November 15, 2001
A weekly feature provided by scientists at the Hawaiian Volcano Observatory.
Day in the Life of an Earthquake Analyst
A "Volcano Watch" article several months ago provided an overview of the recently installed current earthquakes web pages at the Hawaiian Volcano Observatory (HVO). Behind the scenes at HVO are people responsible for maintaining the seismic stations, performing data analysis, and producing catalogs, along with many other data products.
Perhaps one of the more routine, but also most important, jobs at HVO is that performed by our seismic data analyst. It is the data analyst's responsibility to ensure that all automatically triggered earthquakes are reviewed. The job entails many hours at a computer workstation, timing the arrivals of primary (P) and secondary (S) waves recorded on seismograms. Maximum amplitude and signal duration are also measured to estimate the earthquake's magnitude.
Earthquakes are located using a computer program that calculates the difference between the analyst's timed P- or S-wave arrival and the expected arrival time based on the distance the seismic station is from the hypocenter. The hypocenter is the location of the earthquake source in three dimensions: latitude, longitude, and depth. The expected arrival time is computed using a mathematical model of the speed that the waves travel through the varying ground beneath Hawai`i.
The difference that is determined for each arrival time is squared, yielding a positive value. The squared differences for all the arrival times read from the seismograms are added, giving us an indication of how well the data fit the initial hypocenter and origin time (the time the earthquake occurred at the source). The computer program then changes the hypocenter and origin time slightly and recalculates the sum of the squared differences, doing this many times until it finds the hypocenter and origin time that gives the smallest result, or best data fit.
A few arrivals timed by the seismic data analyst may differ significantly from the expected arrival times. This can be caused by a noisy station signal, insufficient seismic energy recorded at the station, differences in actual travel time from the mathematical model, a bad reading by the analyst, or a combination of these. It is up to the analyst to use his/her expertise to decide whether to keep the reading, eliminate it, or make a revision. An experienced analyst is accustomed to the performance of each seismic station in the network and can make an educated guess as to why a certain station's reading does not match the expected arrival time well.
Some of the triggered events recorded at HVO are from sources other than earthquakes in Hawaii. Signals due to ground and air traffic, blasting activities, severe weather, and volcanic activity may be recorded by several seismic stations almost simultaneously, resulting in a trigger on the computer. Arrivals from large, distant earthquakes may also be recorded by our sensitive instruments, fooling our automatic location programs. Recognition of these types of signals is important in becoming an expert seismic data analyst.
Once the earthquakes for the day are located, the reviewed location and magnitude information for those above magnitude 2.0 are fed to the current earthquakes web pages, accessible from the HVO home page (http://hvo.wr.usgs.gov). Many smaller earthquakes are eventually catalogued, too. Erroneous earthquakes and those below magnitude 2.0 are deleted from the web pages. The text messages accompanying the recently edited earthquakes will indicate that they have been reviewed by a seismologist.
For a brief period, our seismic data analyst is caught up and all earthquakes on our web pages have been reviewed. Unfortunately, this does not last for long. Inevitably, earthquakes and other triggered event files begin to refill our computer directories after working hours, and the process starts anew the next working day.
Eruptive activity of Kilauea Volcano continued unabated at the Pu`u `O`o vent during the past week. Lava moves away from the vent toward the ocean in a network of tubes and descends Pulama pali in several separate tubes. Breakouts from the Kamoamoa tube system feed surface flows above the pali and in the coastal flats. A large surface flow, supplied by a breakout from the Kupapa`u tube system, is approaching the seacoast to the west of Kupapa`u. Lava continues to enter the ocean in three locations: Kamoamoa, Kupapa`u and the area east of Kupapa`u.
The public is reminded that the benches of the ocean entries are very hazardous, with possible collapses of the unstable new land. The steam clouds are extremely hot, highly acidic, and laced with glass particles. Swimming at the black sand beaches of the benches can be a blistering or even deadly venture.
Two earthquakes were felt earlier this month but were not reported in this section. A resident of Volcano felt an earthquake at 7:30 p.m. on Sunday, November 4. The magnitude-2.9 earthquake was located 16 km (9.6 mi) southeast of Ho`okena at a depth of 12.3 km (7.4 mi). On Monday, November 5 at 5:09 a.m., residents of Kailua-Kona and Kalaoa felt an earthquake. The magnitude-2.5 temblor was located 2 km (1.2 mi) north of Holualoa at a depth of 25.4 km (15.2 mi).
Updated: November 19, 2001 (pnf)