The underground test of a nuclear weapon in North Korea on Memorial Day was registered by earthquake detectors all around the world - among them many stations of our Berkeley Digital Seismic Network (see blog January 9, 2009). Our station deep inside the Yreka Bluehorn Mine (Siskiyou County) - dubbed YBH - is also part of the International Monitoring System (IMS). It is run by a suborganization of the United Nations in the Austrian capital Vienna, which has what is probably one of the most complicated names on Earth: "The Preparatory Commission for the Comprehensive Nuclear-Test-Ban Treaty Organization."
Underground detonations of nuclear weapons can be detected like earthquakes for a simple physical reason. In both cases - either when rocks rupture in a quake or during the explosion - very strong forces rapidly act inside the Earth. This leads to intensive shaking of the rocks around the hypocenter, which in turn generates elastic waves. They can travel thousands of miles and are detected by sensitive seismometers.
There are, however, major differences between the seismograms of natural tectonic earthquakes and those of explosions. Firstly, the waveforms look very different. While an earthquake generates strong S-Waves, the seismograms of underground nuclear test lack most of these waves. Instead, the P- (or primary or pressure) waves dominate the seismogram from the detonation of an atomic bomb below ground (see figure 1).
A second way to distinguish between the origins of the elastic waves is to analyze the data collected by many stations in what is known as a "Moment Tensor Solution." By performing this computation, seismologists trace the elastic waves back to their origin. That not only pinpoints the precise location of the focus, it also shows the mechanism of the forces initially shaking the rocks. During an earthquake, rock breaks in a shear fracture, which results in the rapid sideways movement of two flanks of a fault. In an explosion, however, the origin is indeed a point, from which elastic pressure waves travel concentrically outward. Berkeley's Moment Tensor Solution for the most recent North Korean test is shown in figure 2. (hra039)
|Figure 1: From "Sleuthing Seismic Signals", Science and Technology Review, March 2009, published by Lawrence Livermore National Laboratory.|
|Figure 2: Moment tensor for the North Korea seismic event of 25 May 2009 calculated by Prof. Doug Dreger using the BSL's complete waveform regional moment tensor code for a source depth of 800m. The seismic waves from this event are consistent with a shallow explosion source.|
When the Earth began shaking under the fertile soils of the Red Basin in China's Sichuan province a year ago today, nobody knew how bad it was going to get. It turned out that with its magnitude of 7.9, this quake was not only the world's strongest temblor last year. It was also the worst earthquake disaster to hit China in more than 30 years. Up to 80,000 people - an official number was never released - died in the rubble of the quake. Many victims were school children, as more than 14,000 classrooms were damaged, half of them collapsed entirely. Almost 400,000 people had to be treated for their injuries, which they sustained under tumbling buildings or in the numerous landslides in the mountainous epicentral region.
The earthquake occured at the Longmen Shan Fault Zone, which marks the tectonic boundary between the eastern Tibetan Plateau and the Sichuan Basin. Its epicenter (yellow ball in Figure 1) was located less than 60 miles from Chengdu, the capital of Sichuan Province with its 11 million inhabitants. Here is what happened during the quake: Tibet, being pushed constantly by the northward movement of the Indian subcontinent, squeezed out a part of itself towards the east, resulting in this major "thrust earthquake" (Figure 2).
But several Chinese scientists speculated that other factors may have contributed to the severity of the quake. In the Chinese journal "Geology and Seismology," Fan Xiao, the chief engineer of Sichuan's provincial geology office, and others even went so far to say that the water in a recently built reservoir might have triggered this devastating earthquake. In 2001 work was started at the Zipingpu Reservoir. Its 470 ft high concrete dam sits just upstream from Dujiangyan. This city is home to a Unesco World Heritage site. The Dujiangyan Irrigation System was built in 256 BC and is still in use today. But the dam was also located just four miles from epicenter and was damaged severely during the quake.
The scientists speculate that the extra pressure of the water in the reservoir may have caused the Longmenshan Fault to slip, thereby releasing the tectonic energy stored in it. That does not mean that the earthquake would not have happened anyway. But the additional hydrostatic pressure from the 320 million tons of water stored behind the dam may have triggered the quake several years before it would have been due without the dam. We will explore how such "induced seismicity" works in one of the future blogs. (hra038)
|Figure 2: A piece of lower crust (red blob) squeezed between the Tibetan Plateau and the lowlands of Sichuan. Image courtesy of MIT Department of Earth, Atmospheric, and Planetary Sciences.|
Fig 1 (oil film). Photo: Horst Rademacher
Earthquakes make waves. That's a trivial, commonplace statement, given the fact that seismic waves can violently shake the ground and topple buildings. But there is another kind of wave generated by temblors, which can only be seen after heavy numerical manipulation of the data from certain types of satellites.
Fig 2 (InSAR), Map courtesy of INGV and ESA (Click to view larger image.)
Here is how it works: About a half dozen of the satellites currently circling the earth continuously send out radar beams to probe the Earth's surface. They have a separate sensor on board which registers the radar beam's reflection off the ground. The primary mission of these "Synthetic Aperture Radars" (SAR) is to exactly map the altitude of the surface with a precision of one half inch or less. To the delight of earth scientists, the radar sensors have an additional benefit. Every few weeks, the satellite path repeats. That means it passes over the same regions of the Earth in the same order. It is then possible to detect slight changes in the elevation of the Earth's surface which have occurred from one fly-over to the next.
Let's apply this to seismology: The European satellite "Envisat," which carries a SAR sensor, flew over the Abruzzo region of Central Italy on February 1, 2009 and then again six weeks later on April 12. In the mean time, a devastating earthquake occurred near the town of L'Aquila, killing more than 260 people (see blog April 6, 2009). Italian scientists compared the radar pictures taken during the two successive orbits using a numerical technique called "Interferometry," essentially subtracting one picture from the other. The result is a map of colored waves, similar to the iridescent patterns one can see on a soap bubble or when an oil film covers a puddle (see Fig 1). Just as a physicist can use these color patterns to determine the thickness of the oil film or the bubble walls, Earth scientists can interpret the maps. Each colored ring is a measure of how much the ground has moved as a result of the earthquake. In this case, each ring represents a movement of just over an inch.
Figure 2 shows this pattern for the Abruzzo quake. The large green square represents the location of the main shock; the smaller green squares show large aftershocks. Along the yellow line east of
L’Aquila geologists found an alignment of surface breaks after the quake, which indicate the orientation of the rupture. The colored wave pattern follows those breaks exactly, indicating that the ground had moved a total of seven inches down to the left side of the yellow line. This movement is also represented by the black and white fault plane solution on the left. More on these "beach balls" in one of the next blogs. (hra037)
Earthquake location (Map courtesy of INGV)
The earthquake which struck the Abruzzo region of Central Italy last night (local time) was the worst temblor to hit this Mediterranean country in almost 30 years. According to news reports, more than 90 people were killed and in the medieval town of L'Aquila alone, up to 10,000 buildings were damaged. The same region about 70 miles northeast of Rome was hit on January 13, 1915, when a magnitude 7.0 quake leveled the town of Avezzano, killing at least 33,000 people. Scientists at the office of the USGS in Golden, CO, calculated the magnitude of the latest quake at 6.3, while Italy's National Institute of Geophysics and Volcanology (INGV) put it at 5.8.
Beside Greece and Romania, Italy has the highest earthquake risk in Europe. During the last 100 years alone, more than 120,000 people have died in the rubble of buildings destroyed by quakes. The cause for this volatility is the slow drift of the African continent towards the North. The whole Mediterranean region is the collision zone between Africa and Europe. In California, we also live in a contact zone between two plates. Here the plate boundary between the Pacific and the North American Plates is limited to a relatively narrow zone defined by the San Andreas Fault and its associated faults like the Hayward and the Calaveras. Along these faults, the two plates slide past each other, the Pacific Plate moving towards the Northwest at a rate of about 2 inches per year.
|This map shows the seismic hazard in Italy. The Abruzzo region (star) is one of the most vulnerable in the country (Map courtesy of Global Seismic Hazard Assessment Program)|
In Italy, however, the movement of the plates is a frontal collision. In a stringent geological sense, Italy's boot is part of the African Plate, ramming into Europe like a nail being hammered into a board. The most obvious consequence of this movement is the folding of the Alps, the highest mountain range in Europe. But this collision also leads to a movement in the shaft of Italy's boot. The Apennine mountains, which are the core of the shaft, are fractured by several major earthquake faults. One of them is the fault west of the Gran Sasso mountains, which was activated in 1915 and then again last night. The strongest quake to have hit Italy in modern times struck on December 28, 1908 at the toe of the boot. The epicenter of the magnitude 7.5 temblor was in the Strait of Messina, a narrow waterway separating the Italian mainland from the island of Sicily. At least 86,000 people were killed by this major temblor. (hra036)
While most of the Bay Area was rattled on Monday morning around 10:40 am by a magnitude 4.3 earthquake near Morgan Hill, another earthshaking event went almost unnoticed by the public - unless you were in Yosemite over the weekend. Early Saturday morning, a huge mass of rock came crashing down from Ahwiyah Point near Half Dome. Greg Stock, the Park Geologist at Yosemite, writes that the rocks "fell roughly 1800 feet to the floor of Tenaya Canyon, striking ledges along the way. Debris extended well out into Tenaya Canyon, knocking down hundreds of trees and burying the southern portion of the Mirror Lake loop trail... Fortunately, due to the event occurring in the early morning, there were no injuries."
But what happens when tons and tons of granite come crashing down unto the valley floor? Such an impact makes the ground vibrate and thereby creates seismic waves very similar to the ones being radiated by an earthquake. Indeed, on Saturday morning seismic stations all over Northern California and Nevada - as far away as 250 miles from Yosemite - registered these waves. The automatic earthquake location computer for Northern California at the offices of the USGS in Menlo Park picked up the recordings and calcuated an epicenter just half a mile to the northwest of Half Dome - which is actually pretty close to Ahwiyah Point. The program even computed a magnitude for the rock fall: Its impact had the same energy as a magnitude 2.4 earthquake.
While the seismic waves generated by a rock fall can be mistaken for the rumblings of an earthquake, the physics behind the two phenomena is completely different. Most earthquakes are the result of tectonic stress, which has accumulated in the rocks due to the movement of the lithospheric plates. A rock fall happens when the rock has been weakened by weathering. Water, which accumulates in cracks, freezes during the winter frosts. As ice occupies a larger volume as the same mass of liquid water, the freezing ice makes the rock expand and burst - similiar to a water bottle left in the freezer for too long. If such cycles of freezing and thawing are repeated often enough, the rock becomes loose and can break.
These rock bursts are by no means rare in granite world of Yosemite. Last October two rock falls hit some of the tents and cabins in Curry Village. In July 1996 more than 162,000 tons of rock cascaded down more than 2,000 feet, killing one visitor and crushing 500 trees. This blast was also recorded on many seismic stations, although it was somewhat smaller than Saturday's rock fall. After the 1996 event, BSL's Bob Uhrhammer analysed the seismic data carefully and reconstructed the details of the fall.