Commit 38112a1c authored by thomas.forbriger's avatar thomas.forbriger
Browse files

librefpar [TASK]: be more specific regarding source time function

The documentation of the source time function in terms of the hard-coded
function f'(t) on the one hand and our understanding of a source-time history
on the other hand was confusing. Hopefully this is resolved now.
parent 6724a900
c this is <sub/refmet_rsource.f>
c ----------------------------------------------------------------------------
c ($Id$)
c
c Copyright (c) 1997 by Thomas Forbriger (IfG Stuttgart)
c
......@@ -34,6 +33,7 @@ c 15/10/00 be more specific about time parameters
c 20/10/01 explain the source amplitude definition in detail
c 24/10/01 incomplete source description
c 12/06/14 zeroes of MB-function
c 25/06/18 be more specific regarding source time function
c
c======================================================================
c
......@@ -60,7 +60,7 @@ c
c
integer typ, outsig, srcsig
character sourcefile*(*), sourcetext*(*), outunits*(*)
real*8 The, Thd, ZQ, M0, Mxx, Myy, Mzz, Mxy, Mxz, Myz
double precision The, Thd, ZQ, M0, Mxx, Myy, Mzz, Mxy, Mxz, Myz
integer cl_vlevel, lev2
logical cl_debug
c
......@@ -128,9 +128,6 @@ c
print 50,'How to build a source configuration file'
print 50,'========================================'
print 50,' '
print 50,'compiled source function is:'
print 50,hdfktstr
print 50,' '
print 51,'line','contents'
print 52,1,'text'
......@@ -140,9 +137,12 @@ c
print 53,'typ','i','=1: source given by moment tensor'
print 55,'=2: source given as single vertical force'
print 55,'=3: source given as single force (not supported)'
print 54,'sig','i','=1: source signal wil be once integrated'
print 55,'=2: source signal will be taken as is'
print 55,'=3: source signal will be once differentiated'
print 54,'sig',
& 'i','=1: source signal will be once integrated: f(t)'
print 55,
& '=2: source signal will be taken as is: f''(t)'
print 55,
& '=3: source signal will be once differentiated: f''''(t)'
print 54,'src','i','=1: take compiled source function'
print 55,'=2: use delta pulse as source'
print 55,'=3: read time series from file (not supported)'
......@@ -167,6 +167,25 @@ c
print 52,10,'units'
print 53,'units','a8','string will be written to output file'
print 50,' '
print 50,'Parameter ''src'' controls the type of function to be'
print 50,'used as the derivative f''(t) of the source-timer-function.'
print 50,'This means the in first place, the time dependency of'
print 50,'its first derivative is specified either to be a delta'
print 50,'pulse, or the hard-coded source-time-function or a'
print 50,'time series to be read from an input file (not yet'
print 50,'supported).'
print 50,' '
print 50,'Parameter ''sig'' controls whether f''(t) is used as'
print 50,'the source-time history or whether it shall be converted'
print 50,'to f(t) by integration or to f''''(t) by derivation.'
print 50,'Both operations are implemented in the frequency domain.'
print 50,' '
print 50,'compiled (hard-coded) source function is:'
print 56,'f''(t)=',hdfktstr
print 50,' '
print 50,'Comments on the Mueller-Bruestle-function'
print 50,'-----------------------------------------'
......@@ -227,8 +246,15 @@ c
print 50,'General comments on the application of the function'
print 50,'---------------------------------------------------'
print 50,' '
print 50,' If the source is defined by a moment tensor'
print 50,' -------------------------------------------'
print 50,' A moment tensor usually is used to describe a shear'
print 50,' fault, a typical earthquake source, for which the'
print 50,' time history is a step function. This is approximated'
print 50,' by f(t), which is the integral of the source-time'
print 50,' function selected by parameter ''src''.'
print 50,' The source time function is given in the form of its'
print 50,' first order derivative. If it is taken as it is'
print 50,' first order derivative f''(t). If it is taken as it is'
print 50,' (sig=2) the seismograms will be velocity waveforms'
print 50,' as response to a unit moment step. If the scalar'
print 50,' moment is given in N*m the seismograms are in m/s.'
......@@ -240,6 +266,24 @@ c
print 50,' be done in the Fourier domain. Notice that the scalar'
print 50,' momentum M0 always refers to the step-function f(t).'
print 50,' '
print 50,' If the source is defined to be an explosion'
print 50,' -------------------------------------------'
print 50,' An explosion is defined by an isotropic moment tensor'
print 50,' (scaled identity tensor). The source-time-function'
print 50,' for the scalar moment is a step function, a sudden'
print 50,' increase in pressure. Further considerations are'
print 50,' similar to those for the earthquake source. If, however,'
print 50,' the explosion happens in a shallow borehole, then'
print 50,' the pressure at the source very likely is released'
print 50,' soon after the explosion. Then considerations are'
print 50,' are similar to those for the hammer blow, with a'
print 50,' one-sided (entirely positive) source-time history.'
print 50,' '
print 50,' If the source is defined by a single force'
print 50,' ------------------------------------------'
print 50,' A hammer blow approximately can be simualted by a'
print 50,' single force with a one-sided (entirely positive)'
print 50,' impulse transfer.'
print 50,' If you want to receive displacement waveforms as a'
print 50,' response to a one-sided force-impulse you may use'
print 50,' f''(t) (sig=2) as source function. However, then you'
......@@ -263,6 +307,18 @@ c
print 50,' In the same way you may calculate the displacement'
print 50,' response to the two-sided pulse f''''(t) of maximum force'
print 50,' Fp when specifying sig=3 and F0=Fp*(T2-T1)**2/8.550.'
print 50,' '
print 50,' General physical considerations regarding single'
print 50,' force sources are discussed by Forbriger (2004.'
print 56,' ',
& 'Dynamics of the Hammer Blow. In: Symposium in Memoriam of '
print 56,' ',
& 'Prof. Gerhard Müller, J. Schweitzer (ed.). Deutsche '
print 56,' ',
& 'Geophysikalische Gesellschaft, ISSN-Nr. 0934-6554, I/2004, 93-97.'
print 56,' ',
& 'doi: 10.5445/IR/1000007818).'
c
return
50 format(1x,a)
......@@ -271,6 +327,7 @@ c
53 format(/6x,a8,1x,1h(,a3,1h),1x,a)
54 format(6x,a8,1x,1h(,a3,1h),1x,a)
55 format(21x,a)
56 format(1x,a,a)
end
c
c ----- END OF sub/refmet_rsource.f -----
Supports Markdown
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