New York seismic record (Palisades broadband station)

This signal was recorded in Palisades, NY at the Lamont-Doherty Earth Observatory. This seismic station has been operating continuously for 50 years. Several phases are visible in the record.

PHASE	P	PP	S	SP	SS	Love waves	Rayleigh waves
ARRIVAL TIME ON RECORDS	799 s	1024 s	1458 s	1555 s	1839 s	2300-2600 s	>2900 s

The first 2000 seconds after the quake
The first 3600 seconds (one hour) after the quake
The first 5000 seconds after the quake
These broad band signals have been filtered between 0.01 and 0.1 Hz to mimic a long period record. The seismograms show a large wiggle at time zero. This is an artifact of the filtering and not part of the earthquake's signal.

The plots show all three components of motion. The top signal is the vertical ground motion (up is positive); the middle is the horizontal ground motion in the north/south direction (north is positive); the bottom signal is the horizontal motion in the east/west direction (east is positive). The reason for recording three compoents of motion is clear from this record. The P waves, which consist of largely vertical motion, are most evident on the vertical component seismogram. The S waves, which contain lots of horizontal motion are more clear on the north and east records. Likewise for PP and SS.

New York City is 10,485 km from the epicenter of this earthquake. That is just over 1/4 of the way around the globe (94.3 degrees). At this distance there is a clear seperation of different phases. Several body waves (P, PP, S, SP, SS) are distinct from one another. Likewise, the surfaces waves (Love and Rayleigh) are clearly seperated. As with the S waves, the long period Love waves are more clear on the horizontal seismograms. This is in keeping with their side to side horizontal motion. The Rayleigh waves, which have a strong up and down motion are clear on all three records but particularly on the vertical component.

Component of motion	Seismic interpretation of body waves (visible on "first 2000 seconds" seismograms)
Vertical: North/South: East/West:	___P_____PP___________SP_________SS________ _________PP________S_____________SS________ _________PP________S_____________SS________

PP and SS are regularly observed at this distance. These phases are just like P and S except they have bounced once off the surface of the earth. The bounce point should be half way between the earthquake and the seismometer.

The clear SP phase is an interesting aspect of this seismogram. This is a wave which travels through the earth and bounces off the surface once on its way to New York (just like PP and SS). However, this phase travels the first leg of its trip as an S wave. When it bounces off the surface, a portion of the energy is converted to a compressional wave (P wave). The PP phase arrives before SS because compressional waves (P waves) travel faster than shear waves (S waves). Not surprisingly then, the SP phase arrives between PP and SS.

Using this same rationale, the astute observer would realize that a PS phase (a wave travelled first as a P wave and then as an S wave) should have the same traveltime as an SP phase and cause confusion on the seismograms. This is true. However, in this case, we are predominately seeing SP. Since SP reaches the seismometer as a P wave, it should be most visible on the vertical record, which it is. PS would logically be more clear on the north and east seismograms.

Records from the global seismic network

The ardekul earthquake was recorded on seismometers around the world. Shown above is a compilation of fifteen seismograms recorded by the Global Seismic Network. The closest station lies 1800 km from the epicenter in Kislovodsk, Russia while the furthest station is nearly half way around the globe, 16,500 km away. These stations have been arranged so that they are each roughly 10 degrees (1110 km) further from the source than the station above.

distance from Ardekul	Seismic record interpretation
10 degrees 20 degrees 30 degrees ... degrees ... degrees 140 degrees 150 degrees	__direct____________other direction___ ___/\/\______________________/\_______ _____/\/\__________________/\_________ _______/\/\______________/\___________ _________/\/\__________/\_____________ ___________/\/\______/\_______________ _____________/\/\__/\_________________

The left side of the plot is the time of the earthquake. Four hours of data is shown so the right side of the plot is the time after 4 hours. Note that some stations far from the source, archive only the data of interest so not all the seismograms begin at time=0. However, they have been aligned so so that they are all on the same time scale. The large waves displayed in the plot are the surface waves. The onset of the body waves is visible but just barely on this scale. As in the local seismograms, the large wiggle at the beginning of each trace is an artifact of the filtering and not part of the seismic record.

It is immediately obvious that the traveltime to each of these stations increases with distance. This is logical given that these waves typically travel between 3.5 and 4 km/s. Close inspection of this plot however will reveal another set of surface wave. Instead of arriving later at distant stations, this set of waves shows up first at the most distant station and last at the closest station. This counterintuitive wave is easily explained by thinking of the Earth as round. Surface waves radiate out in all directions from an earthquake. This second set of arrivals are the waves which travelld the other direction around the globe. The fact that they have travelled a greater distance explains why they are smaller. Together with the first surface waves they form a V on the record.

Larger version of global seismic records plot shown above

Map of stations around the world which recorded this earthquake
image from Incorporated Research Institutions for Seismology (IRIS)

Return toArdekul earthquake main page

Last update: October 1, 1998