Thread: average trace of multiple sac traces

Hi,

I've worked around with this script. Probably there's a more efficient way
but it seems to work.
I am sending my solution in case someone needs.

Best

Aderson


=====
#
# reading all sac files
#
echo "reading all sac files"
sac <<END
r *.SAC
rmean
rtr
w append .rmean.rtrend
q
END
#
# calculating sac average trace
#
echo "calculating sac average trace"
ls -1 *.rmean.rtrend | awk '{if (NR == 1) print "r "$1;else print "addf
"$1} END {print "div "NR"\nw average_trace.sac\nq"}' | sac
#
# calculating the squared differences
#
echo "calculating the squared differences"
sac <<END
r *.rmean.rtrend
subf average_trace.sac
sqr
w append .sqr.difference
q
END
#
# calculating the average of the squared differences
#
echo "calculating the average of the squared differences"
ls -1 *.sqr.difference | awk '{if (NR == 1) print "r "$1;else print "addf
"$1} END {print "div "NR"\nw average_sqr_differences.sac\nq"}' | sac
#
# calculating now the squared root of average of the squared differences =
standard deviation
#
echo "calculating now the squared root of average of the squared
differences = standard deviation"
sac <<END
r average_sqr_differences.sac
sqrt
w std_dev.sac
q
END
#
# now calculating the plus and minus files
#
echo "now calculating the plus and minus files"
sac <<END
r average_trace.sac
addf std_dev.sac
w plus_std_dev.sac
inicm
r average_trace.sac
subf std_dev.sac
w minus_std_dev.sac
q
END
#
# cleaning up
#
\rm *.rmean.rtrend
\rm *.sqr.difference
==========

2016-07-24 19:07 GMT-03:00 Aderson Farias do Nascimento <
nascimentoafd<at>gmail.com>:

Hi All,

I have like a few hundred sac traces and want to calculate the average
(and std dev) trace...
I am positive that writing the data in ascii and then working with scripts
(awk etc...) you could achieve this. But I wonder if I could calculate this
directly from sac more efficiently.

Best,

Aderson Nascimento

Dear All -

See Chapter 8 in the SAC book. In particular, the stacking subprocess, SSS, provides ways to sum multiple traces and provide the mean and standard deviation with different summation methods (linear, Nth root, phase weighting).

On 25 Jul 2016, at 00:08, Aderson Nascimento <nascimentoafd<at>gmail.com> wrote:

Hi All,

I have like a few hundred sac traces and want to calculate the average (and std dev) trace...
I am positive that writing the data in ascii and then working with scripts (awk etc...) you could achieve this. But I wonder if I could calculate this directly from sac more efficiently.

Best,

Aderson Nascimento


----------------------
SAC Help (http://ds.iris.edu/message-center/topic/sac-help/)

Sent via IRIS Message Center (http://ds.iris.edu/message-center/)
Update subscription preferences at http://ds.iris.edu/account/profile/

George Helffrich
george.helffrich<at>bris.ac.uk