Page 1 of 1

Pre-processing in ambient noise

Posted: Thu Jan 29, 2009 4:09 am
by raj
Dear Sir,

Is instrumental correction required before doing the H/V ratio? What are the preprocessing steps for H/V ratio?

Please help.

Posted: Thu Jan 29, 2009 4:22 pm
by admin
There is no instrumental correction required if all three components are measured with the same type of instrument (usually the case).

When you import a signal in Geopsy, you just have to make sure that the three components are properly identified. In a table, you should see Vertical, North and East in 'Component' column. If not, your file format does not contain this information and you must enter it manually.

Go to menu 'Edit/Lock table edition' to allow direct header modifications. Select the correct components directly from the combo box. It might be useful to display the file names attached to each signal. Go to menu 'View/Set table fields'. Insert a new field called 'ShortFileName'.

Once this is done, you can start the H/V tool, select your time windows and processing parameters.

Pre-processing in linear active source records

Posted: Tue Feb 21, 2012 2:26 pm
by hflores
We recorded active source data for two different non synchronized geophone lines. We have some questions.
1.- There was a problem: we plugged our trigger to channel 24 instead of to channel 1 :oops: . Is there a way to modify the headers so we can consider channel 24 as channel 1?
2.- We have a file for each source offset with 10 shots. In order to stack them. Should we cut and form 10 different files for each shot distance? How could we do this? Creating groups of signals?
3.- As I mentioned, we have to different lines of geophones and we want to consider them as a 48 receiver line. Is the correct way to do this by merging the files?

Thank you,

Re: Pre-processing in ambient noise

Posted: Wed Feb 22, 2012 12:10 am
by admin
1. What do you mean exactly? Do you mean that the trigger signal was directly connected to channel 24? Or simply that you shot close to channel 24 instead of channel 1? In the first case, the trigger signal must never be processed with MASW (clipped, body waves, near field,...if not a piezo mounted on hammer). Simply unselect that channel in the pre-pocessing parameters (minimum distance to source). For the second case, active source FK requires the location of the source to be specified, and it does not matter whether you hit on one side or the other, geopsy will take the correct direction of propagation.

2. Can you be more detailled? Do you mean that each of your files contain 10*24 channels? Geopsy does not care about files, process all together, and stack in frequency domain after starting the active FK for all shots separately.

3. Are they all 48 channels in line? If yes there is no need to merge file, just import the two original files in Geopsy and process them together. Make sure that all 48 channels have consistent receiver coordinates and the same source position. The "TimerReference" must be also the same, not necessarily the same if there were two separate files. Edit it with menu "Edit/Set header".

Re: Pre-processing in ambient noise

Posted: Wed Feb 22, 2012 2:47 pm
by hflores
1. I mean the trigger was physically connected to channel 24, as we use an electronic device to "close the circuit" and make sure that we trigger all channels by the cost of loosing one channel.
And I also meant we wrongly plugged it to channel 24 instead of channel 1. However, I think you already answered both questions.

2.- Yes of course. Each file contains 24 channels which recorded 10 shots for each source distance. So my question is if I have to "create" one file for each shot so I would have 10 files for the source distance of 5 m, 10 more for the 15 m source distance and so on.

3.- The 48 channels are in one line. However they were not recorded at the same time. But the sources do have the same positions. That's why we want to consider them as a 48 channels line for each source distance.
In this case how should I proceed with the "TimeReference?

Thank you,

Re: Pre-processing in ambient noise

Posted: Wed Feb 22, 2012 10:15 pm
by admin
2. No there is no need to split the file. Keep it and set distinct TimeReference for each shot. Use for instance menu Edit/Set Header with a formula like this one:

Code: Select all

TimeReference="01/01/2012 00:00:"+justify(((ViewerIndex-1) DIV 24), 2, "0");
Valid if you have less than 61 shots. If not, apply it several times on a lower number of shots or make a more sophisticated formula that increment minutes or even hours.

Code: Select all

sec=(ViewerIndex-1) DIV 24;
min=sec DIV 60;
hour=min DIV 60;
TimeReference="01/01/2012 "+justify(hour, 2, "0")+":"
                           +justify(min, 2, "0")+":"
                           +justify(sec, 2, "0");
3. Set the same source coordinates and TimeReference for the two parts of the records. Geopsy ignore files and check only source coordinates and TimeReference to identify shots. As above, use menu Edit/Set Header:

Code: Select all

TimeReference="01/01/2012 00:00:00"
Or a more realistic date.

Re: Pre-processing in ambient noise

Posted: Fri Feb 24, 2012 9:03 am
by hflores
Ok. Thank you, I will try this and let you know the results or the new questions.