In the Los Angeles Region Seismic Experiment, researchers employed air-guns and explosions to study earthquake hazards. Setting off small sources of noise from the Pacific Ocean to the Mojave Desert, they measured how sound vibrations penetrate or bounce off faults and substrata and constructed images of the subsurface structure that may lead to better prediction of earthquakes in Southern California.
by Gary S. Fuis, Thomas M. Brocher, James Mori, Rufus D. Catchings, Uri S. ten Brink, Kim D. Klitgord, and Robert G. Bohannon, U.S. Geological Survey, 345 Middlefield Rd., Menlo Park, Calif; David A. Okaya, Robert W. Clayton, Thomas M. Henyey, Mark L. Benthien, Paul M. Davis, and Monica D. Kohler, Southern California Earthquake Center; William J. Lutter, University of Wisconsin at Madison, Madison, Wisc.; Trond Ryberg, GeoForschungsZentrum, Potsdam, Germany
The Whittier Narrows earthquake of 1987 and the Northridge earthquake of 1991 demonstrate that serious earthquake hazards lie below the surface in the form of buried faults in the Los Angeles region. To better understand the workings of earthquake-producing machinery in southern California, we need to know more about the subsurface structure: how the crust is subdivided into blocks, how theses blocks rub against one another along faults, and how stress is being applied to this assemblage of blocks. Moreover, knowing the shape of sedimentary basins in the upper part of the crust is essential to predicting where earthquake shaking is likely to be amplified. Enhanced ground motion, which is the most important cause of earthquake damage, often occurs in areas of soft sediments where seismic waves slow down but increase greatly in amplitude, causing strong shaking.
To learn more about the crustal structure of Southern California, geologists and seismologists carried out some noisy experiments on land and off the coast near Los Angeles. These experiments were part of the Los Angeles Region Seismic Experiment (LARSE), a seismic imaging program initiated by the U.S. Geological Survey and the Southern California Earthquake Center. Although the new seismological data were gathered to help the residents of Los Angeles and other towns plan for future earthquakes, the scientists had to make sure that the loud noise from air-guns and explosions would not disturb residents or wildlife.
LARSE93, the "quiet" phase of the program, began in 1993 with the collection of seismic data from distant and local earthquakes along line 1 of the map (see figure). In October 1994, air-gun and explosion surveys along lines 1-3, which cross the Los Angeles region and the offshore Continental Borderland, were carried out (LARSE94). In this second phase, man-made noise sources were recorded on seismographs along line 1, and both images were created from refracted and reflected waves, respectively. Features that we hoped to see included the earthquake faults and bottoms of sedimentary basins.
Obtaining permission for the airgun and explosion surveys in an urban environment was a complex and time-consuming process. The 2-year permitting process required not only an environmental assessment but public addresses to city councils and other governmental bodies; extensive radio, television, and newspaper interviews; and correspondence with many individuals and private groups. The public reception of this high-profile experiment was generally positive because of the recent earthquakes in the Los Angeles region, particularly the 1991 Northridge earthquake. In many cases, however, security for the seismographs required complete burial of the recorder and batteries. To make our relatively weak air-gun signals detectable above all the other man-made noise (LA is a noisy place!), we made up to 6 passes with the ship along the offshore segments of the three lines and added the data together. We also conducted an extensive pre-experiment noise survey, avoided freeways, and only detonated explosions in the hours between 1:30 and 4:30 a.m. We had to take great care to avoid damage—perceived and real—from the explosions.
The ship that we used was the R/V Ewing, of Lamont-Doherty Earth Observatory. It towed an array of 20 airguns that fired about 140 liters of compressed air into the water every 20 seconds. The airguns were recorded by a 4-km-long streamer, pulled by the ship, that contained 160 small seismometers. The airgun signals were also recorded by 10 ocean-bottom seismographs and 170 onshore seismographs. Amazingly, the airgun signals were detectable as far away as the Mojave Desert, 200 km from the ship. This airgun survey was followed by over 60 explosions that involved detonation of 5–3000 kg (~10 to 6,000 lb) of explosive in drill holes along the onshore part of line 1. These explosions were recorded by a linear array of 640 seismographs, installed along line 1 from the coast to the Mojave Desert.
The chief imaging targets along offshore line 1 were the Catalina, San Pedro basin, and Palos Verde Hills faults. Onshore targets along line 1 were the top of basement beneath the Los Angeles basin, which had never before been imaged; the Elysian Park blind thrust fault system, believed to be the causative structure for the M 5.9 Whittier Narrows earthquake of 1987; the Sierra Madre fault system, believed to be the causative structure for the M 5.8 Sierra Madre earthquake of 1991; and the San Andreas fault. LARSE94 was designed to image these features using both refraction and reflection. (For explanation of these features, see figure)
This figure repersents a CAT-scan-like image along line 1 created from measurement of the velocities of seismic P waves (sound waves in rocks) generated by the explosions (see figure ). This image reveals not only the depth and shape of sedimentary basins along line 1, but also the variation of seismic velocity within basement rocks below the basins and within the San Gabriel Mountains. (Basement rocks have seismic P-wave velocities of 5.5 km/s (3.5 miles/s) and higher. This image shows the San Andreas fault and a former branch of the San Andreas fault, the San Gabriel fault (SAF and SGF, respectively), as nearly vertical, low-velocity zones penetrating as deeply as 10 km into the crust. It is interesting that low-velocity zones (located at cracks in the rocks caused by high strain and percolating water) persist in about equal strength and definition along both faults, even though the San Gabriel fault is older and now inactive at this location.
CAT-scan-type image of subsurface structure along part of line 1 obtained from P wave (sound-waves in rock) refraction. Seismic-velocity contour interval, 0.25 km/s; integer velocities are labeled. Hypocenters for three moderate earthquakes have been projected onto the model. Black bar indicates depth of sediment-basement interface obtained independently from a strong reflection. Low-velocity zones extending vertically to 10–15 km depth are centered on the San Gabriel and San Andreas faults. Key: IF, Newport-Inglewood fault; WF, Whittier fault; DF, Duarte fault; SMF, Sierra Madre fault; SGF, San Gabriel fault; P, Punchbowl fault; SAF, San Andreas fault; LLF, Llano fault; and MVF, Mirage Valley fault. Vertical exaggeration 3:1.
To make images of a fault, seismologists record how sound waves generated by explosions at the surface or earthquakes at depths are (a) reflected or (b) transmitted through the subsurface. Using a sonogramlike technique (see part (a)), they construct an image of the fault from sound waves reflected from the fault surface—much like a sonogram image is generated from the sound waves reflected by an unborn baby inside a mother's womb.
LARSE imaged the bottom of the Los Angeles sedimentary basin for the first time (black bar) (Sedimentary rocks generally have seismic velocities less than 5.5–5.75 km/s). The thickness of sedimentary rocks is an important parameter controlling how hard the ground will shake during an earthquake: the thicker the sedimentary rocks are, the harder the shaking. Although the basin depth from the LARSE image agrees with earlier estimates of 8.5 km for the south part of the basin, the 5 km depth in the north at the San Gabriel Valley is surprisingly large and may cause estimates for shaking in this area to be revised. A basement uplift is visible beneath the sediments in the Puente Hills. The 1987 Whittier Narrows M 5.9 earthquake, which occurred in this location, probably represents uplift of basement along a buried thrust fault below the Puente Hills.
On the north side of the San Gabriel Valley, the basement rises abruptly from 5-km depth to the surface within less than 10-km distance. This geometry shows that no significant volume of low-velocity sedimentary rocks is carried beneath the San Gabriel Mountains along the north dipping Sierra Madre fault (SMF). This fact appears to mean that there has been only minor long-term convergence across the Sierra Madre fault zone, although this zone currently represents active thrusting of the San Gabriel Mountain block over the sediments of the Los Angeles basin.
This figure represents a sonogram-type image of the San Gabriel Mountains. Features that reflect seismic energy back to the surface are represented, below 5-km depth, by red colors; non-reflective areas are represented by yellow colors. Above 5-km depth (20-km depth in the Seal Beach-Puente Hills area) the sonogram image has very little detail and has been replaced by the refraction (CAT-scan-type) image which has abundant detail (see figure). The color scheme from the first figure has been changed, however: high velocity is now red instead of blue.) Feature A is a bright reflective zone that probably represents a major young thrust fault beneath the San Gabriel Mountains and northern Los Angeles basin for the following reasons:
1. It occurs at the base of the brittle part of the crust (the part of the crust that produces earthquakes). Therefore, it represents a zone where rocks can deform by flowing.
2. Careful study of this feature reveals that it, like the San Andreas and San Gabriel faults, is a zone of low seismic velocity. Such a zone can be caused by high fluid pressures, by fluids trapped in a fault zone, or by abundant cracks created by high strain in a fault zone.
3. It has a "ramp-and-flat" shape like thrust faults elsewhere in the world.
4. A faint reflection that rises from the top of this zone beneath the southern San Gabriel Mountains (55-km distance on the top scale of the figure) projects to the hypocenter of the 1987 M 5.9 Whittier Narrows earthquake, which occurred on a buried thrust fault. Note that feature A appears disrupted at a point vertically beneath the surface trace of the San Andreas fault (SAF), indicating that the San Andreas fault probably extends at least this deep. (In contrast, feature B appears to extend across the deep projection of the San Andreas fault, although it changes from a diffuse to a sharp feature at the deep projection.)
LARSE's location of the bottom of sedimentary rocks in the Los Angeles basin has enabled estimates of strong ground shaking during future earthquakes can be accurately calculated. In addition, LARSE has imaged several major faults, including a new thrust fault beneath the San Gabriel Mountains that promises to be part of the earthquake-producing "machinery" in the Los Angeles region.
Born in 1944 in Oak Ridge, Tennessee, Gary Fuis has always loved the out-of-doors and geology."The things that have happened in the past on this Earth and in our solar system are more awe-inspiring than any fiction ever written," he said.
Fuis earned a bachelor of science degree at Cornell University in 1966. After completing his Ph.D. at the California Institute of Technology in 1974, he went to work for the U.S. Geological Survey (USGS), where he has been studying earthquakes ever since. When he first joined USGS, he operated and expanded the USGS/Caltech Southern California seismographic network.
In addition to his most recent work exploring the crustal structure and tectonics of the Los Angeles region, he has performed similar studies along the route of the Alaskan oil pipeline, from the Pacific to the Arctic Oceans. In the 1980s, Fuis also used explosions and a newly developed system of portable seismograph to investigate the crustal structure of the Salton Trough region of Southern California and of the Klamath Mountains and Mount Shasta in northern California.
Back to Science for Everyone
