When the Earth rumbled and high rises swayed on Sunday afternoon all over Southern California and parts of Arizona, most people were sure that the "Big One" had hit. This major temblor is expected to rip through the southern end of the San Andreas Fault and race northwest along the fault line causing damage and devastation on its way (see blog November, 10 2008). But when Sunday's swaying motion stopped and seismologists in Southern California Seismic Network's data center in Pasadena checked their records, it very quickly became clear that the "Big One" is still to come. Instead, the shaking was caused by a magnitude 7.2 quake south of the border, with its epicenter near the village of Guadelupe Victoria in the Mexican state of Baja California, about 35 miles south of Mexicali and 100 miles east of Tijuana.
Guadelupe Victoria lies quite a ways south of the southern tip of the San Andreas Fault (see blog November 4, 2008), roughly halfway between the Salton Sea and the northern end of the Gulf of California. From a geologist's point of view, this area is rather unique, because it marks the transition zone between two completely different tectonic regimes. All of California, from Cape Mendocino in the north to Cabo San Lucas at the southern tip of Baja California is dominated by the boundary between the Pacific and the North American Plates. North of the US-Mexican border, this boundary is represented by the San Andreas Fault system, along which the two plates slide past each other with a speed of about 2 inches per year.
South of the border, the picture is completely different. When looking at a map of northwestern Mexico, did you ever wonder about the fingerlike peninsula Baja California? How did this almost 800 mile long, extremely narrow stretch of land come into being? The cause lies under the seafloor of the Gulf of California, which separates Baja from the Mexican mainland. In this area the East Pacific Rise (EPR) rules the boundary between the two plates. Like the Mid Atlantic Ridge or the Red Sea Rift, the Earth's crust splits apart along the EPR. This split causes Baja - located on the Pacific Plate - to move westward, slowly drifting away from the Mexican mainland at less than 2 inches per year.
This movement is also causing the Gulf of California to slowly open up and to extend its shores further and further to the north. The epicenter of Sunday's quake lay exactly in the region where the Gulf is creeping north. This is a slow but steady movement, which has been going on for at least 8 million years. Strong earthquakes like Sunday's temblor have been observed in this area at least four times in the last hundred years, each one cracking the Earth's crust a little bit more. Down there in the salty marshlands where the Colorado River drains into the ocean, the land is slowly ripping apart, like opening a zipper in slow motion, one tooth at a time. (hra054)
|Figure 1: A theodelite.|
Earthquakes do not only generate seismic waves, which can be as devastating like those which recently hit Southern Chile and Haiti. They also cause a lasting movement of the Earth's crust. In fact, it is this very movement which - over millenia - defines the rate at which the tectonic plates drift over the Earth's mantle.
This permanent displacement was first noted a little more than one hundred years ago right on our doorstep. When scientists under the leadership of UC Berkeley's Andrew Lawson fanned out to investigate the effects of the Great San Francisco Earthquake of 1906, one of them, Harry F. Reid from Johns Hopkins University in Baltimore, took his theodelite along. In those days, these instruments (see Figure 1) were used by surveyors to measure the angles between fixed points in the field exactly, be they corner monuments of homesteading parcels or the locations of railroad tracks. Before the 1906 earthquake several such survey lines had been laid across the San Andreas Fault, which was then known among geologists as the "Rift Zone." When Reid revisited these points after the quake and took measurements with his scope, he noted that invariably points on opposite sides of the fault had moved to the right, in some cases by up to 18 feet.
In today's age of satellite navigation, theodelites are an item of the past, having been replaced by highly sensitive, geodetic GPS receivers. These instruments measure movement with much greater precision than your ordinary GPS instrument in a car. They also transmit the data in real-time to research labs, like the Berkeley Seismological Laboratory. In the Bay Area alone, researchers from half a dozen institutions jointly operate a network of 67 such GPS receivers and continuously measure the movement caused by the slip of the Pacific and North American Plates. (See blog October 17, 2008.)
A similar, international cooperation has existed for years in South America under the acronym CAP - Central and Southern Andes GPS Project. This group, led by Mike Bevis from Ohio State University, operates several dozen GPS sites in many countries in Latin American, from Bogota in the North to Punta Arenas in the South. When these researchers compared the GPS data from before and after the Great Chilean Earthquake of February 27th, 2010, they noticed a huge movement. The giant temblor had moved the city of Concepcion in the heart of the epicentral region by more than 10 feet to the west. Chile's capital Santiago moved more than 10 inches, and even in Buenos Aires, more than one thousand miles from the earthquake's focus, the Earth's crust had moved by 1.5 inches towards the Pacific. (hra053)
|Figure 2: Preliminary Coseismic Displacement Field M 8.8 Maule Earthquake, Chile, Feb 27 2010 (courtesy of CAP).|
|This map shows all earthquakes in southern Chile over the last 100 years with magnitudes larger than 5.5. Until Saturday, the area around Concepcion had much fewer quakes than the rest of coastal Chile. The yellow star depicts the epicenter of the most recent 8.8 temblor. (Map courtesy of USGS)|
The huge earthquake, which on Saturday devastated large areas of Chile, was the largest temblor anywhere on Earth in more than five years. With a magnitude of 8.8, it was about 30 percent smaller than the great Sumatra earthquake of December 28, 2004, but also almost one thousand times more energy than the last destructive temblor in our area, the Loma Prieta earthquake of October 1989.
Like the Haiti earthquake of January 12, 2010, this great Chilean earthquake was not unexpected. Here is why: The entire Pacific coast of South America is prone to very large earthquakes, because it defines the collision zone of the westward moving South American plate and the eastbound Nazca Plate. These plates move toward each other at a rate of almost 3.2 inches per year. The South American plate is continental and carries all of Latin America south of the Isthmus of Panama. The Nazca-Plate underlies most of the southeastern Pacific Ocean and, like all oceanic plates, it is much denser than a continental plate. Due to this density difference, the Nazca-Plate is pushed under by the South American Plate by the collision and dives into the Earth's mantle. In the region of the epicenter of Saturday's quake, the angle under which the Nazca Plate subducts beneath South America is about 27 degrees.
As the oceanic plates dives, it tries to take the leading edge of South America with it, thereby bending it downwards. As continental crustal rocks are neither plastic nor elastic, but rather brittle, the bending causes an enormous strain in the rocks of the westernmost South American plate. Finally the bending continues so far that the strain exceeds the strength of the rock. It then breaks and the bending edge of the plate snaps back. This happens extremely quickly and results in huge earthquake - as we saw on Saturday.
However, not all areas affected by the subduction of the Nazca-Plate snap back at the same time. During the last century, regions along most of Chile's 2000 mile long coastline experienced such a snap back in the form of a huge earthquake. Seismologists counted at least four quakes with magnitudes larger than 8 and countless temblors between 7.0 and 7.9. In fact, the largest temblor ever measured occurred just 200 miles south of Saturday's epicenter, when on May 22, 1960, the Earth shook in a magnitude 9.5 earthquake.
The only region along Chile's coast which has been an exception, and had not experienced a major quake in the last hundred years, was the coastal area of Concepcion. And this is exactly, where Saturday's "terremoto" hit with its devastating effects. (hra052)
|Port-au-Prince Harbor after the earthquake (photo by Eduardo Fierro)|
We all have seen the pictures of the dead, the shots of the devastation and moving reports about the suffering after the earthquake, which struck Haiti on January 12. The Haitian government estimates that at least 150,000 people have perished in the disaster and that more than 800,000 people have become homeless. Aid organizations estimate the number of dead to be higher, over 200,000, and put the homeless count at more than one million. Whatever the final numbers, one thing is clear already: This temblor was certainly the most devastating magnitude 7 earthquake in human history. Keep in mind that on average around 15 quakes of similar magnitude occur every year - and only a few of them cause severe damage and deaths (see blog September 14, 2008).
Why, one has to ask, did the quake under the Haiti cause such an unfathomable disaster? "It didn't have to be that bad," says Eduardo Fierro a structural engineer with the firm BFP Engineers, Inc., based in Berkeley. Fierro goes one step further when he postulates: "This was not an earthquake disaster, this was a building disaster." Fierro came to this opinion after spending one week in western Haiti immediately after the earthquake struck. He wanted to find out, why this quake caused such extreme damage to the island's buildings and infrastructure. His conclusion: In most cases, the buildings were neither designed nor constructed to withstand even the slightest lateral shaking. All over the capital, Port-au-Prince, and in the epicentral town of Leogane, he inspected collapsed buildings - and the picture was always the same: the rebars were much too thin, interconnections between them were missing, and concrete was of poor quality. Not only the ramshackle houses in the slums were constructed that way. Two of the biggest public buildings in Haiti, the Presidential Palace and the University in Leogane, were also not designed with earthquakes in mind and consequently collapsed in the shaking more than two weeks ago.
|The cathedral in Leogane was completely destroyed. (Photo by Eduardo Fierro)|
Fierro recently presented his findings to structural and earthquake engineers in a special seminar on Campus. His saddest observation: He saw some people in Port-au-Prince starting to rebuild their home. That they were collecting intact cinderblocks from collapsed buildings and walls was understandable, Fierro said. But that these people were again using flimsy rebars and cement mixed with salty sand from the beach, almost broke his heart. "They are setting themselves up for the next devastating disaster", he said. Fierro's trip was partially funded by the Pacific Earthquake Engineering Research Center (PEER) on UC Berkeley's campus, which also allowed the blogger to use two of his pictures. (hra051)
|Enriquillo Plaintain Garden fault landform SSW of Port-au-Prince (Image from Google Maps)|
One of the reasons, that the massive relief effort after Tuesday's devastating earthquake in Haiti is only very slowly getting into high gear, is the fact, that many roads are impassable and numerous bridges are destroyed. First responders and rescue teams use satellite images to assess the destruction and accessibility of various neighborhoods in Port-au-Prince and in the outlying towns and villages. In addition, the view from space also helps Earth scientists to gain insight into the tectonic movements, which caused the earth crust to break so violently on Hispaniola.
As described in Wednesday's blog, the focus of Tuesday's magntiude 7 temblor lay at a depth of 8 miles on the Enriquillo-Plaintain Garden fault, which cuts through the southern section of Hispaniola. It runs along an almost straight East-West trending line through the mountains southwest of Port-au-Prince. The fault scarp is clearly visible in the satellite picture (Figure 1), which was taken last year from an imaging satellite. If you zoom in on it, you can see that the scarp is actually a linear valley with a stream running through it.
|Figure 2: Three-dimensional elevation model of the region around Port au Prince. The mountains are not snow-capped, rather the color is an indication of the elevation (green low, white high). (Image from Jet Propulsion Laboratory)|
This scarp is also clearly visible on the maps which Nasa created from the data gathered by the Space Shuttle Radar Topography mission. During several flights of the Space Shuttle an automatic radar camera took pictures of the Earth's surface underneath the Shuttle's orbit. For each spot, two pictures were taken from slightly different angles. Various computer codes can be used to combine the two images. Figure 2 shows a three dimensional map of the region around Port-au-Prince under an oblique view from the northwest. The scarp in the mountains is the prominent feature in the center of the image (white arrow). Figure 3 shows a different combination of the two radar images from Haiti. They were merged into an anaglyph. If you happen to have a pair of those crazy 3-D glasses handy, (red for left eye, cyan for the right), you will also clearly see the fault.
In the meantime, Gavin Hayes from the USGS's National Earthquake Information Center in Boulder, CO, has used dozens of seismic recordings to compute the actual movement along the Enriquillo-Plaintain Garden fault. According to his calculations, the fault ruptured exactly underneath the visible scarp over a length of approximately 25 miles. It broke in a left-lateral movement, as indicated by the arrows in figure 3. Within roughly 15 seconds, the northern block had shifted by almost 9 feet to the left, while the southern block had moved a similar distance to the right, resulting in a total fault slip of about 18 feet. Such values are typical for a magnitude 7 quake. During the Bay Area's 1989 Loma Prieta shaker the San Andreas Fault ruptured over a length of approximately 22 miles. (hra050)
|Figure 3: Radar anaglyph image of the Port-au-Prince area. Red arrows indicated the direction of plate movement.(Image from Jet Propulsion Laboratory)|