Oklahoma Geological Survey annotated 24 hour seismograms for year 2007.

2007 JAN scroll 106 eq identified, 5849 seismograms served In the above list "eq identified" gives the number of separate earthquakes identified on the seismograms. If an earthquake has both P/PKP and LR marked, even though they may be an hour apart, it is only counted as one earthquake. The number of seismograms served is the number of seismograms shipped to users during the month. The number 5439 for September means that 5349 seismograns were served to users during September, but they were not all September seismograms. Some could be from as far back as 2000.

Seismogram Index For Year 2006

Seismogram Index For Year 2005

Seismogram Index For Year 2004

Seismogram Index For Year 2003

Seismogram Index For Year 2002

Seismogram Index For Year 2001

Seismogram Index For Year 2000

Back to OGS home page



SHORT EXPLANATION:

The OGS/OU 24 hour seismograms have one main purpose: to show users as many
earthquakes as possible, with details labelled. A large earthquake may
make all or part of the seismogram solid black. That is why exponential
compression (power<1) is used. It can put the large earthquake "on scale",
while still leaving smaller earthquakes clearly visible. When the power >1,
there is a triggering effect. The traces are exaggerated as they rise above
the background earth noise.

The multipliers are used to raise or lower the level of the band (medium frequency,
or low frequency), before the power is applied. You may notice that with the 
Johnso- Matheson seismometer, the medium multiplier is only 0.35 typical, while
the low multiplier may be 250 typical. This is because the seismometer has a high
sensitivity at medium frequencies and a very low sensitivity at low frequencies.

Most of the seismograms were produced from 16 bit 20 sample per second
data from a Johnson-Matheson short period vertical seismometer in the
Leonard walk-down vault. A few were produced from a GS-13 short period
vertical seismometer in the nearby Russian vault. It is so named because
Soviets, later Russians, recorded American underground nuclear blasts in
the Russian vault in the early 1990s. This was part of the verification
of the Threshold Test Ban Treaty. The vault has since been released by
Russia to the United States. The United States released the dollar a year
lease from OU. The GS-13 signals are 24 bit 20 samples per second.

Using eight pole Bessel Bandpass filters, a medium frequency band (0.6 t0 1.4 Hz),
and a low frequency band (0.02-0.05 Hz) are extracted from the 24 hour files.
Each band is multiplied by a constant number (see multiplier in lower left of
seismograms), and raised to a power (lower left). The four numbers are manually
picked for each seismogram. The powers may be less than one, which produces a
non-linear compression of the band, or greater than one, producing a non-linear
"trigger" effect. The trigger effect exaggerates larger amplitudes.

The resulting bands (medium and low) are added and plotted.

If the first plot shows no low frequency signal, the low frequency multiplier
may be set to zero, which causes the low band to not even be raised to a power
or added.

Typical values when the Johnson-Matheson seismometer is usedr: low multiply 250,
low power 0.8, medium multiply 0.35, medium power 1.3.

The multiply power scheme is as follows where X is a series of seismic samples:

 multiply X by multiplier
 save list of signs of X (postive or negative)
 change all X to positive numbers (absolute values)
 take common (base 10) logarithm of X
 multiply log X by power
 take 10 to the power X
 use list of signs to restore original signs to the multiplied powered X

Usually the medium band shows P waves which travel through the earth's mantle,
or PKP which travels through mantle-core-mantle. The medium band also shows
limestone quarry blasts, whose signals are mainly surface waves. The low band
shows almost exclusively Rayleigh (LR) waves. These travel along the crust and upper
mantle similar to ripples on a still water surface. Rayleigh waves (LR) are
retrograde, pond ripples are prograde. 

P or PKP are always a few minutes to an hour or more earlier than LR for the
same earthquake. Other phases like PP, S, SKS, SS, Pdiffracted are sometimes
seen and labelled on these seismograms.

Distances (eg D=65) are in great circle degrees. Hypocenter depths (eg h=510 km)
are given only when the depth exceeds 50 kilometers.

The seismograms were developed gradually over a period beginning in 2000 by
Jim Lawson. Although obvious, the exponential compression/expansion may be original.
It differs greatly from arcsinh compression published decades ago. The data processing
depends on freeware/openware including:
Operating systems: SuSE Linux 9.x 10.x, in the past Solaris, in the future we
                   may convert partially to UBUNTU Linux.
Languages: PERL, NAWK, EXPECT, BASH shell.
Prograns: ImageMagick, SAC (Seismic Analysis Code), PERL, NAWK, EXPECT, BASH scripts
written by Jim Lawson.

Most of our computers were surplused from BP/Amoco and others. Amie Gibson and
Jim add memory and disks, and install Linux. These old slightly modified computers
are in no way considered junk. They run 24/7 for years, some handling over 1000
seismic data points per second in near real time. With Linux typical times between
reboot are 100 days.
 
The epicenters used in annotation are from these three sources:

1. SED = Swiss Seismological Service/Schweizerischer Erdbebendienst
   seismo.ethz.ch/redpuma/redpuma_ami_list.html
2. USGS = U.S. Geological Survey, National Earthquake Information Center.
   neic.usgs.gov/neis/bulletin/
3. KNET = Kyrgyzstan Real-time Broadband Array
   epicenter.ucsd.edu/~quake_KN/bulletin.txt

SED epicenters include USGS and KNET epicenters, as well as epicenters from
several other sources.