinematic model for this evolution is discussed ina subsequent section. Between the Los Angeles trough and the top of the structural shelf, thebase of the Pliocene has approximately 3200 m (10,500 ft) of structural relief. Offset ofthe base of the Delmontian Stage on the Las Cienegas fault is only approximately 610m (2000 ft) (Plate 6). It is possible that because the Delmontianwas rotated prior to being cut, the offset of the base of the Delmontian Stage doesnot reflect the total reverse slip on the Las Cienegas fault. Because the Mohnian Stagestrata were deposited and cut by the fault prior to reactivation, the offset of the base of the Middle Memberof the Puente formation should yield a maximum reverse slip value. The base of thisinterval is offset 850 m (2800 ft) on the Las Cienegas fault. Therefore,at most, reverse motion on the Las Cienegas fault (sensu stricto) accounts for only about 30% of the total structural growth represented by the monocline.

Summary

In summary, the Mohnian Stage strata have similar thicknesses fromeast to west on the north side of the San Vicente fault, suggesting that during this time, the amount of sediment accommodation space and thus the amount of normal fault slipwas similar along the length of the fault. The San Vicente and Las Cienegas faultswere reactivated in a reverse sense near the end of the Delmontian Stage. Reverse slipon the San Vicente fault, as mapped by reverse offset of the base of the Delmontian Stage, indicates that the slip decreases from west to east. It is possible that this decrease is due to shortening being transferred from the San Vicente fault to the Las Cienegas fault. Neither the San Vicente fault nor the Las Cienegas fault is completely responsible for the total structural relief shown by the base of the Pliocene from the Los Angeles central trough to the top of the structural shelf. Also, the San Vicente fault was inactive by the end of the Repetto and the Las Cienegas anticlinewas inactive by the Middle Pico, even though structural growth continued into the Upper Pico. Therefore, monoclinal folding and structural growth continued due to movement on a deeper unconstrained fault or shear zone here named the Monocline fault. In the following section, various models are considered to predict the geometry of the Monocline fault. 24

STRUCTURAL KINEMATICS

Key Structural Observations

The principal observation is that the northern Los Angeles shelf is predominantly a south-vergent monocline that has been complicated by smaller scale folding and break-through faulting. The South Salt Lake anticline has formed inresponse to motion on the San Vicente fault. Similarly, the Las Cienegas anticlinegrew during motion on the Las Cienegas fault. Both of these faults show much lessreverse slip than would be required to form the observed structural relief of the monocline andare thus interpreted as secondary structures. The East Beverly Hills anticline is interpreted as an out-of-the-syncline or rabbit-ear fold as presented by Brown (1988). This means that the formation of the East Beverly Hills anticline is directly tiedto, and is synchronous with, the formation of the synclinal lower fold of the monocline. Because the monocline appears to still be active, to assess the seismic hazard associated with this blind structure we must define the dip and slip rate of the Monocline fault.

Problems with Classical Methods

The kinematic models for fault and fold interaction were developed in thin- skinned structural provinces such as the Canadian Rocky Mountain foothills (Bally et al., 1966; Dahlstrom, 1969) and the Appalachian Mountains (Rich, 1934; Gwinn, 1964). Suppe (1983) developed quantitative geometric models to reconstruct fault locations based on fold shape. These models are based on parallel folding of the rock. Parallel folding requires three assumptions: (1) preservation of layer thickness, (2) conservation of bed length, and (3) no net distortion where the layers are horizontal or non-elasticity (Suppe, 1983). These assumptions are generally valid within bedded sediments where bedding is the predominant anisotropy within the rock, such that folding is principally accommodated by bedding-parallel slip. However, structures that involve igneous or metamorphic rocks, such as the Las Cienegas anticline, do not have a sub-horizontal bedding-plane anisotropy and therefore do not necessarily meet parallel behavior assumptions. The next section discusses the reasons why fault-bend 25

(Suppe,1983)and fault-propagation (Suppe and Medwedeff,1990)models are not admissible solutions to describe the kinematic evolution of the monocline.

Fault-Bend Fold Model

Shaw(1993)interprets the monoclinal flexure on the north side of the Los Angeles basin to be the forelimb of a fault-bend fold. He supports this interpretation by showing that the growth of the forelimb of the monoclinal flexure (he termed the Whittier Narrows trend) and the backlimb of the Compton-Los Alamitos (to the south parallel with the Newport-Inglewood fault zone) structure occurred simultaneously. Using this correlation, Shaw(1993)suggests that a sub-horizontal detachment within the basement links and transfers slip from the Whittier Narrows fault-bend fold to the Compton-Los Alamitos fault-bend fold. This interpretation is problematic for the East Beverly Hills and Las Cienegas areas, farther to the west, because Pliocene growth strata maintain thickness or thin moderately onto the forelimb of the monocline. This observed geometry does not fit the geometry predicted by fault-bend fold theory of an active anticlinal hinge (Figure8).At the East Beverly Hills (Plates 2 and3)anticline, the forelimb shows a gradual thinning, rather than having the full thickness on the forelimb. At Las Cienegas (Plates 5, 6, and 7), however, the full thickness does move onto the forelimb, opposite of that predicted by fault-bend fold theory (Figure8).Fault-bend fold theory predicts that the synclinal hinge is inactive and the thin growth strata on the crest of the fold roll through the active anticlinal hinge onto the forelimb. Therefore, the fault-bend fold model is not admissible. The fact that the East Beverly Hills anticline is forming at the same time as the monocline influences the thickness of growth strata on the forelimb. In addition, the Adamson CH (Plate 2) and Dublin CH (Plate 6) both show a gradual steepening of dips down stratigraphic section. Fault-bend fold theory predicts that folds grow in a self similar manner such that the forelimb dip is established at the inception of folding and does not change during the growth of the fold. The observed "fanning" of dips in the Pliocene growth strata suggest that the monocline evolved by progressive limb rotation rather than constant dip growth as predicted by fault-bend folding. 26

Growth fault-bend fold geometry a) V Growth Hinge

Growth Strata

Pre-growth .-- Active Hinge strata \

Sheared Pin Line Inactive Hinge41%4411441%.44%%%.

Growth geometry observed at East BeverlyHills b)

Growth Strata

Pre-growth strata

Growth geometry observed at Las Clenegas c) Growth Hinge

Growth Strata

Pre-growth strata I Active Hinge

Figure 8.Comparison of growth strata geometry. a) Fault-bendfold geometry. Notice that the active hinge is at the top of the anticlinesuch that thin growth strata roll onto the forelimb (Suppeet al. 1992).b) Growth strata geometry observed at the East Beverly Hillsarea. The dip of the growth strata increases with depth. Here it isunclear whether the synclinal or the anticlinal axis is active.c)Growth strata geometry observed at the Las Cienegasarea.Thickness is maintained from the basin onto the fore limb indicating that the synclinalhinge is active. 27

Fault-Propagation Fold Model

A key observation that discounts the use of a fault propagation fold modelto describe the monocline is that there is no backlimb to the monocline observed within the study area. Fault-propagation fold theory predicts the formation ofa backlimb (Suppe and Medwedeff, 1990). Davis et al., (1989) describe the northern Los Angeles monocline as the forelimb of a fault-propagation fold. This interpretation requires that the north-dipping strata that form the north flank of the Santa Monica Mountains represent the backlimb (Davis, et al., 1989). However this interpretation is problematic because there is no evidence of a backlimb within the studyarea analogous to that within the Montebello Oil Field as shown by Davis et al. (1989; their Figure 9). Also, fault-propagation folds grow self-similarly. The steepening of dips in the growth strata suggest progressive limb rotation not self-similar growth and thus preclude the use of a fault-propagation model. Because neither fault-bend or fault-propagation fold models are admissible, I propose that the monocline is forming by shear above a fault within the basement that does not shallow into a de,collement, but rather, maintains constant dip into the ductile lower crust.

Basement Shear Deformation

Non-parallel behavior was investigated by Narr (1991) to explain some of the features associated with Laramide basement-involved compressive structures in Wyoming. Narr (1990) notes that monoclinal geometry is one of the principal characteristics of the Laramide structures that he examined. Narr suggests that faults propagating within crystalline rocks will exploit anisotropies such as foliation or pre- existing faults. Such an anisotropy would most probably have a different orientation than the propagating fault. When the propagating fault encounters an anisotropy, a triple-junction will form (Figure 9; Narr, 1990) that can deform either the footwall or hangingwall by penetrative distributed shear described by the angular shear strain (v). The cover sequence of bedded sedimentary rocks responds passively by bedding- parallel shear to the deformation of the basement (Figure 10). If the triple-junction geometry is such that the length of the basement surface decreases as the active axial 28

a

Footwa II-shear, Nw fault-bend anticline Displacement vectors

b

FIN Hanging wail-shear, fault-bend anticline

ION Displacement vectors Wts"64. FY/

Hanging wall-shear, fault-bend syncline Displacement vectors

Figure 9. Four types of folding that could result from of a fault-fault- axial surface(a-a') triple junction. The shaded areas are sheared by the triple junction and the angular shear is represented by w. Depending on the geometry of the triple-junction (TJ), either the hangingwall (HW) or the footwall (FW) can be sheared. S denotes the part of the basement that has been sheared and the region W will not be sheared except in (a). (From Nam 1990) 29

a

b

c

Figure 10. Kinematic development of a footwall- shear, fault-bend fold anticline (Narr, 1990). The triple junction (t*) moves along with the hangingwall parallel to the main fault at depth causing shear folding of the footwall. Stages a through c cause shortening in the cover sequence which is represented by shearing of the forward pin line. Stage d requires layer- parallel extension as new basement surface is created. Extension is represented by thinning of the beds in the forelimb. 30 surface sweeps through the footwall wedge (W) then the coversequence experiences net shortening. In this example, this shortening results in the forward pin line being displaced to the left at the cover-basement interface. If the lower fault dippedsteeper, then overall shortening would be less. Shortening occurs in the first stages of shear when the angle between the shear direction and the dip of the monoclinal fold limb is less than 90° (Figure 11). At 90° the basement surface length is a minimum. As shear continues, the basement surface length increases. This length increase is accommodated by either draping of thecover sequence by rolling through the active synclinal axis, or development of extensional features in the fold limb such as normal faults or thinned bedding (Narr, 1990). This kinematic model suggests progressive limb rotation during fold growth which is consistant with fanning dips of the Pliocene growth strata. Narr's footwall-shear, fault-bend anticline model (Figure 10) may be applicableto describing the kinematic evolution of the Las Cienegas anticline (Plates 5, 6, and 7). At the Murphy and Fourth Avenue drillsites, the Las Cienegas fault cuts into the Miocene and Pliocene cover sequence (Figure 10d). However, to the west at the Pacific Electric drillsite, the forelimb is not cut by the Las Cienegas fault and the Lower Member of the Puente Formation sits directly on schistose basement. This is analogous to the situation depicted in figure 10c. One problem with adopting the basement shear model is that there isno control on the dip of the shear plane within the basement. Without knowing the dip of the shear plane there is no way of calculating the shortening involved in structural growth of the monocline. However, the presence of the East Beverly Hills anticline suggests that perhaps the excess shortening that is moved away from the monocline along a detachment as shown in figure 8, is actually taken up by detachment folding on the limb of the monocline. The folding of the base of the Delmontian Stage strata accounts for 1800 ft (550 m) of shortening or approximately 23% of the total shortening due to folding. Again, however, without knowing the dip of the shear zone, there is no way to determine if all of the shortening is accounted for by folds within the cross sections. Basement Surface

Figure 11. Deformational geometry of thebasement surface. The shearzone dips 50°. During the first stages of shear (1), the lengthof the basement surface decreases until it attains a minimum at step (2) such that the dip ofthe limb and the shear plane makean angle of 90° and Lmin= cos(90-a) x L. Continuing shear increases the length again untilthe limb is as long as the original basement length (3). 32

GROWTH STRATA ANALYSIS

Introduction

As stated in the introduction to the paper, the variation in growth strata thicknesses reflects the structural growth rate. Specifically, if sedimentation rate exceeds growth rate, then the thickness of strata adjacent to a structure minus the thickness of growth strata over the crest of the structure equals the vertical component of structural growth (Figure 6). Conversely, if growth rate exceeds sedimentation rate and erosion has truncated the crest of the structure, the structural growth equals the thickness of growth strata adjacent to the structure plus the thickness of strata, growth or pregrowth, that has been eroded from the crest. Growth strata also contain a record of shortening due to folding. If we assume that cross-sectional bed lengths between two pin lines are conserved, then the bed length at the bottom of a sequence of growth strata minus the bed length at the top of the growth strata equals the horizontal shortening associated with the structural growth (Figure 12). The shortening recorded in the growth strata reflects the shortening due to folding.If the monocline is forming due to a shallow (20°-40°) dipping basement shear zone, then the amount of shortening accommodated by a detachment below the growth strata pin line can be significant. Thickness variation and bed length variation allow us to determine the timing and rate of growth of individual structures. Determination of both the horizontal and vertical components of structural growth allows the calculation of the dip of the fault responsible for folding. Because the estimate of shortening is a minimum, the resulting calculated fault dip will be a maximum. The Delmontian through Pleistocene sedimentary package that fills the Los Angeles trough thins dramatically to the north onto the crest of the monoclinal uplift, suggesting that the monocline was growing during deposition of these sediments. In the following section, I use the cross-sectional geometry of these growth strata to describe the timing of the formation of individual structures and to calculate the amount of growth, shortening, and therefore fault dips for both the East Beverly Hills and Las Cienegas areas. 33

Pin Lines Shortening a old bed length-young bed length. Relative subsidence a Trough thickness -Crest thickness Mir young bed length

Fault dipTan-1( Relative Subsidence) Shortening

Pre-Growth Strata

shortening 1

I4iI Fault i -

Figure 12. Calculation of shortening and fault dip. The faultdip calculation assumes that all of the shortening is accommodated by folding. This assumption may be invalid so the calculated fault dip isa maximum. 34

East Beverly Hills area

A cross section through the East Beverly Hills and South Salt Lake anticlines shows the dramatic thickening of the Lower Repettian through Pleistocene (Qmg) growth package (Figure 13; Plate 3). Shortening, growth, and fault dipwere calculated for six time intervals within the growth package as wellas for the package as a whole. Table la summarizes these calculations. Because we do not know the maximum thickness of the Delmontian Stage in the Los Angeles trough, itwas excluded from the calculations.

Structural Observations The gross structure at this location is a south vergent monocline that is complicated by two smaller folds: the East Beverly Hills anticline and the South Salt Lake anticline. Upper Delmontian Stage sandstones thin and lens-out onto both the East Beverly Hills and the Salt Lake anticlines, indicating that these two folds began to form at this time. The age of the top of the upper Delmontian is approximately 5.0 Ma (Blake, 1991), which suggests that compression began earlier than 2.2-4.0 Maas suggested by Davis et al. (1989). The East Beverly Hills anticline is interpreted as an out-of-the syncline or "rabbit-ear" fold as described by Brown (1988) and is directly tied to, and synchronous with, the formation of the synclinal lower fold of the monocline. The East Beverly Hills anticline influenced depositional thicknesses through the Upper Repettian Stage and became inactive by the Lower Pico Stage. The history of the South Salt Lake anticline is clouded by a time-transgressive unconformity that truncates the fold (Figure 13, inset box; Plate 3). The South Salt Lake anticline formed in response to reverse motion on the San Vicente fault. As stated above, Delmontian Stage sandstones lens out onto the south flank of the Salt Lake anticline, marking the beginning of its formation. The South Salt Lake anticline is erosionally truncated by a time-transgressive unconformity so that the exact timing of the end of motion on the San Vicente fault is unknown. The latest that the San Vicente fault could have been active was the Upper Pico as suggested by thickening of strata in the syncline between the Las Cienegas anticline and the South Salt Lake anticline (Plate 5). B East Reconstructid Base of Lower Repettian Stage NORTH Beverly Genesee Hills Packard ------Yavneh Dri ------EH-1 anticline CH-1 Oal alevel South Salt Luke anticline

0)/... ta i , r r / Luisian Growth r /' Topangs Formation Strata `/ / )% /,/ v I,/ % ,a, /, %.....ar ., %2 ._--- ., %w LA // Basement , //// 10 ....., / \ CD /? northern Los Angeles ,I /, / Y/ shelf

...... / I ? ,s0 Union Seibu . .-tx - CH-11 . . . op., I G. CH-1 Metro . . . . . Local Pln Rail -' M ohnian , 9. - .-° ... borings ..-- Pre-growth ...- ...... L .....--...- ...... 1 Striate ...-..- ...... ""' __,..- ...-...- ..- "'...... -... % -4- 9 1

0 1 kilometers no vertical exaggeration

Figure 13. Cross section through the East Beverly Hillsarea, showing the position of the pin lines used in the shortening calculations. The location of this cross section is shownon Plate 1. The base of the Repettian Stage strata is reconstructed into the air above and north of the South Salt Lake anticline. The baseof the Qmg and the Upper Pico strata are projected to the pin line in the Los Angeles trough and shown by dashed lines. 36

Growth Calculations The maximum thicknesses are constrained by the Unocal Genesee EH-1 in the synclinal axis of the Los Angeles central trough. Thicknesses in this wellare compatible with those shown in isopachs for the central Los Angeles trough(Yeats and Beall (1991). On the crest of the monocline, Pleistocenestrata unconformably overlie Upper Mohnian Stage strata so that the true minimum thickness of Pliocene growth deposition at this location cannot be determined. Therefore, thickness measurements were taken on the flank of the monocline at the South Salt Lake anticline (at the local pin lines shown in Figure 13, inset) andare considered to be maximum values for the minimum thicknesses. Because the minimum thickness values are not known, the calculated growth varies froma minimum (when the measured minimum thickness is used) to a maximum (when the minimum thickness is assumed to be zero). The maximum growth equals maximum thickness measured in the Los Angeles trough, because the minimum thickness iszero. Because the growth strata are thinning before being eroded, the assumption that these strata thin tozero is taken as the most valid for the growth and fault dip calculations. The unconformity at the crest of the structure is time transgressive. Middle Repettian Stage strata are truncated by and onlap this unconformity, suggesting that erosion began during this stage (Figure 13, inset). This erosion must be considered in the growth calculation. The growth that occurred during the Middle Repettian Stage equals the maximum thickness of growth strata in the Los Angeles trough (1500 ft) minus the thickness of growth strata over the structure (100 ft) plus the thickness that has been eroded (approximately 300 ft of Middle Repettian, 500 ft of Lower Repettian, and approximately 800 ft of Delmontian Stage for a total of 1600 ft within the unconformity). The Middle Repettian vertical component of growth is therefore 3000 ft. The growth strata are truncated by the Pleistocene strata on the south flank of the South Salt Lake anticline. Because of this, it is not possible to see whenor if the Pliocene unconformity stopped forming. Thus it is not possible to ascribe the amount of growth within the Pliocene unconformity to any specific time interval. Therefore the growth calculations for the individual time intervals should be considered minimum values. The only way to include all of the growth that is within the Pliocene unconformity is to consider the growth package as a whole. When considering the growth package as a whole, the amount of growth that is within the unconformity at the crest of the monocline depends on the amount that the growth strata are assumed 37

to thin. If the growth strata thin to zero over the top of the fold belt, then the unconformity only includes the amount of growth equal to the thickness of the Delmontian Stage strata, which are assumed to maintain constant thickness, plusthe thickness of Mohnian Stage strata that have been eroded. The other end member is the assumption that the Pliocene thickness measurement on the flank of thestructure represents the thickness of growth strata over the monocline so that the unconformity represents erosion of this thickness in addition to the Delmontian and Mohnian Stage strata. In either case, the total amount of growth is the same (Table la, bottom two rows); the only difference is the timing of growth. The Upper Pico is erosionally thinned to the north by the base of the Pleistocene. It is possible that the Upper Pico Stage strata did not thin to the degree observed for the Middle Pico. This would mean that growth during the Upper Pico could be much less than the value calculated when assuming that the thickness of these strata thinto zero over the structure. Because the calculation of growth is directly related to growth-strata thickness, it is important to consider the effect of compaction on resultant thicknesses. The sediment column was sequentially backstripped to obtain pre-compaction thicknesses of the growth strata using the program SUBSIDE (Hsui, 1989) which is basedon equations derived by Sclater and Christie (1980). The effect of decompaction is significant. Notice in Table la that the decompacted maximum thickness for the Upper Repettian Stage strata is 42% greater than the compacted thickness. Decompaction increases the maximum thickness of the entire growth package from 9165 ft to 12,272 ft (34%) but has relatively little effect on the thin strataon the flank of the fold belt.

Shortening The shortening calculations are shown in Table la. The basinal pin line is taken at the synclinal axis of the Los Angeles trough (Figure 13). Two pin lines were used on the flank of the monocline due to erosion by the Pliocene and Pleistocene unconformities. One is located on the south flank of the South Salt Lake anticline at the location where minimum thicknesses were measured. This pin line is used for individual time interval shortening calculations because bedlengths are unknown to the north where the units are eroded. The pin line for the Lower Repettian Stage is located below the unconformity (Figure 13, inset). A third pin line is located to the north to account for shortening associated with the South Salt Lake anticline. This pin Table 1: a) Shortening, Growth, and Fault Dip Calculations for the East Beverly Hillsarea Age Top Bottom ShorteningMaximum Minimum GrowthDip Dip if Decomp. Decomp. Decomp. Dip Dip if Length Length Thickness Thickness Min. =>0MaximumMinimum Growth Min. => 0 U. Pico 10.950 11,225 275 1650 375 1275 78° ':!:::: 4' 1772 394 1372 79° 81 M. Pico 11,225 12,050 825 2025 100 1925 66° 2494 100 2394 71° 472. L Pico 12,050 12,500 450 900 100 800 60° 12g° 100 1180 69° U. Rep. 12,500 14,100 i tr 1600 1890 190 1700 47° 5 2690 216 2487 57° 591# M. Rep 14,100 15,550 1450 1500 500/100 3000* 64° 2067 525/141 3526+ 68° tor L. Rep. 15,000 15,875 875 1200 700/0 500 30° 1969 804/0 1969 66° 66* Ave. 58* 64° Ave. 68° 70° U. Pico -18,300 26,000 7700 9165 1965 7200 55° ." 12,272 2139 10,133 61° L Rep. Within +3865 within +4.039 Unconfonn.11,065 Unconfonn. 14,172 . A Maximum 9165 Maximum 12,272 Growth Within ±.1290 Growth Within ±1.244 61, Calculation Unconfonn. 11,065 CalculationUnconfonn. 14,172 The Middle and Lower Repettian Stages are eroded at the location of the pin line. Therefore, the growthcalculation for the Middle Repettian includes 1600 ft of relative subsidence that is within the unconformity. For the U.Pico- L Repettian calculation the minimum thickness values are 700 ft for the L Repettian and 500 ft for the M. Repettian. + The Middle and Lower Repettian Stages are eroded at the location of the pin line. Therefore, the decompacted growth calculationfor the Middle Repettian includes 1600 ft relative subsidence that is within the unconformity. For the U.Pico- L Repettian calculation, the minimum thickness values are 804 ft for the L Repettian and 525 ft for the M. Repettian Stage. b) Shortening, Growth, and Fault Dip Calculations for the Las Cieneas area Age Top Bottom ShorteningMaximum Minimum GrowthDip Dip if Decomp. Decomp. Decomp. Dip Dip if Length Length Thickness Thickness Min. =>0Maximum Minimum Growth Min. => 0 U. Pico 16,050 16,450 400 2300 425 1875 78° A SG 2510 433 2077 79° M. Pico Sr 16,450 18,625 2175 3000 100 2900 53° 34° 3740 105 3635 L Pico 59° illr 18,625 19,575 950 875 50 825 41° 43.- 1345 72 1273 53° 5S U. Rep. 19,575 20,950 1375 1550 300 1250 42° 4* 2356 325 2031 56° 410- M. Rep 20,950 22,850 1900 1490? 290 1200 32° '311r 2106 308 1798 43° 4th` L. Rep. 22,850 24,550 1700 1625? 200 1425 40° 44' 2648 230 2418 55° $7 Ave. 48* 51° Ave. 58* ° U. Pico - 24,200 33,050 8850 10,840 1365 9475 55° 14,705 1473 13,232 62° L Rep. Within +3165 Within = +3.273 Unconfonn. 12,640 Unconfonn. 16,505 Maximum 10,840 Maximum 14,705 Growth = Within ±11-39-0 $ Growth Within +1.800 Calculation Unconfomi. 12,640 CalculationUnconfonn. 16,505 ... 39

line is used only to calculate the shortening of the growth packageas a whole. For this measurement, the base of the Lower Repettian Stage is projected into the air, assuming that it is parallel to the base of the Delmontian Stage anduncut by the San Vicente fault, such that folding takes up all of the shortening. The bed length measurements are based on the assumption thatno bedding parallel slip occurs through the pin lines. By placing the basinal pin line in the synclinal axis of the Los Angeles trough, we minimize the possibility of layer parallel slip through this line (Woodward et al., 1989). However the synclinal axis isnot planar. The axis moves to the north within the Middle Pico Stage (Figure 13). Because of this, the beds are projected to the south such that they intersect the axis formed by the Lower Pico -Lower Repettian Stage strata. Thereare no data to constrain the shape of the synclinal axis within the Middle Pico stageso this projection is a simplification. This reconstruction introduces minimalerror in the shortening calculation because of the low dip. The local pin line on the south flank of the South Salt Lake anticlinewas constructed perpendicular to bedding. The Middle Repettian through Upper Pico Stage strata are dipping 20°, so these strata have undergonean angular shear of 19° where [angular shear strain = tan (angular shear)] (Suppe, 1983). Thisamount of bedding-parallel slip is negligible for individual time intervals, and only introducesup to 150 ft (2%) of error in the shortening measurement for the whole growth package. The Lower Repettian Stage has been rotated to vertical, requiring 63° of angular shear. Thus, the pin line is not perpendicular to bedding but is inclined 63° from the perpendicular to bedding. The northernmost pin line is perpendicular to bedding because the bedsare dipping less than 10° over the crest of the structure, requiring an angular shear of 10°. The top of the Delmontian Stage strata are constructed into the air parallel to the base of the Delmontian. This construction allows for an estimation of the bedlength of the base of the Lower Repettian Stage strata.

Fault Dip None of the growth strata used in the growth analysis have been cut by a fault. Therefore all of the shortening is being taken up by folding, although it is possible that some shortening is taking place along a detachment below the growth and pre-growth strata. If we assume that folding is taking up all of the shortening, then we can measure the amount of shortening from the change in bed lengths.The shortening 40

represents the horizontal component of fault slip and growth represents the vertical component of fault slip. Thus, we can calculate the dip of the faultor shear plane over which the folding occurs (Figure 12). The fault dip calculations are shown in Table la. The shaded columns show the fault dip if the growthstrata thin to zero over the crest of the structure. The values average 64° for non-decompactionand 70° including decompaction. The values for the growth packageas a whole are the most valid because these calculations include all of the growth that is withinunconformities and the shortening associated with the South Salt Lake anticline. The faultdip ranges from 55° without decompaction to 61° including decompaction. This shallowerdip indicates that the South Salt Lake anticline accounts fora significant amount of shortening that is not incorporated into the individual time interval calculations because of the position of pin lines. Figure 14a illustrates cumulative shorteningversus cumulative relative subsidence (growth). These curves essentially show the path thata particle in the Los Angeles trough would take relative to a fixed particleover the crest of the monocline through time. The data points are cumulative in that the shortening and relative subsidence that occurred in the Middle Repettian Stage is addedto that of the Lower Repettian Stage. Therefore, the slope of a line from the origin throughany data point equals the dip of a deeply-buried fault causing the observed shortening. The different curves reflect the minimum and maximum growth for both the non-decompacted and decompacted thicknesses. Notice that the Middle Repettian Stage shows large relative subsidence relative to the later stages. This high growth rate is due to motionon the San Vicente Fault and growth of the South Salt Lake anticline relative to the rest of the monocline. Also notice the data points for the growth packageas a whole show only a little more additional growth but almost 2000 ft of additional shortening associated with the South Salt Lake anticline. This is because most of the relative subsidence is recorded in the Middle Repettian unconformity and was used in the Middle Repettian growth calculation, but none of the shortening was used. Cumulative Shortening (ft) Cumulative Shortening (ft) 0 ,2000 40006000 8000 10,000 Minimum Growth 0 2000 4000 6000 8000 10,000 oj Remontril r without 0 L Reoetttan 1 Decompaction

O Maximum Growth without q 4000 Decompaction U. Rethan

Minimum 6000 It Pico Decompacted Growth

O Maximum Decompacted Growth L1

10000 10000

12000 12000 4 -4 14000 14000

16000 16000 5km East Beverly Hills area 5km Las Cienegas area 18000 18000 Figue 14. Cumulative shortening vs. cumulative relative subsidence for the East Beverly Hillsand Las Cienegas areas. Growth equals the thickness of the growthstrata in the Los Angeles basin minus the thickness of of the monocline. growth strata on the crest Four points are shown for each timestep representing different assumptions of growthstrata thicknesses. Maximum growthoccurs if growth strata are assumed to thin tozero on the crest of the monocline. Minimum growth assumes that the measured minimumthickness continues onto thecrest of the monocline. These end members are also calculated using decompactedthicknesses. Shortening is calculated bysubtracting the bedlength of the top of a growth-strata interval from the bedlengthat the bottom of the interval.Shortening and relative subsidence for each time step are added to the previous timestep to form the cumulative curve. The growthpackage as a whole is also shown. This analysis assumes that all shortening istaken up by folding. The slope ofa line drawn from the origin to a data point equals the dip of a blind fault thatcould cause the observed folding. Thescatter in the data is partly due to errors in reconstruction and measurements. 42

Las Cienegas area

The cross section used for growth analysis of the Las Cienegasarea crosses the Las Cienegas anticline at the Unocal 4th Avenue drillsite and the San Vicentefault zone between the Unocal CH 29 and the Chevron Hobart CH (Figure 15).

Structural Observations The growth strata in the Las Cienegas area are structurally simpler than the East Beverly Hills area because there is no lower Pliocene unconformity. Thegross structure is a steeply-dipping south-vergent monocline that is complicated by the Las Cienegas anticline (Figure 15). The Las Cienegas anticlinewas active from the Delmontian Stage through the Lower Pico Stage as evidenced by stratigraphic thickening adjacent to the structure. Growth of the monocline during the Middle and Upper Pico Stage, after the Las Cienegas anticline became inactive, has eroded the crest of the fold. Two minor folds occur north of the Las Cienegas anticline; one at the location of the Union CH 29 and one just to the north at the Chevron Wilton CH #1. These folds have associated Delmontian and possibly Lower Repettian Stage growth strata; however the amount of growth is very small.

Growth Calculations The maximum thicknesses within the Los Angeles trough are constrained by the Chevron Dublin CH. The well only reaches the top of the Middle Repettian Stage strata so the Middle and Lower Repettian Stage strata are constructed with regional thicknesses. These thicknesses agree with Los Angeles trough isopachs as presented by Yeats and Beall (1991). Just as at East Beverly Hills, the growth strataare truncated by the base of the Pleistocene over the top of the monocline. At the Chevron Hobart CH, Pleistocene strata rest unconformably on Mohnian Stage strata. If the growth strata thinned to zero over the monocline, then this unconformity contains the amount of growth equal to the thickness of the Delmontian Stage strata (1000 ft) plus the thickness of Mohnian Stage strata (800 ft) that have been eroded (1800 ft; Table lb). If, however, the growth strata were deposited over the top of the monocline with the thicknesses measured at Union CH 29, then the unconformity also includes growth equal to the growth strata (1365 ft, Table lb). In either case the total amount of growth of the Las Cienegas anticline is approximately 12,600 ft, or 16,500 ft if decompaction is considered. Paramount \ U-14-1 Union Reconstructed Base of Lower Repettian Stage Standard Union CH-29 Dublin CH Reconstructed Fourth Ave. Union Wilton Pectic Telephone CH #1 eroded beds Drillske CH 3825 CH-1 L

U. Mohnian

A Las Cienegas San Vicente fault zone Luisian anticline Topanga Formation

/0 northern Los Angeles shelf

1000 ft

1 kilometer No Vertical Exaggeration

Pre -growh strata IFigure 15. Cross section through the Las Cienegasarea, showing the position of the pin lines used in Los Angeles the shortening calculations. The location of this cross section is different than cross section E-E' in trough that from the Fourth Avenue drillsite the section goes northeast to Union CH 29 and the Hobart CH.. The base of the Repettian Stage strata is reconstructed into the air above and north of the Wilton CH- 1. The base of the Qmg and the Upper Pico strata are projected to the pin line in the Los Angeles 41. trough and shown by dashed lines. w 44

Shortening Calculations The basinal pin line is located along the Delmontianto basal Middle Pico Stage synclinal axis. The synclinal axis shifts position to the north in theMiddle Pico through Pleistocene strata. For the purposes of measuring bedlengths,the top of the Middle and Upper Pico contacts are projected withconstant dip south to the basinal pin line (Figure 15). This simplification is neededso that the bedlength of the top of the Upper Pico Stage can be compared to the bedlength of the base of theLower Repettian Stage to derive the shortening associated with the growth packageas a whole. While this construction is a simplification, it has little effecton the measured shortening for the Upper and Middle Pico Stages because both bedlengthsare increased nearly the same amount. The shortening valuesare shown in Table lb. Figure 14b shows the cumulative shorteningversus cumulative relative subsidence curves from the growth data. The curve shows that the monoclineat the Las Cienegas area is growing fairly uniformly until the Middle Pico. At thistime there is a marked increase in shortening and relative subsidence. This increase in relative subsidence is also seen at East Beverly Hills but it isnot associated with as great an increase in shortening.

Fault Dip Calculations None of the growth strata used in the analysis are faulted,so the assumption that all shortening is accommodated by folding is valid, assuming that there isno decollement beneath the monocline. The calculated fault dipsrange from 48° to 81° with a mean of 60°, assuming maximum growth and decompaction ofstrata (Table lb). For the growth package as a whole the resultant fault dip is 62°. A fault dip of 62° is very similar to that calculated for the East Beverly Hills (61°) suggesting that the monocline at both localities is controlled by a single continuous fault at depth

Fault Slip and Shortening Rates

The slip rates for both areas are very similar. At East Beverly Hills, because most of the growth associated with the unconformity was during the Repetto,average slip rate can be determined for both the Repetto and Pico. The top of the Delmontian is age correlated at 4.95±0.15 Ma (Blake, 1991) and the top of the Repetto is estimated at 2.5 Ma (Blake, 1991), thus the Repetto lasted between 2.3 and 2.6 m.y. The 45

Repetto maximum decompacted growth was 8326 ft including 1600 ftwithin the unconformity and the shortening was 3925 ft resulting ina total fault slip of 9205 ft (2806 m). Therefore, the average Repetto fault sliprate (slip/time) is 1.1-1.2 mm/yr. The top of the Pico is age correlated to 0.9-1-0.1 Ma, thus the Picolasted 1.5-1.7 m.y. The shortening during the Pico was 1550 ft and the decompactedgrowth was 5846 ft (including 300 ft within the unconformity). Therefore, the total faultslip was 6048 ft (1843 m) and results in an average Pico fault sliprate of 1.1-1.2 mm/yr which is in remarkable agreement with the Repetto. At Las Cienegas, the growth associated with the unconformityon the crest of the monocline cannot be assigned to specific time intervals. Thereforeonly a Repetto- Pico average slip rate can be calculated. The decompacted growth is16,505 ft and the shortening is 8850 ft resulting in a net fault slip of 18,728 ft (5708 m). Fromthe age assignments given above, the Repetto-Pico time interval lasted between 3.8and 4.3 m.y. thus yielding an average fault slip rate of 1.3-1.5 mm/yr. Assuming that the shear zone extends into the ductile crust,as is suggested by earthquakes, the horizontal shortening rate (S)= fault slip rate x cos (dip of fault). At East Beverly Hills, the fault dip is 61° assuming maximum decompactedgrowth. Therefore, the horizontal shortening rate is .5-.6 mm/yr. At Las Cienegas, the fault dip is 62° and results in a similar horizontal shortening rate of .6-.7 mm/yr.

Summary

The Los Angeles trough was submarine during the entire growth packageexcept the Pleistocene. The basin was deepest during the Repettian Stage, approximately 5000 ft (1500 m), as evidenced by lower bathyal foraminiferal assemblages (Blake, 1991). During the Pico and Pleistocene, the basin shoaled tosea level. The East Beverly Hills area contains a minimum of 11,000 ft (3353 m) of growth strata, and the Las Cienegas area contains a minimum of 12, 500 ft (3810 m) of growth strata. Thus, approximately 12,000 ft (3658 m) of growth strata, minus 5000 ft (1524 m) due to shoaling of the basin results in approximately 7,000 ft (2134 m) of relative subsidence during the last 5.0 m.y. The East Beverly Hills area experienced 7700 ft (2347 m) of shortening between the Lower Repettian Stage and the Upper Pico Stage. The San Vicente faultcan only account for 2350 ft (716 m) of vertical separation or 21% of the 11,065 ft (3373 m) 46

(Table la) of structural relief shown by the Top of the Delmontian Stagestrata. This suggests that the majority of shortening and relative subsidence occurredon the Monocline fault. Similarly, the Las Cienegas fault shows approximately 2500ft (762 m) of vertical offset of the base of the Delmontian Stage strata whichaccounts for approximately 20% of the structural relief shown by the base of the Lower Repetto strata. Based on shortening values calculated for unfaulted horizons, the maximum dip of this fault or shear zone is 55°-62°. These valuesare a maximum because we do not know how much shortening has been accommodated by a horizontal detachment below the growth strata. The growth strata provide a detailed record of the structural evolution of the northern Los Angeles fold-and-thrust belt for the last 5m.y.. At approximately 5 Ma, a monoclinal flexure began to form at both the East Beverly Hills and Las Cienegas areas. The East Beverly Hills anticline began forming at the same time in response to monoclinal folding. Also, at this time, the Mohnian Stage Las Cienegas normal fault was reactivated as a reverse fault, as evidenced by the formation of the Las Cienegas anticline. The San Vicente fault north of Las Cienegaswas also active at this time, however, the amount of reverse motion was very small. At approximately 3.5 Ma, the San Vicente fault was reactivated north of the East Beverly Hills anticline, based on rapid growth of the South Salt Lake anticline. Growth exceeded the sedimentation rate such that at least 1600 ft (488 m) of the crest of the South Salt Lake anticline was eroded prior to the end of the Middle Repettian Stage. By approximately 2.5 Ma, the end of the Upper Repettian Stage, the East Beverly Hills anticline had stopped growing, but the monocline continued togrow based on presence of thick Pico growth strata. Similarly, the Las Cienegas anticline stopped growing by approximately 2.0 Ma (base of Middle Pico), signaling the end of Las Cienegas fault propagation. However, at the Las Cienegas area, there is a thick Middle Pico growth package indicating that growth on a deeper fault or shearzone continued after the Las Cienegas fault stopped. This history is shown graphically in Figure 16. The growth record for the Upper Pico is obscured by the unconformity at the base of the Pleistocene, however, the presence of the unconformity implies that growth occurred during this period. The fact that the base of the Pleistocene is folded attests to recent growth, although not necessarily on the same blind fault (Hummon et al., in press). Pleistocene Monocline fault Las Cienegas anticlinal folding Upper Pico Las Cienegas Reverse Fault East Beverly Hills Middle Pico anticlinal folding

Lower Pico

Upper Repettian Western San Vicente Middle Repettian revs fault Eastern San Vicente reverse fault Lower Repettian L Delmontian Western San Vicente Normal Fault Eastern San Vicente Div. "CID" Noma! Fault Las Cienegas Normal Fault c 0 Div. "E" 7)c a) Luisian x i1 W

Figure 16. Relative timing of structures in the northern Los Angeles basin. Boththe Las Cienegas fault and the San Vicente faults show reactivation. The Monocline fault isactive into the Upper Pico after the secondary structures have becomeinactive. 48

DISCUSSION AND CONCLUSIONS

Detailed subsurface structure contour maps have shown the northern Los Angeles basin to be underlain by a south vergent monocline that is complicated by secondary faults and folds. The monocline forms a structural shelf that marks the northern boundary of the Los Angeles central trough. Isopach maps show that during the Miocene, the predominant structural style was extension. Thick accumulations of volcanic and volcaniclastic rocks controlled by normal faults hada very different depositional pattern than during the Pliocene. At approximately the beginning of the Pliocene extension changed to compression resulting in the reactivation of the Miocene normal faults in a reverse sense and the beginning of the formation of the monocline and secondary structures. Thick growth sequenceswere deposited to the south of the growing monocline toward the Los Angeles central trough. Although a balanced solution to explain the kinematic evolution of the northern Los Angeles shelf was not reached, the timing and relative uplift rates of the monocline have been determined. The monocline, the East Beverly Hills, Las Cienegas, and East Beverly Hills anticline all started forming at approximately the beginning of the Pliocene. The Las Cienegas anticline and therefore the Las Cienegas fault stopped growing by the Upper Pico. The South Salt Lake anticline and the San Vicente fault were inactive by the Lower to Middle Repetto. The East Beverly Hills anticline stopped growing by the Lower Pico. Thick accumulations of Upper Pico growth strata attest to the continuing growth of the monocline and motion on the Monocline fault, after the secondary structures had stopped growing. The Pliocene-Pleistocene growth strata in the limb of the monocline exhibit gradually steepening dips. This geometry indicates that progressive limb rotation occurred during the growth of the monocline. This observation precludes the use of fault-bend or fault-propagation models of fold evolution as presented by previous investigators. Progressive limb rotation can be accomodated by a basement shear model. There is no evidence that the fault responsible for growth of the northern Los Angeles monocline shallows at depth into a horizontal detachment as presented by Davis et al. (1989).Davis et al. (1989) suggest that the basal detachment is at 13 km coincident with the regional base of seismicity at 10-15 km. However, the 1994 Northridge earthquake and aftershock sequence delineated a 40° dipping fault that extended to at least 18 km depth (unpublished California Institute of Technology, 49 seismology report, 1994). If indeed we see all of the shortening within the growth strata, and the fault dip of 60-61° is correct, then shallowing this steeply dipping fault into a subhorizontal decollement at 18 km would be improbable. Additionally, shortening above a subhorizontal detachment at the brittle-ductile transition would produce overall uplift of structures rather than the observed overall subsidence that occurred throughout the Pliocene. The fault slip rates for the Monocline fault are 1.1-1.2 mm/yr in the East Beverly Hills and 1.3-1.5 mm/yr. in Las Cienegas. The resulting horizontalconvergence rates are also similar, .5-.6 mm/yr and .6-.7 mm/yr. However, this value is much less than that calculated for the Pleistocene based on modeling the fault responsible for deforming the base of the Quaternary marine gravels (Schneider, 1993; Hummon, 1994). The Wilshire fault (see Appendix), as modeled, hasa Pleistocene horizontal convergence rate of 1.4-1.7 mm/yr and has a much shallower dip of 35°. The Hollywood basin, as mapped by the deformed base of the Quaternary marine gravels, truncates the pre-Quaternary structures (Plate 2). This implies that the fault responsible for the Wilshire arch is younger than the Monocline fault which formed the pre-Quaternary structure. The formation of the Wilshire fault may marka change in tectonic style within the Los Angeles from overall subsidence to uplift and shallow and non-marine deposition in The Pleistocene and Holocene. 50

BIBLIOGRAPHY

Atwater, T., Implications of plate tectonics for the Cenozoic tectonic evolutionof western North America, Geol. Soc. Am. Bull., 81, 3513- 3536, 1970.

Bally, A.W., P.L. Gordy, and G.A. Stewart, Structure, seismic data, and orogenic evolution of the southern Canadian Rocky Mountains, Bull. Can. Pet. Geol., 14, 337-381, 1966.

Barbat, W. F., The Los Angeles basin area, California, in Habitat of oil- a symposium, edited by L. G. Weeks, pp. 62-77, American Association of Petroleum Geologists, Tulsa, Oklahoma, 1958.

Blake, G. H., Review of the Neogene biostratigraphy and stratigraphy of the Los Angeles basin and implications for basin evolution, in, Active Margin Basins,pp. 135-184, edited by Kevin. T. Biddle, American Association of Petroleum Geologists Memoir 52, 1991.

Brown, W. G., Deformational style of Laramide uplifts in the Wyoming foreland, Mem. Geol. Soc. Am. 171, 1-25, 1988.

Campbell, R. H., R. F. Yerkes, Cenozoic evolution of the Los Angeles basinarea - relation to plate tectonics, Am. Assoc. Pet. Geol., Misc. Pap., Pac. Sec., 24, 541- 558, 1976

Dahlstrom, C.D.A., Balanced cross sections, Can. J. Earth Sci., 6, 743-757, 1969.

Davis, T.L., Namson, J., and Yerkes, R.F., A cross section of the Los Angelesarea: seismically active fold and thrust, the 1987 Whittier Narrows earthquake, and earthquake hazard, J. Geophys. Res., 94, 9644-9664, 1989.

Dibblee, Jr. T.W., Geologic Map of the Beverly Hills and Van Nuys (South 1/2) Quadrangles, Los Angeles County, California, Dibblee Geological Foundation Map #DF-3 I, Dibblee Geological Foundation, Santa Barbara, California, 1991a

Dibblee, Jr. T.W., Geologic Map of the Hollywood and Burbank (South 1/2) Quadrangles, Los Angeles County, California, Lucy E, Birdsall Memorial Map, Dibblee Geological Foundation Map #DF-31, Dibblee Geological Foundation, Santa Barbara, California, 1991a. 51

Dolan, J.F. and Sieh, K.E., Tectonic geomorphology of the northern Los Angeles basin: seismic hazards and kinematics of young fault movement, in, Engineering Geology Field Trips: Orange County, Santa Monica Mountains, and Malibu: 35th annual meeting of the Association of Engineering Geologists, Field trip guide, B20-B26, 1992.

Eldridge, G. H. and Arnold, R., The Santa Clara Valley, Puente Hills, and Los Angeles oil districts, southern California: U.S. Geol.S.urv. Bull., 309, 266pp., 1907.

Feigl, K.L., D.C. Agnew, Y. Bock, D. Dong, A. Donnellan, B.H. Hager, T.A. Herring, D.D. Jackson, T.H. Jordan, R.W. King, S. Larsen, K.M. Larson, M.H. Murray, Z. Shen, and F.H. Webb, Space geodetic measurement of crustal deformation in central and southern California, Eos, Trans., AGU, 74, 59, 1993.

Gardett, P. H., Petroleum Potential of Los Angeles Basin, in Future Petroleum Provinces of the United States-Their Geology and Potential, edited by I. H. Cram, pp. 298-308, Mem. Am. Assoc. Pet. Geol. 15,1971.

Gwinn, V.E., Thin-skinned tectonics in the Plateu and northwestern Valley and Ridge Provinces of the central Appalachians, Geol.Soc. Am. Bull., 75, 863-900, 1964.

Hanks, T. C., H. Kanamori, A moment magnitude scale, J. Geophys. Res., 84, 2348- 2350, 1979.

Hauksson, E., Earthquakes, faulting and stress in the Los Angeles basin, J. Geophys. Res., 95, 15,365-15,394, 1990.

Hsui, A.T., Computer program: SUBSIDE, Rockware, Inc., Wheatridge, Colorado, 1989.

Hummon, C., Subsurface Quaternary and Pliocene structures of the northern Los Angeles basin, California., unpublished M.S. thesis, Oregon State University, Corvallis, Oregon, pp. 109, 1994.

Hummon, C., C.L. Schneider, R.S. Yeats, J.F. Dolan, K.E. Sieh, and G.J. Huftile, The Wilshire fault: Earthquakes in Hollywood?, Geology, 22, 291-294, 1994.

Isaaks, C.M., Diagenesis in the Monterey Formation examined laterally along the coast near Santa Barbara, California, unpublished Ph.D. thesis, Stanford University, Palo Alto, California, pp. 329, 1980. 52

Jacobson, J. G. and R. G. Lindblom, Geology of the Beverly Hills Oil Field, in Petroleum geology of coastal southern California, edited by T. L. Wright,R. Heck, pp. 32-39, Pacific Section, American Association of PetroleumGeologists, Los Angeles, 1987.

Kamerling, M.J. and B.P. Luyendyk, Tectonic rotations of the Santa Monica Mountains region, Western Transverse Ranges, California, suggested by paleomagnetic vectors, Geol. Soc. Am. Bull., 90, 331-337, 1979.

King, G. C. P., Stein, R. S., and Rundle, J. B. The growth of geologicstructures by repeated earthquakes: 1. Conceptual framework, J. Geophys. Res., 93, 13,307- 13,318, 1988.

Kleinpell, R. M., Miocene Stratigraphy of California, 450pp., American Association of Petroleum Geologists, Tulsa, Oklahoma, 1938.

Kleinpell, R. M., The Miocene Stratigraphy of California revisited, 182, American Association of Petroleum Geologists, Tulsa, Oklahoma, 1980.

Lin, J., R. S. Stein, Coseismic folding, earthquake recurrence, and the 1987source mechanism at Whittier Narrows, Los Angeles basin, California, J. Geophys. Res., 94, 9614-9632, 1989.

Luyendyk, B.P., Kamerling, M.J., and Terms, R., Geometric model for Neogene crustal rotations in southern California, Geol. Soc. Am. Bull., Part 1, 91, 211-217, 1980.

Mefferd, M. G., Murphy Area of Las Cienegas oil field, Summary of Operations: California Oil Fields, 56, pp. 5-13, California Division of Oil and Gas, 1970.

Mitra, S., Geometry and Kinematic Evolution of Inversion Structures, Am. Assoc. Pet. Geol. Bull., 77, 1159-1191, 1993.

Narr, W., Deformational behavior and Kinematics of basement-involved structures and Joint spacing in sedimentary rocks, unpublished Ph.D. dissertaion, Princeton University, Princeton, New Jersey, pp. 154, 1990.

Natland, M. L., Pleistocene and Pliocene stratigraphy of southern California, PhD, pp. 165, University of California, Los Angeles, City, 1952. 53

Normark, W.R., H. Posamentier, and E. Mutti, Turbidite systems: state of the art and future directions, Reviews of Geophysics, 31, 91-116, 1993.

Obradovich, J.D., and C.W. Naeser, Geochronology bearing on the age of the Monterey Formation and siliceous rocks of California, in, The Monterey Formation and related siliceous rocks of California, edited by R.E. Garrison and R.G. Douglas, Special Publication, Pacific Section, Society of Economic Paleontologists and Mineralogists, p. 87-96, 1981. Pisciotto, K.A., R.E. Garrison, Lithofacies and depositional environments of the Monterey Formation, California, in, The Monterey Formation and Related Siliceous Rocks of California, edited by R.E. Garrison and R.G. Douglas, Special Publication, Pacific Section, Society of Economic Paleontologists and Mineralogists, 97-122, 1981.

Ponti, D.J., J.A. Barron, A.M. Sarna-Wojcicki, M.L. Cotton, C. Hummon, and C.L. Schneider, Benthic foram biostratigraphy and the age of the Pico Formation in the northern Los Angeles basin: Potential problems for evaluating activity of blind thrust faults, Eos, Trans., AGU, 74, 434, 1993.

Ramsay, J.G., Development of Chevron Folds, Geol. Soc. Am. Bull., 85, 1741-1754, 1974.

Redin, T., Oil and gas production from submarine fans of the Los Angeles basin, in Active Margin Basins, edited by K. T. Biddle, pp. 239-259, American Association of Petroleum Geologists Memoir 52, 1991.

Rich, J. L., Mechanics of low-angle overthrust faulting as illustrated by Cumberland thrust block, Virginia, Kentucky, and Tennessee, Bull. Amer. Assoc. Pet. Geol., 18, 1584-1596, 1934.

Rundle, J.B., Static elastic-gravitational deformation of a layered half-space by point couple stresses, J. Geophys. Res. 85, p.5355-5363, 1980.

Rundle, J. B., Viscoelastic-gravitational deformation by a rectangular thrust fault in a layered earth, J. Geophys. Res., 87, 7787-7796, 1982.

Schneider, C.L., C. Hummon, R.S. Yeats, and G. Huftile, Locating the Wilshire fault: elastic and non-elastic approaches, Eos, Trans. AGU, 74, 434, 1993.

Sclater, J.G., and P.A.F. Christie, Continental stretching: an explanation of the post- mid-Cretaceous subsidence of the central North Sea Basin, J. Geophys. Res., 85, 3711-3739, 1980. 54

Shaw, J.H., Active blind-thrust faulting and strike-slip fault-bend folding in California, unpublished Ph.D. dissertation, Princeton University, Princeton, New Jersey, 216 p., 1993. Sorensen, S.S., Petrology of basement rocks of the California Continental Borderland and the Los Angeles basin, Ph.D. dissertation, 423 pp., Univ. of California, Los Angeles, 1984.

Stein, R.S., and G.C.P. King, Seismic potential revealed by surface faulting: the 1983, Coalinga, California, earthquake, Science, 224, 869-872, 1984.

Stein, R.S., G.C.P. King, and J.B. Rundle, The growth of geologic structures by repeated earthquakes 2, Field examples of continental dip-slip faults, J. Geophys. Res., 93, 13,319-13,331, 1988.

Suppe, J., Geometry and Kinematics of fault-bend folding, Am. J. Sci., 283, 684-721, 1983.

Suppe, J. and D.A. Medwedeff, Geometry and kinematics of fault-propagation folding, Eclogae geol. Helv., 83, 409-454, 1990.

Suppe, J., G.T. Chou, and S.C. Hook, Rates of folding and faulting determined from growth strata, in Thrust Tectonics, edited by K.R. McClay, pp. 105-121, Chapman and Hall, London, 1992

Turner, D.L., Potassium-argon dating of Pacific Coast Miocene foraminiferal stages, Geological Society of America Special Paper 124, p. 91-129, 1970

Weigand, P.W., Middle Cenozoic volcanism of the Western Transverse Ranges, in Geology and Mineral Wealth of the California Transverse Ranges, edited by Donald L. Fife and John A. Minch, Mason Hill volume, Annual Symposium and Guidebook Number 10, South Coast Geological Society, Santa Ana, California, 1982.

Williams, G.D., C.M. Powell, and M.A. Cooper, Geometry and kinematics of inversion tectonics, in Inversion Tectonics, edited by M.A. Cooper, and G.D. Williams, Geol. Soc. Special Publications Number 44, pp. 3-15, London, 1989.

Wissler, S. G., Stratigraphic formations of the producing zones of the Los Angeles basin oil fields, in Geologic formations and economic development of the oil and gas fields of California, edited by 0. P. Jenkins, pp. 209-234, California Division of Mines and Geology, 1943. 55

Woodring,W.P. and M.N. Bramlette, Geology and paleontology of the Santa Maria district, California, U. S. Geol. Surv. Prof Pap. 222, 185 pp., 1950.

Woodward, N.B., S.E. Boyer, and J. Suppe, Balanced Geological Cross-Sections: An Essential Technique in Geological Research and Exploration, AGU Short Course in Geology, 6, 1989.

Wright, T.L., Structural geology and tectonic evolution of the Los Angeles Basin, California, in Active Margin Basins, edited by K.T. Biddle, p. 35-134, American Association of Petroleum Geologists Memoir 52, 1991.

Yeats, R.S., Rifting and rafting in the southern California borderland, in Proceedings of conference on geologic problems of the San Andreas fault system, edited by W.R Dickinson and A. Grantz, Stanford University Publications, Geological Sciences, 11, 307-322, 1968.

Yeats, R.S., Newport-Inglewood Fault Zone, Los Angeles Basin, California, Bull. Am. Assoc. Pet. Geol.,57, 117-135, 1973.

Yeats, R.S., Changing tectonic styles in Cenozoic basins of southern California, in Cenozoic basin development of coastal California, edited by R. V. Ingersoll, W. G. Ernst, pp. 284-298, Prentice-Hall, Inc., Englewood Cliffs, NJ, 1987.

Yeats, R. S., and J. M. Beall, Stratigraphic controls of oil fields in the Los Angeles basin: a guide to migration history, in Active Margin Basins, edited by K. T. Biddle, pp. 221-237, American Association of Petroleum Geologists Memoir 52, 1991.

Yerkes, R. F., et al., Geology of the Los Angeles basin, Californiaan introduction, U. S. Geol. Surv. Prof Pap. 420-A, 57 pp., 1965. APPENDIX 56

APPENDIX

LOCATING THE WILSHIRE FAULT: ELASTIC AND NON-ELASTIC APPROACHES

Introduction

The Wilshire arch in the northern Los Angeles basin is defined by the deformed base of the Quaternary marine gravels which are estimated at 0.8-1.0 Ma (Figure A.1; Hummon, 1994). The arch is generated by the Wilshire fault,a blind reverse fault which is beneath subsurface well control; thus its location can be determined only by indirect means. To understand the seismic hazard associated with this fault, I estimate the size, depth, and slip rate of the Wilshire fault based on both non-elastic, geometric reconstruction and elastic dislocation methods. The non-elastic methodassumes parallel folding, conservation of cross-sectional area, constant thickness and bedding length, and non-elastic behavior of the upper crust. Two theories of foldingare consistent with these assumptions: fault-bend (Suppe, 1983) and fault-propagation folding (Suppe and Medwedeff, 1990). The elastic dislocation method involves matching the observed surface deformation with that produced by a dislocationor fault in a homogeneous elastic medium (Rundle, 1980, 1982; Stein and King, 1984). It also assumes the deformation resulting from one seismic event can be repeated to form the cumulative geologic structure (King, Stein and Rundle, 1988, Stein, King, and Rundle, 1988). A model of the Wilshire arch needs to match the geometric shape of the observed deformation. In cross-section the Wilshire arch is symmetrical in that both limbs dip nearly equally. The fault-normal (north-south) wavelength is 10 km, the along-strike 1/2 wavelength is 8 km and the total amplitude is 400m. Microseismic events below the Wilshire arch appear to delineate a plane dipping 28-40° toward N18E which may illuminate the fault. A further constraint on an acceptable model is the requirement that the width of the fault be equal to or greater than the total net slip required to form the fold. 34°7'30" kilometers Hollywood k oil well with dip direction(s) basin . water well with e-log SantaMonca Mountans B'. x LA Metro Rail boring e tiourJOus'U01% Ae topographic fault scarp ***"" Wilshire arch I feu 0 -50 -100

-so

.41111111k1q1Iry A19-/56,100 4. -2S0k0S A 091)"9e, es Baldwin . ItS \ ".411 34° 00' 18° 30' (Beverly Hills 7.5' quad) 118° 22' 30" (Hollywood 7.5' quad) 118° 15'

Figure A.1. Structure contours and dip data showing deformation of the base of Quaternary marine gravels (Hummon, 1994). The base of Qmg is correlated using electric logs (oil wellsand water wells D), and from Metro Rail borings (x). The stippled areas on the contour map are interpreted as faults scarps by Dolan and Sieh (1992). The dashed lines A-A' and B-B' locate the cross- sections shown in Figures A.2 and A.10 respectively. LBTP denote the location of the La Brea Tar Pits 58

NON-ELASTIC MODELS

Fault-Bend Fold

Fault-bend folds are formed when motion along a fault forces rock arounda bend in the fault surface. A fault-bend fold requires that the fault surface exists beforeany slip occurs and that slip is constant along the fault. The fault-bend fold model of the Wilshire arch is shown in the figure A.2. The base of the Pleistocene marine gravels shown in the fault-bend fold model is based on points taken along line A-A' shown in Figure A.1. These points were fit to straight lines to form dip panels. The model assumes that motion is within the plane of the section. In this solution, the backlimb length is less then the ramp length and thus equals total slip (Figure A.2). This is a minimum slip solution and results ina net slip of 1.5 km and a slip-rate of 1.5-1.9 mm/yr. The maximum slip solution results when the length of the backlimb represents the total length of the ramp. The total displacement equals the distance from the base of the ramp to the position where the forward anticlinal axis intercepts the fault. The maximum slip solution results ina net slip of 6.3 km and prdicts a slip rate of 6.3-7.9 nun/yr. Because geodetic estimates of the crustal shortening rate across the Los Angeles basin, which includes numerous structures that accommodate shortening, in the direction the the cross section is 5± 1 mm/yr (Feigl, et al. 1993), the minimum slip solution is regardedas the most valid.

Fault-propagation fold

As slip tapers to zero at the tip of a propagating fault, the strain must be accommodated by folding assuming that there is no layer parallel shortening. Folds of this type are termed fault-propagation folds (Suppe and Medwedeff, 1991). A fault- propagation model is not consistant with the structure of the Wilshire arch for several reasons. First, the Wilshire arch is symmetric and the fault-propagation model predicts an asymmetric fold. Second, the shallow dip of the limbs predicts a fault tip at a depth of approximately 21 km. The maximum depth of seismic events in the basin appears to be 18-20 km with very few events beneath 20 km (Hauksson, 1990). If 20 km is the brittle-ductile transition then it would be inappropriate to apply a brittle fault-propagation model below this depth. Los Angeles Hollywood trough axis Basin A Wilshire Arch 0 km

Control point from structure contour map Net Slip = 1.5 km Fault surface Slip Rate = 1.5 -1.9 mm/yr I Wern.M.*Approximate lower limit of well control al im:Pleistocene marine gravels No Vertical Exaggeration

Figure A.2. Fault-bend fold model of the Wilshire arch. Thecross section was constructed from the structure contour map of the base of Quaternay marine gravels along line A-A' shown in Figure A.1. Thepoints on the cross section indicate where the cross-section line crosses a contour-line. The points were fit to straight linesto form dip panels. The dip of the backlimb (approximately 15°) reflects the dip of the fault. The length of the backlimb reflectsthe net fault slip (1.5 kin). 60

ELASTIC MODELS

The elastic dislocation model consists of a simple elastic halfspace in which the free surface represents the base of Pleistocene marine gravels. The elasticparameters are Poisson's Ratio (.25) and Young's Modulus (2.5 x 1010 Nm-2) from King, Stein, and Rundle (1988). Introduction of a displacement dislocation into this halfspace produces deformation of the free surface. The dislocation parameters (Figure A.3)are: burial depth of the fault tip, fault dip, amount of slip, and downdip length (width). The width is defined as the length of fault area along which slip (seismicor aseismic) occurs. Burial depth and width have the most influence on the resultingstructure. An increase in burial depth results in an increase in wavelength anda decrease in amplitude (Figure A.4). An increase in width results inan increase in both wavelength and amplitude (Figure A.5). King, Stein and Rundle (1988) showed that small ( +1- 10°) changes in dip have little effect on the shape of the resultant deformation. The relationship between burial depth and width affects thecross- sectional symmetry of the structure. A fault at the surface will cause the most asymmetry. When the burial depth approaches the width of the fault the resulting structure becomes more symmetric. A further constraint on an acceptable model is the requirement that the width of the fault be equal to or greater than the total net slip required to form the fold. I use 1.0 m of coseismic slip in the models because thatwas the coseismic slip during the 1987 (ML=5.9) Whittier Narrows earthquake (Lin and Stein, 1989), which occurred on an analogous north-dipping blind fault in the northern Los Angeles basin, the Elysian Park thrust (Davis et al., 1989). The amount of slip only affects the amplitude and not the wavelength.

2D Elastic models

The 2D program "2D.for", (written by Sergio Barrientos and Ross Stein, unpublished fortran programs, used with permission from the United States Geological Survey),calculates the vertical displacement of the free surface caused by slip on a dislocation (Figures A.4 and A.5), assuming that the dislocation is infinite along strike. Figure A.6 shows the range of possible 2D solutions. The net slip is determined by dividing the amplitude of the Wilshire arch (400 m) by the amplitude of deformation produced during each 1.0 m slip increment (which are shown by the 61

Free Surface

aortal Depth

01100*11011 Or fault Dip aidth

Poisson's Ratios IS Yormsts Alockaus =ZS x1010 Um41

Figure A.3. Dislocation Parameters used in elastic dislocation models. The width of the fault is defined as the length down dip over which slip seismic or aseismic occurs. Elastic Parameters used in the modelsare Poission's Ration = .25 and Young Modulus = 2.5 x 1010 Nm-2. 62

Vertical Displacement (cml 1.0 m of slip on fault 40 Fault Burial Depth = 2.0 km

30

20 Fault Burial Depth = 4.0 km 10

0 cm 0 km -5 -2 Fault {Width = 2.0 4 Dip = 45° -6 Depth 0 km 10 15 20

Figure A.4. Effect of dislocation burial depth on free surface deformation. Increasing depth results in a decrease of amplitude and an increase in wavelength.

Vertical Disolacement (cm) 1.0 m slip on fault 40 I Width= 4.0 kmI 30

20 LI.Nidth = 2.0 kill 10

0 cm 0 km -5 =MM. 2

Burial Depth = 4.0 4 Dip = 45° Fault `s. 6 Depth 0 km 5 10 15 20

Figure A.5. Effect of dislocation width on free surface deformation. Increasing depth results in an increase of amplitude and wavelength. 63

Vertical Displacement (cmi 1.0 m slip on fault 40

30

20

10

0 c 0 km -5

-6 Depth 0 km 5 10 15 20 Distance 1.0 m slip on faultX 400 m total amplitude Net Slip uplift (m) Minimum Net Slip = 1.1 km Maximum Net Slip - 1.8 km

Figure A.6. Minimum and maximum two dimensional elastic dislocation models of the Wilshire fault. The minimum burial depth (2 km) is controlled by oil wells. This solution results in a net slip of 1.1 km. The maximum burial depth solution results in the minimum coseismic amplitude and thus the maximum slip solution (1.8 km). 64 models). The fault is modeled with a dip of 35°so that it is consistent with the dip of the observed zone of microseismicity. The burial depth of the fault tip controls the amplitude of the resultingstructure. Increasing the burial depth decreases the resultant amplitude. Therefore, the minimum slip solution results from the shallowest burial depth which is constrained by oil wells to be below 2.0 km. Conversely, the maximum slip solution results from the maximum allowable burial depth which is 2.8 km. If the fault dip is lower,the wavelength of the fold becomes too great. The range of fault widths is 2.0-3.0 km. These end members yield a range of possible net slips from 1.1-1.9 km. The zone of microseismicity beneath the Wilshire arch begins ata depth of approximately 3 km. The depth suggests that the best fit solution is the maximum depth solution. This model shows that every meter of slip generates approximately 17 cm of uplift and 4 cm of subsidence adjacent to the to the fold for a total amplitude of 21 cm. Therefore, generation of the Wilshire arch (amplitude of 400 m) requires 1900 events like the one above or 1.9 km of net slip. This solution yields a slip rate of 1.9- 2.4 mm/yr based on the 0.8-1.0 Ma age estimate of the base of the Quaternary marine gravels. The 2D solution gives a first estimate of the fault geometry. However, the 2D model assumes an infinate fault along strike and the contourmap shows that the fold flattens to the west suggesting that the fault ends. Also, to estimate the seismic hazard we need to determine the maximum size of the fault in three dimensions.

3D Elastic Models

Thre dimensional modeling of the Wilshire arch uses the same methodology as in 2D modeling except that the along strike length of the fault must be considered. Figure A.7 shows the 3D deformation produced by the maximum burial depth solution shown in figure A.6 with an along-strike length of 9 km (produced using the unpublished fortran program "ando.for" written by Masataka Ando and Grant Marshall and used with permission from the United States Geological Survey). The deformation is shown in map view and the contours represent lines of equal uplift or subsidence in centimeters. This model has an amplitude of 20 cm resulting from 1 m of dip slip on the fault. The amplitude is a bit lower than the 2D model because the fault is not infinate along strike. This model requires a net slip of 2.0 km. The solution is symmetrical in plan view in that the strikes of the forelimb and 65

Net Slip = 2.0 km Burial Depth = 2.8 km Width = 2.0 km Length = 9.0 km .. '"1-'1.... A Dip = 35° Total Amplitude= 20 cm

0 10 km Contour Interval = 5 cm

Figure A.7. Three dimentional elastic dislocation model for thesame fault geometry as the 2D maximum slip solution shown in Figure A.6. This figure showsa map view of the deformation associated with 1.0 m slip on the fault. Thecontours are lines of equal uplift or subsidence in centimeters. The surface projection of the blind Wilshire faultis shown by the shaded box. The deformation is symmetrical and results formusing dip slip alone. The net slip is greater than the 2D model because the fault isnot infinate along strike. 66 backlimb are parallel. The symmetry of this model results from theuse of dip slip only. However, the observed deformation is asymmetrical. The Hollywood basinand the Los Angeles trough axis are not parallel, but rather, forman angle. Therefore, we must consider the possibility of oblique slip. Figure A.8 shows that the asymmetry of the Wilshire archcan be modeled by including a right-lateral slip component on the Wilshire thrust. The 50° angle between the Hollywood basin and the Los Angels trough requiresa right-lateral slip component equal to 1.1 times the dip slip component. This model, with thesame fault geometry as the dip slip solution, uses 1.0 m of dip slip and 1.1 m of right-lateral slip resulting in 1.5 m right-lateral oblique slip. However, inclusion of the right-lateral slip component inreased the wavelength of the deformation from 10.0 km to approximately 12 km (Figure A.8), thus a smaller fault width must be considered. The best-fit right-lateral slip solution uses a fault width of 1.8 km to match the 10 km wavelength of the Wilshire arch (Figure A.9). The best-fit model dips 35° and has a fault tip depth of 2.8 km to match the observed microseismicity and results in a net oblique slip of 2.6 km. Because we seek to use this model to determine slip rates of the Wilshire fault, the right-lateral oblique slip component must equal 1.0m (the characteristic coseismic slip from the 1987 Whittier Narrows earthquake). Therefore, the dip slip component for the best fit model is 0.67 m and the right-lateral slip component is 0.74 m. The amplitude of this best-fit solution is 15.6 cm. Thus approximately 2600 one-meter oblique-slip earthquakes of the type modeled in figure A.9 would be required to form the 400 m amplitude of the Wilshire arch. Therefore, 2.6 km of right-reverse oblique slip has occurred on the best-fit Wilshire faultover the past 0.8-1.0 m.y., yielding a slip rate of 2.6-3.2 mm/yr. If the microseismicity is ignored than a minimum slip solution resulting from a minimum fault tip burial depth of 2.0 km results in a net oblique slip of 1.4 km. This minimum slip solution results in an oblique slip rate of 1.4-1.8 mm/yr. The principal stress direction in the northern Los Angeles basin is approximately 013° (Hauksson, 1990). Therefore, the Wilshire fault must strike at least 103° or more to the south to allow a right-lateral oblique solution. This orientation is shown in figure A.9. Dip Slip Component= 1.0 m Right-Lateral Slip Component= 1.1 m Net Slip=2.4 kml Oblique Slip = 1.5 m

Wavelength= 12 km

Too Wide !!1

Burial Depth = 2.8 km 0 Width =2.0 km Dip = 35° Contour Interval = 5 cm Total Amplitude = 25 cm

Figure A.B. Three dimensional elastic dislocation model of the Wilshire arch ausing right-lateral strike slip component.. The 50° angle between the Hollywood basin and the Los Angeles central trough requiresa right-lateral slip component equal to 1.1 times the dip slip component. The resulting oblique slip= 1.5 m. The right-lateral slip component increases the wavelength to approximately 12 km. 68

10 km , ,. N al -1 Hollywood'` / basin \ Net Oblique Slip = 2.6 km / -I 1 -ar I ) +1

f Oblique Slip Rate = fault geometry: 2.6 - 3.2 mm/yr dip 35° NE lc? strike N75°W rake 42° .0 width 1.8 km length 9.0 km -re/ depth2.8 km o

Figure A.9. Best fit three dimensional elastic dislocation model of the Wilshire arch. Reducing the width of the fault from 2.0 to 1.8 km reduces the wavelength to 10 km, in agreement with the observed deformation. The slip components were normalized to an oblique slip value of 1.0 m. The principal stress direction is approximately 013°. Therefore, to allow a right-lateral slip component, the fault is oriented striking 105°. 69

DISCUSSION AND CONCLUSIONS

Using the fault parameters determined from the dislocation modelingwe can estimate the seismic hazard associated with the blind Wilshire fault. Themoment magnitude (Mw) of a possible Wilshire fault earthquake is 5.7, basedon Mw=(2/3 log M0)-10.7, where M0=1.41A, p=shear modulus of elasticity (3x 1011 dyne/cm2), u=coseismic slip, and A= area of slip (Hanks and Kanamori, 1979). Forour calculation, y =1.0 m (from the analogous 1987 Whittier Narrows earthquake;Lin and Stein, 1989),A =1.8 km x 9.0 km=16.2 km2. The moment magnitude is not very sensitive to the amount of coseismic slip. If the coseismic slip were 0.5 m the resulting moment magnitude would be 5.5. If the coseismic slip were 2.0 m the moment magnitude would be 5.9. Also, the best-fit solution was the maximum slip solution for a 35° dipping fault. The minimum slip solution, in which the fault width was 3.0 km, would result ina moment magnitude of 5.9. For this fault geometry, varying the coseismic slip from .5-2.0m results in a moment magnitude range of 5.7 - 6.1. Therefore, given the range of acceptable 3D elastic models and the uncertainty in the coseismic slip, the Wilshire thrust could produce an earthquake with a moment magnitude ranging from 5.56.1. Both elastic and non-elastic models have been considered. The fault-bend fold model represents the maximum net-slip solution due to its lack of elastic strength, whereas the 2D elastic solution represents the minimum. Figure A.10 compares the cross sectional geometry of the Wilshire thrust solutions. The elastic model utilizes oblique slip and therefore results ina greater net slip (2.6 km) than the non-elastic dip slip solution (1.5 km). However, the amount of fault normal horizontal shortening is the same for both models; 1.4 km. However, microseismictiy suggests that the fault dips greater than than the 15° suggested by the non-elastic fault-bend fold solution. On the basis of an 0.8-1.0 Ma estimatedage of the base of the Quaternary marine gravels, the resulting fault-normal shortening rate for both models is 1.4-1.8 mm/yr or 28-36% of the total horizontal shorteningacross the Los Angeles basin as suggested by geodetic data (Feigl, 1993). A limitation of this study is the fact that the effect of other known active structures within the field area, such as the Newport-Inglewood fault, the Hollywood fault and the MacArthur fault, have been ignored. The principal reason these structures were disregarded was to simplify the elastic model. Also, these structures Hollywoodnorth Santa B Wilshire arch basin \ B' Oa! ,--mZin 0 km r..- tstca 7/./7/Z/Zow Oal z base- Omg /w,,t 0 ment

1 1

1.=

2

Oil Wells

3 Best Fit Elastic Dislocation Model Net Oblique Slip = 2.6 km Shortening Rate = 1.4-1.8 mm/yr

44.'"t40 44 4 * likok Fault-Bend Fold Model Net Dip Slip = 1.5 km Shortening Rate = 1.4-1.8 mm/yr

Qmg - Quaternary 2 Alluvium 9 1 V AQmg Quatemary kilometers marine gravels No Vertical Exaggeration

Figure A.10. Comparison of elastic and non-elastic solutions. The location of thecross section is shown in Figure A.1. The shaded area shows the location of the north-dippingzone of microseismicity. The horizontal shortening rates for the two models are equal. 71

are much larger than the modeled Wilshire fault. Therefore, the deformation associated with these structures would have a much larger wavelength than that observed for the Wilshire arch. These models are end members. The fault-bend fold modelassumes that the rock has no elastic strength. A weakness of the non-elastic solution is that it istwo dimensional and thus, can only examine the dip slip component. Also, itassumes the rock has no elastic strength. The weakness of the elastic solution is that the rocksare assumed to have perhaps too much strength in that elastic behavior is assumedeven after the elastic limit has been reached. Another shortcoming is that the elastic model predicts a rootless fault; a fault that does not continue at depth. This is problematic because the fault would have to maintain the same geometry and notpropagate throughout the evolution of the fold. The truth probably lies somewhere between these end members. The strength of the elastic model is that it allowsus to model the length of the fault and to investigate the possibility of oblique slipon the Wilshire thrust. An oblique slip component on the Wilshire fault suggests that the right-lateral slip on the Whittier fault may be accommodated to the northwest by right-lateral, oblique reverse faults. (Figure A.11). 72

118° 30' 118° 00' RHF

WN rier V"- Possible Wilshire 34° thrust Earthquake -00' Mw = 5.7 PscHl© Caosn Loy Angoso

33° Long 45' Beach

0 10 20 kilometers

Figure A.11. Location and focal mechanism of the possible earthquake on the Wilshire fault. The oblique-slip solution of the elastic dislocation model suggests the Wilshire fault might accommodate right-lateral slip on the Whittier fault (WF).