Blantied USGS Q&A August 17 2007


QUESTION: Do those signals correspond to small earthquakes in the USGS or Berkeley catalogs? If so, are those earthquakes located close to the later epicenters such that they'd be classified as foreshocks? If not, what distinguishes them from earthquakes?
REFERENCE: See Page 10, Paragraph 3 of Science Horizons Internal Report #IR-07-0005.
ANSWER: We have not identified the precursors with any specific physical event, e.g. earthquake or acoustic emission. What was important was the identification of the physical characteristics of the precursor. We have done that.

QUESTION: Since the array processing technique pinpoints an azimuth but not a distance, there's a wide swath of earth that could potentially have hosted the precursor source. For example, can you rule out the possibility that you're seeing the P-wave signal from an Aleutian earthquake? Fortunately the relevant earthquake catalogs are easily searched for such information.
ANSWER: One way to solve this problem is to use multiple arrays to determine the distance of the precursor.

QUESTION: What is the seismic magnitude (or equivalent magnitude) of the precursor event? (This could be determined by comparing the signal strength to that of the mainshock.) Given that magnitude (i.e., strength of p wave at the source), would one expect to see it recorded on high-gain seismic stations close to its epicenter?
ANSWER: We are working on the magnitude issue. It's not as easy as one might think because the only relevant magnitude data that SeismicRadar® gives us is the SNR. However, what we need to do is to back out the signal strength of the precursor from the raw seismic data.

I intend to install a seismic array in a tectonically active area and I expect the number of precursors to increase substantially.

QUESTION: What is the frequency of the wave train that crossed the Lajitas array? Given known attenuation of P-wave energy, would one expect to those amplitudes to be observed at 1000 miles? Asking it the other way around, given those amplitudes, what would the amplitudes have had to be at the source? (This is similar to the previous set of questions, asked a different way.)
ANSWER: We obtained the best results for the P-wave at frequencies between 0.3 Hz to 1.2 Hz. For the precursors, the best results were obtained between 0.3 Hz to 0.75 Hz. The P-wave was easy to identify visually (See Figures 9 and 18 of Science Horizons' Internal Report #IR-07-0005). Only SeismicRadar® was able to identify th precursors

QUESTION: The Lajitas array has presumably collected data before from numerous M6 events, including many off the Mexican and Central American coast that are located closer than Parkfield (some with much larger magnitude). Also the 2003 M6.4 San Simeon earthquake was located very near Parkfield. Can precursors be identified for any or all of those earthquakes?
ANSWER: Southern Methodist University is the operator of the Lajitas array. They should answer this question. All they need is the SeismicRadar® processing tool

QUESTION: Is there something unique about the 1000-mile distance, or can the precursor signals be identified for earthquakes at many thousands of miles distance?
ANSWER: There is nothing unique about the 1000 mile distance. I chose the Gulf of Mexico earthquake because it was felt in Melbourne, Florida. I chose the Parkfield earthquake because there allegedly were no precursors. I chose TXAR because that was an AQUINAS® System arra that Science Horizons delivered and we have worked well with colleagues at SMU. It was a coincidence that both earthquakes were about 1,000 miles from TXAR.

QUESTION: A devil's advocate would ask whether the precursors might be either randomly correlated noise or reflective of waves passing through the array whether or not there's a mainshock coming. It would be interesting to "invent" some mainshocks that did NOT occur, and search for precursors. This is commonly done (and necessary, really) in pattern-recognition studies of earthquake signals.
ANSWER: So far we've only used a random 7 degree arc, and we haven't been able to find any precursors in that interval.