Commit f72bbefe authored by thomas.forbriger's avatar thomas.forbriger
Browse files

[WP] (stf_issue12): revise test case descriptions

parent 6e60f2e3
......@@ -47,6 +47,10 @@ all:
clean:
-/bin/rm -vf flist *.xxx *.xxx.* *.sff *.TZ *.TR *.grn *.su
.PHONY: simulclean
simulclean:
-/bin/rm -vf *.sff *.su stf.fil.xxx
#
# ======================================================================
# 1. Produce synthetic shot gathers by using refmet
......@@ -271,14 +275,17 @@ additionalfiltered.sff: additional.sff stf.fil.xxx
# =============
#
# test case 1
# -----------
# 6a. Test case 1
# ---------------
define TESTCASE1
synthetics are calculated with a standard source
time function which is band limited and has a
duration of $(SRCDUR) seconds. The recorded data
is simulated from this by applying low-pass filter
and a time delay.
Synthetics are calculated with a standard source time function which is band
limited and has a duration of $(SRCDUR) seconds. The recorded data is
simulated from this by applying a low-pass filter, a scaling factor, and a
time delay. Noise with an rms-amplitude of $(NOISEAMP) additionally is added
to the simulated recordings. Because simulated recordings and synthetics have
the same wave propagation characteristics, the simulated field data can
exactly be reproduced by the source time function correction filter, at least
in the case where no noise is added to the data.
endef
export TESTCASE1
......@@ -337,7 +344,7 @@ soutifu1_add_%.ps: soutifu1_add.su additional.su additionalfiltered.su
soutifu1_stf.ps: soutifu1_stf.su
stuplox -ty su -d $@/ps -E -m 0. -Y 0.9 -S -$(SOUTIFU1SHIFT) \
-R 0.6 -L 0.1 -T "soutifu test case 1: $(SOUTIFU1)" -l 3,3,3\
-R 0.7 -T "soutifu test case 1: $(SOUTIFU1)" -l 3,3,3\
-u "source time function" $<
#----------------------------------------------------------------------
......@@ -353,13 +360,18 @@ soutifu1_add_report.pdf: soutifu1_description.ps soutifu1_section.ps\
gs -sDEVICE=pdfwrite -dNOPAUSE -dBATCH -sOutputFile=$@ $^
#----------------------------------------------------------------------
# test case 2
# -----------
# 6b. Test case 2
# ---------------
define TESTCASE2
We take the same data like in test case 1.
This time, however, data and synthetics are exchanged.
Since synthetics are narrow band with respect to data
an appropriate water level is required.
We take the same data like in test case 1. This time, however, data and
synthetics are swapped. Since synthetics are narrow band with respect to
data an appropriate water level is required.
As a consequence of swapping the role of data and synthetics, the source time
function correction necessarily is entirely acausal. Since simulated data and
synthetics are both based on the same wave propagation characteristics,
soutifu should be able to almost reproduce the simulated data, at least in the
case where no noise is added.
endef
export TESTCASE2
......@@ -408,21 +420,22 @@ soutifu2c_report.pdf:
/bin/mv -fv soutifu2x_report.pdf $@
#----------------------------------------------------------------------
# test case 3
# -----------
# 6c. Test case 3
# ---------------
define TESTCASE3
The model for which synthetics are calculated is too
slow in its wave propagation properties. Therefore
a compromise has to be made between far and near
offset traces.
The model for which synthetics are calculated produced a slightly smaller wave
propagation. It is no longer possible to repoduce the data with a single
source time function correction filter for all traces. Therefore a compromise
has to be made between far and near offset traces. This test case can be used
to study the effect of waterlevel and offset dependent weights.
Signal description:
Synthetics are calculated with a standard source
time function which is band limited and has a
duration of $(SRCDUR) seconds. The recorded data
is simulated from this by applying low-pass filter
and a time delay. The corner period of the low-pass is
so small that it pracically has no effect:
Synthetics are calculated with a standard source time function which is band
limited and has a duration of $(SRCDUR) seconds. The recorded data is
simulated from this by applying a low-pass filter, a scaling factor, and a
time delay.
Noise with an rms-amplitude of $(NOISEAMP) is added to the data.
endef
export TESTCASE3
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment