From: John Hernlund hernlund@............

Date: Mon, 01 Jul 2002 16:18:23 -0700

In practice, the FFT can be used to very easily obtain the integral or derivative of a time series (i.e. division or multiplication of all spectral points by i*frequency respectively). Since this is already built into Winquake, this would be the most efficient method, and is the one I always assumed was being used by Winquake. In this case there is no finite integration step, since the FFT is a sinusoidal fit, which is then played with. Randall Pratt wrote: >Arie, > >I don't have an answer but I have wondered about this also. I've considered >summing as you thought and also the possibility of maybe trapazoidal areas >with several points. Another question I have is how the result can begin at >other than zero displacement with a velocity sensor and zero time at file >start? I wonder if the algorithym starts in the middle or from some average >point and works to the ends? > >Randy > >----- Original Message ----- >From: "Arie Verveer">To: >Sent: Tuesday, June 25, 2002 6:28 AM >Subject: Integration and Winquake > > > > >>Hi, Just a small question; When integrating data with Winquake >>does anyone now the integration period.? I would assume the integrated >>value is the sum of the y values times 1/ ( sample rate ) over the length >>of the integration period? . I'm probably wrong, any idea's. >> >> >>Cheers >> >>Arie >>__________________________________________________________ >> >>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. > > In practice, the FFT can be used to very easily obtain the integral or derivative of a time series (i.e. division or multiplication of all spectral points by i*frequency respectively). Since this is already built into Winquake, this would be the most efficient method, and is the one I always assumed was being used by Winquake. In this case there is no finite integration step, since the FFT is a sinusoidal fit, which is then played with.

Randall Pratt wrote:

Arie, I don't have an answer but I have wondered about this also. I've considered summing as you thought and also the possibility of maybe trapazoidal areas with several points. Another question I have is how the result can begin at other than zero displacement with a velocity sensor and zero time at file start? I wonder if the algorithym starts in the middle or from some average point and works to the ends? Randy ----- Original Message ----- From: "Arie Verveer" <ajbv@............> To: <PSN-L@..............> Sent: Tuesday, June 25, 2002 6:28 AM Subject: Integration and WinquakeHi, Just a small question; When integrating data with Winquake does anyone now the integration period.? I would assume the integrated value is the sum of the y values times 1/ ( sample rate ) over the length of the integration period? . I'm probably wrong, any idea's. Cheers Arie __________________________________________________________ Public Seismic Network Mailing List (PSN-L) To leave this list email PSN-L-REQUEST@SEISMICNE T.COM 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@SEISMICNE T.COM with the body of the message (first line only): unsubscribe See http://www.seismicnet.com/maillist.html for more information.