NAME

     sample1d - Resampling of 1-D data sets


SYNOPSIS

     sample1d infile [ -Fl|a|c ]  [  -H[nrec]  ]  [  -Ixinc  ]  [
     -M[flag]  ]  [  -Nknotfile  ]  [  -Sxstart  ]  [  -Tx-col] [
     -bi[s][n] ] [ -bo[s] ]


DESCRIPTION

     sample1d reads a multi-column ASCII  [or  binary]  data  set
     from   file   [or   standard   input]   and   resamples  the
     timeseries/profile at locations where  the  user  needs  the
     values.   The  user  must  provide  the column number of the
     independent (monotonically increasing or  decreasing)  vari-
     able.   Equidistant  or  arbitrary sampling can be selected.
     All columns are resampled based on the new  sampling  inter-
     val.  Several interpolation schemes are available.

     infile
          This is a multi-column ASCII [of binary, see  -b]  file
          with  one  column  containing  the independent variable
          (which must be monotonically  in/de-creasing)  and  the
          remaining  columns  holding  misc.  data values.  If no
          file is provided, sample1d reads from standard input.


OPTIONS

     No space between the option flag and  the  associated  argu-
     ments.

     -F   Choose  from  l  (Linear),  a  (Akima  spline),  and  c
          (natural  cubic  spline)  [Default  is  -Fa].   You may
          change the default interpolant; see INTERPOLANT in your
          .gmtdefaults file.

     -H   Input file(s) has Header record(s).  Number  of  header
          records  can  be  changed  by editing your .gmtdefaults
          file.  If used, GMT default is 1 header record.

     -I   xinc defines the sampling  interval.  [Default  is  the
          separation  between  the first and second abcissa point
          in the infile]

     -M   Multiple segment file.  Segments  are  separated  by  a
          record  whose  first  character  is  flag.  [Default is
          '>'].

     -N   knotfile is an optional ASCII file with the x locations
          where  the  data  set  will  be  resampled in the first
          column

     -S   For equidistant sampling, xstart indicates the location
          of  the  first  output  value. [Default is the smallest
          even multiple of xinc inside the range of infile]

     -T   Sets the column  number  of  the  independent  variable
          [Default is 0 (first)].

     -bi  Selects binary input.  Append s  for  single  precision
          [Default  is  double].   Append  n  for  the  number of
          columns in the binary file(s).  [Default is  2  (or  at
          least the number of columns implied by -T].

     -bo  Selects binary output.  Append s for  single  precision
          [Default is double].


EXAMPLES

     To  resample  the  file   profiles.tdgmb,   which   contains
     (time,distance,gravity,magnetics,bathymetry) records, at 1km
     equidistant intervals using Akima's spline, try

     sample1d profiles.tdgmb -I1 -Fa -T1 > profiles_equi_d.tdgmb

     To resample the file depths.dt at positions  listed  in  the
     file  grav_pos.dg,  using  a cubic spline for the interpola-
     tion, try

     sample1d depths.dt -Ngrav_pos.dg -Fc > new_depths.dt


SEE ALSO

     gmt(l), filter1d(l)