Dear Dr. Lars
Thank you very much. Pardon my understanding of English. I need to properly
understand this and i hope it is beneficial to all of us. Please answer this.
So basically, even after using gr v command on screen (to obtain velocity) and
outputing it in ASCII format using OUTw(T) , the values are still counts not
velocity?
And i need to convert it into physical units (velocity)?
Right?
>as far as i can se there is nothing wrong. if you plot velocity on the
>screen, you get 35000 nm/s as max on SUVO B Z. that is the same in the
>waveform.out file, which you can also plot with mulplt. since both Mw and
>Ml comes out similar, i think thye seisan values are correct.
Thanks
Atul
________________________________
From: "[email protected]" <[email protected]>
To: [email protected]
Sent: Wednesday, July 3, 2013 3:30:04 PM
Subject: seisan Digest, Vol 65, Issue 2
Today's Topics:
1. Re: SEISAN ASCII format help (Lars Ottemoller)
2. Re: SEISAN ASCII format help (Jens Havskov)
3. Re: SEISAN ASCII format help (Santosh Kumar Sundriyal)
4. Re: SEISAN ASCII format help (Jens Havskov)
----------------------------------------------------------------------
Message: 1
Date: Tue, 02 Jul 2013 16:37:50 +0200
From: Lars Ottemoller <[email protected]>
Subject: Re: [seisan] SEISAN ASCII format help
To: [email protected]
Dear Atul,
Seisan corrects for the complete transfer function (PAZ and gain
factor), and you need to do the same, multilplying with g is not
correct.
The data are in units of counts also when you convert to ascii, and
you need to translate it to physical units.
Regards,
Lars
Quoting atul jhajhria <[email protected]>:
> Hello Everyone
>
> Good morning. I recently received seismic dataset in SEISAN format
> with response files from ISR (Indian seismological research
> institute, Gandhinagar). I have following issue with this:
>
> when i use l (locate command in SEISAN) one particular event gives
> me mw=3.1 (this is correct).
>
> I use gr v command and out the velocities in seisan format. I then
> use seiasc command to convert the file to ascii format (for
> subsequent processing).
>
> I understand that ascii file numbers are in nano unit (n-m or nano-g)?
>
>
> I do subsequent processing in matlab with well tested code, and it
> gives mw~2.45.
>
> However, if i multiply ascii output velocity numbers by g (9.8), and
> then do processing in matlab i get correct mw of 3.1.
>
> It has been written in SEISAN manual that if i use seiasc command i
> need to multiply the output velocity/acceleration values by a
> constant. Is multiplying by g is correct or how to calculate that
> constant. I need output file in ascii format only.
>
> I have attached the seisan file of the event, response files, and
> ascii output. Is multiplying with g is correct?
>
> Thanks for the help.
>
> Atul
-----------------------------------------------------
Lars Ottemöller
Department of Earth Science, University of Bergen
http://folk.uib.no/lot081
currently at
Lamont-Doherty Earth Observatory, Columbia University
mobile +1(201) 301-5634 office +1(845) 365-8323
-----------------------------------------------------
------------------------------
Message: 2
Date: Tue, 2 Jul 2013 21:51:42 +0200 (CEST)
From: Jens Havskov <[email protected]>
Subject: Re: [seisan] SEISAN ASCII format help
To: atul jhajhria <[email protected]>
Cc: "[email protected]" <[email protected]>
atul
On Tue, 2 Jul 2013, atul jhajhria wrote:
> Hello Everyone
>
> Good morning. I recently received seismic dataset in SEISAN format with
> response files from ISR (Indian seismological research institute,
> Gandhinagar). I have following issue with this:
>
> when i use l (locate command in SEISAN) one particular event gives me mw=3.1
> (this is correct).
>
> I use gr v command and out the velocities in seisan format. I then use
> seiasc command to convert the file to ascii format (for subsequent
> processing).
your input file seems to already be in velocity, unfiltered as hown on top
of plot. so doing one more time would give funny results
>
> I understand that ascii file numbers are in nano unit (n-m or nano-g)?
>
that is what the file should be
> I do subsequent processing in matlab with well tested code, and it gives
> mw~2.45.
>
you seiasc file has too small numbers, should be around 40000, is around 2
so hard to see what you did
> However, if i multiply ascii output velocity numbers by g (9.8), and then do
> processing in matlab i get correct mw of 3.1.
9.8 does not make sense and is not used
>
> It has been written in SEISAN manual that if i use seiasc command i need to
> multiply the output velocity/acceleration values by a constant.
i do not see this in manual. it is only a problem if values are less than
1.0. some programs automatically take care of this by multiplying with a
factor whci i sthen used automatically when you read. the out function in
mulplt is doing that. however you seiasc has a gain factor which should
not be there, i do no tget it if i do the same with your data. maybe you
should use latest seisan version. in tha tyou will directly get an asci
file with real number and no gain factor, option is OutW (T)
jens
Is
> multiplying by g is correct or how to calculate that constant. I need output
> file in ascii format only.
>
> I have attached the seisan file of the event, response files, and ascii
> output. Is multiplying with g is correct?
>
> Thanks for the help.
>
> Atul
>
>
>
>
>
------------------------------
Message: 3
Date: Wed, 3 Jul 2013 12:08:09 +0530
From: Santosh Kumar Sundriyal <[email protected]>
Subject: Re: [seisan] SEISAN ASCII format help
To: Jens Havskov <[email protected]>, [email protected]
Cc: "[email protected]" <[email protected]>, atul jhajhria
<[email protected]>
Dear Atul
I found below lines numbered by me from the manual of Seisan ver 8.1 on
page 91, 2nd paragraph under header *Output of binary waveform file*,
O(Out) by reading these there may be some confusion.
*"1.1 The note could be e.g. ? Displacement 1.250- 20.0 Hz? indicating that
output has been filtered and converted to displacement (nm). Note that
numbers have been scaled so only if the SEISAN file is read with standard
SEISAN routines, will the numbers be correct. *
*
*
*1.2 If the output file is converted to ASCII by SEIASC, the number shown
must be multiplied with a given scaling factor, see SEISAN binary format
description (Appendix 2). "*
*
*
For 1.1, I understand that if there are accurate response files using
poles and zero there in CAL (Seismo), then GRD, GRV, GRA and subsequently *
SEIASC* command can be used. Output data file will be in nm
(displacement), nm sec (velocity) or nm sec sec (acceleration). There is no
need of multiplying by any factor.
For 1.2, which should be read as separate case, if output file is in
counts and converted in ASCII without response information then ASCII files
are needed to be scaled using some factor. The scaling factor will depend
upon the parameters like sensitivity of Sensor, digitizer gain etc.
*
*
*
*
*Lars, Jens, please correct my understanding, if wrong. *
* *
*
*
*regards*
Santosh Kumar
On Wed, Jul 3, 2013 at 1:21 AM, Jens Havskov <[email protected]>wrote:
> atul
>
>
> On Tue, 2 Jul 2013, atul jhajhria wrote:
>
> Hello Everyone
>>
>> Good morning. I recently received seismic dataset in SEISAN format with
>> response files from ISR (Indian seismological research institute,
>> Gandhinagar). I have following issue with this:
>>
>> when i use l (locate command in SEISAN) one particular event gives me
>> mw=3.1
>> (this is correct).
>>
>> I use gr v command and out the velocities in seisan format. I then use
>> seiasc command to convert the file to ascii format (for subsequent
>> processing).
>>
>
> your input file seems to already be in velocity, unfiltered as hown on top
> of plot. so doing one more time would give funny results
>
>
>> I understand that ascii file numbers are in nano unit (n-m or nano-g)?
>>
>
> that is what the file should be
>
>
> I do subsequent processing in matlab with well tested code, and it gives
>> mw~2.45.
>>
>>
> you seiasc file has too small numbers, should be around 40000, is around 2
> so hard to see what you did
>
>
> However, if i multiply ascii output velocity numbers by g (9.8), and then
>> do
>> processing in matlab i get correct mw of 3.1.
>>
>
> 9.8 does not make sense and is not used
>
>
>
>
>> It has been written in SEISAN manual that if i use seiasc command i need
>> to
>> multiply the output velocity/acceleration values by a constant.
>>
>
> i do not see this in manual. it is only a problem if values are less than
> 1.0. some programs automatically take care of this by multiplying with a
> factor whci i sthen used automatically when you read. the out function in
> mulplt is doing that. however you seiasc has a gain factor which should not
> be there, i do no tget it if i do the same with your data. maybe you should
> use latest seisan version. in tha tyou will directly get an asci file with
> real number and no gain factor, option is OutW (T)
>
> jens
>
>
> Is
>
>> multiplying by g is correct or how to calculate that constant. I need
>> output
>> file in ascii format only.
>>
>> I have attached the seisan file of the event, response files, and ascii
>> output. Is multiplying with g is correct?
>>
>> Thanks for the help.
>>
>> Atul
>>
>>
>>
>>
>>
> _______________________________________________
> seisan mailing list
> [email protected]
> http://mailman.uib.no/listinfo/seisan
>
>
--
with warm regards
Santosh Kumar Sundriyal
Scientist 'D' & Incharge SeiDAC
ISR, Raisan,
Near PDPU
Gujarat, Gandhinagar.
079-66739023
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: text/html
Size: 6378 bytes
Desc: not available
URL:
<http://mailman.uib.no/private/seisan/attachments/20130703/2771d5b6/attachment.txt>
------------------------------
Message: 4
Date: Wed, 3 Jul 2013 10:37:02 +0200 (CEST)
From: Jens Havskov <[email protected]>
Subject: Re: [seisan] SEISAN ASCII format help
To: atul jhajhria <[email protected]>
Cc: "[email protected]" <[email protected]>
all
On Wed, 3 Jul 2013, atul jhajhria wrote:
> Dear Dr. Jens
>
> Thankyou for your quick response. Here is the follow-up of my last email and
> the email sent by Mr. Santosh Kumar from ISR. Please consider these attached
> file. The station name is SUV for this example. In my last email the seisan
> format file was already converted to velocity, so please discard that and
> consider these raw files.
>
> I have attached original raw data for one particular event and all stations:
>
> Attached:
> File #1, #2 from WAV directory. (contains all station data, no gr v command
> or any filter applied)
> File #3(02-0758), from REA directory.
>
> File #4, #5 having station info in WOR directory.
> File #6: ASCII file as i obtained (for station SUV only) using latest
> version of SEISAN(10.0) and using OUTW(T) command as suggested by you.
>
> File #7... are the response files for 4 stations.
>
> By just using command l (locate, we get Mw=3.0)
>
> However, the ascii output still contains small values as also pointed by
> you.
>
> As pointed by Santosh Kumar, the manual says ascii output needs to be
> multiplied by a constant.
not if you use outw command
> Is that constant a generator constant = 10.1 as calculated on page: 325 of
> manual version 9.1. or can it be just g (9.8).
>
as far as i can se there is nothing wrong. if you plot velocity on the
screen, you get 35000 nm/s as max on SUVO B Z. that is the same in the
waveform.out file, which you can also plot with mulplt. since both Mw and
Ml comes out similar, i think thye seisan values are correct.
jens
> where i am wrong?
>
> This file (mulplt.wav) is practically same as i obtained from last seiasc
> (seiasc).
>
> Kindly help me in this.
>
> Thanks
> Atul
>
>
>
> ____________________________________________________________________________
> From: Santosh Kumar Sundriyal <[email protected]>
> To: Jens Havskov <[email protected]>; [email protected]
> Cc: atul jhajhria <[email protected]>; "[email protected]"
> <[email protected]>
> Sent: Wednesday, July 3, 2013 12:08:09 PM
> Subject: Re: [seisan] SEISAN ASCII format help
>
>
>
> Dear Atul
>
> I found below lines numbered by me from the manual of Seisan ver 8.1 on
> page 91, 2nd paragraph under header Output of binary waveform file, O(Out)
> by reading these there may be some confusion.
>
> "1.1 The note could be e.g. ? Displacement 1.250- 20.0 Hz? indicating that
> output has been filtered and converted to displacement (nm). Note that
> numbers have been scaled so only if the SEISAN file is read with standard
> SEISAN routines, will the numbers be correct.
>
> 1.2 If the output file is converted to ASCII by SEIASC, the number shown
> must be multiplied with a given scaling factor, see SEISAN binary format
> description (Appendix 2). "
>
> For 1.1, I understand that if there are accurate response files using poles
> and zero there in CAL (Seismo), then GRD, GRV, GRA and subsequently SEIASC
> command can be used. Output data file will be in nm (displacement), nm sec
> (velocity) or nm sec sec (acceleration). There is no need of multiplying by
> any factor.
>
> For 1.2, which should be read as separate case, if output file is in
> counts and converted in ASCII without response information then ASCII files
> are needed to be scaled using some factor. The scaling factor will depend
> upon the parameters like sensitivity of Sensor, digitizer gain etc.
>
>
> Lars, Jens, please correct my understanding, if wrong.
>
>
> regards
>
> Santosh Kumar
>
> On Wed, Jul 3, 2013 at 1:21 AM, Jens Havskov <[email protected]>
> wrote:
> atul
>
> On Tue, 2 Jul 2013, atul jhajhria wrote:
>
> Hello Everyone
>
> Good morning. I recently received seismic dataset
> in SEISAN format with
> response files from ISR (Indian seismological
> research institute,
> Gandhinagar). I have following issue with this:
>
> when i use l (locate command in SEISAN) one
> particular event gives me mw=3.1
> (this is correct).
>
> I use gr v command and out the velocities in seisan
> format. I then use
> seiasc command to convert the file to ascii format
> (for subsequent
> processing).
>
>
> your input file seems to already be in velocity, unfiltered as hown on
> top of plot. so doing one more time would give funny results
>
> I understand that ascii file numbers are in nano unit (n-m
> or nano-g)?
>
>
> that is what the file should be
>
> I do subsequent processing in matlab with well tested
> code, and it gives
> mw~2.45.
>
>
> you seiasc file has too small numbers, should be around 40000, is
> around 2
> so hard to see what you did
>
> However, if i multiply ascii output velocity numbers by g
> (9.8), and then do
> processing in matlab i get correct mw of 3.1.
>
>
> 9.8 does not make sense and is not used
>
>
>
> It has been written in SEISAN manual that if i use seiasc
> command i need to
> multiply the output velocity/acceleration values by a
> constant.
>
>
> i do not see this in manual. it is only a problem if values are less
> than 1.0. some programs automatically take care of this by multiplying
> with a factor whci i sthen used automatically when you read. the out
> function in mulplt is doing that. however you seiasc has a gain factor
> which should not be there, i do no tget it if i do the same with your
> data. maybe you should use latest seisan version. in tha tyou will
> directly get an asci file with real number and no gain factor, option
> is OutW (T)
>
> jens
>
> Is
> multiplying by g is correct or how to calculate that
> constant. I need output
> file in ascii format only.
>
> I have attached the seisan file of the event, response
> files, and ascii
> output. Is multiplying with g is correct?
>
> Thanks for the help.
>
> Atul
>
>
>
>
>
> _______________________________________________
> seisan mailing list
> [email protected]
> http://mailman.uib.no/listinfo/seisan
>
>
>
>
> --
> with warm regards
>
> Santosh Kumar Sundriyal
> Scientist 'D' & Incharge SeiDAC
> ISR, Raisan,
> Near PDPU
> Gujarat, Gandhinagar.
> 079-66739023
>
>
>
>
----------------------------------------------------------------------
Send seisan mailing list submissions to
[email protected]
To subscribe or unsubscribe via the World Wide Web, visit
http://mailman.uib.no/listinfo/seisan
or, via email, send a message with subject or body 'help' to
[email protected]
You can reach the person managing the list at
[email protected]
When replying, please edit your Subject line so it is more specific
than "Re: Contents of seisan digest..."
_______________________________________________
seisan mailing list
[email protected]
http://mailman.uib.no/listinfo/seisan
End of seisan Digest, Vol 65, Issue 2
*************************************
_______________________________________________
seisan mailing list
[email protected]
http://mailman.uib.no/listinfo/seisan