by John Dvorak
But there came a time when the western edge of the North American plate drifted far enough west to ride over the spreading centers separating the Farallon and Pacific plates, causing a sliver of the North American to detach itself from the rest of the continent and be captured and move with the Pacific plate. Ever since, a succession of transform faults has formed, the latest and currently the most active one being the San Andreas Fault.
The first transform fault developed along the western edge of the North American plate 25 million years ago in the broad offshore region of southern California where the Channel Islands stand today, a region commonly referred to among seismologists as “the borderlands.”* These islands—San Clemente and Santa Catalina south of Los Angeles, and San Miguel, Santa Rosa, and Santa Cruz south of Santa Barbara—are all comprised of continental rocks and stand above sea level as the highest ridges of a complex basin-and-ridge area that lies off the coast of southern California. This offshore continental area is crisscrossed by a host of strike-slip faults. Which one might represent the ur–San Andreas Fault has yet to be determined. But this much is known: For the first few million years after the North American plate began to ride over the Pacific plate, whatever transform faults developed, developed in borderlands. There is no evidence of a San Andreas–type fault on land until about 20 million years ago.
These earliest known faults are no longer a single strand but have been cut into segments, which have been moved around by earthquakes that occurred along younger fault strands. But at least three segments of an early strand can still be found. One, known as the San Francisquito Fault, runs through the western end of the San Gabriel Mountains. Another, the Fenner Fault, also in the San Gabriel Mountains, lies a few miles east of Valyermo. And another, the Clemens Well Fault, lies far to the east in the Orocopia Mountains near the Salton Sea. The complex work of restoring geologic units to their original positions by undoing the sliding and rotation of blocks and straightening out folded and compressed sedimentary layers—a technique known as palinspastic reconstruction—shows that these three short fault segments once formed a single long strand. Furthermore, it shows that the single long strand was once the primary transform fault in California, that it was active until about 12 million years ago, and that 60 miles of horizontal displacement accumulated along it. But nothing is permanent, certainly not in the geology of California. And the San Francisquito-Fenner-Clemens Well Fault was eventually cut up and replaced by younger, longer strands as the primary movers and shakers in the region.
When the North American plate began to drift over the Farallon-Pacific’s spreading central region, a transform fault formed, and then a peculiar feature developed at either end of that fault. The feature, known as a triple junction, is a place where the boundaries of three tectonic plates meet. In this case, two of the plates are the North American and Pacific plates; the third, which is actually what remains of the Farallon plate, has been given a different name depending on whether it is north or south of the transform fault. At the north end, the surviving part of the Farallon plate is now known as the Gorda plate and the point where the three plates meet is the Mendocino triple junction, because the point is currently located near Cape Mendocino. At the south end is the Cocos plate—a remnant of the Farallon plate—and the Rivera triple junction. What is important here is that, because of the directions in which the various plates are moving, neither the Mendocino nor the Rivera triple junction is stationary; both migrate. And they migrate in opposite directions, the Mendocino triple junction to the north and the Rivera to the south. As time progresses, the transform boundary between the Pacific and North American plates lengthens. And that brings us back to the San Francisquito-Fenner-Clemens Well Fault.
Twelve million years ago, when the San Francisquito-Fenner-Clemens Well Fault became inactive and a new fault strand formed, the northern triple junction was near San Francisco and the southern triple junction was somewhere south of San Diego. In northern California, the strike-slip movement between the Pacific and the North American plates was taken up by the San Gregorio-Hosgri Fault, which would slice through the conglomerate that contained the fossilized bivalve mollusk Hillichnus lobosensis and transport the western part northward to where it now resides at Point Reyes while the southern part remained at Point Lobos. In southern California, movement was primarily along the San Gabriel Fault, which runs along the western and southern boundaries of the San Gabriel Mountains. Then, after several million years, seismic activity shifted again in both northern and southern California.
In the north, activity shifted from the San Gregorio-Hosgri Fault eastward, perhaps first to the east side of what is now San Francisco Bay, then back to the west side, where it is today. In fact, the segment of the San Andreas Fault that runs through San Andreas Valley—the namesake of the fault and where Lawson first recognized the fault—may be the youngest strand of the entire fault, having formed one or two million years ago. It also has a relatively small amount of accumulated movement; in the San Andreas Valley, the fault has displaced rocks only about 20 miles.
If the position of the fault has shifted before, it will almost certainly shift again. A future candidate for taking up much of the motion between the Pacific and North American plates and replacing the current active trace of the San Andreas Fault is the Hayward Fault on the east side of San Francisco Bay. A look at a map of fault locations in the Bay Area shows that the Hayward Fault is nearly aligned with the Calaveras Fault to the south and the Rodgers Creek, Healdsburg, and Maacama Faults that continue north of San Francisco Bay. And so it seems natural that these individual fault strands might coalesce into a single long strand and become the primary fault strand that runs through northern California.
But the shift may not stop there. There is reason to believe that, millions of years from now, the primary fault strand between the Pacific and North American plates could shift hundreds of miles and be on the east side of the Sierra Nevada. To understand why this might happen, one has to understand how the southern segment of the San Andreas Fault has evolved.
Almost everything in southern California is of recent origin—including the geology.
The San Gabriel Fault was the main fault strand through southern California from 12 to about 4 million years ago and it may have extended farther east than it does today along what is known as the Cajon Valley Fault. Exactly what was the geologic layout before 4,000,000 years ago has been difficult to determine because almost every square foot of land in southern California has since been disturbed. Almost every mountain, every hill, every river, every canyon, and every fault has either been shoved away from its original location or disrupted or come into existence in the last 4,000,000 years. If one wanted to attribute this regional chaos to one entity, it would be the erratic migration of the Rivera triple junction at the southern tip of the San Andreas Fault.
The Mendocino triple junction at the north end of the San Andreas Fault has had a simple history: It has migrated northward at a steady rate, leaving behind, as evidence of its migration, a progression of volcanic activity. South of San Jose are the Quien Sabe volcanics in the Diablo Range, which erupted about 12,000,000 years ago. North of San Jose, in the Berkeley Hills east of Oakland, are the Moraga volcanics, dated at 10,000,000 years. Near Calistoga, north of San Francisco Bay, are the Sonoma volcanics, where the heavy fall of volcanic ash produced a petrified forest of redwood trunks 6,000,000 years ago. And farther north, 80 miles south of the Mendocino triple junction, is the Geysers geothermal field, where the most recent volcanic activity was a mere 10,000 years ago. The Rivera triple junction has not left such a trail of volcanics—nor has its migration been as simple.
In fact, there may have been multiple triple junctions at the southern end of the developing transform fault that eventually coalesced into one. This much is certain: About 4,000,000 years ago, what is today known as the Rivera triple junction jumped a few hundred miles east to a point under the North America
n plate, shearing off a long slab of continental material that became the peninsula of Baja California and forming the Gulf of California.
The Gulf of California formed because of the development of a series of short spreading centers connected by a series of equally short transform faults, creating new ocean crust. After 4,000,000 years, the opposite sides of the gulf have spread apart 200 miles. And the seaway would extend all the way north as far as Indio, California, except for a surprising geologic consequence—the creation of the Grand Canyon.
As the Gulf of California spread open, the crust thinned and the land surface dropped. In California, this produced a long broad feature known as the Salton Trough, which runs as far north as Indio. Studies show that this trough should be as much as three miles deep, except that it is filled with river sediments. And it is filled with river sediments because the formation of the Gulf of California and the lowering of the land surface steepened the gradient of the Colorado River.
The steepening of the gradient led to a brief period of rapid erosion and the formation of the Grand Canyon. Much of what was removed now fills the Salton Trough. And then, there is a modern twist to the story.
When western settlers first arrived in California, the Salton Trough was a desert, barely able to support farming. But in 1900, a canal system was constructed to bring water from the Colorado River to the Salton Trough. To attract more farmers, the area was advertised as the Imperial Valley. The canals were indeed successful and brought the needed water, but they soon silted up, so more canals were constructed. The construction of the second round of canals was streamlined by bypassing control gates. Then a series of floods occurred, so that by October 1905, virtually the entire Colorado River was flowing into the Salton Trough. The flow of river water was finally stopped in 1907, but by then a saline lake had formed: the Salton Sea.
The twist is that, in the past, water from the Colorado River has been diverted naturally into the Salton Trough, forming a large lake. At those times, paleoseismological studies show that earthquakes along the San Andreas were frequent; that is, the weight of the lake water may have modulated the earthquakes. If so, the question can be asked: Would the refilling of the Salton Trough trigger an earthquake along this segment of the San Andreas, which has not had a major earthquake since 1690? Moreover, could this paradoxically be a way to control earthquakes?
The filling of water reservoirs behind dams has triggered some major earthquakes, the most serious case being the magnitude-7.9 Sichuan earthquake in south-central China in May 2008 that killed 80,000 people. This might have been avoided if we knew more about how the weight of water in a reservoir actually triggered the earthquake and whether a partial filling of a reservoir could insure the occurrence of a series of moderate earthquakes and not a single large destructive one. And so I return to consider how the formation of the Salton Trough and the northwest drift of Baja California have affected the evolution of the San Andreas Fault.
For 4,000,000 years, after its separation from North America, Baja California has been ramming into southeastern California. The net result is that the greater Los Angeles area is caught in a vise. The area is being squeezed, and that has caused the crust to buckle and throw up an east-west line of mountains—the Transverse Ranges, which run from Santa Barbara to San Bernardino—and is responsible for the “Gordian knot” of geology declared by Josiah Whitney when he began to examine California geology.
At the same time, the Los Angeles area is also being dragged to the northwest by the Pacific plate. All of this tectonic action has created a complex network of active faults that, when plotted on a map, looks like a shattered pane of glass. And along the northern and the eastern edges of the pane is the San Andreas Fault.
The Mojave segment of the San Andreas Fault runs along the northern edge of the San Gabriel Mountains. This segment is remarkably straight and came into existence when the mountains did, about 4,000,000 years ago—replacing the old San Gabriel Fault with the modern San Andreas strand. But to the east, after it passes through Cajon Pass, the modern San Andreas Fault loses its straight-line simplicity, and even today geologists are unsure exactly where the San Andreas Fault runs.
The general opinion is that it separates into two strands—the Mission Creek and the Banning Faults—that run along the north side of San Gorgonio Pass, which connects the greater Los Angeles basin and the Salton Trough. The southern strand—the Banning Fault—passes a few miles east of the luxury resorts and celebrity mansions at Palm Springs, then reconnects with the Mission Creek Fault at the southern end of the Indio Hills. From there, the San Andreas is again a single straight strand and is easy to follow another 40 miles to its end at the east edge of the Salton Sea.
But it is the complexity between Cajon Pass and Palm Springs that has geologists pointing to that segment of the San Andreas Fault as going through a rapid evolution.
If, instead of following the evidence of surface faulting that leads one to the Mission Creek and Banning Faults and through San Gorgonio Pass, one looks at a map of seismicity and follows the most prominent line of recent earthquakes, one would pass through the city of San Bernardino, south to Colton, then through Moreno Valley, over the San Jacinto Mountains, down through Borrego Valley, and end at the Superstition Hills on the west side of the Salton Sea. This is the San Jacinto Fault and it is the most seismically active fault in southern California, meaning it has had more large historic earthquakes than any fault—including the San Andreas—in southern California.*
To add another level of complexity—and intrigue—there is yet another well-developed, horizontally slipping fault in southern California. The Elsinore Fault runs between the San Jacinto Fault and the coast. Furthermore, when the southern segment of the San Andreas Fault is considered, there is a progression in age and activity among the three faults: The Elsinore Fault is the oldest and least active; the southern segment of the San Andreas that runs through San Gregorio Pass and close to Palm Springs is middle in age and in activity; and the San Jacinto is the most active and the youngest, probably forming in the last 1,000,000 or 2,000,000 years.
So the continual opening of the Gulf of California has had the following effect: It caused a transfer of activity from the Elsinore Fault to the San Andreas Fault, then from the San Andreas Fault to the San Jacinto Fault.
At first glance this seems of no consequence, until one considers it on a large scale—a continent-wide scale—and tries to understand exactly how the western edge of the North American plate is being disrupted by the northwest motion of the Pacific plate. And to do this, one finds the final crucial element by leaving the hot, arid land of Los Angeles and the Salton Trough and traveling 1,000 miles north to the cool Ponderosa pines at the edge of Lake Tahoe. At that point, one is standing at the northern end of another trail of earthquakes—the little-known Walker Lane seismic zone.
Located along the border between California and Nevada, Lake Tahoe is the largest alpine lake in North America and the second-deepest lake in the United States after Crater Lake in Oregon, which was formed by the collapse of Mount Mazama during a volcanic eruption 8,000 years ago. The formation of Lake Tahoe was decidedly different.
Lake Tahoe sits in a basin that formed in two stages. Beginning about 2,000,000 years ago, it grew as a result of an east-west pull that produced the Basin and Range province, which covers most of Nevada and that created the high peaks of the Carson Range on the east side and the Sierra Nevada Mountains on the west side of Lake Tahoe. Then, about 1,000,000 years ago, a set of horizontally slipping faults began to develop in a broad zone that runs along the California-Nevada border.
The zone begins somewhere north of Lake Tahoe—perhaps, as some geologists say, at Pyramid Lake north of Reno, which is also in a basin, or, as seismologists who locate earthquakes argue, as far north as south-central Oregon. In any case, the zone runs south of Lake Tahoe and includes Owens Valley in California and the Yucca Mountains in Nevad
a—where the infamous Area 51 is located, the source of many conspiracy theories about captured UFOs, but which in actuality is a supersecret test site for military aircraft.
The zone is a place of persistent earthquakes known as the Walker Lane seismic zone. It was named for Joseph Walker, leader of a military expedition that in 1833 found what became known as the California trail, the primary route used by immigrants to reach the gold fields, and, that same year, probably the first easterner to gaze upon Yosemite Valley.
Though the seismic activity along the Walker Lane is less than along the San Andreas and its associated faults, it is not insignificant. The most recent flurry of activity near Lake Tahoe occurred in April 2008 when more than 600 earthquakes shook the region, the largest a magnitude-4.7 event on April 15. The last significant earthquake was a magnitude-6.1 tremor in 1914.
The largest historical earthquake in the Walker zone was the 1872 Owens Valley earthquake, studied by Josiah Whitney and Grove Karl Gilbert and felt by Muir in Yosemite Valley. As Gilbert noted, there was a large amount of horizontal displacement in the same direction as slip for large earthquakes along the San Andreas Fault—that is, the Sierra Nevada Mountains moved northward with respect to the floor of Owens Valley and the rest of North America. And that is not a coincidence. It shows that the effect of the northward motion of the Pacific plate is being felt at least as far east as the Walker Lane seismic zone, and in some way is responsible for the formation of the basin that holds Lake Tahoe. But that is not the limit of the effect of the Pacific plate on North America, or at least that is not the limit of whatever forces inside the Earth are driving the motion of the Pacific plate.
