UNIVERSITY OF CALIFORNIA
RIVERSIDE
Tectonic Geomorphology of Coastal Mountain Ranges along a Transform Plate Boundary: Geomorphic Evolution of Fluvial Terraces with Implications for Defining Rates of Crustal Displacement and Earthquake Recurrence Intervals
A Dissertation submitted in partial satisfaction of the requirement for the degree of
Doctrine of Philosophy
in
Geological Sciences
by
Patrick Eugene Smith
March 2009
Dissertation Committee: Dr. Alan E. Williams, Chairperson Dr. Lewis A. Owen Dr. David D. Oglesby Dr. Tin-Chang Lee
CHAPTER 1
INTRODUCTION
1.1. Introduction
Throughout the last century earthquake scenarios in southern California have focused on the possible effects of a great earthquake generated by the San Andreas fault. The Whittier Narrows (1987 Mw 6.0) and Northridge (1994 Mw 6.7) blind thrust faults in Los Angeles, California, have clearly demonstrated the seismic hazards associated with urban faults other than the San Andreas fault (Fig. 1).
Due to the continued urban expansion in the Los Angeles and Orange County areas, it is now generally accepted that moderately large to large earthquakes (Mw 7.0 to 7.5) may cause at least as much, and possibly more damage than a much larger earthquake occurring on the San Andreas fault. From the triple junction near Mendocino, California to the triple junction near Puerto Vallarta, Mexico, a series of transform faults dissect the West Coast of the North American continent causing Baja California and West Coast California to be located on the Pacific plate rather than the North American continental plate. Current global positioning system (GPS) measurements indicate that the Pacific plate is moving past the North American plate at about 56 mm/yr, but the rate of motion on the San Andreas fault (the primary transform fault striking through California) is only about 35 mm/yr (Keller and Pinter, 1996;Sieh and Wallace, 1987; Sieh, 1994). Ultra Long Baseline measurements indicate that 11 mm/yr of convergent movement is accommodated in the Eastern California Shear Zone (Argus and Gordon, 1991; Jordan and Minster, 1989). The remaining 10 mm/yr (18%) of movement is probably taken up by deformation west of the San Andreas fault. Current work in southern California is focused on establishing slip rates and motions for faults west of the San Andreas fault system. For example, Dolan et al. (1997) uses a rate of 0.35 mm/yr of oblique northward movement on the Hollywood fault. Rockwell et al. (1992) establishes a slip rate of 1.07 mm/yr for the Rose Canyon fault in San Diego. Grant et al. (1999) suggests that 0.42-0.79 mm/yr of movement may be occurring on the blind thrust faults beneath the San Joaquin Hills. The movement across associated smaller faults such as the Cristianitos and Mission Viejo faults is unknown but could in total be as significant as the larger faults in this area.
1.2. Research Problem
The research problem is to determined slip rates and general recurrence intervals for the Cristianitos and Mission Viejo faults by mapping, correlating, and dating fluvial terraces of the San Juan Creek drainage basin. Such information will represent a pertinent addition to the understanding of southern California seismicity, and an aid to understanding the total western U.S. tectonic regime. An identification of faults in southern California that are currently taking up the remaining 10 mm/yr of motion is critical both for establishing a working model of plate motions and fault activity throughout this region, and for establishing seismic risk and hazard parameters based on projected earthquake magnitudes and recurrence intervals for faults occurring within this densely populated region.
1.3. Hypothesis
This research focuses on the geomorphic evolution of terraces along the San Juan Creek drainage basin. the research is based on the hypothesize that terraces throughout the San Juan Creek study area can be correlated into former longitudinal stream profiles. Further, it is hypothesized that a variety of dating techniques can establish the chronology of the former longitudinal stream profiles, including correlation with marine terraces, correlation with the glacio-eustatic sealevel fluctuation curve, and CRN dating. Finally, it is hypothesize that an analysis of the former longitudinal stream profiles with their chronology can reveal and assess seismic slip rates and recurrence intervals of the Cristianitos and Mission Viejo faults, and determined rates of crustal displacement within this coastal regime. Understanding rates and processes of landscape evolution in this portion of southern California is essential for developing an understanding of processes relating to tectonic geomorphology of coastal mountain ranges along a transform plate boundary.
1.4. Project Objectives
The objectives of this project are as follows:
- Establish detailed fluvial terrace maps for portions of the San Juan Creek drainage area within the 3 detailed study zones proposed in this study area.
- Correlate the fluvial terraces throughout the study area into former longitudinal stream profiles.
- Date the fluvial terraces and former longitudinal stream profiles throughout the San Juan Creek drainage basin and assess the amount of fault displacement on the terraces and profiles.
- Calculate rates of crustal displacement for the Cristianitos and Mission Viejo faults.
- Calculate recurrence intervals for earthquakes on the Cristianitos and Mission Viejo faults.
- Define the geologic and geomorphic history of the San Juan Creek study area to establish a model for this active coastal mountain range.
1.5. Significants
Throughout the last century, researchers in southern California have focused almost exclusively on the earthquake scenarios and seismic hazards associated with the San Andreas fault which takes up a majority of the motion between the North American and Pacific plates. Yet, identification of the faults in southern California that are currently taking up the remaining 18% of motion west of the San Andreas fault is critical for establishing a working model of plate motions and fault activity along transform plate boundaries, as well as for establishing seismic risk and hazard parameters in this densely populated area. Establishing the seismic slip rates and recurrence intervals of the Cristianitos and Mission Viejo faults is important because this fills a critical gap between research on faults located in Orange County (Dolan et al., 1995; Grant et al., 1999) and faults located in San Diego County (Rockwell et al., 1992; Eisenberg, 1992). This information is necessary for a more complete understanding of the seismic hazards still remain in Southern California. The research also establishes a geomorphic history for fluvial terrace formation using constructed former longitudinal stream profiles. Ultimately this research will provide an increased understanding of rates of crustal displacement, earthquake recurrence intervals, and rates of geomorphic evolution that will aid in ascertaining a more complete understanding of seismic interactions and motions along an actively deforming transform plate boundary.
1.6. Setting of the Study Area
This coastal zone consists of formations that range in age from late Cretaceous to Tertiary. Formations, ages, and rock types are noted in Figure 1.2.
In general, most of the formations are composed of marine siltstone and sandstone, with some occurrence of terrestrial sandstone and conglomerate (Jenkins, 1966; Jennings, 1977; Morton, 2004). The oldest formations are exposed to the east of the study area and progressively become younger westward. The San Juan Creek drainage basin, flowing from the northeast to the southwest, is superimposed on the Late Cretaceous to Tertiary formations, and contains fluvial terraces generally assumed to be Pleistocene in age (Hanson et al., 1994; Peska, 1986). The fluvial terraces extend to the mouth of San Juan Creek where they merge with marine terraces along the southern California coast. Two fault zones strike primarily north-south across the drainage basin, including the eastern Mission Viejo fault zone, which consists of several branches that crosscut San Juan Creek and Bell Canyon, and the western Cristianitos fault zone, also consisting of several branches that crosscut San Juan Creek and Trabuco Arroyo (Fig. 1.2). Three detailed locations have been selected based on terrace mapping and field reconnaissance to evaluate the relationship between seismic activity and fluvial terrace formation. Each of these detailed study zones (Fig. 1.3) possesses geomorphic features that are indicative of seismic activity interacting with the fluvial terraces.
Detailed Study Zone No. 1 Within the stream canyon of the southern San Juan Creek (SSJ), numerous high elevation terraces occur downstream of the Mission Viejo fault zone that have been offset and uplifted from terraces located upstream of the fault zone (Fig. 1.4).
Detailed Study Zone No. 2 This study zone includes Bell Canyon where the Mission Viejo fault strikes across the canyon, offsetting terraces on both sides. Located in this canyon are some of the highest terraces in the study area.
Detailed Study Zone No. 3 In northern San Juan Creek (NSJ) medium and lower elevation terraces upstream of the Mission Viejo fault zone have been uplifted from terraces located downstream (SSJ) of the fault zone (Fig. 1.4).
1.7. Previous Work
The southern California fault regime is shown in Figure 1.1. The San Joaquin Hills form the southern margin of the Los Angeles Basin Faults System (Dolan et al., 1995). The hills are formed by a northeast-vergent anticline that uplifts and deforms marine terraces. Grant et al. (1999) proposed that this anticlinal fold developed above an active, southwest-dipping blind thrust that slips at a rate of 0.42-0.79 mm/yr based on uplift rates of 0.21-0.27 mm/yr. Rivero et al. (2000) confirms the existence of the blind thrust, and suggests that the San Joaquin Hills anticline is the northern onshore extension of the Oceanside thrust detachment, having an average southwest dip value of 23o. Grant et al. (1999) interprets the San Joaquin Hills anticline and blind thrust to be the product of partitioned strike-slip and compressive shortening across the southern Newport-Inglewood fault zones. This anticline may represent a compressional fold feature in a strike-slip shear zone on a dextral shear couple (Fig. 1.1). The Cristianitos and Mission Viejo fault zones crosscut the San Juan Creek drainage basin and strike southeastward toward the offshore Oceanside thrust detachment. In addition, these faults trend southward subparallel to, and in a converging direction with the San Joaquin Hills anticline (Grant et al., 1997; 1999; Fig. 1.1). Seismic activity along the Cristianitos and Mission Viejo faults may represent normal faults of the dextral shear couple. When the dextral shear of the strain ellipse in Figure 1.1 is aligned with the San Andreas fault, the Cristianitos and Mission Viejo faults are represented by normal faults.
1.8. Methodology
Aerial photographs obtained from collections at Whittier College have been used in the construction of initial base maps. Fluvial terraces, faults, and bedrock geology were identified from the aerial photographs to aid in mapping. Aerial photographic data was transferred onto Geographic Information System (GIS) generated base maps. The study area lies within property owned by the Mission Viejo Company, Orange County (Casper County Regional Park), and the Audubon Society. The 3 property owners made their properties available for some aspects of dissertation research. Field mapping included terraces, faults, and the bedrock geology. United States Geological Survey 1:24,000 scale 7.5 minute series topographic quadrangle based maps were used to compile overall geomorphology/geology for the study area. Base maps for this project included the Canada Gobernadora, Dana Point, and San Juan Capistrano 1:24,000 quadrangles in the area of the San Juan Creek drainage basin. Dissertation fieldwork included the construction of detailed field maps for the 3 detailed study areas at a scale of 1:6,000. The enlarged detailed study area maps were used to delineate terraces and plot data and sample locations obtained with Global Positioning System (GPS) handheld receivers. Sites were identified for the collection of samples for numerical dating during the field mapping phase of this project. Geologic and geomorphic maps form the basis for chronological correlation of the terraces and have been used to evaluate fault slip rates, recurrence intervals, and rates of landscape evolution.
1.8.1. Sedimentary Sections and Terrace Chronology
Sedimentary sections of fluvial terraces have been logged in the field by walking along available exposures in ravines and canyons of the study area. Fluvial terrace logs established the nature of terrace material (fill, as opposed to cut or strath), which in turn was used to indicate formative processes and terrace history. Terrace surveys and sample collection was completed using hand-held GPS receivers obtained from Mt. San Jacinto College (MSJC). This research used the Wide Area Augmentation System (WAAS) correction technique to enhance the GPS accuracy. With this correction technique, satellite signals received at an unknown location are simultaneously compared with those received from a known base station yielding accuracy to within 30 cm (Messina and Stoffer, 2001). In addition, existing survey benchmarks were surveyed with the hand-held GPS receivers each day after data collection. Benchmark data was used to correct GPS errors, enhancing data accuracy. Data was plotted using the ArcView 3.2a GIS.
Correlative morphostratigraphy was established for fluvial terraces by using the lateral, elevational, and sequential position of geomorphic features, as well as by comparison with the modern stream profiles. To establish terrace chronology, and to quantify age and rates of geomorphic features and processes, numerical dating was used. Twenty-one prospective samples were collected from various locations within the 3 studies zones, and of these, 5 test samples have thus far been analyzed using the CRN surface exposure method in an attempt to establish ages of terraces at critical locations (see Chapter 3). Sample locations were recorded in the field using hand-held GPS receivers. Parameters (depth, location, rock/soil type, date, time, etc.) of sample collection were noted, as well as factors that could shield the sample locations from cosmic rays (such as the surrounding hills). Chemical analysis of the CRN samples was completed at the University of Cincinnati, and the dating of the 5 samples was completed at the Accelerator Mass Spectrometer (AMS) facility at the Lawrence Livermore National Laboratory.
1.8.2. Fluvial/Marine Terrace Correlation
To assess the chronology and evolutionary setting of fluvial terraces, elevational correlations between fluvial and marine terraces near the mouth of San Juan Creek were established. Establishing the date of formation and rate of uplift on marine terraces is often accomplished by a comparison of terrace elevations with glacio-eustatic sea level changes (Fig. 1.5), a technique first proposed by Alexander (1953).
Various workers have mapped locations and determined the chronology of marine terraces in southern California from San Diego to Newport Beach (Ehlig, 1980; Shlemon, 1987; Barrie et al., 1992; Eistenberg, 1992; Kern and Rockwell, 1992; Munro, 1992; Dolan et al., 1995). Timing of formation between upstream fluvial terraces and marine terraces may vary, but in the area of the mouth of a river debauching into the ocean, fluvial and marine terrace chronology are inextricably linked. At this location during Pleistocene, as today, the marine wave-cut platform (now marine terraces) and fluvial floodplain (now fluvial terraces) merged into the same elevation at 0 ft Mean Sea Level (MSL). This study established the elevational equivalency between fluvial and marine terraces at the mouth of the San Juan Creek. The establishment of fluvial and marine terrace relationships allowed the determination of the chronology of the fluvial terraces at this location. Construction of former longitudinal stream profiles then links various upstream fluvial terrace levels within the drainage basin to a specific base level chronology. Once the chronologies of the fluvial and marine terraces were related, the fluvial terraces were placed within the proper glacio-eustatic sea level and oxygen isotope chronological setting.
1.8.3. Calculating Slip Rates and Recurrence Intervals
Terrace maps and surveys in other studies have indicated possible horizontal and vertical offsets (Harms et al., 1984) on individual terraces and on former longitudinal stream profiles that are crossed by faults. In this study, slip rates and recurrence intervals were calculated for the Cristianitos and Mission Viejo faults (Fig. 1.6) from the fluvial terrace correlation and geochronologic analysis. Where terraces are offset by faults, slip rates can be calculated from the relation:
R=D/T
where R is the slip rate, D is the amount of displacement across the terrace or across the former longitudinal stream profile, and T is the age of the terrace. Slip rates of selected time periods have been calculated by using the same formula, and by substituting the displacement and ages between selected sequential terrace levels for T and D. Original terrace deposits were assumed unbroken by faults until stream incision and floodplain abandonment (terrace formation) began to occur.
Recurrence intervals were calculated from the equation:
t=d/R
where t is average recurrence time, R is slip rate, and d is average displacement-per-earthquake. The Cristianitos and Mission Viejo faults are ~35 and 25 km in length respectively (Shlemon, 1992). The average displacement per earthquake can be calculated using the empirical regression relationship between fault length and maximum displacement-per-earthquake established by Wells and Coppersmith (1994). Applying this regression yields a maximum displacement of 1.57 m and 1.11 m for these 2 faults respectively.
1.8.4. Paleoseismicity and Rate Assumptions
The calculation of a slip rate on a fault should be based on the net-slip which is a compilation of the dip-slip and strike-slip components (rake) of the fault. For this study, it is assumed that minimal movement has occurred in the strike direction on the Mission Viejo and Cristianitos faults. This assumption is based on the strain ellipse (Fig. 1.1) which, when aligned with the dextral shear component parallel to the San Andreas fault, indicates that both the Mission Viejo and Cristianitos faults represent normal faults. Additionally, no evidence of strike-slip movement was noted along these faults during the field reconnaissance phase of this study.
A second assumption used in this study is that both of these faults have dips that are near vertical. The sinuous nature of the faults along with their poor outcrop exposures precludes calculation or direct measurement of the dip direction. Therefore, for this study, net-slip will be approximated with vertical slip (vertical throw). It should be noted that this slip calculation will represent a minimum slip motion, and any fault motion with significant dip inclination, or motion in the strike direction, will significantly increase the slip rates and decrease the recurrence intervals calculated in this study.
Thirdly, the calculation of recurrence intervals in this study is based on the assumption of uniform recurrence and characteristic slip. The Wells and Coppersmith (1994) regression model will yield a maximum displacement, and this maximum displacement is used as the average displacement-per-earthquake to calculate the average recurrence interval. It must be noted that this average displacement-per-earthquake is the maximum that the fault can yield, and therefore, this average recurrence interval should be viewed as the maximum possible. If smaller displacement-per-earthquake ruptures are commonplace on this fault, then a shorter recurrence interval (more frequent) will be required in order to achieve the total displacement indicated by the offset terraces.
Additionally, the Wells and Coppersmith (1994) regression is only a statistical model, and a wide range of variation may actually be present on the faults involved in this study. Determination of average fault displacement per seismic episode could be better estimated by fault trenching. Trenching was not permitted by the property owners, however, so this technique is beyond the scope of the current research proposal.
Lastly, when discussing the offset and tilt (uplift rate gradient) of the terraces, the geographic orientation is important. The Mission Viejo fault bisects both the SSJ and Bell Canyon within a distance of ~2 km. In order to discuss the uplift or tilting of blocks on various sides of the fault, spatial uniformity across the 2 km is assumed.
1.8.5. Establishing Geologic and Geomorphic History
The relationship between alluvial deposits, bedrock, and the hillside fluvial terraces throughout an idealized river system is shown in Figure 1.7.
The fluvial terrace formation and degradation can result from a variety of interacting influences including tectonic activity, climatic conditions, and autocycling (Fig. 1.8), and the type of terrace formed indicates the formative influence. Terraces controlled by tectonic uplift will generally consist of tread platforms cut directly onto a bedrock substrate reflecting the exposure of newly uplifted bedrock to stream processes. In addition, tectonically controlled terraces will tend to show symmetry with the modern stream channel profile, while terraces formed by climatic and autocycling processes will not (Keller, 1996). Terraces resulting solely from changing climate conditions will consist of paired aggradations of fluvial deposits, and will be sequentially asymmetric, with the upstream segment of terraces being proportionally higher above the modern stream channel (Keller, 1996). Autocycling refers to aggradation and degradation adjustments that occur in a stream channel in response to natural stream processes. Terraces resulting from autocycling may have treads that consist of either bedrock or alluvium, and are relatively small and unpaired. Such adjustments are usually the result of internal variables (discharge, velocity, etc.) fluctuating across the critical threshold (Threshold of critical power is the ratio of stream power to resisting power; Bull, 1991; Keller, 1996). For example, Bull (1991) describes fluctuations across the critical threshold in Charwell River (New Zealand) terraces noting that the resisting power increased during times of streambed armoring and riparian plant growth, which increased hydraulic roughnesses and the shear stress needed to entrain streambed materials. Each postulated major flood discharge that disrupted streambed armor caused an abrupt decrease in resisting power. Resisting power gradually increased again during the next period of renewed degradation and concurrent renewal of streambed armor.
Preliminary research indicates that both the fluvial and marine terraces of this area have been primarily influenced by tectonic uplift, however, a relative percent of formative influences (tectonic vs. climatic) can be established by comparing the vertical spacing of the terraces to eustatic sea level. Once former longitudinal stream profiles were constructed, the profiles were compared with oxygen isotope climactic data (eustatic sea level highstands) as is typical for marine terrace analysis. When marine terrace strandlines (tectonically elevated above sea level) are compared with eustatic sea level, the slope of the line drawn between the peak sea level highstand and the abscissa represents the rate of tectonic uplift. For this research, fluvial terraces were similarly compared with eustatic sea level (elevated from height above current active longitudinal stream profile rather than above current sea level). The resulting fluvial “uplift rate” 0.29 mm/yr, or incision rate, compares favorably to the locally established uplift rate of 0.21-0.27 mm/yr (Grant, 1999). This tectonic uplift rate is comparable to other rates that have been established in the area (0.26 mm/yr in Dana Point, 0.14-0.19 mm/yr in Laguna Beach, and 0.12-0.19 mm/yr in San Clemente; Peska, 1986). Compatible rates of tectonic uplift indicates that the fluvial incision rate throughout the San Juan Creek basin may actually represent the tectonic uplift rate throughout the study area.
1.9. Summary
The research outlined above will establish rates of plate motions and a geomorphic evolutionary history by mapping, correlating, and dating fluvial terraces throughout the San Juan Creek drainage basin. Establishing slip rates and recurrence intervals for the Cristianitos and Mission Viejo fault zones will be a pertinent addition to the understanding of southern California seismicity. Identifying fault activity and establishing rates of crustal displacement are critical both for establishing a working model of plate motions throughout southern California, and for establishing seismic risk and hazard parameters based on earthquake magnitudes and recurrent intervals for faults occurring within this densely populated region. The dissertation is divided into 5 chapters, the first being this introduction and the last being a discussion of the results and geological conclusions. The intermediate 3 chapters are presented in scientific journal format, with each representing progressive studies. Chapter 2 presents the field mapping and correlation of 124 terraces along San Juan Creek and Bell Canyon. Research in this chapter establishes the extent and sequential nature of San Juan Creek former longitudinal stream profiles. Chapter 3 uses a variety of methods to date the former longitudinal stream profiles establishing the timing of the terraces and profiles, and revealing the geomorphic history and timing of regional tectonic uplift. Chapter 4 explores the implications of the interaction of the terraces and former longitudinal stream profiles with the Mission Viejo and Cristianitos faults that crosscut Bell Canyon and San Juan Creek.