The evolution of the southern California uplift, 1955 through 1976

Abstract

The southern California uplift culminated in 1974 as a 150- km-wide crustal swell that extended about 600 km eastward and east-southeastward from Point Arguello to the Colorado River and Salton Sea, respectively; it was characterized by remarkably uniform height changes between 1959 and 1974 of 0.30-0.35 m over at least half of its 60,000-70,000 km 2 area. At its zenith, the uplift included virtually the entire Transverse Ranges geologic province and parts of the Coast Ranges, San Joaquin Valley, Sierra Nevada, Basin and Range, Mojave Desert, Peninsular Ranges, and Salton Trough provinces. The alinement of the western part of the uplift closely paralleled the east-trending Transverse Ranges, whereas the southern flank of the eastern lobe roughly coincided with the west-northwest-trending San Andreas fault. The position and configuration of the uplift associate it with a singularly complex section of the boundary between the North American and Pacific plates that has certainly sustained major modification during the past 5 million years and probably during the past 1 million years. Surface deformation can be categorized as tectonic or nontectonic. Nontectonic vertical displacements associated with the activities of man have overwhelmed natural compaction and areally significant soil expansion in the southern California area. Because tectonic displacements are implicitly defined as those that cannot be otherwise explained, those vertical movements that can be reasonably attributed to artificial processes have been subtracted from our reconstructed configurations of the uplift. Hence this reconstruction has necessarily included the assembly and evaluation of an enormous volume of data on oil-field operations, changes in ground-water levels, and measured subsidence (or rebound) associated with changes in the underground fluid regimen. Measured changes in height at various stages in the evolution of the uplift have been based chiefly on first-order levelings carried out between 1953 and 1976. Exceptions to this generalization consist largely of the results of pre-1953 surveys through the western Transverse Ranges and the eastern Mojave Desert. Errors in measured height differences derive from blunders, systematic survey errors, random survey errors, improperly formulated orthometric corrections, and intrasurvey movement; the last of these has created the most serious problems encountered in our reconstruction of the basic data. A variety of independent tests indicate that survey error associated with the utilized levelings was generally small and fell largely within the predicted random-error range. Moreover, the redundancy and coherence displayed by the entire data set provide convincing evidence of survey accuracy and the virtual absence of height- and slope-dependent error in particular. Our reconstructions of the changing configuration of the uplift derive chiefly from comparisons among sequentially developed observed elevations along the same route. Most of the observed elevations from which the vertical displacements were computed have been reconstructed with respect to bench mark Tidal 8, San Pedro, as invariant in height. Because the San Pedro tide station has been characterized by a history of modest relative uplift, vertical displacements referred to this station are biased slightly toward the appearance of subsidence. Where the observed elevations cannot be conveniently tied to Tidal 8, they have been referred to secondary control points whose history with respect to Tidal 8 can be independently established. Each of the lines of observed elevation changes provides, accordingly, a section athwart or along the axis of the uplift from which the changes in the configuration of the uplift can be roughly generalized. Because relatively few surveys were run in 1955, which we choose as a representative temporal datum, we have commonly incorporated the results of earlier or of somewhat later levelings as the equivalents of 1955 surveys. Although this procedure introduces a certain subjectivity, the probable equivalence between the results of these earlier or later surveys with those that would have been obtained had this leveling been carried out in 1955, usually can be independently tested. Wherever the calculated vertical displacements are based on comparisons between the results of levelings over different routes, the observed elevations have been orthometrically corrected to agree with those that would have been produced had each of these surveys been along the same route. The growth of the southern California uplift consisted of two well-defined spasms of positive movement, the second of which was closely followed by partial collapse. Our reconstruction, although it clearly errs in detail, indicates that the uplift, together with marginal and apparently ephemeral tectonic subsidence, nucleated in the west-central Transverse Ranges near Ozena, sometime between the spring of 1959 and the spring of 1960. The uplift expanded rapidly eastward (and probably westward as well), and by the fall of 1961 much of the Transverse Ranges and the Mojave Desert at least as far east as Twentynine Palms had risen by as much as 0.25 m. Between 1962 and 1972 the area included by the initially developed (1959-61) uplift sustained additional but clearly decelerating uplift accompanied locally by oscillatory displacements. Between 1972/ 73 and 1974 a second crustal spasm extended the uplift eastward to the Colorado River and elevated much of the eastern Mojave Desert by values that equaled or exceeded those developed within the western lobe. Between 1974 and 1976, at least the central part of the uplift sustained partial collapse that nowhere amounted to less than 50 percent of the cumulative uplift since 1959. Whether this collapse affected the entire uplift is conjectural, but we now recognize well-defined evidence of major down-to-the-north tilting that must have occurred within the eastern part of the uplift at some time between 1974 and 1976. Accumulating evidence indicates that nearly all the area included with the southern California uplift underwent similar uplift and partial collapse during the early part of the 20th century. Thus we infer that the recent uplift represents but a single event in an ongoing, more or less cyclic deformational process characterized by a period of about 50 years. Even though less than two full cycles are expressed in the geodetic record, the cumulative rate of uplift near the center of the recent uplift probably has averaged about 5 mm/yr, a value that is roughly consistent with the uplift rates that have been deduced for the late Quaternary emergent marine terraces along the south flank of the Transverse Ranges. Although the evolution of the recent uplift is relatively well defined, its correlation with the regional seismicity is poorly defined. A comparison between the occurrence of southern California earthquakes of magnitude ≥4 during the period 1932 to 1976 with the 1974 configuration of the uplift demonstrates the existence of (1) relatively aseismic areas within the western lobe of the uplift (in the western Transverse Ranges), in the central part of the uplift (in the western Mojave Desert), and along an east-trending zone that extends into the eastern Mojave athwart the south flank of the uplift (north of the Salton Sea) and (2) localized concentrations of seismic activity along the flanks of the uplift. Moreover, 9 of the 10 largest earthquakes recorded within or around the area of the southern California uplift during the period 1932 to 1976 (the 1933 Long Beach, the 1941 Santa Barbara, the 1946 Walker Pass, the 1947 Manix, the 1948 Desert Hot Springs, and the four major 1952 Kern County shocks) occurred before the inception of the uplift in 1959 or 1960. The area embraced by the southern California uplift has been identified with geodetically defined horizontal strain, part of which may have accumulated as a major north-south contractional event that roughly coincided with the first spasm of uplift. Nonetheless, continuing contractional strain associated with regionally developed partial collapse argues that the uplift cannot be fully explained simply as the vertical expression of continuing north-south compression. Consideration of the two well-defined historical episodes of uplift and partial collapse indicate that the southern California uplift may be the product of decoupling and viscous flow beneath the seismogenic zone, presumably driven by continuing motion between the irregularly margined plates south of the great bend of the San Andreas fault. Because the magnitude of the maximum uplift associated with each episode was approximately the same, there may be some threshold value above which collapse (viscous flow) may ensue; the absence of total collapse may be a function of precollapse strain hardening within the postulated subseismogenic viscoelastic layer.