tramp.doc

利用fortran写的实现2D射线追踪的源代码！运行于linux下！

                                      

            DOCUMENTATION FOR TRAMP AND RELATED PROGRAMS
            ----------------------------------------------

Date:       Nov 1992
Version:    1.3

Written by: C.A. Zelt
Address:    Geological Survey of Canada
            1 Observatory Crescent
            Ottawa, Ontario, Canada  K1A 0Y3
phone:      [phone]
fax:        [phone]
e-mail:     [e-mail]


            Table of Contents
            -----------------


TRAMP
     Program Description
     Files
     Input Parameters
         Plotting Parameters
         Axes Parameters
         Ray tracing Parameters
         Amplitude Parameters
     Additional Notes
     Array Sizes
     Warning and Error Messages
Other programs
     PLTSYN
     COMBSEC


TRAMP: a program to  trace rays in  2-D media for  rapid forward 
        modeling of refraction  and reflection travel  times and 
        amplitudes

                        Program Description
  
  A 2-D (x,z) isotropic medium is assumed. The velocity  model is 
composed  of  a  sequence  of   layers  separated  by  boundaries  
consisting of  linked  linear segments  of  arbitrary dip.  Layer  
boundaries must  cross  the  model  from  left  to  right.  Layer  
thicknesses may be reduced to zero to model pinchouts or isolated 
bodies. The velocity within a layer is defined by velocity values 
specified at arbitrary x-coordinates along the  top and bottom of 
the layer. The x-coordinates at which  layer boundaries and upper 
and lower velocities are specified can  be completely general and 
independent within and  between layers.  Velocity discontinuities  
across layer  boundaries are  allowed but  not required.  For the  
purposes of ray tracing, the model is automatically broken up into 
an irregular network of  trapezoids, each with  dipping upper and 
lower  boundaries  and  vertical   left  and  right   sides.  The  
velocities at  the four  corners  of the  trapezoid  are used  to  
interpolate a  velocity field  within the  trapezoid so  that the  
velocity  varies  linearly  along  its   four  sides.  Therefore,  
horizontal as well as vertical velocity gradients may exist within 
a trapezoid. A simulation of smooth  layer boundaries is possible 
in which the incident and emergent ray angles are calculated using 
the slope of the smoothed boundary.
  The source(s) may be positioned anywhere in the model  and rays 
may be directed any angle. The receivers are always assumed to be 
at the top of  the model. Both  P- and S-wave  propagation can be 
considered including (multiple)  conversions. A  unique Poisson's  
ratio may be assigned to each trapezoid  of the model. Refracted, 
reflected and head waves may be  traced, each possibly containing 
multiple and/or surface reflections and conversions. Ray take-off 
angles are determined automatically by the  program for those ray 
groups specified by the user using an iterative shooting/bisection 
search mode. Ray tracing is performed  by numerically solving the 
ray tracing equations for 2-D media, a pair of first order ODE's, 
using a Runge Kutta method. The ray  step length is automatically 
adjusted at each  step to  maximize efficiency  while maintaining  
accuracy. Travel  times are  calculated by  numerical integration  
along ray  paths using  the  trapezoidal rule.  Travel times  and  
amplitudes are linearly  interpolated to the  observed seismogram 
locations since two-point ray tracing is not used.  A plot of the 
model and all rays  traced may be  produced along with  a plot of 
reduced  travel  time  versus  distance   for  the  observed  and  
calculated data as well as other plots.
  The  amplitudes  of  refracted  and  reflected  rays,  possibly  
multiply reflected and/or converted, are  calculated according to 
zero-order  asymptotic  ray  theory.   The  in-plane  geometrical  
spreading is evaluated  by fitting  a cubic  spline to  the curve  
defined by range  versus take-off angle  for each ray  group. The 
amplitudes  of  head  waves  are   calculated  using  first-order  
asymptotic ray theory assuming  the layer along the  top of which 
the wave  propagates  has  zero  vertical velocity  gradient.  An  
explosive source  with  uniform  directional  characteristics  is  
assumed. Both  P- and  S-wave Q  values may  be assigned  to each  
model layer and/or block to allow for an approximate Q attenuation 
calculation. The receivers  may represent vertical  or horizontal 
component geophones.
  The model parameterization is  specifically suited to  modeling 
crustal refraction/reflection data  since realistic  earth models  
can be represented by a minimum number of model parameters, i.e.,. 
the number and position of parameters specifying each layer can be 
adapted to a data set's particular subsurface ray coverage. Layer 
boundaries,  including  the  surface,  may   be  horizontal  (one  
parameter) or consist of numerous straight line segments. A layer 
may have  a constant  velocity  (one parameter)  or the  velocity  
structure may be defined  by many upper and  lower layer velocity 
points. Different  velocity  points may  be  specified above  and  
below a layer  boundary if  a velocity discontinuity  is required  
across the boundary, or  a single row  of velocity points  may be 
specified if an interface with no discontinuity is needed.

                              Files

Executable file: TRAMP

Input file: r.in contains program input parameters in five parts:
   (1) the PLTPAR namelist contains arameters related to plotting
   (2) the AXEPAR namelist contains parameters related to axes
   (3) the  TRAPAR namelist  contains parameters  related  to ray  
   tracing
   (4)  the  AMPPAR  namelist  contains   parameters  related  to  
   amplitudes
   (5) after skipping three lines (column headings), the velocity 
   model is specified as follows:
     (a) the  layer number,  the x-coordinates  (km)  of a  layer 
        boundary entered  from  left to  right  (format: I2,  1X,  
        10F7.2)
     (b) the z-coordinates (km) of a layer boundary corresponding 
        to the x-coordinates listed above (format: 3X, 10F7.2)
     (c) a blank line
     (d) the layer number, the  x-coordinates (km) of the  points 
        at which the  upper layer  velocity is  specified entered  
        from left to right (format: I2, 1X, 10F7.2)
     (e) the upper layer  P-wave velocities (km/s)  corresponding 
        to the x-coordinates listed above (format: 3X, 10F7.2)
     (f) a blank line
     (g) the layer number, the  x-coordinates (km) of the  points 
        at which the  lower layer  velocity is  specified entered  
        from left to right (format: I2, 1X, 10F7.2)
     (h) the lower layer  P-wave velocities (km/s)  corresponding 
        to the x-coordinates listed above (format: 3X, 10F7.2)
     (i) a blank line
  The above sequence  of nine lines  is repeated  for each model  
   layer, the  top-most layer  specified  first, the  bottom-most  
   last, and is ended by specifying the  bottom layer boundary of 
   the model as  in (a) and  (b) above.  If the number  of points  
   defining a boundary or the upper or  lower velocity of a layer 
   must exceed 10 then the points can be continued onto subsequent 
   lines of the  file as  follows: line  (b), (e)  or (h)  of the  
   particular parameter to be extended is modified to include a 1 
   in the second column so the complete format of the line becomes 
   I2, 1X, 10F7.2. The sequence of three lines (a)-(c), (d)-(f) or 
   (g)-(i) is then repeated  as many times as  is necessary using 
   the same format described above.

Input file: v.in contains  the velocity model in  the same format  
   as described in part (3) above for r.in.

Input file:  tx.in  contains  the  observed travel  time-distance  
    pairs in the following format:
   (1) the  x-coordinate  (km)  of  the  shot  point,  1  if  the 
      receivers are to the right of the  shot point or -1 if the  
      receivers are to the left, 0, and 0 (format: 3F10.3, I10)
   (2)  the  x-coordinate   (km)  of  the   observed  data,  the   
      corresponding unreduced travel time (s), the estimated 
      uncertainty of the travel time pick  (s), and a  non-zero 
      integer used to  identify the type of arrival  to allow for 
      the appropriate  comparison with the rays traced 
      (format: 3F10.3, I10)
   Line (2) is repeated for each  pick corresponding to the shot  
   point in line (1).  The sequence (1) and  (2) is repeated  for 
   each shot point of the data  set. The file is terminated with  
   the following line:
   (3) 0, 0, 0, -1 (format: 3F10.3, I10)

Output file: tx.out contains the  calculated travel time-distance 
    pairs in the same format as described above for tx.in.

Output file: amp.out  contains the  calculated amplitude-distance  
    pairs in the following format:
   (1) the  x-coordinate  (km)  of  the  shot  point,  1  if  the
      receivers are to the right of the  shot point or -1 if the
      receivers are to the left, 0, and 0 (format: F10.3, 2E10.3,
      I10)
   (2)  the  x-coordinate   (km)  of  the   observed  data,  the
      corresponding amplitude, the estimated  uncertainty of the
      amplitude, and a non-zero integer used to identify the type
      of arrival to allow for the appropriate comparison with the
      rays traced (format: F10.3, 2E10.3, I10)
   Line (2) is repeated for each  pick corresponding to the shot
   point in line (1).  The sequence (1) and  (2) is repeated  for
   each shot point of the data  set. The file is terminated with
   the following line:
   (3) 0, 0, 0, -1 (format: F10.3, 2E10.3, I10)

Output file:  r1.out contains  summary information  for  each ray  
    traced including  the  shot  number,  take-off and  emergent  
    angle, range, reduced  travel time,  number of  points (step  
    lengths) defining the ray and the ray code.

Output  file:  r2.out  contains  detailed   parameters  for  each  
    trapezoid of the velocity model, a one-dimensional equivalent 
    (average) velocity  model and  summary information  for each  
    point of each ray traced.

Output file: a1.out contains the range,  amplitude, phase and ray 
    code for rays reaching the surface.

Output file:  a2.out  contains  the  deatails  of  the  amplitude  
    calculations for each ray traced.


Intput file: rec.in contains the observed receiver locations at
    which the seismograms are to be calculated in the following 
    format:
   (1) the  x-coordinate  (km)  of  the  shot  point,  1  if  the 
      receivers are for this shot point only, or 2 if these
      receivers are for all shots (format: F10.3, I10)
   (2)  the  x-coordinate   (km)  of  the   receiver locations, 0   
      (format: F10.3, I10)
   Lines (2) is repeated for each receiver location for that 
   particular shot, and the sequence (1) and (2) is repeated for 
   each shot point (if necessary). The file is terminated with  
   the following line:
   (3) 0, -1 (format: F10.3, I10)

Output file: sect.out contains  the time, amplitude  and phase of  
    each arrival of the synthetic sections to be used as input by 
    the plotting program PLTSYN.

Output file: p.out  contains all plot  commands for the  run used  
    for input by the program RAYPLOT.

Output file: n.out contains the namelist parameter values.

Output file: m.out contains one or  more of the following: (1) 
    isovelocity contours, (2)  velocity-depth profiles,  (3) RMS  
    velocity variations across the model, (4) the velocity model 
    sampled on  a uniform  grid, and/or  (5) the  velocity model  
    converted to  density  for input  into  a  gravity modelling  
    program.

                         Input Parameters

1) Plotting parameters (PLTPAR namelist):

a) Switches  (usually 0 = off, 1 = on):

iroute - equals 1 to plot to the screen, 2 to create a postscript
   file, 3 to create a plot file for the VERSATEC plotter, or 4
   to create a colour postscript file; if iroute does not equal 1 
   there is no plotting to the screen  (default: 1)

iseg - create a Uniras segment(s) (default: 0)

iplot - generate the plot  during the run (1), or  write all plot 
   commands to the file p.out (0), or do both (2) (default: 1)

imod - plot model boundaries (default: 1)

ibnd - plot vertical model boundaries (default: 1)

idash - plot model boundaries as dashed lines (default: 0)

ivel - plot the P-wave (1) or  S-wave (-1) velocity values (km/s) 
   within each trapezoid of the model (default: 0)

icntr -  plot  the velocity  structure  as  isovelocity contours,  
   icntr=1 for  P-wave  velocity contours,  icntr=-1  for  S-wave 
   velocty contours; the isovelocity contours join points of equal 
   velocity at  the  minimum depth  at  which it  occurs  at any  
   particular x-coordinate; therefore, low-velocity zones will not 
   be contoured; the isovelocity contours are output to the file  
   m.out if idump=1 (default: 0)

iray - plot  all rays traced  (1) or  only those which  reach the  
   surface (2) (default: 1)

irays - plot the rays traced in the search mode (default: 0)

irayps - plot  the P-wave  segments of  ray paths as  solid lines  
   and the S-wave segments as dashed lines (default: 0)

idot - plot a symbol at each point (step length) defining each ray