Geoff,

If you placed your calibrated 1 hz sensor, amp, digitizer combination at
the quietest place in the world the PDF point at 1 Hz would be on the
lower gray line, the low noise line.  Recorded earthquake or any ground
motion would lift of the point towards the high noise line.  The
digitizer bits and all that fancy stuff only affects the how accurate
the measurement was.

The point of the PDF is exactly to be able to tell as you put it "Help a
person know if their system is functioning correctly." without all the
scientific mumbo jumbo.  Once it is setup you can at a glance see if a
station is ok across the entire frequency range.  For the GSN and the
ANSS a live person looks at the PDFs every day and in about 30 minutes
knows the state of the hundreds of stations in the system.

I think that the scary word for a lot of folks in all of this is the
word "calibrated".  All I mean by calibration is that you could relate
counts or volts to displacements.  x volts from the sensor = x
nanometers of ground motion. 

You can do a PDF on an uncalibrated station you just can't relate it to
the noise models.  You could though easily see daily and seasonal
variations and you could at a glance see if your station was working
like it had been working in the past.

I am real dunce when it comes to a lot of this scientific stuff but it
would hard to make seeing if your system was working correctly much
simpler.  Especially when all I do is send the raw data to IRIS and they
do it for me.

Saludos,

Angel



On Fri, 2010-04-09 at 14:09 -0700, Geoffrey wrote:
> I hate to say this.... but some of this stuff
> really quacks me up :-)
> 
> I find it hard to translate this Montage of scientific data
> into what I want to see.
> 
> GIVEN:
> If you take a 1Hz geophone and plant it in a
> relatively quiet place free from human artifacts,
> you run the signal straight through filtering
> out everything above 1/2 the sample frequency
> as well as everything greater than 40 seconds of period.
> You amplify the signal until noise is like +/- two counts
> in a twelve or more bit A/D converter.
> Then you record this Z channel data as RAW
> single floating point 4 byte data.
> Say its on the Colorado plateau??(spelling unsure)
> OR Arizona basin range area ??
> 
> I should think such results would be regional in nature.
> Help a person know if their system is functioning correctly.
> After making such a data base of regional noise profiles
> you should not need fancy equipment to calibrate your system
> with meaningful values.
> 
> FIND:
> The frequency profile exhibited from a FFT
> ENERGY (FREQ(Y)/TIME(X))display or
> waterfall ( three dimensional (time)Y/energy(MAG)/frequency (X)) display 
> 
> Communicating in English with people trained Outside
> your own discipline is very difficult since basically
> you do not speak the same language or have the same
> meaning given the very same words.
> 
> I am just hoping you can understand me here.
> 
> But you are on the right track.
> 
> Thanks,
> geoff
> 
> 
> ----- Original Message ----- 
> From: "sismos" 
> To: 
> Sent: Friday, April 09, 2010 12:59 PM
> Subject: Re: Question to Techno-Philes
> 
> 
> > Geoff,
> > 
> > I am not quite sure that I understand what you are looking for but maybe
> > this will help
> > 
> > Take a look at this which talks about the NLNM which is sort of a ideal
> > and impossible model for any one site.  As I understand it they chopped
> > up the thing by frequency and them picked the quiets place on earth and
> > made a composite
> > 
> > http://www.geophys.uni-stuttgart.de/oldwww/seismometry/man_html/node28.html
> > 
> > then take a peek at this article about power density functions
> > 
> > http://geohazards.cr.usgs.gov/staffweb/mcnamara/PDFweb/Noise_PDFs.html
> > 
> > and then take a look at what a PDF looks like for my station BRU2.  The
> > article below is very good and has some good examples of what you can
> > see on a PDF plot.
> > 
> > http://www.iris.edu/servlet/quackquery/pdfDrill.do?station=BRU2&yyyyDDD=2010.032&length=31&location=30&channel=HHZ&network=PA
> > 
> > For all this to be done the station need to be calibrated so that counts
> > or volts or whatever can be converted to real ground motion.  But is
> > give you a very good idea of how a whole station is working.  Some
> > station PDF are very interesting.  It is also easy to see daily and
> > seasonal variation at a station.  Most sites will be between the two
> > gray lines, the high noise and low noise models.
> > 
> > I the PDF of my station BRU2 you can see that between 1 Hz and 10 hz the
> > there is a high (light blue) probability that it is one of two noise
> > levels.  Mostly likely one is daytime and one is night time, in the of
> > the frequencies my station is mostly likely in a known place.  
> > 
> > Any instrument aberation would show up on a PDF.
> > 
> > We would all like a station like ISCO in an abandoned mine tunnel
> > station in Idaho Springs CO
> > 
> > I hope this helps a  little.
> > 
> > Angel
> > 
> > 
> > 
> > 
> > 
> > 
> > 
> > 
> > On Fri, 2010-04-09 at 11:02 -0700, Geoffrey wrote:
> >> Hello PSN;
> >> 
> >> I have been looking at your noise profiles and begin to wonder
> >> about technical specs for amateur hardware.
> >> 
> >> Can any of you tell/show me what the noise profile
> >> should look like coming from the ground, through
> >> velocity transducer, through anti-alized filtered hardware
> >> Through 12 bit A/D into single float raw data display using
> >> like Winquake FFT  for the following types of data:
> >> 
> >> 0. acceleration    inches/second^2
> >> 1. Velocity            inches/second
> >> 2. Displacement  inches
> >> 
> >> I invite you to create a representation of your
> >> ideas and draw them in paint or whatever program
> >> then send it to me as a BMP or JPG or GIF file
> >> as an attachment.
> >> 
> >> You do not need to show detailed squiggles just
> >> the general profile of what to expect.
> >> 
> >> It seems to me your hardware is overfiltering
> >> or your sample rates are wrong. Something
> >> does not look right to me in your various
> >> noise profiles but I cant put a finger on
> >> any of it. Possibly you are converting from
> >> other kinds of source data than RAW to make
> >> your PSN files.
> >> 
> >> Thanks Ahead of time, for your responses.
> >> regards,
> >> geoff
> >> 
> >> __________________________________________________________
> >> 
> >> Public Seismic Network Mailing List (PSN-L)
> >> 
> >> To leave this list email PSN-L-REQUEST@.............. with 
> >> the body of the message (first line only): unsubscribe
> >> See http://www.seismicnet.com/maillist.html for more information.
> >> 
> > 
> > 
> > __________________________________________________________
> > 
> > Public Seismic Network Mailing List (PSN-L)
> > 
> > To leave this list email PSN-L-REQUEST@.............. with 
> > the body of the message (first line only): unsubscribe
> > See http://www.seismicnet.com/maillist.html for more information.
> >
> __________________________________________________________
> 
> Public Seismic Network Mailing List (PSN-L)
> 
> To leave this list email PSN-L-REQUEST@.............. with 
> the body of the message (first line only): unsubscribe
> See http://www.seismicnet.com/maillist.html for more information.
> 


__________________________________________________________

Public Seismic Network Mailing List (PSN-L)