
      File : ETERNA34.HLP, Status 1997.09.21  4746 records
                                                                               
      #######    ########    #######     #######      ##    ##        ####     
      ##            ##       ##          ##    ##     ###   ##       ##  ##    
      ##            ##       ##          ##    ##     ####  ##      ##    ##   
      ######        ##       ######      #######      ## ## ##      ##    ##   
      ##            ##       ##          ##  ##       ##  ####      ########   
      ##            ##       ##          ##   ##      ##   ###      ##    ##   
      #######       ##       #######     ##    ##     ##    ##      ##    ##   
                                                                               
                                                                               
      Prof. Dr.-Ing. H.-G. Wenzel                                              
      Black Forest Observatory                                                 
      Universitaet Karlsruhe                                                   
      Englerstr. 7                                                             
      D-76128 KARLSRUHE                                                        
      Germany                                                                  
      Phone   : ++49-0721-6082307                                              
      Telefax : ++49-0721-694552                                               
      e-mail  : wenzel@gik.bau-verm.uni-karlsruhe.de                           
                                                                               
               ***************************************************             
               *                                                 *             
               *    EARTH TIDE DATA PROCESSING PACKAGE ETERNA    *             
               *                                                 *             
               *                   Version 3.40                  *             
               *                                                 *             
               *                Manual ETERNA34.HLP              *             
               *                                                 *             
               ***************************************************             
                                                                               
      Contents of the manual ETERNA34.HLP:                                     
      ------------------------------------                            record
                                                                               
       1.:   General description of the ETERNA package...............    123 
        1.1: How to start............................................    240
        1.2: Distribution of the ETERNA package......................    402   
        1.3: Platforms...............................................    418   
        1.4: Contents of the CD-rom..................................    436 
 
       2.:   Description of program INSTALCD.........................    685
 
       3.:   Description of program HELPME34.........................    710
 
       4.:   Description of the ETERNA standard format...............    736
        4.1: File header.............................................    747
        4.2: Parameter section.......................................    762
        4.3: File body...............................................    770
 
       5.:   References..............................................    867 
 
       6.:   List of enclosures......................................   1043
 
       7:    Description of program RECTIDE..........................   1144 
        7.1: Hardware requirements...................................   1169 
        7.2: Program description.....................................   1209
        7.3: Description of the project name file "project"..........   1308
        7.4: Description of control parameter file "pn".ini..........   1319 
        7.5: Description of output files.............................   1496 
        7.6: Program tests...........................................   1559 
 
       8.:   Description of program ETSTEP...........................   1576 
                                                                               
       9.:   Description of program IERS.............................   1845
 
      10.:   Description of program BENCHMAR.........................   1877
 
      11.:   Description of program PREDICT..........................   1894
       11.1: Description of the project name file "project"..........   1911
       11.2: Description of the control parameter file "pn".ini......   1922
       11.3: Description of the output file "pn".out.................   2145
       11.4: Execution time for program PREDICT......................   2196
 
      12.:   Description of program PREGRED .........................   2223 
 
      13.:   Description of program DECIMATE.........................   2783
       13.1: Decimation of data......................................   2796
       13.2: Description of the project name file "project"..........   2839
       13.3: Description of the control parameter file "pn".INI......   2850
 
      14.:   Description of program DETIDE...........................   2894
       14.1: Calibration of data.....................................   2984   
       14.2: Computation of model tides and residual signal..........   3025   
       14.3: Description of the project name file "project"..........   3056
       14.4: Description of the control parameter file "pn".ini......   3067
       14.5: Description of the calibration parameter file...........   3212
       
      15.:   Description of program DESPIKE .........................   3325
       15.1: Destepping, despiking and degapping.....................   3364
       15.2: Description of the project name file "project"..........   3455
       15.3: Description of the control parameter file "pn".INI......   3466
       15.4: Examples for data preprocessing.........................   3514
 
      16.:   Description of program PREPLOT..........................   3573
 
      17.:   Description of program ANALYZE..........................   3604
       17.1: Description of the project name file "project"..........   3811
       17.2: Description of the control parameter files "pn".ini.....   3835
       17.3: Description of the data input files "pn".dat............   4225
       17.4: Description of the residual output files "pn".res.......   4318
       17.5: Description of the residual amplitude spectrum file               
             "pn".far................................................   4354
       17.6: Description of examples.................................   4368
       17.7: Execution time for the example files....................   4442
 
      18.:   Description of program TRANS............................   4464
 
      19.:   Description of program PLOTDATA.........................   4485
 
      20.:   Description of program PLOTFILT.........................   4542
 
      21.:   Description of program PLOTHIST.........................   4585
 
      22.:   Description of program PLOTSPEC.........................   4634
 
      23.:   Description of program PRINTGL..........................   4677
 
      24.:   Description of program LOAD89...........................   4719
 
      25.:   Description of program WPAREX...........................   4731
 
      Section 1:   General description of the ETERNA package                   
      ------------------------------------------------------                   
                                                                               
      The Earth tide data processing package ETERNA 3.40  (e.g. Wenzel 1996)
      allows the recording, preprocessing and analysis of earth tide  obser-
      vations, the prediction of earth tide signals and the  computation  of
      ocean tide  loading  under  MS-DOS  operating  system  on  a  personal 
      computer 80386/387 upwards. The  computing  kernels  DETIDE,  DESPIKE,
      DECIMATE, ANALYZE, PREDICT and OCELOAD are written in standard Fortran
      77/90 and can also be compiled and executed  on  a  workstation  under 
      UNIX operating system. The new international standard format  for  the
      storage and exchange of high resolution earth tide data  (Wenzel 1995)
      is used in all parts of the ETERNA 3.40 package.
                     
      Compared to previous versions, we have included into the ETERNA  earth
      tide data processing package the most accurate tidal  potential  cata-
      logue of Hartmann and Wenzel (1995a,b), and the DUT1 correction due to
      the Earth's variable rotation. The DUT1 values provided by the  Inter-
      ational Earth  Rotation  Service  (IERS)  are  used  in  the  relevant 
      programs. These upgrades enable the  computation  of  synthetic  earth 
      tide signals with a model accuracy better than 1  ngal  (1 ngal = 0.01
      nm/s**2). 
 
      Several parts of the ETERNA earth tide data  proceesing  package  have
      been rewritten, several programs have been added, and a  few  programs
      have been renamed. For the computation of the earth  tide  signals  we
      have implemented recursion formulas which reduce the computation  time
      for some data sets up to 35% compared  to  previous  versions  without 
      loss of accuracy. The ETERNA package has been given a new  and  better
      structure (see below) in order  to  avoid  multiple  storage  of  data 
      files. We have added an online manual  with  program  HELMPE33,  which
      enables the reading of the manual, searching and  extraction  of  text
      from the online manual. The former program ETERNA has been renamed  to
      ANALYZE, the former  program  PRETERNA  has  been  splitted  into  the 
      programs DETIDE, DESPIKE and DECIMATE, and the former  program  ETGTAB
      has been rewritten and renamed to PREDICT. O. Francis from the  Inter-
      national Center for Earth Tides (ICET) was  so  kind  to  provide  his
      program LOAD89 (and associated  data  files  for  ocean  tide  loading
      computation), which has been added to the ETERNA package in  order  to
      provide the users  the  possibility  to  compute  ocean  tide  loading 
      effects by themselves. Thanks to O. Francis !

      We believe that a substantial improvement with  respect  to  accuracy,
      flexibility and operational comfort  has  been  achieved  compared  to
      previous versions of the package. ETERNA 3.40 is  currently  the  only
      earth tide data processing package with a model accuracy  better  than
      1 ngal !
 
      The ETERNA 3.40 package is delivered on a CD-rom together with a short
      manual printed on paper. Because the total  package  needs  about  158 
      MByte when installed, there is no  diskette  version  of  the  package 
      available. The package is divided into  several  pieces  and  will  be 
      installed in 7  directories,  if  you  use  the  installation  program 
      SETUP34 given on the CD-rom:
                                                                               
      +-------------+       +------------------+          
      |  eterna34   |---+---| eterna34/rectid  |  data acquisition
      +-------------+   |   +------------------+
                        |             
                        |   +------------------+          
                        +---| eterna34/bin     |  executable files
                        |   +------------------+
                        |             
                        |   +------------------+
                        +---| eterna34/commdat |   common data files
                        |   +------------------+
                        |    
                        |   +------------------+                       
                        +---| eterna34/1mindat |   1 min data files
                        |   +------------------+ 
                        |
                        |   +------------------+
                        +---| eterna34/hourdat |   hourly data files
                        |   +------------------+
                        |
                        |   +------------------+
                        +---| eterna34/oceload |   ocean tide loading
                        |   +------------------+
                        |
                        |   +------------------+
                        +---| eterna34/iers    |   earth orientation 
                            +------------------+
                         
 
      The ETERNA 3.40 package contains the following 21 programs (including
      the necessary data files to run these programs plus examples):
 
      name          length    purpose
                    [records]
 
      ANALYZE:        6374    analysis of earth tide data.
      BENCHMAR:       3913    comparison of model tides with benchmark series. 
      DETIDE:         4079    calibration and detiding of earth tide data.
      DESPIKE:        2166    despiking of earth tide data.
      DECIMATE:       1034    decimation of earth tide data.
      ETSTEP:         1028    evaluation of a recorded stepresponse.
      IERS:            179    transformation of earth rotation data.
      SETUP34:         267    installation of the ETERNA 3.40 package.
      HELPME34:        356    display of the online manual. 
      LOAD97:          554    ocean tide loading computation.
      PREDICT:        3718    computation of synthetic model tides.
      PREGRED:        4067    graphical editing of earth tide data.
      PREPLOT:         361    visualization of earth tide data.
      PRINTGL:                hard copy of a plot.
      PLOTDATA:       2054    plot of earth tide data. 
      PLOTFILT:       1568    plot of numerical filter gain.
      PLOTSPEC:       1519    plot of spectra of earth tide residuals.
      PLOTHIST:       1570    plot of histograms of earth tide residuals.
      RECTIDE:        2511    recording of earth tide data. 
      TRANS:           983    transformation of old international format 
      WPAREX:          231    interpolation of gravity tide parameters 
                         
      in total       38525 program records   
 
      Please note that - except for very few programs - we do not supply the
      source code for the programs of the ETERNA 3.40 package. The experien-
      ce experience with previous versions of the  package  has  shown  that 
      few users only are able to recompile the programs,  because  generally 
      the users do not own the necessary compilers. Additionally, we want to
      avoid user generated modifications of the programs.  
    
      Section 1.1: How to start                                                
      -------------------------                                                
                                                                               
      Attention: 
 
      Your personal computer should  have  an  80386/387  upwards  processor 
      (i.e. an 80486 or PENTIUM) with at least 8 MBytes  of  RAM  adressable 
      the programs. Although the Fortran programs have been  compiled  using 
      the Lahey virtual memory manager, these programs can be executed  with
      a few MBytes of RAM only. The computation speed  is  however  severely
      degraded under such conditions. From the ETERNA 3.40 package, only the
      programs  RECTIDE  and  PREPLOT  can  be  executed  on  an   80286/287 
      processor.   
 
      You should have more than 157 MByte empty  space  on  your  hard  disk
      available when  starting  the  installation  program  INSTALCD  for  a  
      complete installation of the ETERNA 3.40 package. All program and data
      files in directory oceload need about 120 MByte of your hard disk.  If
      you have less free disk space on your hard  disk  available,  you  may 
      decide during the execution of program INSTALCD on the installation of
      directories "rectide" and "oceload", which saves about 126 MByte  disk 
      space.
 
      To perform the  installation  of  the  ETERNA  3.40  package,  use  an  
      80386/387, 80486 or PENTIUM personal computer operating  under  MS-DOS
      5.0...6.20 operating system. Insert the CD-rom into the  CD-rom  drive
      (supposed this is e:), and make the CD-rom drive your active drive  by
      entering
     
                   <e:>                         
                     
      You have to define the hard disk  partition  on  which  the  directory
      eterna33 and all subdiretories shall be installed  (supposed  this  is 
      d:). Attention: If there exists already a directory  named  "eterna33"
      on the chosen hard disk, the next step will probably ne dangerous  and
      you should carefully read the manual before continuing!
 
      Start the installation program INSTALCD  provided  on  the  CD-rom  by 
      entering
 
                   <instalcd d:>
 
      and follow the instructions. The installation  program  INSTALCD  will
      establish the diretory eterna33 at  your  chosen  hard  disk  and  all
      necessary subdiretories under this directory (for some subdirectories,
      the installation program will ask for your advice to  install  a  sub-
      directory or not). At its last step,  the  installation  program  will
      execute an earth tide analysis  for  five  different  data  sets  with
      program ANALYZE. This will take  several  minutes  (depending  on  the 
      speed of your processor). You should see on your screen the  following
      printout (except for the execution time on the right column):
 
         ******************************************************
         *                                                    *
         *  Program ANALYZE, version 3.40 970921 Fortran 90.  *
         *                                                    *
         *        Analysis of earthtide observations.         *
         *                                                    *
         *    The  Black  Forest  Observatory  Schiltach      *
         *    wishes you much success when using ANALYZE.     *
         *                                                    *
         ******************************************************


      1 HAL29901: Analysis started                                     0.00 s 
      1 HAL29901: Tidal potential catalogue:Hartmann+Wenzel (1995)     0.00 s 
      1 HAL29901: Input of observations completed                      1.76 s 
      1 HAL29901: Normal equations computed                            3.63 s 
      1 HAL29901: Normal equations solved                              3.68 s 
      1 HAL29901: Residuals computed                                   4.34 s 
      1 HAL29901: Spectrum of residuals computed                       4.84 s 
      1 HAL29901: Analysis finished,    stdv.=     6.945 nm/s**2       4.84 s 

      2 BFL24903: Analysis started                                     0.00 s 
      2 BFL24903: Tidal potential catalogue:Hartmann+Wenzel (1995)     0.00 s 
      2 BFL24903: Input of observations completed                      2.64 s 
      2 BFL24903: Normal equations computed                           12.31 s 
      2 BFL24903: Normal equations solved                             12.42 s 
      2 BFL24903: Residuals computed                                  14.72 s 
      2 BFL24903: Spectrum of residuals computed                      15.49 s 
      2 BFL24903: Analysis finished,    stdv.=     0.746 nm/s**2      15.49 s 

      3 BFD00801: Analysis started                                     0.00 s 
      3 BFD00801: Tidal potential catalogue:Hartmann+Wenzel (1995)     0.00 s 
      3 BFD00801: Input of observations completed                      1.92 s 
      3 BFD00801: Normal equations computed                            9.55 s 
      3 BFD00801: Normal equations solved                              9.61 s 
      3 BFD00801: Residuals computed                                  10.49 s 
      3 BFD00801: Spectrum of residuals computed                      11.20 s 
      3 BFD00801: Analysis finished,    stdv.=     1.554  nstr        11.20 s 

      4 BFET1907: Analysis started                                     0.00 s 
      4 BFET1907: Tidal potential catalogue:Hartmann+Wenzel (1995)     0.00 s 
      4 BFET1907: Input of observations completed                      5.21 s 
      4 BFET1907: Normal equations computed                           34.16 s 
      4 BFET1907: Normal equations solved                             34.27 s 
      4 BFET1907: Residuals computed                                  45.09 s 
      4 BFET1907: Spectrum of residuals computed                      47.07 s 
      4 BFET1907: Analysis finished,    stdv.=     0.558 nm/s**2      47.07 s 

      5 BHTT4003: Analysis started                                     0.00 s 
      5 BHTT4003: Tidal potential catalogue:Hartmann+Wenzel (1995)     0.00 s 
      5 BHTT4003: Input of observations completed                     10.49 s 
      5 BHTT4003: Normal equations computed                           46.25 s 
      5 BHTT4003: Normal equations solved                             46.63 s 
      5 BHTT4003: Residuals computed                                  76.07 s 
      5 BHTT4003: Spectrum of residuals computed                      83.00 s 
      5 BHTT4003: Analysis finished,    stdv.=     6.884 nm/s**2      83.00 s 

          **********************************************
          *   Program ANALYZE finished the execution.  *
          **********************************************
 
          Total execution time:    161.870  seconds
 
      When the installation program INSTALCD has finished its execution, you
      should modify in  the  "autoexec.bat"  file  of  your  pc  the  record 
      containing  the  path  statement  by  adding  the  path  to  directory 
      \eterna34\bin. Change the already existing path statement to e.g.
 
                   path c:\dos;d:\eterna34\bin
 
      After rebooting your pc, you should be able to  execute  all  programs
      of the ETERNA 3.40 package from  the  directories  which  contain  the
      relevant data files.
 
      After  the  installation  of  the  ETERNA  3.40  package,  you  should 
      carefully study the manual and the relevant publications given in  the
      list of references. If you feel familiar  with  the  capabilities  and
      restrictions of the programs, you should e.g.  start  program  ANALYZE
      from directory "\eterna33\hourdat" by entering
 
                   <analyze>
 
      You should receive a small part of the  printout  on  the  screen  and
      the complete printouts in  the  files  "hal2901.prn",  "bfl24903.prn",
      "bfd00801.prn" and "bhtt4003.prn"  in  directory  "\eterna33\hourdat".
      The contents of these files should be  almost  the  same  as  in  this 
      manual under section 10 (except  for  the  printed  jobtimes,  because 
      this depends on the speed of your computer). 
    
      If you want to run programs from the ETERNA package on  a  workstation 
      under UNIX operating system, you should at first install  the  package 
      on a pc under MS-DOS operating system using the  installation  program
      INSTALCD and afterwards transfer the  relevant  files  to  your  work- 
      station. You have to enable at the Fortran 90 programs those  records,
      which are currently disabled by comment   C HP-UX   and  you  have  to 
      disable the relevant statements for the MS-DOS operating  system.  You 
      have to re-compile the programs with  your  own  Fortran  90  compiler
      (most of the programs can be compiled with  a  Fortran  77  compiler). 
      You should not transfer the unformatted files with  suffix  ".uft"  to 
      your workstation, but  have  these  files  established  by  the  first 
      execution of the relevant programs. 
 
      There is little hope  that  you  will  be  able  to  re-compile  under 
      operating system UNIX e.g. the  powerfull  graphical  editor  PREGRED, 
      which is written for the Microsoft Visual  C++  compiler.  Instead  of 
      spending much time to transform some of the  programs  of  the  ETERNA 
      3.40 package to a workstation, the better procedure is to buy a  cheap 
      PENTIUM 100 MHz pc which most probably executes  the  programs  faster 
      than your workstation !
 
      Section 1.2: Distribution of the ETERNA package                          
      ------------------------------------------------                        
                                                                              
      The programs including  data  files  and result  files are distributed  
      on a CD-rom, together with sections 1 and 2  of  this  program  manual
      printed on paper. Please remind that the programs  may  only  be  used 
      within your institution, but should not be copied or given in loan  to
      anybody outside your institution. Requests  for  the  program  package 
      ETERNA should be forwarded directly to  the  program  author.  If  you 
      experience any error in the programs or failure of the programs during
      execution, which are not due to an error in input data or a misuse  of
      the programs, please report these errors to the program author. If you
      have any proposal for a significant improvement of  the  handling  and 
      performance of the ETERNA package, please do not hesitate to contact
      the author.                                                    
                                                                               
      Section 1.3: Platforms                                                   
      ----------------------                                                   
                                                                               
      All programs have been installed  and  tested  on  different  personal 
      computers (80486 DX2 66 MHz, 80486 DX4 100 MHz, PENTIUM 90 MHz ... 166 
      MHz), using a standard VGA graphics adapter and  different  super  VGA 
      graphics adapters (Spea V7 Mirage, Spea V7 Mercury, Miro Crystal  20SD
      PCI). Some  programs have been executed for timing purpose on  several
      different platforms. The programs DETIDE, DESPIKE, DECIMATE,  ANALYZE,   
      ETSTEP, PREDICT and LOAD89 are written in standard Fortran,  and  have   
      been tested additionally on a SUN sparc 2 workstation  under operating   
      system SOLARIS 2.3. However, it is recommended to use ETERNA 3.40 on a
      high speed PENTIUM processor under MS-DOS  operating  system,  because
      the overall performance of the ETERNA package is much  superior  under   
      MS-DOS  than under UNIX or SOLARIS operating systems. This  is  mainly   
      because the ETERNA package does  not  supply  programs  for  graphical   
      visualization under UNIX or SOLARIS operating systems.                   
 
      Section 1.4: Contents of the CD-rom                                   
      -----------------------------------                                   
                                                                                 
      file name        kbytes     date   description                             
                                                                                 
      instalcd.exe    286.612  22.08.96  installation program                 
 
      directory \eterna33\rectide (15 files, 772.367 kB):
 
      file name        kbytes      date  description
 
      autoexec.bat        473  13.08.96  sample AUTOEXEC.BAT file 
      etcpot.dat      112.983  22.01.96  tidal potential catalogues
      etstep.exe      304.949  14.08.96  program ETSTEP, executable file
      etstep.ini          571  13.08.96  control parameter file for ETSTEP
      etstep.prn       62.841  14.08.96  print file of program ETSTEP
      n20m1s01.nlf     25.832  04.09.96  numerical lowpass filter 1s
      n20m5s02.nlf      6.172  04.09.96  numerical lowpass filter 5s
      project               9  22.01.96  project file
      recsimul.ini      2.775  14.08.96  control parameter file for RECTIDE
      kal68601.ini      2.779  14.08.96  control parameter file for RECTIDE
      rectide.bas     132.950  11.09.96  program RECTIDE, source code
      rectide.exe     108.385  11.09.96  program RECTIDE, executable file
      st299v81.dat      8.977  14.08.96  step response sample data file
      plotdata.ini      1.658  22.08.96  control parameters for PLOTDATA
      plotfilt.ini      1.108  22.08.96  control parameters for PLOTFILT
       
      directory \eterna33\bin (27 files, 5 012.852 kB):
                                                                             
      name             kbytes     date   description                         
                                                                          
      analyze.exe     401.925  09.09.96  executable file of program ANALYZE
      benchmar.exe    352.117  09.09.96  executable file of program BENCHMAR 
      decimate.exe    301.429  06.09.96  executable file of program DECIMATE
      detide.exe      351.493  09.09.96  executable file of program DETIDE
      despike.exe     324.053  06.09.96  executable file of program DESPIKE
      iers.exe        280.645  15.08.96  executable file of program IERS
      instalcd.exe    269.028  03.09.96  executable file of program INSTALCD
      helpme33.exe    419.381  11.08.96  executable file of program HELPME33
      hpno.bin            927  05.08.94  driver for HPGL plot files
      hpyes.bin         1.167  02.08.94  driver for HPGL plot files 
      hplsr3.bin        1.123  08.08.94  driver for HP-laser printer
      eps-p.bin         1.764  30.06.93  driver for Postcript plotfiles
      eps-l.bin         1.796  30.06.93  driver for Postcript plotfiles
      lf90.eer         40.736  19.10.95  error message file from Lahey
      oem08.fon         4.752  16.09.94  font file for PREGRED
      oem10.fon         5.264  16.09.94  font file for PREGRED
      plotdata.exe    336.661  22.09.96  executable file of program PLOTDATA
      plotfilt.exe    329.637  22.09.96  executable file of program PLOTFILT
      plothist.exe    334.101  08.09.96  executable file of program PLOTHIST
      plotraus.bat         26  05.08.96  batch file for hardcopy of a plot
      plotspec.exe    327.045  22.09.96  executable file of program PLOTSPEC
      predict.exe     346.437  09.09.96  executable file of program PREDICT
      pregred.exe     110.874  06.09.96  executable file of program PREGRED
      preplot.bas      15.295  27.08.96  sourcecode file of program PREPLOT
      preplot.exe      67.059  27.08.96  executable file of program PREPLOT
      printgl.exe      75.376  01.09.91  executable file of program PRINTGL
      trans.exe       312.725  27.08.96  executable file of program TRANS
 
      directory \eterna33\commdat (50 files, 7 486.977 kB):
                                                                             
      name             kbytes     date   description                         
                                                                             
      buellehw.dat     75.609  28.06.95  Buellesfeld catalogue, ASCII
      buellehw.uft     74.005  24.05.96  Buellesfeld catalogue, binary
      cted73hw.dat     61.186  28.06.95  Cartwright et al. catalogue, ASCII
      cted73hw.uft     60.107  20.03.96  Cartwright et al. catalogue, binary
      doodsehw.dat     46.036  28.06.95  Doodson catalogue, ASCII
      doodsehw.uft     45.265  28.03.96  Doodson catalogue, binary
      eopc04.96        17.200  13.08.96  IERS ftp file, ASCII
      etddt.dat        11.316  23.03.96  TDT-UTC, ASCII
      eterna33.hlp    325.009  11.09.96  ETERNA33 manual, ASCII 
      etpolut1.dat    603.131  13.08.96  IERS polar motion and DUT1, ASCII
      etpolut1.uft    718.560  13.08.96  IERS polar motion and DUT1, binary
      hw95s.dat     1 401.137  04.08.96  Hartmann + Wenzel catalogue, ASCII 
      hw95s.uft     1 363.319  04.08.96  Hartmann + Wenzel catalogue, binary
      ratgp95.dat     700.818  13.03.96  Roosbeek catalogue, ASCII
      ratgp95.uft     681.677  24.05.96  Roosbeek catalogue, binary
      tamurahw.dat    133.823  25.05.96  Tamura catalogue, ASCII
      tamurahw.uft    130.667  25.05.96  Tamura catalogue, binary
      xi1989hw.dat    319.604  13.03.96  Xi catalogue, ASCII
      xi1989hw.uft    311.175  24.05.96  Xi catalogue, binary
      n14h5m01.nlf      4.425  29.08.96  lowpass (old) 5 min / 60 min old
      n2h1m001.nlf      3.748  29.08.96  lowpass (old) 1 min /  5 min old
      n01s1m01.nlf     32.878  29.08.96  lowpass  1 s / 1 min
      n02s1m01.nlf     17.556  29.08.96  lowpass  2 s / 1 min
      n03s1m01.nlf     17.578  29.08.96  lowpass  3 s / 1 min
      n05s1m01.nlf      8.397  29.08.96  lowpass  5 s / 1 min
      n10s1m01.nlf      5.275  29.08.96  lowpass 10 s / 1 min
      n20s1m01.nlf      3.653  29.08.96  lowpass 20 s / 1 min
      n30s1m01.nlf      3.083  29.08.96  lowpass 30 s / 1 min
      n02s5m02.nlf     79.053  29.08.96  lowpass  2 s / 5 min
      n03s5m02.nlf     53.548  29.08.96  lowpass  3 s/ 5 min
      n05s5m02.nlf     33.143  29.08.96  lowpass  5 s / 5 min
      n10s5m02.nlf     17.705  29.08.96  lowpass 10 s / 5 min
      n20s5m02.nlf     10.199  29.08.96  lowpass 20 s / 5 min
      n30s5m02.nlf      7.423  29.08.96  lowpass 30 s / 5 min
      n60s5m02.nlf      4.729  29.08.96  lowpass 60 s / 5 min
      n1h1h001.nlf      1.128  29.08.96  lowpass for ANALYZE, 1 h sampl. 
      n1h1h002.nlf      1.362  29.08.96  lowpass for ANALYZE, 1 h sampl. 
      n1h1h003.nlf      2.927  29.08.96  lowpass for ANALYZE, 1 h sampl.
      n1h1h004.nlf      4.557  29.08.96  lowpass for ANALYZE, 1 h sampl.
      n1h1h005.nlf      1.328  29.08.96  lowpass for ANALYZE, 1 h sampl.      
      n1h1h006.nlf      1.334  29.08.96  lowpass for ANALYZE, 1 h sampl.
      n1h1h007.nlf      2.965  29.08.96  lowpass for ANALYZE, 1 h sampl.
      n1h1h008.nlf      4.598  29.08.96  lowpass for ANALYZE, 1 h sampl.
      n05m05m2.nlf     36.342  06.09.96  lowpass for ANALYZE,  5 min sampl.
      n10m10m2.nlf     19.342  05.09.96  lowpass for ANALYZE, 10 min sampl.
      n20m20m2.nlf     10.839  05.09.96  lowpass for ANALYZE, 20 min sampl.
      n30m30m2.nlf      8.017  05.09.96  lowpass for ANALYZE, 30 min sampl.
      n60m60m2.nlf      5.249  05.09.96  lowpass for ANALYZE, 60 min sampl.
 
      directory \eterna33\1mindat (26 files, 10 856.083 kB):
                                                                           
      name             kbytes     date   description                         

      preplot.bas      15.295  27.08.96  source code file of PREPLOT
      kal299tt.cal      1.045  04.09.96  calibration data        kal299tt
      kal299tt.raw    129.480  04.09.96  raw 1 min data          kal299tt
      kal299tt.ini      3.995  06.09.96  control parameter       kal299tt
      kal299tt.tid    199.696  06.09.96  detided 1 min data      kal299tt
      kal299tt.cor    199.861  06.09.96  despiked 1 min data     kal299tt
      kal299tt.fil     40.076  06.09.96  preprocessed 5 min data kal299tt
      kal299tt.dec      4.583  06.09.96  preprocessed 1 h   data kal299tt   
      sg102tes.ini      4.149  06.09.96  control parameter       sg102tes
      sg102tes.raw   1102.905  06.09.96  raw data                sg102tes
      sg102tes.cal      1.059  05.09.96  calibration data        sg102tes
      sg102tes.tid   1560.448  06.09.96  detided data            sg102tes    
      sg102tes.pro      6.322  06.09.96  PREGRED protocol        sg102tes
      sg102tes.cor   1560.408  06.09.96  manually edited data    sg102tes
      sg102tes.des   1560.488  07.09.96  despiked data           sg102tes  
      sg102tes.fil    311.194  07.09.96  preprocessed 5 min data sg102tes
      sg102tes.dec     25.982  09.09.96  preprocessed 1 h   data sg102tes
      bfhw9502.rrr   1109.811  07.09.96  raw synthetic data      bfhw9502
      bfhw9502.ini      8.184  07.09.96  control parameter       bfhw9502
      bfhw9502.cal      1.465  07.09.96  calibration data        bfhw9502
      bfhw9502.tid   1398.255  07.09.96  detided data            bfhw9502
      bfhw9502.cor   1398.340  07.09.96  manually edited data    bfhw9502
      bfhw9502.fil    278.199  07.09.96  preprocessed 5 min data bfhw9502 
      bfhw9502.dec     23.341  07.09.96  preprocessed 1 h   data bfhw9502
      plotdata.ini      1.493  07.09.96  control parameters for PLOTDATA
      project               9  05.09.96  project file
 
      directory \eterna33\hourdat (54 files, 23 364.605 kB):
                                                                             
      name             kbytes     date   description                         

      analyze.prn       3.879  09.09.96  logfile of program ANALYZE
      bfde403f.dat  1 898.359  08.09.96  benchmark gravity tides bfde403f
      bfde403f.dif  3 296.611  08.09.96  differences bfde403f - HW95 
      bfde403f.ini      5.427  08.09.96  control parameter for     bfde403f
      bfde403f.prn     44.810  08.09.96  print file of ANALYZE for bfde403f
      bfde403f.res  2 602.404  08.09.96  residuals  of ANALYZE for bfde403f
      bfde403f.far     25.094  08.09.96  spectrum of residuals for bfde403f
      bfde403f.par      3.840  08.09.96  adjusted tidal params.for bfde403f
      bfd00801.ini      5.123  07.09.96  control parameters for    bfd00801
      bfd00801.dat     73.375  06.09.96  observed hor. strain      bfd00801
      bfd00801.prn     35.285  07.09.96  print file of ANALYZE for bfd00801
      bfd00801.res     88.664  07.09.96  residuals  of ANALYZE for bfd00801
      bfd00801.far     19.436  07.09.96  spectrum of residuals for bfd00801
      bfd00801.par      1.920  07.09.96  adjusted tidal params.for bfd00801
      bfet1907.ini      6.141  07.09.96  control parameters for    bfet1907
      bfet1907.dat    259.448  06.09.96  observed gravity data     bfet1907
      bfet1907.prn     38.436  07.09.96  print file of ANALYZE for bfet1907
      bfet1907.res    322.370  07.09.96  residuals  of ANALYZE for bfet1907
      bfet1907.far     23.946  07.09.96  spectrum of residuals for bfet1907
      bfet1907.par      2.720  07.09.96  adjusted tidal params.for bfet1907
      bfhw9501.ini      6.920  08.09.96  control parameters for    bfhw9501
      bfhw9501.prd  3 738.177  08.09.96  synthetic gravity signal  bfhw9501 
      bfhw9501.dat  3 738.275  08.09.96  synthetic gravity signal  bfhw9501
      bfhw9501.prn     37.393  08.09.96  print file of ANALYZE for bfhw9501
      bfhw9501.res  3 739.268  08.09.96  residuals  of ANALYZE for bfhw9501
      bfhw9501.far     20.666  08.09.96  spectrum of residuals for bfhw9501
      bfhw9501.par      3.840  08.09.96  adjusted tidal params.for bfhw9501 
      bfl24903.ini      5.581  05.09.96  control parameters for    bfl24903
      bfl24903.dat    111.915  06.09.96  oberserved gravity data   bfl24903
      bfl24903.prn     38.971  09.09.96  print file of ANALYZE for bfl24903
      bfl24903.res    125.406  09.09.96  residuals  of ANALYZE for bfl24903
      bfl24903.far     21.814  09.09.96  spectrum of residuals for bfl24903
      bfl24903.par      2.240  09.09.96  adjusted tidal params.for bfl24903 
      bhtt4003.ini      6.892  07.09.96  control parameters for    bhtt4003
      bhtt4003.dat    895.331  06.09.96  observed gravity data     bhtt4003
      bhtt4003.prn     37.258  09.09.96  print file of ANALYZE for bhtt4003
      bhtt4003.res  1 618.588  09.09.96  residuals  of ANALYZE for bhtt4003
      bhtt4003.far     20.748  09.09.96  spectrum of residuals for bhtt4003
      bhtt4003.par      3.520  09.09.96  adjusted tidal params.for bhtt4003   
      deafult.ini       6.293  07.08.96  default control parameter file
      hal29901.ini      5.141  07.09.96  control parameters for    hal29901
      hal29901.dat    119.234  05.09.96  observed gravity data     hal29901
      hal29901.prn     33.124  09.09.96  print file of ANALYZE for hal29901
      hal29901.res     72.446  09.09.96  residuals  of ANALYZE for hal29901
      hal29901.far     20.584  09.09.96  spectrum of residuals for hal29901
      hal29901.par      1.920  09.09.96  adjusted tidal params.for hal29901 
      plotdata.ini      1.484  07.09.96  control parameters for PLOTDATA
      plotspec.ini      2.010  07.09.96  control parameters for PLOTSPEC
      plothist.ini      1.127  08.09.96  control parameters for PLOTHIST
      project              51  07.09.96  project file
      trans.dat        46.387  22.01.96  example input file for TRANS
      trans.prn         8.813  27.08.96  print  file of TRANS
      trans.ini         3.950  27.08.96  output file of TRANS
      trans.out       111.950  27.08.96  output file of TRANS 
 
      directory \eterna33\oceload (42 files, 126.135.507 kB):
                                                                             
      name             kbytes     date   description                         
       
      load89.f         15.541  08.05.96  source code file of program LOAD89
      load89.for       15.604  08.05.96  source code file of program LOAD89
      mascor.csr          312  08.05.96  mass cons. param.s for CSR3.0
      mascor.scw          114  08.05.96  mass cons. param.s for Schwiderski
      para.bxl            431  08.05.96  
      prem              2.001  08.05.96  Green function for PREM earth model
      scw80.mf      1 439.393  08.05.96  Schwiderski model for wave MF
      scw80.q1      1 442.308  08.05.96  Schwiderski model for wave Q1
      scw80.o1      1 467.149  08.05.96  Schwiderski model for wave O1
      scw80.p1      1 446.210  08.05.96  Schwiderski model for wave P1
      scw80.k1      1 476.612  08.05.96  Schwiderski model for wave K1
      scw80.m2      1 520.343  08.05.96  Schwiderski model for wave M2
      scw80.n2      1 454.624  08.05.96  Schwiderski model for wave N2
      scw80.s2      1 485.849  08.05.96  Schwiderski model for wave S2
      scw80.k2      1 446.142  08.05.96  Schwiderski model for wave K2
      csr3.q1       5 344.414  07.05.96  CSR3.0      model for wave Q1
      csr3.o1       5 325.820  07.05.96  CSR3.0      model for wave O1
      csr3.p1       5 323.618  07.05.96  CSR3.0      model for wave P1
      csr3.k1       5 362.966  07.05.96  CSR3.0      model for wave K1
      csr3.n2       5 331.767  07.05.96  CSR3.0      model for wave N2
      csr3.m2       5 406.137  07.05.96  CSR3.0      model for wave M2
      csr3.s2       5 373.547  07.05.96  CSR3.0      model for wave S2
      csr3.k2       5 330.499  07.05.96  CSR3.0      model for wave K2     
      fes952.q1     4 447.496  08.05.96  Grenoble 95 model for wave Q1
      fes952.o1     4 477.472  08.05.96  Grenoble 95 model for wave O1
      fes952.p1     4 403.049  08.05.96  Grenoble 95 model for wave P1
      fes952.k1     4 513.484  08.05.96  Grenoble 95 model for wave K1
      fes952.phi    5 114.332  08.05.96  Grenoble 95 model for wave PHI1
      fes952.pi1    5 023.486  08.05.96  Grenoble 95 model for wave PI1
      fes952.j1     4 675.572  08.05.96  Grenoble 95 model for wave J1
      fes952.sgi    4 884.838  08.05.96  Grenoble 95 model for wave SIGMA1
      fes952.n2     4 416.500  08.05.96  Grenoble 95 model for wave N2
      fes951.m2     4 682.982  08.05.96  Grenoble 95 model for wave M2
      fes952.s2     4 548.218  08.05.96  Grenoble 95 model for wave S2
      fes952.k2     4 400.315  08.05.96  Grenoble 95 model for wave K2
      fes952.t2     4 573.182  08.05.96  Grenoble 95 model for wave T2
      fes952.2n2    4 491.611  08.05.96  Grenoble 95 model for wave 2N2
      wparm.dat     5 174.490  07.09.96  world wide grid of Timmen and Wenzel
      wparex.for       10.947  07.09.96  interpolation program WPAREX
      wparex.exe      283.429  07.09.96  interpolation program WPAREX
      wparex.dat          155  07.09.96  input data file for program WPAREX
      wparex.prn        2.548  07.09.96  print file of program WPAREX 
 
      Section 2:   Description of program INSTALCD
      --------------------------------------------
 
      The automatic installation program INSTALCD is provided in the root of
      the CD-rom. If you insert the CD-rom into the CD-rom drive  and  start
      program INSTALCD from the root of the CD-rom,  program  INSTALCD  will 
      install the ETERNA 3.40 package on  your  hard  disk.  The  hard  disk 
      partition ond which the ETERNA 3.40 package shall be installed can  be
      input into  program  INSTALCD.  Program  INSTALCD  will  ask  for  the 
      installation of subdirectories "rectide" and "oceload",  in  order  to
      save hard disk space. The use of program INSTALCD is selfexplaining.
 
      After having installed the ETERNA 3.40 package,  you  should  add  the
      path to \eterna33\bin to the path  statement  of  your  "autoexec.bat" 
      file, e.g.
 
         path=c:\dos;c:\eterna33\bin;
 
      and define the MS-DOS environment variable MSCFONTS by adding in  your  
      "autoexec.bat" file the line
 
         set MSCFONTS=c:/eterna33/bin
 
      provided you have installed the ETERNA 3.40 package on drive c:
 
      Section 3:   Description of program HELPME33
      --------------------------------------------
 
      The program HELPME33.FOR is supplied in order to on-line  display  the
      manual for the ETERNA 3.40 earth tide data processing  package,  given
      in  file  "eterna34\commdat\eterna34.hlp".  The  executable program is
      given in file "\eterna34\bin\helpme34.exe". The program is  written in 
      non-standard Fortran 90 (it uses  special  statements  for  the  Lahey 
      starter-interacter kit LISK and can only be compiled using  the  Lahey 
      2.00 Fortran 90 compiler  with  option  -lisk). After entering
 
                   <helpme34>
 
      from any subdirectory, the program displays the first  20  records  of
      this manual on the screen. To exit the program, you can  either  press
      the escape-key or function key 4. You can use function key  1  to  see
      the on-line help window of the program. You can use function key 2  to
      to search for a certain string within the manual. You can use function
      key 3 to export the contents of the  manual  currently  shown  on  the 
      screen to file  "helpme34.prn",  which  will  be  established  in  the 
      current directory. You can use the up- and down-cursor keys to  scroll
      through the file. You can use the next and previous page keys  to  see
      the next or previous page of the manual. 
  
      Section 4: Description of the ETERNA standard format
      ----------------------------------------------------
                                                                               
      The data files of the ETERNA 3.40 package use the same standard format   
      which is described in Wenzel (1995) and has been adopted for  the  ex-   
      change of high precision and high rate earth tide data by the  Working   
      Group on High Precision Tidal Data Processing at its meeting  in  Bonn   
      1994. Data files to be used within ETERNA consist of one file  header,
      and at least one parameter section and one file body. A file body  may
      be followed by another parameter section plus another file body.
                                                                             
      Section 4.1: File header
      ------------------------
                                                                               
      A data file used in ETERNA  must  have  an  alphanumeric  file  header 
      containing some information to describe the file  and  the  processing
      which has eventually been done with  these  files.  The  length  of  a 
      header record is restricted to 80 characters (FORMAT A80). The  number 
      of header records is not restricted.  Please  try  to  enter  as  much  
      information as possible in the file header, because this might  be  of 
      some value for you when looking into the  file  after  some  time  has  
      gone, and for other people (e.g. your successor).   
                                                          
      The alphanumeric file header ends with a record having  C*********  in
      columns 1...10.                                                       
      
      Section 4.2: Parameter section
      ------------------------------
       
      A parameter section is allowed before each file  body.  The  parameter
      section ends with flag 77777777 in cols.  1...8. A  parameter  section
      is currently used with program ANALYZE only (see  the  description  of 
      program ANALYZE). 
                                                               
      Section 4.3: File body                                                           
      ----------------------                                                           
                                                                               
      The file body starts with an offset record (flag 77777777  in  columns 
      1...8), with offsets at all data  channels.  These  offsets  might  be 
      added to all subsequent data records. Offset records are allowed  also
      also within the file body at all positions in  the  data  set.  Offset
      records use the standard data format (I8,7X, 4F10.3) and  are  defined 
      by flag 77777777 in columns 1...8. The  first  offset  record  defines 
      initial offsets for all data channels. These offsets are added to  the
      samples for all subsequent records when reading the data file, and may
      be used to glue two or more data sets together  by  removing  eventual 
      steps between the data sets. The units of the offsets are identical to
      the units of the samples in the corresponding channels.  
                     
      Data records:                                                            
      -------------                                                            
                                                                               
      All subsequent records have the same format (I8,1X,I6,4F10.3):           
      IDAT    = data of the record, 19931125 means year = 1993,                
                month = 11 = November, day = 25.                               
      ITIM    = time of the sample in UTC, 140800 means 14 hours,  08  min.,   
                00 seconds. Please remind that the samples have to be  given   
                at full minute if e.g. the sample interval is 60 s.            
      DCIN(1..NC) = samples for channels 1...NC. The number NC of channels is
                   restricted to 7.                                     
                   Channel 1      is generally the earth tide observation.
                   Channel 2...NC may contain samples of meteorological
                                  parameters.    
                   Channel NC+1   may contain the model tide computed by
                                  e.g. program DETIDE.
                   Channel NC+2   may contain the residual signal computed 
                                  by e.g. program DETIDE.
 
                   The residual signal is computed by linear multiple 
                   regression:                                                 
                   ch(NC+2)= ch1 - ch(NC+1) + sum{FAC(i) * ch(i)}              
                                                                               
      The end of a the data block is defined by IDAT = 999999999.              
                                                                               
      A step correction may be carried out by using an offset  record  (flag   
      IDAT = 777777777) between two data records.  In  this  case,  ITIM  is   
      ignored in the offset record,  and  the  samples  for  the  individual   
      channels are interpreted as corrections to the offset, which  will  be   
      added for all subsequent data records.                                   
                                                                               
      A gap in the data may be defined by putting the undefined data  values   
      of the channels to 99999.999. It is allowed to have gaps in  all  data
      channels.
                                                                               
      We list in the following the first 50 records of  file  "kal299tt.tid"
      given in directory "\eterna33\1mindat"  for a better understanding  of
      the structure of ETERNA data files:     
                                                                       
      File              : kal299tt.tid                                         
      Status            : 19941101                                             
      Start             : 19921125 14:08                                       
      End               : 19921227 23:59                                       
      Contents          : Detide 1 min data from BFO-TLD1 data                 
                          acquisition, station Karlsruhe.                     
                          Channel 1 is LCR-G299-SW02    in nm/s**2,            
                          Channel 2 is barometer BMS5   in hPa,
                          Channel 3 is model tide       in nm/s**2,
                          Channel 4 is residual signal  in nm/s**2.
 
                        Chan. 1   Chan. 2   Chan. 3   Chan. 4 
      C**************************************************************
      77777777            0.000     0.000     0.000     0.000
      19921125 140800  1226.128     6.870   803.679   467.678
      19921125 140900  1227.583     6.903   805.466   467.568
      19921125 141000  1228.986     6.941   807.252   467.432
      19921125 141100  1230.650     6.954   809.037   467.401
      19921125 141200  1232.677     6.935   810.819   467.520
      19921125 141300  1234.756     6.899   812.599   467.579
      19921125 141400  1236.523     6.867   814.377   467.360
      19921125 141500  1237.978     6.855   816.152   466.962
      19921125 141600  1239.538     6.875   817.924   466.877
      19921125 141700  1241.461     6.931   819.693   467.402
      19921125 141800  1243.540     7.012   821.458   468.247
      19921125 141900  1245.359     7.089   823.220   468.810
      19921125 142000  1246.867     7.133   824.978   468.852
      19921125 142100  1248.374     7.125   826.731   468.552
      19921125 142200  1250.037     7.062   828.480   468.052
      19921125 142300  1251.752     6.973   830.224   467.441
      19921125 142400  1253.676     6.907   831.963   467.186
      19921125 142500  1255.651     6.887   833.697   467.298
      19921125 142600  1257.522     6.899   835.425   467.519
      19921125 142700  1259.289     6.912   837.148   467.651
      19921125 142800  1261.004     6.912   838.864   467.647
      19921125 142900  1262.720     6.897   840.574   467.557
      19921125 143000  1264.487     6.871   842.277   467.451
      19921125 143100  1266.254     6.842   843.973   467.332
      19921125 143200  1268.073     6.824   845.662   467.341
      19921125 143300  1270.204     6.824   847.344   467.791
      19921125 143400  1272.284     6.838   849.018   468.286
      19921125 143500  1273.947     6.854   850.684   468.389
    
      Section 5: References                                                   
      ---------------------                                                   
                      
      Beavan, J., R. Bilham, D. Emter and G. King  (1979):  Observations  of 
           strain enhancement across a fissure. In:  Bonatz,  M.:  Beitraege 
           zur Erdgezeitenforschung des Arbeitskreises Geodaesie / Geophysik
           der Bundesrepublik Deutschland, Deutsche Geodaetische Kommission,
           Reihe B, Heft Nr. 231, 47-58, Muenchen 1979.
                                                          
      Buellesfeld, F.-J. (1985): Ein Beitrag  zur  harmonischen  Darstellung   
           des   gezeitenerzeugenden   Potentials.   Deutsche   Geodaetische   
           Kommission, Reihe C, Heft Nr. 314, Muenchen 1985.                   
                                                                               
      Cartwright, D.E. and  R.J. Tayler (1971):  New  computations  of  tide   
           generating potential. The  Geophysical  Journal,  vol. 23, no. 1,   
           Oxford 1971.                                                        
                                                                               
      Cartwright, D.E. and A.C. Edden (1973):  Corrected  tables  of   tidal   
           harmonics. The Geophysical Journal, vol. 33, no. 3, Oxford  1973.   
                                                                               
      Chojnicki, T. (1973):  Ein  Verfahren  zur  Erdgezeitenanalyse  in An-   
           lehnung an das Prinzip der kleinsten  Quadrate, Mitteilungen  aus   
           dem Institut fuer Theoretische Geodaesie der   Universitaet  Bonn   
           Nr. 15, Bonn 1973;                                                  
                                                                               
      Dehant, V. (1987): Tidal parameters for an inelastic Earth. Physics of   
           the Earth and Planetary Interiors, 49, 97-116, 1987.                
                                                                               
      Doodson, A.T. (1921): The harmonic development of  the tide generating   
           potential. Proceedings  of  the  Royal  Society,  Series  A  100,   
           306-328, London 1921. Reprint in International Hydrographic Revue   
           vol. 31 no. 1, Monaco 1954.                                         
 
      Hartmann, T. and H.-G. Wenzel (1994a): Catalogue  of  the  earth  tide
           generating potential due to the planets. Bulletin  d'Informations
           Marees Terrestres, vol. 119, 8847-8880, Bruxelles 1994.
                                                             
      Hartmann, T. and H.-G. Wenzel (1994b): The harmonic development of the
           earth tide generating potential due to the direct effect  of  the
           planets. Geophysical Research Letters, vol. 21, 1991-1993, 1994.  
 
      Hartmann, T. and  H.-G.  Wenzel  (1995a):  The  HW95  tidal  potential
           catalogue. Geophysical Research Letters, vol. 22, no. 24,  3553 -
           3556, 1995. 
 
      Hartmann, T. and H.-G. Wenzel (1995b):  Catalogue  HW95  of  the  tide
           generating potential. Bulletin d'Informations Marees  Terrestres,
           vol. 123, 9278 - 9301, Bruxelles 1995. 
                    
      Pertsev, B. (1957): On the calculation of drift curve  in  observation   
           of bodily tides. Bulletin d' Informations Marees Terrestres, no.    
           5, 71-72, Bruxelles 1957.                                           
                                                                               
      Pertsev,  B. (1959): Ob  outchetie  spolzaniya  nulia  pir nabloudenij   
           ouprougikh prilivov, Izv. Akad. Naouk SSR, no. 4, 1959.             
                                                                               
      Richter, B. (1987): Das supraleitende  Gravimeter,  Deutsche Geodaeti-   
           sche Kommission, Reihe C, Nr. 329, Muenchen 1987.                   
                                                                               
      Richter, B. and H.-G. Wenzel (1991): Precise  instrumental  phase  lag   
           determination by the step response  method.  Bulletin  d'Informa-   
           tions Marees Terrestres, vol. 111, 8032-8052, Bruxelles 1991.       
                                                                               
      Roosbeek, F. (1996): RATGP95: A harmonic  development  of  the  tide -
           generating potential  using  an  analytical  method.  Geophysical
           Journal International, vol. 126, 197 - 204, 1966.
  
      Schueller, K. (1976):  Ein Beitrag zur Auswertung  von  Erdgezeitenre-   
           gistrierungen,  Deutsche  Geodaetische  Kommission, Reihe  C  Nr.   
           227, Muenchen 1976;                                                 
                                                                               
      Schueller, K.  (1986):  Simultaneous  tidal  and  multi-channel  input   
           analysis as implemented in  the  HYCON-method.  Proceedings  10th   
           International Symposium on Earth Tides, 515-520, Madrid 1986.       
                                                                               
      Tamura, Y. (1987):  A harmonic development  of  the  tide - generating   
           potential.  Bulletin  d'Informations  Marees  Terrestres  no. 99,   
           6813-6855, Bruxelles 1987.                                          
                                                                               
      Timmen, L. and H.-G. Wenzel (1994): Worldwide synthetic  gravity  tide   
           parameters available on INTERNET. Bulletin d'Informations  Bureau   
           Gravimetrique International, vol. 75, 32-40, Toulouse 1994.         
                                                                               
      Timmen, L. and H.-G. Wenzel (1995): Worldwide synthetic  gravity  tide   
           parameters. Submitted to Proceedings of the Joint Meeting of  the   
           International Gravity  Commission  and  the  International  Geoid   
           Commission, September 11-17, Graz 1994.                             
                                                                               
      Vetter, M. and H.-G. Wenzel (1994): PREGRED - an interactive graphical   
           editor for digitally recorded tidal data. Submitted  to  Bulletin   
           d'Informations Marees Terrestres, Bruxelles 1994.                   
                                         
      Wahr, J.M. (1974): Body tides on anelliptical, rotating,  elastic  and
           oceanless Earth. Geophysical Journal of  the  Royal  Astronomical
           Society, vol. 64, 677 - 703, 1981.           
                                        
      Wenzel, H.-G. (1976a): Zur Genauigkeit von  gravimetrischen  Erdgezei-
           tenbeobachtungen, Wissenschaftliche Arbeiten der Lehrstuehle fuer
           Geodaesie,  Photogrammetrie und  Kartographie an der  Technischen   
           Universitaet Hannover Nr. 67, Hannover 1976.                        
 
      Wenzel,  H.-G.  (1976b):  Some  remarks  to  the  analysis  method  of 
           Chojnicki. Bulletin d'Informations Marees  Terrestres,  vol.  73,
           4187-4191, Bruxelles 1976.
                                                                           
      Wenzel, H.-G. (1977):  Estimation  of  accuracy  for  the  earth  tide
           analysis  results.  Bulletin  d'Informations  Marees  Terrestres,
           vol. 76, 4427 - 4445, Bruxelles 1977.
 
      Wenzel, H.-G. (1993): Tidal data processing on a pc.  Proceedings  XII
           International Symposium on Earth  Tides,  Beijing  1993.  Science
           Press, 235 - 244, Beijing 1995.                             
                                                                           
      Wenzel, H.-G. (1994a): PRETERNA - a preprocessor for digitally  recor-
           ded tidal data. Bulletin d'Informations Marees Terrestres,   vol. 
           118, 8722-8734, Bruxelles 1994.                                
                                                                              
      Wenzel, H.-G. (1994b): Accurate instrumental phase  lag  determination
           for feedback gravimeters. Bulletin  d'Informations Marees Terres-
           tres, vol. 118, 8735-8752, Bruxelles 1994.
                               
      Wenzel, H.-G. (1994c): Gezeitenpotential. Seminar der  Deutschen  Geo-
           physikalischen Geselleschaft, Oberwolfach 17.-21.  Oktober  1994.
           Mitteilungen  der   Deutschen   Geophysikalischen   Gesellschaft,
           Sonderband II/1995, 1 - 18, 1995.
 
      Wenzel, H.-G. (1994d): Erdgezeitenanalyse. Seminar der Deutschen  Geo-
           physikalischen Geselleschaft, Oberwolfach 17.-21.  Oktober  1994.
           Mitteilungen  der   Deutschen   Geophysikalischen   Gesellschaft,
           Sonderband II/1995, 1 - 18, 1995.
 
      Wenzel, H.-G. (1994e): Earth tide data processing package ETERNA 3.20.
           Bulletin d'Informations Marees Terrestres, vol. 120, 9019 - 9022,
           Bruxelles 1994.
                                                                            
      Wenzel, H.-G. (1995): Format and structure for the  exchange  of  high    
           precision tidal  data. Bulletin d'Informations Marees Terrestres, 
           vol. 121, 9097-9101, Bruxelles 1995.                       
 
      Wenzel, H.-G. (1996a): Accuracy assessment for tidal  potential  cata-
           logues. Bulletin  d'Informations  Marees  Terrestres,  vol.  124,
           9394 - 9416, Bruxelles 1996.
                                                                           
      Wenzel,  H.-G.  (1996b):  The  nanogal  software:   Earth   tide  data 
           processing package ETERNA 3.30. Bulletin d'Informations Marees 
           Terrestres, vol. 124, 9425-9439, Bruxelles 1996.
 
      Wenzel, H.-G. (1996c): Zum Stand der  Erdgezeitenanalyse.  Zeitschrift
           fuer Vermessungswesen,  vol. 121,  Heft 6,  242 - 255,  Stuttgart
           1996.
       
      Wenzel, H.-G., H. Otto, W. Zuern and B. Richter (1995):  Comparison of   
           a cryogenic and a spring gravimeter between 0.2 and 96 cpd at BFO   
           Schiltach. Bulletin d'Informations Marees Terrestres,  vol.  122,   
           9163 - 9172, Bruxelles 1995.                                            
 
      Wilhelm, H. and  W. Zuern (1984):  Tidal  forcing  field. In: Landolt-   
           Boernstein, Zahlenwerte und  Funktionen  aus  Naturwissenschaften   
           und Technik, New series, group V, Vol. 2, Geophysics of the Solid   
           Earth, the Moon and the Planets, Berlin 1984.                       
 
      Xi, Q. (1989): A new complete  development  of  the  tide - generating 
           potential for the epoch J2000.0. Bulletin  d'Informations  Marees
           Terrestres, vol. 99, 6766 - 6812, Bruxelles 1989.
                                                                               
      Zschau, J. and R. Wang (1981):  Imperfect  elasticity  in  the Earth's   
           mantle. Implications for Earth tides  and  long  period  deforma-   
           tions. Proceedings of the 9th International  Symposium  on  Earth   
           Tides, New York 1981, pp. 605-629, editor  J.T.  Kuo,  Schweizer-   
           bartsche Verlagsbuchhandlung, Stuttgart 1981.                       
                                                                               
      Zuern, W. and  H. Wilhelm  (1984):  Tides  of  the  solid  Earth.  In:   
           Landolt-Boernstein, Zahlenwerte und  Funktionen  aus Naturwissen-   
           schaften und Technik, New series, group V, Vol. 2, Geophysics  of   
           the Solid Earth, the Moon and the Planets, Berlin 1984.             
                                                                               
      Section 6: List of enclosures
      -----------------------------
 
      001: Gain of filter n20m1s01.nlf for online decimation 1 s / 1 min
      002: Gain of filter n20m5s02.nlf for online decimation 5 s / 1 min
      003: Step response recorded in file st299v81.dat
      004: Differences benchmark series bfde403f - HW95
      005: Histogram of differences bfde403f - HW95
      006: Predicted gravity variation "bfhw9501.prd"
      007: Predicted gravity pole tides "bfhw9501.prd"
      008: Predicted gravity LOD  tides "bfhw9501.prd"
      009: Gain of filter n01s1m01.nlf for decimation  1 s / 1 min 
      010: Gain of filter n02s1m01.nlf for decimation  2 s / 1 min 
      011: Gain of filter n03s1m01.nlf for decimation  3 s / 1 min
      012: Gain of filter n05s1m01.nlf for decimation  5 s / 1 min
      013: Gain of filter n10s1m01.nlf for decimation 10 s / 1 min 
      014: Gain of filter n20s1m01.nlf for decimation 20 s / 1 min  
      015: Gain of filter n30s1m01.nlf for decimation 30 s / 1 min 
      016: Gain of filter n02s5m02.nlf for decimation  2 s / 5 min
      017: Gain of filter n03s5m02.nlf for decimation  3 s / 5 min
      018: Gain of filter n05s5m02.nlf for decimation  5 s / 5 min
      019: Gain of filter n10s5m02.nlf for decimation 10 s / 5 min
      020: Gain of filter n20s5m02.nlf for decimation 20 s / 5 min
      021: Gain of filter n30s5m02.nlf for decimation 30 s / 5 min
      022: Gain of filter n60s5m02.nlf for decimation 60 s / 5 min
      023: Gain of filter n2h1m001.nlf for decimation  1 min /  5 min (old)
      024: Gain of filter n14h5m01.nlf for decimation  5 min / 60 min 
      025: Observed gravity signal kal299tt.tid (1 min)
      026: Detided  gravity signal kal299tt.tid (1 min)
      027: Despiked detided gravity signal kal299tt.cor (1 min)
      028: Despiked gravity signal kal299tt.cor (1 min)
      029: Preprocessed gravity signal kal299tt.fil (5 min)
      030: Preprocessed detided gravity signal kal299tt.fil (5 min)
      031: Observed gravity signal sg102tes.tid (1 min)
      032: Detided  gravity signal sg102tes.tid (1 min)
      033: Detided  gravity signal sg102tes.tid 19940607
      034: Detided and edited gravity signal sg102tes.cor 19940607
      035: Detided gravity signal sg102tes.tid (1 min)
      036: Detided despiked gravity signal sg102tes.des (1 min)
      037: Detided despiked gravity signal sg102tes.des (1 min)
      038: Manually edited detided gravity signal sg102tes.des (1 min)
      039: Despiked gravity signal sg102tes.des (1 min)
      040: Preprocessed detided gravity signal sg102tes.fil (5 min)
      041: Preprocessed gravity signal sg102tes.fil (5 min)
      042: Synthetic gravity signal bfhw9502.tid (1 min)
      043: Detided synthetic gravity signal bfhw9502.tid (1 min)
      044: Despiked detided synthetic gravity signal bfhw9502.cor (1 min)
      045: Preprocessed detided synthetic gravity signal bfhw9502.fil (5 min)
      046: Gain of ANALYZE highpass filter n1h1h001.nlf (Pertzev 1957)
      047: Gain of ANALYZE highpass filter n1h1h002.nlf (Pertzev 1959)
      048: Gain of ANALYZE highpass filter n1h1h003.nlf (Wenzel 143 h)  
      049: Gain of ANALYZE highpass filter n1h1h004.nlf (Wenzel 239 h)
      050: Gain of ANALYZE highpass filter n1h1h005.nlf (HYCON-MC)
      051: Gain of ANALYZE highpass filter n1h1h006.nlf (Wenzel 49 h)
      052: Gain of ANALYZE highpass filter n1h1h007.nlf (Wenzel 145 h)
      053: Gain of ANALYZE highpass filter n1h1h008.nlf (Wenzel 241 h)
      054: Gain of ANALYZE highpass filter n05m05m2.nlf  5 min sampling int.
      055: Gain of ANALYZE highpass filter n10m10m2.nlf 10 min sampling int.
      056: Gain of ANALYZE highpass filter n20m20m2.nlf 20 min sampling int.
      057: Gain of ANALYZE highpass filter n30m30m2.nlf 30 min sampling int.
      058: Gain of ANALYZE highpass filter n60m60m2.nlf 60 min sampling int.
      059: Plot of observed gravity tide data hal29901.dat
      060: Analysis results for file          hal29901.dat
      061: Residuals after analysis           hal29901.res
      062: Histogram of residuals             hal29901.res
      063: Amplitude spectrum of residuals    hal29901.far
      064: Plot of observed gravity tide data bfl24903.dat  
      065: Analysis results for file          bfl24903.dat
      066: Residuals after analysis           bfl24903.res
      067: Histogram of residuals             bfl24903.res
      068: Amplitude spectrum of residuals    bfl24903.far
      069: Plot of observed strain tide data  bfd00801.dat  
      070: Analysis results for file          bfd00801.dat
      071: Residuals after analysis           bfd00801.res
      072: Histogram of residuals             bfd00801.res
      073: Amplitude spectrum of residuals    bfd00801.far
      074: Plot of observed gravity tide data bfet1907.dat  
      075: Analysis results for file          bfet1907.dat
      076: Residuals after analysis           bfet1907.res
      077: Histogram of residuals             bfet1907.res
      078: Amplitude spectrum of residuals    bfet1907.far
      079: Plot of observed gravity tide data bhtt4003.dat  
      080: Analysis results for file          bhtt4003.dat
      081: Residuals after analysis           bhtt4003.res
      082: Histogram of residuals             bhtt4003.res
      083: Amplitude spectrum of residuals    bhtt4003.far
      084: Plot of synthetic gravity tides    bfhw9501.dat
      085: Analysis results for file          bfhw9501.dat
      086: Residuals after analysis           bfhw9501.res
      087: Histogram of residuals             bfhw9501.res
      088: Amplitude spectrum of residuals    bfhw9501.far 
      089: Plot of benchmark gravity tides    bfde403f.dat
      090: Analysis results for file          bfde403f.dat
      091: Residuals after analysis           bfde403f.res
      092: Histogram of residuals             bfde403f.res
      093: Amplitude spectrum of residuals    bfde403f.far
      094: Printout of program WPAREX 
 
      ****** This ends the startup section of the ETERNA 3.40 manual.
      ****** The rest of the manual is available on thd CD-rom only.
 
      Section 7: Description of program RECTIDE                              
      -----------------------------------------                              
                                                                               
      The program RECTIDE is a pc-based earth tide data acquisition  program   
      for gravimeters, which enables simultaneously  a  high  sampling  rate   
      and high precision data recording and the  graphical  display  of  the   
      recorded data, reduced for computed model tides. The  source  code  of 
      the program is given in file "\eterna33\rectid\rectid.bas",  the  exe-
      cutable program is given in file "\eterna33\rectid\rectid.exe".
       
      The data sampling and the model tide computation is carried out at 1 s
      interval using one of four different available tidal  potential  cata-
      logs (Doodson 1921, Cartwright-Tayler-Edden  1973,  Buellesfeld  1985, 
      Tamura 1987). Because the computed model tides are only used to online
      display the difference between the observation and the model tide,  we
      dont need the highest accuracy for the model tides (i.e. we dont  need
      the Hartmann and Wenzel 1995 catalogue). With RECTIDE it  is  possible 
      to display on-line the reduced data with a very high resolution of  up
      to 1 nm/s**2 per cm on the screen of the pc (depending  on  the  short
      periodic noise level). The 1 s samples are numericall filtered on-line
      by a symmetric FIR (finite impulse  response) lowpass  filter  with  5 
      min cutoff period and decimated to 1 min interval. The data  decimated
      to 1 min interval are stored on the hard disk of the pc.  The  program 
      can also be used to record data from other sensors, e.g. tiltmeters.
                                                                              
      Section 7.1: Hardware requirements                                       
      ----------------------------------                                       
                                                                               
      The hardware you need to record earth tide data with  program  RECTIDE  
      consists of                                                              
                                                                               
      - a personal computer with a built-in AD-converter card,                 
      - a clock providing a 1 pulse per second TTL signal.                     
                                                                               
      The current version of the RECTID earth tide data acquisition  program   
      with a sampling interval of 1 s needs a high  speed  80486 66  MHz  or   
      PENTIUM P60 processor, a EGA or VGA color screen, and a hard  disk  of   
      about 200 MByte capacity. Priciply, the program can be  executed  (and   
      has been with the BFO-TLD-1 and BFO-TLD-2 data acquisition system  for   
      several years) on an 80286/287 pc, with a larger sampling interval  of   
      e.g. 5 s. To increase the sampling interval to 5 s, a clock with a 5 s   
      TTL pulse must be available, and some modifications have to be made in   
      program RECTID. For this purpose, the lowpass filter N20M5S02 for 5  s   
      sampling interval is still available with file N20M5S02.NLF.             
                                                                              
      The currently used AD-converter card RTI870 built  by  Analog  Devices   
      has a capacity of 4 channels and a resolution of 22 bits. The  AD-con-   
      verter may be purchased by                                               
                                                                               
         Analog Devices                                                        
         One Technology Way                                                    
         P.O. Box 9106                                                         
         Norwood, MA 02062-9106, U.S.A.                                        
         Fax: ++1-617-326-8703                                                 
                                                                               
      If you want to use a different AD-converter card, you have  to  modify   
      the routines RTI870TE and RTI870AD within program RECTID and you  have   
      to re-compile program RECTID.                                            
                                                                               
      The triggering of the AD-conversion has to be provided by a 1 pps  TTL   
      clock pulse (between 0 and +5 V) at digital input terminal D1  of  the   
      AD-converter. The AD-conversion is triggered by  the  downgoing  pulse   
      flank, which should be  synchronized  to  Universal  Time  Coordinated   
      (UTC).                                                                   
                                                                               
      Section 7.2: Program description                                         
      --------------------------------                                         
                                                                               
      The data recording is done with  program RECTIDE.BAS  under  operating   
      system MS-DOS 5.0 using Microsoft Quick Basic  4.0. The  raw  sampling   
      interval is 1 s,  using  the  1  s  TTL  pulse  of  the  PZF535  radio   
      controlled clock at digital input channel 1 of the RTI870 AD-converter   
      card to trigger the analog to digital conversion of up to 4  channels.   
      The range of the analog input channels is set to +/- 5  V  within  the   
      program, the resolution of the analog input channels is set to 22  bit   
      which corresponds to about 2 microvolt.                                  
                                                                               
      With program RECTIDE, the graphic screen and the text screen is  avai-
      lable simultaneously. The program is usually switched to  the  graphic   
      screen, but may be switched by the user to the text screen by entering   
                                                                               
          <text>                                                               
                                                                               
      from keyboard during the operation of the program. You have  available   
      the on-line option to print the raw data or the filtered data  to  the   
      text screen. For a description of  the  available  keyboard  commands,   
      hit function key no. 1 to print the help-screen:                         
                                                                               
                                                                               
                 Help-screen of Program RECTIDE                                
                 ------------------------------                                
                                                                               
                 to print a list of data files enter    <files>                
                 to print raw 1 s samples enter         <praw>                 
                 to print filtered 1 min samples enter  <pfil>                 
                 to switch to graphic screen enter      <graph>                
                 to switch to text    screen enter      <text>                 
                 to increase clock reading enter         <+>                   
                 to decrease clock reading enter         <->                   
                 to stop the program enter              <end>                  
                                                                               
      After reading the help-screen, you have to enter                         
                                                                               
              <graph>                                                          
                                                                               
      to continue with the graphic screen.                                     
                                                                               
      The data sampled at 1 s interval  are  smoothed  by  taking  a  moving   
      average over 60 s and the smoothed  data  are  displayed  on  the  VGA   
      colour graphic screen with a horizontal extension of 1 day (ticks have   
      an x-spacing of 1 hour). The screen starts at 00:00:00  UTC.  For  the   
      gravimeter channels, the model  tide  of  the  station  computed  with   
      routine TIDE5 (one of four different tidal potential catalogs  may  be  
      chosen by the user with observed/predicted tidal parameters)  is  sub-   
      tracted. The computed model tide is displayed in red  on  the  graphic   
      screen with +/- 1250 nm/s**2 vertical range. The  residual  signal  of   
      the gravimeters (recorded signal minus computed model  tide)  is  dis-   
      played with  a  user-defined  vertical  range;  we  recommend  +/- 100   
      nm/s**2 vertical range. The recorded  barometer  signal  is  displayed   
      on the screen with a user-defined vertical range; we recommend  +/- 10   
      hPa vertical range. In the plot program, a screen stepper  takes  care   
      that the recorded or residual signal will always  be  visible  on  the   
      screen of the pc.                                                        
                                                                               
      The data sampled at 1 s  interval  are  numerically  lowpass  filtered   
      using a symmetrical FIR lowpass filter with zero phase shift with 1225   
      coefficients and a length of about 20 minutes.  The  filter  has  been   
      supplied by Dr. J. Neumeyer from GFZ Potsdam. The filter  coefficients   
      are defined with file "n20m1s01.nlf" and  the  filter  gain  is  given 
      below. Note that the filter gain can be plotted with program PLOTFILT.
      The filter gives a damping of more than  78  dB  for  periods  shorter 
      than 120 s. For periods larger than 1 hour, the gain of the filter  is
      1 with a maximum deviation of 1.4*10**-6.
                                                                               
      The gain of numerical filter "n20m1s01.nlf" (see encl. 1) is:
                                                                               
       Frequ.    Gain     Frequ.    Gain      Frequ.    Gain                   
       [cps]     [dB]     [cps]     [dB]      [cps]     [dB]                   
                                                                               
       0.002      0.00     0.02  -121.64       0.22   -141.08                  
       0.003     -0.02     0.03  -124.98       0.23   -144.53                  
       0.004     -0.59     0.04  -137.90       0.24   -170.16                  
       0.005     -3.95     0.05  -135.68       0.25   -144.22                  
       0.006    -13.29     0.06  -132.34       0.26   -141.60                  
       0.007    -32.73     0.07  -136.92       0.27   -144.77                  
       0.008    -77.88     0.08  -150.08       0.28   -173.40                  
       0.009   -133.53     0.09  -135.36       0.29   -145.91                  
       0.010   -118.68     0.10  -133.68       0.30   -142.99                  
       0.011   -113.08     0.11  -137.24       0.31   -146.26                  
       0.012   -118.43     0.12  -157.27       0.32   -176.47                  
       0.013   -121.07     0.13  -140.71       0.33   -146.08                  
       0.014   -119.58     0.14  -137.92       0.34   -143.30                  
       0.015   -130.61     0.15  -141.68       0.35   -146.41                  
       0.016   -121.61     0.16  -162.46       0.36   -179.71                  
       0.017   -123.83     0.17  -141.12       0.37   -147.06                  
       0.018   -124.32     0.18  -138.77       0.38   -144.09                  
       0.019   -129.52     0.19  -142.05       0.39   -147.24                  
                           0.20  -166.62       0.40   -183.30                  
                           0.21  -143.95       0.41   -147.15                  
                                                                               
      After numerical filtering, the data are resampled at 1 min  resampling   
      interval. The data resampled at 1 min interval  are  recorded  on  the   
      hard disk in ETERNA standard format with 1 microvolt resolution.       
                                                                               
      Section 7.3: Description of the project file "project"
      ------------------------------------------------------
 
      The ASCII project name file "project"  for  program  RECTIDE  contains
      one single variable only, the project name  which  is  abbreviated  as
      "pn" in this manual. The project  name  is  an  alphanumerical  string 
      consisting of 8 characters at maximum. The project name is  read  from 
      the first 8 characters of  the  first  record  of  project  name  file 
      "project". The project name is used to define  the  control  parameter 
      file name "pn".ini for program RECTIDE. 
 
      Section 7.4: Description of control parameter file "pn".ini    
      -----------------------------------------------------------    
                                                                               
      The control parameter file "pn".ini (where "pn" stands for the project  
      name defined in file "project")  defines  several  control  parameters 
      necessary for the operation of program RECTIDE. This control parameter
      file has an identical structure to the control parameter files used by
      programs ANALYZE, DETIDE, DESPIKE, DECIMATE, PREGRED and PREDICT. Some
      of the variables are identical to  those  used  by  programs  ANALYZE, 
      DETIDE, DESPIKE,  DECIMATE,  PREGRED  and  PREDICT.  Thus  it  may  be 
      possible to use the same control parameter file  for  data  recording, 
      data preprocessing and data analysis.                      
                                                                               
      All control parameter statements must begin with the control parameter   
      name (column 1...10) which ends with an equal sign. The control parame-  
      ter names must be spelled exactly with uppercase  letters  as  defined   
      below. Control parameters  with  unknown  names  will  be  ignored  by   
      program RECTIDE. The control parameter variables must be  input  after   
      the equal sign following the control parameter name, with  appropriate   
      format. The control parameter statements  must  not  have  a  specific   
      sequence. A record starting with # in column 1 will be ignored and may   
      be used to comment the control parameters.  All  characters  following   
      # in a record will be ignored and may be used to comment  the  control   
      parameters.                                                              
                                                                               
      STATIONAME= station name, 10 characters (col. 12...21)  following  the   
                  equal sign. The station name  will  be  displayed  on  the   
                  graphics screen during data recording.                       
      STATLATITU= ellipsoidal latitude of the station in degree referring to   
                  WGS84 reference system (col.12...26).                        
      STATLONITU= ellipsoidal longitude of the station in  degree  referring   
                  to WGS84 reference system (col.12...26).                     
      STATELEVAT= ellipsoidal height of the station in meter referring to      
                  WGS84 reference system (col.12...26).                        
      STATGRAVIT= gravity of the station in m/s**2, necessary for tidal tilt   
                  only (col. 12...26,  real  with  decimal  point).  If  the   
                  gravity is unknown, use a value of less than 1.0  and  the   
                  program will  compute  and  subsequently  use  the  normal   
                  gravity value referring to GRS80 reference system.           
      STATAZIMUT= azimuth of the  instrument  in  degree  decimal,  reckoned   
                  clockwise from north (col. 12 ... 26,  real  with  decimal   
                  point). This parameter is used for tidal tilt,  horizontal   
                  displacement and horizontal strain only.                     
      TIDALPOTEN= parameter for the tidal potential catalog to be used.        
                  1 = Doodson (1921) tidal potential catalog,                  
                  2 = Cartwright-Tayler-Edden (1973) tidal potential catalog   
                  3 = Buellesfeld (1985) tidal potential catalog.              
                  4 = Tamura (1987) tidal potential catalog,                   
      TIDALCOMPO= defines component of model tides which will be computed.     
                  0 = gravity tides in nm/s**2,                                
                  1 = tilt tides in mas in azimuth STATAZIMUT.                 
      RECTIDNUMC= number of analog input channels which will be recorded.      
                  The number of analog channels is restricted to 4.            
      RECTIDSIMU= simulation parameter. For RECTIDSIMU=0 signals from the      
                  analog input channels will be recorded. For RECTIDSIMU=1,    
                  a program test with simulated data (computed model tides)    
                  will be carried out.                                         
      RECTIDLOWP= name of the data file, from which the numerical lowpass      
                  filter for the decimation of the recorded data will be       
                  read. RECTIDLOWP is restricted to 12 characters.             
      RECTIDCHAN= definition of analog input channels. For each analog input   
                  channel, the following variables have to be input:           
                  Col.12..14: IC%  = analog input channel number               
                  Col.15..17: ICOL%= screen plot color for the channel         
                               0 = black               1 = dark blue           
                               2 = green               3 = light blue          
                               4 = dark red            5 = dark pink           
                               6 = brown               7 = light grey          
                               8 = dark grey           9 = bright dark blue    
                              10 = bright green       11 = bright light blue   
                              12 = bright red         13 = bright pink         
                              14 = yellow             15 = bright white        
                  Col.18..20: IRED%= 0 for plot of calibrated signal,          
                              IRED%= 1 for plot of calibrated  signal  minus   
                              computed model tides (residual signal).          
                  Col.21..30: CAL#= scale factor for calibration of recorded   
                               signal before being plotted.                    
                  Col.31..40: SR#= Range of screen plot of calibrated signal   
                              or residual signal. Plot range is +/-SR#.        
      TIDALPARAM= wave group selection for a priori model tide  computation.   
                  The number of wavegroups is restricted to  85.  The  tidal   
                  parameters may be taken from an  earth  tide  analysis  of   
                  another data set in the same station,  or  from  a  neigh-   
                  bouring station, or from synthetic gravity tide parameters   
                  (Timmen and Wenzel 1994, 1995), or from a guess only.        
                  Col.12..21: DFRA#=  start frequency of the wave  group  in   
                                      cpd (real).                              
                  Col.22..31: DFRE#=  end   frequency of the wave  group  in   
                                      cpd (real).                              
                  Col.32..41: DGAM0#= amplitude factor  of  the  wave  group   
                                      (real).                                  
                  Col.42..51: DPHI0#= phase lead of the wave  group  in  deg   
                                      (real).                                  
                                                                               
      The wave group selection for model tide  computation  should  be  made   
      according to the wave group selection for the analysis of  Earth  tide   
      data (see the program manual for ANALYZE); the list given below  is  a
      proposal only). The frequencies are given in cycles per day (cpd).      
                                                                               
      group        > 1 month            > 6 months          > 1 year           
                   from      to         from      to        from      to       
                  [cpd]   [cpd]        [cpd]   [cpd]       [cpd]   [cpd]       
                                                                               
      SA             -       -            -       -       0.001379  0.004107   
      SSA            -       -      0.004108  0.020884    0.004108  0.020884   
      MM      0.020885  0.054747    0.020885  0.054747    0.020885  0.054747
      MF      0.054748  0.091348    0.054748  0.091348    0.054748  0.091348
      MTM     0.091349  0.501369    0.091349  0.501369    0.091349  0.501369   
      Q1      0.501370  0.911390    0.501370  0.911390    0.501370  0.911390   
      O1      0.911391  0.947991    0.911391  0.947991    0.911391  0.947991   
      M1      0.947992  0.981854    0.947992  0.981854    0.947992  0.981854
      P1             -       -      0.981855  0.998631    0.981855  0.998631   
      S1             -       -            -       -       0.998632  1.001369   
      K1      0.981855  1.023622    0.998632  1.023622    1.001370  1.004107   
      PSI1           -       -            -       -       1.004108  1.006845   
      PHI1           -       -            -       -       1.006846  1.023622   
      J1      1.023623  1.057485    1.023623  1.057485    1.023623  1.057485   
      OO1     1.057486  1.470243    1.057486  1.470243    1.057486  1.470243   
      2N2     1.470244  1.880264    1.470244  1.880264    1.470244  1.880264   
      N2      1.880265  1.914128    1.880265  1.914128    1.880265  1.914128   
      M2      1.914129  1.950419    1.914129  1.950419    1.914129  1.950419   
      L2      1.950420  1.984282    1.950420  1.984282    1.950420  1.984282   
      S2      1.984283  2.451943    1.984283  2.002736    1.984283  2.002736  
      K2             -       -      2.002737  2.451943    2.002737  2.451943   
      M3M6    2.451944  7.000000    2.451944  7.000000    2.451944  7.000000
                                                                               
      In the following we show the control parameter file kal68601.ini given
      on the CD-rom. For your actual recording setup, you should modify  the 
      parameters according to your instruments and to your station!       
 
      # this is file KAL68601.INI version 1996.08.06 for program RECTID 3.40
 
      STATIONAME= KARLSRUHE      #stations name
      STATLATITU=  49.0121       #stations latitude  in degree
      STATLONITU=   8.4126       #stations longitude in degree
      STATELEVAT=  114.000       #stations ellipsoidal height in m
      TIDALPOTEN=  4             #Tamura (1987) tidal potential catalogue
      TIDALCOMPO=  0             #tidal component
      RECTIDNUMC=  4             #number of channels for RECTID
      RECTIDSIMU=  0             #simulation without A/D conversion if = 1 
      RECTIDLOWP=N20M1S01.NLF    #numerical lowpass filter
 
      # plot colors: 0 = black                     1 = dark blue
      #              2 = green                     3 = light blue
      #              4 = dark red                  5 = dark pink
      #              6 = brown                     7 = light grey
      #              8 = dark grey                 9 = bright dark blue
      #             10 = bright green             11 = bright light blue 
      #             12 = bright red #             13 = bright pink
      #             14 = yellow                   15 = bright white
 
      # next lines RECDTIDCHAN: col. 12..14: channel number
      #                              15..17: plot color
      #                              18..20: 1 for reduction of model tides
      #                              21..30: scale factor
      #                              31..40: plot range 
 
      RECTIDCHAN=  1 14  1    4785.4    100.00             #gravimeter  
      RECTIDCHAN=  2 11  0     10.00     10.00             #barometer  
      RECTIDCHAN=  3 10  0   1000.00      2.00             #beam position
      RECTIDCHAN=  4  9  0     10.00      2.00             #voltage standard
 
      TIDALPARAM=  0.000000  0.249951   1.15000    0.0000 LONG    
      TIDALPARAM=  0.721500  0.906315   1.13980   -0.1700 Q1       
      TIDALPARAM=  0.921941  0.940487   1.14790    0.1560 O1      
      TIDALPARAM=  0.958085  0.974188   1.14340    0.5290 M1      
      TIDALPARAM=  0.989049  1.011099   1.13520    0.2560 P1K1    
      TIDALPARAM=  1.013689  1.044800   1.14960   -1.6760 J1      
      TIDALPARAM=  1.064841  1.216397   1.14560    0.8600 OO1     
      TIDALPARAM=  1.719381  1.872142   1.15850    4.4470 2N2     
      TIDALPARAM=  1.888387  1.906462   1.17330    3.0360 N2      
      TIDALPARAM=  1.923766  1.942754   1.18100    1.9920 M2      
      TIDALPARAM=  1.958233  1.976926   1.18990    1.5590 L2      
      TIDALPARAM=  1.991787  2.182843   1.18440    0.6590 S2K2    
      TIDALPARAM=  2.753244  3.081254   1.05680    0.9260 M3      
      TIDALPARAM=  3.791964  3.937897   1.02000    0.0000 M4      
 
      Section 7.5: Description of output files                                 
      -----------------------------------------                                
                                                                               
      The data output files of  program  RECTIDE  use  the  ETERNA  standard
      format for exchange of  high  resolution  and  high  rate  tidal  data   
      (Wenzel 1995). The format of these files is                              
                                                                               
      Col.01...08: date in year, month, day; e.g. 900328  means  year  1990,   
                   month 03 = March, day 28th.                                 
      Col.10...15: time in hour, minute, second,  given  in  Universal  Time   
                   Coordinated (UTC); e.g. 231200 means 23 hours 12  minutes   
                   00 seconds. The seconds  are assumed to be exact, e.g. 00   
                   seconds means 0.000 seconds.                                
      Col.16...25: Sample of channel 1 in Volt.                                
      Col.26...35: Sample of channel 2 in Volt.                                
      Col.36...45: Sample of channel 3 in Volt.                                
      Col.46...55: Sample of channel 4 in Volt.                                
        ...
 
      As an example, a part of the output  file  96081500.10  recorded  with  
      program RECTIDE at  station  Karlsruhe  August  15th,  1996  (filtered 
      feedback output of gravimeter LCR-G686 at channel 1, barometer BS05 at
      channel 2, beam position of gravimeter  LCR-G686  at  channel  3,  4 V 
      voltage reference at channel  4)  using  the  control  parameter  file 
      "kal68601.ini"given on the CD-rom are shown below:                 
                                                                               
      File 96981500.10:                                                        
                                                                               
      Date     Time    Channel1  Channel2  Channel3                            
      
 
      19960814      0 -0.759007  0.078377 -0.000543 -4.000112
      19960815    100 -0.759616  0.078571 -0.000554 -4.000112
      19960815    200 -0.759109  0.078542 -0.000581 -4.000112
      19960815    300 -0.758253  0.078165 -0.000581 -4.000112
      19960815    400 -0.758292  0.077714 -0.000555 -4.000112
      19960815    500 -0.758696  0.077609 -0.000553 -4.000111
      19960815    600 -0.758454  0.077102 -0.000572 -4.000111
      19960815    700 -0.757854  0.075946 -0.000579 -4.000111
      19960815    800 -0.757493  0.075523 -0.000571 -4.000112
      19960815    900 -0.757573  0.075894 -0.000558 -4.000112
      19960815   1000 -0.757887  0.075801 -0.000558 -4.000112
      19960815   1100 -0.757769  0.075235 -0.000572 -4.000111
      19960815   1200 -0.757349  0.074995 -0.000569 -4.000111
      19960815   1300 -0.757452  0.074869 -0.000555 -4.000111
      19960815   1400 -0.757757  0.074162 -0.000557 -4.000111
      19960815   1500 -0.757678  0.073639 -0.000564 -4.000111
      19960815   1600 -0.757610  0.074012 -0.000560 -4.000111
      19960815   1700 -0.757729  0.074505 -0.000557 -4.000111
      19960815   1800 -0.757810  0.074733 -0.000561 -4.000112
      19960815   1900 -0.757751  0.075035 -0.000567 -4.000113
      19960815   2000 -0.757467  0.075400 -0.000570 -4.000113
      19960815   2100 -0.757117  0.075520 -0.000566 -4.000113
      19960815   2200 -0.757027  0.075070 -0.000562 -4.000114
      19960815   2300 -0.757072  0.074139 -0.000563 -4.000115
      19960815   2400 -0.756957  0.073641 -0.000564 -4.000115
      19960815   2500 -0.756690  0.073838 -0.000567 -4.000115
      19960815   2600 -0.756294  0.073682 -0.000572 -4.000114
      19960815   2700 -0.756127  0.072732 -0.000566 -4.000114
      19960815   2800 -0.756247  0.071956 -0.000565 -4.000113
      19960815   2900 -0.755843  0.072568 -0.000583 -4.000112
      19960815   3000 -0.755032  0.074007 -0.000585 -4.000112
 
      Section 7.6:  Program tests                                               
      ---------------------------                                                
                                                                                
      Previous versions of program RECTIDE have  been  used  since  1990  to    
      record gravity tides with the transportable  data  acquisition  system    
      BFO-TLD-1 (80286/287 pc with Strawberry Tree Inc. 16 bit/8 channel AD-    
      converter card in stations Karlsruhe, Potsdam, Hannover, Berggiesshue-    
      bel and Bonn. There is built into the program the test variable ISIM%,    
      which can be enabled by using RECTIDSIMU= 1 in the  control  parameter    
      file "pn".INI. This internal program test has been used intensively to    
      check and to increase the performance of program RECTIDE.                  
                                                                                
      Long term program tests with program RECTIDE version 3.00 ... 3.40 has    
      been performed with data acquisition BFO-TLD-3 (80486 66 MHz  pc  with    
      RTI870 22 bit/4 channel ADC-card) using feedback gravimeter  LCR-G686F    
      and an electronic barometers between August 8th and 29th, 1996. 
 
      Section 8:   Description of program ETSTEP                               
      ------------------------------------------                               
                                                                               
      The program ETSTEP computes  the  frequency  transfer  function  of  a   
      linear dynamic causal  system  with  constant  parameters  by  Fourier   
      transform of differentiated step response (e.g.  Wenzel 1976,  Richter   
      and Wenzel 1991, Wenzel 1994).                                           
                                                                               
      The recommended method to carry out a step response experiment  is  to   
      input an electronic step into a parallel  adder  to  the  gravimeter's   
      feedbeck (e.g. Wenzel 1994) and to record the gravimeter's output with   
      a high resolution and high rate data acquisition. The  time  given  in   
      the sample files has to be counted in seconds from the step.             
                                                                               
      The principal method of program ETSTEP is that the  sample  data  (not   
      necessarily given at equidistant instants) are normalized to  a  unity   
      step and subsequently used to fit a 2nd  degree  least  squares  poly-   
      nomial to the data collected within a certain  time  window  from  the   
      complete set of sample data. The least squares  polynomial  is  subse-   
      quently used to differentiate the recorded step response.  The  diffe-   
      rentiated step response is finally transformed into  frequency  domain   
      by Discrete Fourier Transform (DFT), and the complex transfer function   
      of the gravimeter is given by the real and imaginary part of the  DFT.   
      The system's gain is given by the  SQRT(real**2 +  imag**2),  and  the   
      system's phase lag is given by ATAN(imag/real).                          
                                                                               
      The recorded step response may be visualized and plotted on paper with
      program PLOTDATA (see encl.003). Remember that program ETSTEP does not
      subtract the tides from the recorded step response. The tides  contai-
      ned always in signals recorded with earth tide instruments can  signi-
      ficantly influence the step response evaluation, if the  tidal  signal
      during the step response experiment is significant. However,  you  can
      remove to a large part the tidal signal contained in the recorded step
      response data using program DETIDE, provided you  know  the  necessary
      parameters.
      
      To evaluate the step response, you have to record  the  step  response  
      with high rate (e.g. 1 s) using a suitable data acquisition in  ETERNA 
      standard format on a file. The samples are assumed to  be  equidistant
      at exact full seconds. The program ETSTEP reads the necessary  control 
      parameters from a file called "etstep.ini". This file must be  located
      in the same directory as file "etstep.exe" and the file containing the
      recorded step response. 
  
      All control parameter statements given in file "etstep.ini" must begin
      with the control parameter name (column 1...10)  which  ends  with  an 
      equal sign. The control parameter names must be spelled  exactly  with 
      uppercase letters as defined below. Control  parameters  with  unknown
      names will be ignored by program ETSTEP. The control  parameters  must  
      be input after the equal sign following the  control  parameter  name, 
      with appropriate format. The control  parameter  statements  must  not 
      have a specific sequence. A record starting with # in column 1 will be
      ignored and may  be  used  to  comment  the  control  parameters.  All 
      characters following # in a record will be ignored and may be used  to
      comment the control parameters. The control parameter file for program
      ETSTEP has the same structure as other control parameter files for the
      ETERNA 3.40 package.
                                                              
      The following control parameters have to be  defined  in  the  control 
      parameter file "etstep.ini":
 
      DATFILNAME= alphanumeric name of the file, in which the recorded  step
                  response is given.
      STEPCHANIS= Integer channel number at which the step response has been 
                  recorded. Remember that the  data  acquisition may  record
                  parameters in more than one channel.
      STARTTIMIS= Integer start time of the data which will be used for  the 
                  step  response   evaluation  in  hours,  minutes,  seconds 
                  (hhmmss). E.g. 162000  means 16 hours,  20 min, 0 sec.
      STEPTIMEIS= Integer time at which the input step  to  the  system  has 
                  been generated, in hours, minutes, seconds  (hhmmss). E.G.
                  162000 means 16 hours, 20 minutes, 0 seconds.
      STOPTIMEIS= Integer stop time of the data which will be used  for  the
                  step response evaluation in hours, minutes,  seconds. E.g.
                  162500 means 16 hours, 25 minutes, 0 seconds.
      STEPFITINT= Intervall in seconds, at which the step response  will  be
                  fitted by the polynomial  adjustments.  The  fit  interval
                  should be in the order of the average sampling interval.
      STEPCOLRAD= Collection radius in seconds for the data, which  will  be
                  used around the fit epoch to adjust  the  coefficients  of
                  the polynomial. This interval should be 2 to 3  times  the
                  the average sampling interval.
      We list below as an example the control parameter file "etstep.ini"
      provided on the CD-rom:
 
      # this is file etstep.ini version 1996.08.13 containing control 
      # parameters for program  ETSTEP
 
      DATFILNAME=st299v81.dat    # input data file name
      STEPFITINT=   0.500        # fit interval in secs
      STEPCOLRAD=   2.500        # collection radius in sec
      STARTTIMIS=  162000        # start time of the data set in hh mm ss  
      STEPTIMEIS=  162000        # step  time                 in hh mm ss
      STOPTIMEIS=  162500        # end   time of the data set in hh mm ss
      STEPCHANIS=       1        # data channel at which the step was recorded
 
      # end of file etstep.ini
 
      The data file to be used for the step response evaluation  (e.g.  file
      "st299v81.dat") has to  be  defined  in  the  control  parameter  file 
      "estep.ini" and to be provided in the  directory  from  which  program
      ETSTEP will be executed. We give  in  the  following  the  first  part
      of file "st299v81.dat" as example. The step response  data  file  must
      contain a file header, which gives an alphanumeric description of  the 
      file contents, and a file body, which must contain the data. The  file 
      header ends with  'C*********'  in  columns 1 ... 10.  The  file  body 
      contains the date, time  and  recorded  channels  in  standard  ETERNA 
      format. The file body must end with 99999999 in columns 1 ... 8.
                                                                               
      File: ST299V81.DAT           Status: 1996.08.14
 
      Step response of gravimeter LCR-G299-SW02 feedback,
      step by electronic step into parallel adder, computer controlled
      by BFO-TLD1 data acquisition, program STEPRES.BAS. 
      !!! Modified version of program STEPRES.BAS, feedback voltage at ADC
      !!! channels 4 and 5.
 
      Re-sampled at full second from original file ST299V80.DAT, 
      which was sampled  at irregularly intervals opf about 0.3 s.
 
      Calibration function of the gravimeter is 
                  linear calibration factor:    51902.71 nm/s**2 per V
                  quadratic calibration factor:    36.39 nm/s**2 per V**2
                  cubic calibration factor:        -0.22 nm/s**2 per V**3 
 
      Step at 930329 16:20:00 UTC, 20. experiment
      C***********************************************************************
      77777777         0.000000
      19930329 162000  0.062601
      19930329 162001  0.062568
      19930329 162002  0.062575
      19930329 162003  0.062578
      19930329 162004  0.062476
      19930329 162005  0.062122
      19930329 162006  0.060227
      19930329 162007  0.055921
      19930329 162008  0.045564
      19930329 162009  0.027073
      19930329 162010 -0.004299
      19930329 162011 -0.050774
      19930329 162012 -0.114854
      19930329 162013 -0.196773
      19930329 162014 -0.294235
      19930329 162015 -0.404705
      19930329 162016 -0.521605
      19930329 162017 -0.640638
      19930329 162018 -0.753298
      19930329 162019 -0.857758
      19930329 162020 -0.948130
      19930329 162021 -1.022431
      19930329 162022 -1.080996
      19930329 162023 -1.123435
      19930329 162024 -1.153067
      19930329 162025 -1.171148
      19930329 162026 -1.180805
      19930329 162027 -1.185414
      19930329 162028 -1.186197
      19930329 162029 -1.185607
      19930329 162030 -1.184291
         .
         .
      The  computed  gain,  phase  lag and time  lag  of  the  system  under 
      consideration is provided in file "etstep.prn." We give below  a  part
      of the printout for sample file "st299v81.dat" provided on the CD-rom:
                                                                               
      ####################################################
      # Program ETSTEP, gain and phase lag of the system #
      # Normalization factor:  0.999981                  #
      ####################################################
 
      frequency  real part imag.part      gain     phase  time lag
           [cps]                                [degree]       [s]
 
 
        0.000001  1.000000  0.000103  1.000000     0.006    16.381
        0.000002  1.000000  0.000206  1.000000     0.012    16.381
        0.000003  1.000000  0.000309  1.000000     0.018    16.381
        0.000004  1.000000  0.000412  1.000000     0.024    16.381
        0.000005  1.000000  0.000515  1.000000     0.029    16.381
        0.000006  1.000000  0.000618  1.000000     0.035    16.381
        0.000007  1.000000  0.000720  1.000000     0.041    16.381
        0.000008  1.000000  0.000823  1.000000     0.047    16.381
        0.000009  1.000000  0.000926  1.000000     0.053    16.381
        0.000010  0.999999  0.001029  1.000000     0.059    16.381
        0.000020  0.999998  0.002058  1.000000     0.118    16.381
        0.000030  0.999995  0.003088  1.000000     0.177    16.381
        0.000040  0.999991  0.004117  0.999999     0.236    16.381
        0.000050  0.999986  0.005146  0.999999     0.295    16.381
        0.000060  0.999980  0.006175  0.999999     0.354    16.381
        0.000070  0.999972  0.007205  0.999998     0.413    16.381
        0.000080  0.999964  0.008234  0.999998     0.472    16.381
        0.000090  0.999954  0.009263  0.999997     0.531    16.381
        0.000100  0.999943  0.010292  0.999996     0.590    16.381
        0.000200  0.999773  0.020583  0.999985     1.179    16.381
        0.000300  0.999490  0.030871  0.999967     1.769    16.381
        0.000400  0.999093  0.041155  0.999941     2.359    16.381
        0.000500  0.998584  0.051433  0.999908     2.948    16.380
        0.000600  0.997961  0.061705  0.999867     3.538    16.380
        0.000700  0.997226  0.071968  0.999819     4.128    16.380
        0.000800  0.996377  0.082221  0.999764     4.717    16.380
        0.000900  0.995417  0.092464  0.999702     5.307    16.380
        0.001000  0.994344  0.102694  0.999633     5.897    16.379
        0.002000  0.977492  0.204045  0.998561    11.791    16.376
        0.003000  0.949717  0.302793  0.996818    17.684    16.374
        0.004000  0.911401  0.397713  0.994398    23.575    16.372
        0.005000  0.863102  0.487619  0.991321    29.465    16.369
        0.006000  0.805503  0.571466  0.987627    35.354    16.368
        0.007000  0.739326  0.648266  0.983286    41.245    16.367
        0.008000  0.665453  0.717035  0.978246    47.137    16.367
        0.009000  0.584979  0.776970  0.972565    53.024    16.365
        0.010000  0.498994  0.827528  0.966332    58.910    16.364
        0.020000 -0.406528  0.772176  0.872651   117.765    16.356
        0.030000 -0.733380  0.043879  0.734691   176.576    16.350
        0.040000 -0.325935 -0.472524  0.574032   235.403    16.347
        0.050000  0.169341 -0.376246  0.412598   294.232    16.346
        0.060000  0.265793 -0.033230  0.267862   352.874    16.337
        0.070000  0.098245  0.117282  0.152994   410.048    16.272
        0.080000 -0.016037  0.073423  0.075154   462.321    16.053
        0.090000 -0.028304  0.018132  0.033614   507.356    15.659
        0.100000 -0.014532 -0.000944  0.014563   543.718    15.103
        0.200000  0.000167 -0.000033  0.000170   708.942     9.846
        0.300000  0.000015  0.000052  0.000054   433.630     4.015
        0.400000  0.000164  0.000085  0.000185   387.336     2.690
        0.500000  0.000015  0.000051  0.000053   433.625     2.409
        0.600000  0.000109 -0.000051  0.000120   695.049     3.218
        0.700000  0.000012  0.000029  0.000032   427.361     1.696
        0.800000  0.000062  0.000011  0.000063   370.085     1.285
        0.900000 -0.000853  0.000054  0.000855   536.362     1.655
 
      ####################################################
      # Program ETSTEP, gain and phase lag of the system #
      # Normalization factor:  0.999981                  #
      ####################################################
 
      wave     freq.     freq.      gain     phase  time lag
             [deg/h]     [cpd]               [deg]       [s]
 
 
      MM    0.544375  0.036292  1.000000    0.0025    16.381
      MF    1.098033  0.073202  1.000000    0.0050    16.381
      MTM   1.642408  0.109494  1.000000    0.0075    16.381
      Q1   13.398661  0.893244  1.000000    0.0610    16.381
      O1   13.943036  0.929536  1.000000    0.0634    16.381
      M1   14.496694  0.966446  1.000000    0.0660    16.381
      P1   14.958931  0.997262  1.000000    0.0681    16.381
      S1   15.000002  1.000000  1.000000    0.0683    16.381
      K1   15.041069  1.002738  1.000000    0.0684    16.381
      PSI1 15.082135  1.005476  1.000000    0.0686    16.381
      PHI1 15.123206  1.008214  1.000000    0.0688    16.381
      J1   15.585443  1.039030  1.000000    0.0709    16.381
      OO1  16.139102  1.075940  1.000000    0.0734    16.381
      2N2  27.968208  1.864547  1.000000    0.1273    16.381
      N2   28.439730  1.895982  1.000000    0.1294    16.381
      M2   28.984104  1.932274  1.000000    0.1319    16.381
      L2   29.528479  1.968565  1.000000    0.1344    16.381
      S2   30.000000  2.000000  1.000000    0.1365    16.381
      K2   30.082137  2.005476  1.000000    0.1369    16.381
      M3   43.476156  2.898410  1.000000    0.1978    16.381
      M4   57.968208  3.864547  0.999999    0.2638    16.381
 
 
      ######################################
      # Program ETSTEP finished execution. #
      ######################################
 
      The instrumental time lag of the system is 16.381 s, the  instrumental   
      phase lag of the system is 0.0634 deg for tidal wave O1 and 0.1319 deg   
      for tidal wave M2.                                                       
 
      Section 9: Description of program IERS
      --------------------------------------
 
      The program IERS.FOR is  supplied  in  order  to  transform  an  earth
      orientation parameter file, as e.g.  "eopc04.96"  available  from  the
      International Earth Rotation Service (IERS) via ftp  into  the  format
      necessary for the ETERNA package. The source code of  the  program  is 
      provided in file "\eterna33\sourcef\iers.for",  the executable program
      is provided in file "\eterna33\bin\iers.exe". The program  is  written 
      in standard Fortran 90. 
      
      The IERS earth rotation parameters can be transferred from IERS 
      Central Bureau at Paris via FTP:
 
      adress:     149.238.2.21
      login:      anonymous
      password:   your e-mail adress
      directory:  cd iers cd old
      files:      eopc04.xx where xx stands for the year.
  
      The program IERS assumes  the  input  data  in  IERS  format  on  file 
      "eopc04.96", and provides the output on  the  print  file  "iers.prn". 
      The data from this output have to be added at the  end  of  the  ASCII 
      file  "etpolut1.dat"  and  the  unformatted  file   "etpolut1.uft"  in
      directory "eterna33\commdat" has to be deleted in order to update  the
      pole coordinates and DUT1 corrections  for  the  ETERNA  package.  The 
      program IERS has to be updated and re-compiled after 1996 or when  a
      new leap second has been introduced in UTC. When a new leap second has
      been introduced, you should also update file "etddt.dat" in directory
      "eterna33\commdat".
 
                                                                               
      Section 10: Description of program BENCHMARK
      --------------------------------------------
   
      The Fortran 90 program BENCHMAR reads  data  from  a  tidal  benchmark
      series (e.g. bfde403f.dat, see  Wenzel  1996),  computes  model  tides
      from a tidal potential catalogue and computes the differences  between
      the benchmark series and the model tides. Program BENCHMAR is a  modi-
      fication of program PREDICT. There can be used seven  different  tidal
      potential  catalogues  (Doodson  1921,  Cartwright-Tayler-Edden  1973, 
      Buellesfeld 1985, Tamura 1987, Xi 1989, Roosbeek  1996,  Hartmann  and 
      Wenzel 1995). Program BENCHMAR reads the project name "pn"  from  file
      "project", the control parameters from file "pn.ini and the  benchmark
      series from file  "pn.dat".  Program  BENCHMAR  writes  the  benchmark 
      series, the computed model tide and the difference in standard  ETERNA
      format on file "pn.dif". For a description of the  control  parameters
      see the description of program PREDICT.
  
      Section 11: Description of program PREDICT                              
      ------------------------------------------                              
                                                                               
      The Fortran 90 program PREDICT can  be  used  for  the  computation of
      earth tide signals with constant time  interval  for  one  station  in 
      order to generate a table of synthetic model tides  (tidal  potential, 
      gravity  tides,  tilt  tides,  vertical  or  horizontal  displacement,
      vertical strain, horizontal strain, areal strain, shear strain, volume   
      strain, and oceanic tides). There can be  used seven  different  tidal   
      potential  catalogs   (Doodson  1921,  Cartwright-Tayler-Edden   1973,   
      Buellesfeld 1985, Tamura 1987, Xi 1989, Roosbeek  1996,  Hartmann  and
      Wenzel 1995) together with observed  tidal  parameters.  Additionally,
      gravity pole tides (gravity variation due to polar motion,  see  encl.
      007) and gravity LOD tides (gravity  variation  due  to  variation  of 
      length of day, see encl. 008) can be computed for those periods, where
      the necessary data are given in file "etpolc.dat".    
 
      Section 11.1: Description of the project file "project"
      ------------------------------------------------------
 
      The ASCII project name file "project"  for  program  PREDICT  contains
      one single variable only, the project name  which  is  abbreviated  as
      "pn" in this manual. The project  name  is  an  alphanumerical  string 
      consisting of 8 characters at maximum. The project name is  read  from 
      the first 8 characters of  the  first  record  of  project  name  file 
      "project". The project name is used to define  the  control  parameter 
      file name "pn".ini for program PREDICT.  
 
      Section 11.2: Description of control parameter file "pn".ini    
      -----------------------------------------------------------    
                                                                               
      The control parameter file "pn".ini (where "pn" stands for the project  
      name defined in file "project")  defines  several  control  parameters 
      necessary for the operation of program PREDICT. This control parameter
      file has an identical structure to the control parameter files used by
      programs RECTIDE, DETIDE, DESPIKE, DECIMATE, PREGRED and ANALYZE. Some
      of the variables are identical to  those  used  by  programs  RECTIDE, 
      DETIDE, DESPIKE,  DECIMATE,  PREGRED  and  ANALYZE.  Thus  it  may  be 
      possible to use the same control parameter file  for  data  recording, 
      data preprocessing and data analysis.                      
                                                                               
      All control parameter statements must begin with the control parameter   
      name (column 1...10) which ends with an equal sign. The control parame-  
      ter names must be spelled exactly with uppercase  letters  as  defined   
      below. Control parameters  with  unknown  names  will  be  ignored  by   
      program RECTIDE. The control parameter variables must be  input  after   
      the equal sign following the control parameter name, with  appropriate   
      format. The control parameter statements  must  not  have  a  specific   
      sequence. A record starting with # in column 1 will be ignored and may   
      be used to comment the control parameters.  All  characters  following   
      # in a record will be ignored and may be used to comment  the  control   
      parameters.                                                              
                                                                               
      STATIONAME= station name, 10 characters (col. 12...21)  following  the   
                  equal sign. The station name  will  be  displayed  on  the   
                  graphics screen during data recording.                       
      STATLATITU= ellipsoidal latitude of the station in degree referring to   
                  WGS84 reference system (col.12...26).                        
      STATLONITU= ellipsoidal longitude of the station in  degree  referring   
                  to WGS84 reference system (col.12...26).                     
      STATELEVAT= ellipsoidal height of the station in meter referring to      
                  WGS84 reference system (col.12...26).                        
      STATGRAVIT= gravity of the station in m/s**2, necessary for tidal tilt   
                  only (col. 12...26,  real  with  decimal  point).  If  the   
                  gravity is unknown, use a value of less than 1.0  and  the   
                  program will  compute  and  subsequently  use  the  normal   
                  gravity value referring to GRS80 reference system.           
      STATAZIMUT= azimuth of the  instrument  in  degree  decimal,  reckoned   
                  clockwise from north (col. 12 ... 26,  real  with  decimal   
                  point). This parameter is used for tidal tilt,  horizontal   
                  displacement and horizontal strain only.                     
      TIDALPOTEN= parameter for the tidal potential catalog to be used.        
                  1 = Doodson (1921) tidal potential catalog,                  
                  2 = Cartwright-Tayler-Edden (1973) tidal potential catalog   
                  3 = Buellesfeld (1985) tidal potential catalog,            
                  4 = Tamura (1987) tidal potential catalog,
                  5 = Xi (1989) tidal potential catalog,
                  6 = Roosbeek (1996) tidal potential catalog,
                  7 = Hartmann and Wenzel (1995) tidal potential catalog.
                  As default, TIDALPOTEN= 7 is used.                   
      SAMPLERATE= Data sample interval in s (col. 12 ... 26, integer).         
      INITIALEPO= Initial epoch, used to compute the Fourier development  of   
                  the specific earth tide component. The initial epoch  con-   
                  sists  of  the  integer  year (col. 12 ... 16, e.g. 1992),   
                  integer month (col. 17 ... 21), and integer day (col.        
                  22 ... 26).                                                  
      PREDICSPAN= Time span for the  prediction  in  hours  (col. 12 ... 26,   
                  integer). The model tide series will start at the  initial   
                  epoch INITIALEPO and the  time  span  will  be  PREDICSPAN   
                  hours.                                                       
      TIDALCOMPO= Earth tide component (col. 12 ... 26, integer).              
                  = -1 for tidal potential in m**2/s**2.                       
                  =  0 for tidal gravity in nm/s**2.                           
                  =  1 for tidal tilt in mas, at azimuth STATAZIMUT.           
                  =  2 for tidal vertical displacement in mm.                  
                  =  3 for tidal horizontal displacement in mm at              
                       azimuth STATAZIMUT.                                     
                  =  4 for tidal vertical   strain in 10**-9 = nstr.           
                  =  5 for tidal horizontal strain in 10**-9 = nstr,           
                       at azimuth STATAZIMUT.                                  
                  =  6 for tidal areal  strain in 10**-9 = nstr.               
                  =  7 for tidal shear  strain in 10**-9 = nstr.               
                  =  8 for tidal volume strain in 10**-9 = nstr.               
                  =  9 for ocean tides in mm.                                  
                  The computed model  tides  will  be  given  in  the  units   
                  defined above.                                               
      AMTRUNCATE= Amplitude threshold for the tidal potential  catalogue  in
                  m**2/s**2. Only tidal waves with amplitudes exceeding  the
                  amplitude threshold are used  for  the  computation.  This 
                  reduces the execution time, but also the accuracy  of  the
                  computed tidal signales. For mean latitudes, the  relation
                  between amplitude threshold and gravity tide accuracy   is
                  for the Hartmann and Wenzel (1995) tidal potential catalog
 
                           threshold              rms error             
                                                 [nm/s**2]          
 
                             1.D-01                88.40              
                             1.D-02                14.40              
                             1.D-03                 2.25              
                             1.D-04                 0.44              
                             1.D-05                 0.068             
                             1.D-06                 0.011             
                             1.D-07                 0.002             
                             1.D-08                 0.001             
                             1.D-09                 0.001              
                             1.D-10                 0.001
             
      POLTIDECOR= Amplitude factor for gravity pole  tide. If  the amplitude
                  factor is greater zero, gravity pole tides will be  compu-
                  ted using the IERS daily pole  coordinates  and  stored as 
                  additional channel of the ouput file "pn".prd.
      LODTIDECOR= Amplitude factor for gravity LOD tide.  If  the  amplitude
                  factor is greater zero, gravity LOD tides will be computed
                  using  the IERS  daily  pole  coordinates  and  stored  as 
                  additional channeld of the output file "pn".prd.    
      TIDALPARAM= wave group selection for a priori model tide  computation.   
                  The number of wavegroups is restricted to  85.  The  tidal   
                  parameters may be taken from an  earth  tide  analysis  of   
                  another data set in the same station,  or  from  a  neigh-   
                  bouring station, or from synthetic gravity tide parameters   
                  (Timmen and Wenzel 1994, 1995), or from a guess only.        
                  Col.12..21: DFRA#=  start frequency of the wave  group  in   
                                      cpd (real).                              
                  Col.22..31: DFRE#=  end   frequency of the wave  group  in   
                                      cpd (real).                              
                  Col.32..41: DGAM0#= amplitude factor  of  the  wave  group   
                                      (real).                                  
                  Col.42..51: DPHI0#= phase lead of the wave  group  in  deg   
                                      (real).                                  
                                                                               
      The wave group selection for model tide  computation  should  be  made   
      according to the wave group selection for the analysis of  Earth  tide   
      data; the list given below is a  proposal  only. The  frequencies  are 
      given in cycles per day (cpd).            
                                                                               
      group        > 1 month            > 6 months          > 1 year           
                   from      to         from      to        from      to       
                  [cpd]   [cpd]        [cpd]   [cpd]       [cpd]   [cpd]       
                                                                               
      SA             -       -            -       -       0.001379  0.004107   
      SSA            -       -      0.004108  0.020884    0.004108  0.020884   
      MM      0.020885  0.054747    0.020885  0.054747    0.020885  0.054747
      MF      0.054748  0.091348    0.054748  0.091348    0.054748  0.091348
      MTM     0.091349  0.501369    0.091349  0.501369    0.091349  0.501369   
      Q1      0.501370  0.911390    0.501370  0.911390    0.501370  0.911390   
      O1      0.911391  0.947991    0.911391  0.947991    0.911391  0.947991   
      M1      0.947992  0.981854    0.947992  0.981854    0.947992  0.981854
      P1             -       -      0.981855  0.998631    0.981855  0.998631   
      S1             -       -            -       -       0.998632  1.001369   
      K1      0.981855  1.023622    0.998632  1.023622    1.001370  1.004107   
      PSI1           -       -            -       -       1.004108  1.006845   
      PHI1           -       -            -       -       1.006846  1.023622   
      J1      1.023623  1.057485    1.023623  1.057485    1.023623  1.057485   
      OO1     1.057486  1.470243    1.057486  1.470243    1.057486  1.470243   
      2N2     1.470244  1.880264    1.470244  1.880264    1.470244  1.880264   
      N2      1.880265  1.914128    1.880265  1.914128    1.880265  1.914128   
      M2      1.914129  1.950419    1.914129  1.950419    1.914129  1.950419   
      L2      1.950420  1.984282    1.950420  1.984282    1.950420  1.984282   
      S2      1.984283  2.451943    1.984283  2.002736    1.984283  2.002736  
      K2             -       -      2.002737  2.451943    2.002737  2.451943   
      M3M6    2.451944  7.000000    2.451944  7.000000    2.451944  7.000000
                                                                               
  
      As an example for the control parameter file for program  PREDICT,  we 
      list below file "bfhw9501.ini", which is provided on the CD-rom:
 
      # This file BFHW9501.INI status 1996.08.15 containing control parameters
      # for program package ETERNA 3.40
 
      # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
      # ! NOTE: The datalines have to start with their names.       !
      # !       An additional comment may follow after the values,  !
      # !       delimited by a whitespace                           !
      # ! Values of 0 or less causes PREGRED to calculate the       !
      # ! range(s) automatically resp. to use default values        !
      # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
 
      # a commentline starts with an '#', it may appear at any position
      # in this file. Empty lines may appear too
 
      TEXTHEADER=  1.15000   0.0000 long   1.14673   -0.2474 Q1    
      TEXTHEADER=  1.14882   0.0804 O1     1.13070   0.2132 M1   
      TEXTHEADER=  1.13899   0.2062 P1     1.13000   0.0000 S1      
      TEXTHEADER=  1.13600   0.2070 K1     1.25000   0.5000 PSI1    
      TEXTHEADER=  1.18000   0.3000 PHI1   1.15354   0.0801 J1   
      TEXTHEADER=  1.14851  -0.0251 OO1    1.15205   2.4463 2N2  
      TEXTHEADER=  1.17054   2.5425 N2     1.18705   2.0327 M2   
      TEXTHEADER=  1.22450   4.1630 L2     1.18963   0.6271 S2
      TEXTHEADER=  1.18947   0.8226 K2     1.06234   0.3783 M3
      TEXTHEADER=  1.02000   0.0000 M4     1.01500   0.0000 M5M6   
 
      SENSORNAME=PREDICT         # earth tide sensor name
      SAMPLERATE=   3600         # sampling interval in seconds 
      STATLATITU=     48.3306    # stations latitude  in degree
      STATLONITU=      8.3300    # stations longitude in degree
      STATELEVAT=    589.000     # stations elevation in meter
      STATGRAVIT=      0.        # stations gravity in m/s**2
      STATAZIMUT=      0.        # stations azimuth in degree from north
      INITIALEPO= 1987    1    1 # start epoch for PREDICT
      PREDICSPAN=  65544         # time span in hours for PREDICT
      TIDALCOMPO=      0         # tidal component, see manual
      TIDALPOTEN=      7         # HW95 tidal potential development
      AMTRUNCATE=  1.D-10        # amplitude threshold 
      POLTIDECOR=   1.16         # poletide amplitude factor
      LODTIDECOR=   1.17         # LOD tide amplitude factor   
      POLTIDEREG=      1         # ANALYZE pole tide regression adjustment
 
      TIDALPARAM=  0.000000  0.501369   1.15000    0.0000 long   #tidal param. 
      TIDALPARAM=  0.501370  0.911390   1.14673   -0.2474 Q1     #tidal param.
      TIDALPARAM=  0.911391  0.947991   1.14882    0.0804 O1     #tidal param.
      TIDALPARAM=  0.947992  0.981854   1.13070    0.2132 M1     #tidal param.
      TIDALPARAM=  0.981855  0.998631   1.13899    0.2062 P1     #tidal param.
      TIDALPARAM=  0.998632  1.001369   1.13000    0.0000 S1     #tidal param. 
      TIDALPARAM=  1.001370  1.004107   1.13600    0.2070 K1     #tidal param.
      TIDALPARAM=  1.004108  1.006845   1.25000    0.5000 PSI1   #tidal param. 
      TIDALPARAM=  1.006846  1.023622   1.18000    0.3000 PHI1   #tidal param. 
      TIDALPARAM=  1.023623  1.057485   1.15354    0.0801 J1     #tidal param.
      TIDALPARAM=  1.057486  1.470243   1.14851   -0.0251 OO1    #tidal param.
      TIDALPARAM=  1.470244  1.880264   1.15205    2.4463 2N2    #tidal param.
      TIDALPARAM=  1.880265  1.914128   1.17054    2.5425 N2     #tidal param.
      TIDALPARAM=  1.914129  1.950419   1.18705    2.0327 M2     #tidal param.
      TIDALPARAM=  1.950420  1.984282   1.22450    4.1630 L2     #tidal param.
      TIDALPARAM=  1.984283  2.002736   1.18963    0.6271 S2     #tidal param.
      TIDALPARAM=  2.002737  2.451943   1.18947    0.8226 K2     #tidal param.
      TIDALPARAM=  2.451944  3.381478   1.06234    0.3783 M3     #tidal param.
      TIDALPARAM=  3.381379  4.347615   1.02000    0.0000 M4     #tidal param.
      TIDALPARAM=  4.347616  7.000000   1.01500    0.0000 M5M6   #tidal param.
 
      # End of file BFHW9501.INI
 
      Section 11.3: Description of output file "pn".prd
      -------------------------------------------------
 
      The data output file of  program  PREDICT  uses  the  ETERNA  standard
      format for exchange of  high  resolution  and  high  rate  tidal  data   
      (Wenzel 1995). The output file is divided into a file header giving
      an alphanumeric description of the file (provided with control 
      parameters TEXTHEADER of the "pn".ini file) and the file body
      containing the computed tidal signal. The format of the file body is
                                                                               
      Col.01...08: date in year, month, day; e.g. 19900328  means year 1990,   
                   month 03 = March, day 28th.                                 
      Col.10...15: time in hour, minute, second,  given  in  Universal  Time   
                   Coordinated (UTC); e.g. 231200 means 23 hours 12  minutes   
                   00 seconds. The seconds  are assumed to be exact, e.g. 00   
                   seconds means 0.000 seconds.                                
      Col.16...25: Channel 1 in units of the tidal component. This channel
                   contains the sum of the tidal signal and the gravity
                   pole tide and LOD tide signals.  
      Col.26...35: Channel 2 in units of the tidal component. This channel
                   usually conatins the gravity tide.                  
      Col.36...45: Channel 3 in units of the tidal component. This channel
                   contains the gravity pole tide, if computed.
      Col.46...55: Channel 4 in units of the tidal component. This channel
                   contains the gravity LOD tide, if computed.
 
      The file ends with 99999999 as date.
 
      As an example, a part of the output file "bfhw9501.prd" computed with
      program PREDICT using the control parameter file "bfhw9501.ini" is 
      given below. This files contains gravity tides, gravity pole tides and
      gravity LOD tides. 
                                                                               
      77777777            0.000     0.000     0.000     0.000
      19870101      0 -1630.338 -1649.708    18.906     0.464
      19870101  10000 -1544.612 -1563.988    18.909     0.467
      19870101  20000 -1253.944 -1273.327    18.913     0.469
      19870101  30000  -811.681  -831.070    18.917     0.472
      19870101  40000  -295.698  -315.093    18.921     0.474
      19870101  50000   208.593   189.191    18.925     0.477
      19870101  60000   626.498   607.090    18.929     0.479
      19870101  70000   909.354   889.939    18.933     0.482
      19870101  80000  1043.335  1023.914    18.937     0.485
      19870101  90000  1049.655  1030.227    18.941     0.487
      19870101 100000   976.094   956.660    18.944     0.490
      19870101 110000   881.969   862.529    18.948     0.492
      19870101 120000   820.439   800.992    18.952     0.495
 
      The data stored in the output file of program PREDICT can be plotted
      using program PLOTDATA (see encl. 006...008).
 
      Section 11.4: Execution time of program PREDICT
      -----------------------------------------------
  
      The execution time of program PREDICT depends mainly on the number of
      waves of the chosen tidal potential catalogue, on the amplitude
      threshold of the tidal potential, and on the number of samples to be
      computed. The execution time given below has been measured on a 100 
      MHz PENTIUM pc with operating system MS-DOS 6.22 for 8760 hourly
      gravity signals including pole tide and LOD tide. The rms error has
      been determined by comparison with gravity tide benchmark series
      BFDE403F.DAT.
 
      catalogue           threshold     nwave    rms error    exec.time
                                                 [nm/s**2]       [s]
 
      Tamura (1987)          1.D-06       1200      0.070        11.86
      Hartmann+Wenzel (1995) 1.D-01          9     88.40          3.90
      Hartmann+Wenzel (1995) 1.D-02         42     14.40          4.06
      Hartmann+Wenzel (1995) 1.D-03        155      2.25          4.94
      Hartmann+Wenzel (1995) 1.D-04        434      0.44          7.20
      Hartmann+Wenzel (1995) 1.D-05       1248      0.068        13.51
      Hartmann+Wenzel (1995) 1.D-06       3268      0.011        33.01
      Hartmann+Wenzel (1995) 1.D-07       7761      0.002        83.82
      Hartmann+Wenzel (1995) 1.D-08      11462      0.001       132.86
      Hartmann+Wenzel (1995) 1.D-09      12000      0.001       139.95
      Hartmann+Wenzel (1995) 1.D-10      12011      0.001       140.11
  
      Section 12: Description of program PREGRED
      ------------------------------------------
                                                                               
      PREGRED is an interactive graphical editor for digital recorded  earth   
      tide data (Vetter and Wenzel 1995), written in  Microsoft  Visual  C++
      1.52 language to supplement the batch  programs  DETIDE,  DESPIKE  and
      DECIMATE (see below). Problems which cannot automatically be solved by
      program DESPIKE (e.g. several steps within short time interval,  small 
      slow glitches) can be managed with the graphical editor PREGRED  under
      graphical control of the user. PREGRED uses the standard ETERNA format
      (Wenzel 1995). The creation of the graphical editor PREGRED was initi-
      ated by Dr. J. Neumeyer, Geoforschungszentrum  Potsdam,  who  supplied 
      his programs ALPHA and BETA to us.          
                                                                         
      The graphical editor PREGRED is able to plot simultaneously data  from
      two channels for a predefined time window (e.g. one day) on the screen
      of the pc. Vertical and/or horizontal zooming is  possible.  The  user 
      can manually delete disturbed data, interpolate gaps and correct steps
      in all channels. PREGRED has originally been designed for 1 min  data,
      but can be applied to data with  1 s ... 1 h  sampling  interval  (the 
      sampling interval has to be an integer number of seconds equal to 3600
      divided by an integer n).  Recommended  sampling  intervals  are  1 s, 
      2 s, 3 s, 5 s, 10 s, 20 s, 30 s or 1 min. PREGRED can  be  applied  to 
      data from 9 channels at maximum, but can display simultaneously on the
      screen data from two channels only.                         
                                                                               
      PREGRED runs on an IBM-compatible personal computer with at least  640   
      kB RAM, preferably with a color  screen.  Although  PREGRED  could  in   
      principle be executed with a monochromatic screen,  it  will  be  very   
      difficult to distinguish the traces of  the  different  channels  (for   
      this situation, PREGRED has the built-in option to display one channel   
      only). The graphics adapter should be  EGA-, VGA-standard  or  higher.   
      Due to the high processing effort of  graphical  editing,  a   PENTIUM   
      cpu is recommended.                                  
                                                                               
      PREGRED needs the two font files "oem08.fon" and "oem10.fon", supplied
      on the CD-rom. These two font files have either to be  copied  to  the
      working directory from which PREGRED is executed, or the path to these
      files has to be defined with the MS-DOS environment variable  MSCFONTS
      by adding in  your  "autoexec.bat" file the line
 
         set MSCFONTS=c:/eterna34/bin
 
      provided you have installed the ETERNA 3.40 package on drive c:
 
      The combination of programs DETIDE, DESPIKE, DECIMATE and PREGRED is
      shown in the following diagram:                                         
                                                                               
                 ---------------                                               
                 I  Raw data   I                                               
                 ---------------                                               
                        Y                                                      
                 ---------------                                               
                 I   PREGRED   I      <---- find and flag gaps                 
                 ---------------                                               
                        Y                                                      
                 ---------------                                               
                 I   DETIDE    I      <---- calibrate and detide data          
                 ---------------                                               
                        Y                                                      
                 ---------------                                               
                 I   PREGRED   I      <---- flag disturbances, destep data     
                 ---------------                                               
        --------------->Y                                                      
        I        ---------------                                               
        I        I   DESPIKE   I      <---- destepping, despiking, degapping   
        I        ---------------                                               
        ----------------Y                                                      
                 ---------------                                               
                 I   PREGRED   I      <----- final check                       
                 ---------------                                               
                        Y                                                      
                 ---------------                                               
                 I   DECIMATE  I      <---- decimation to 5 min samples        
                 ---------------                                               
                        Y                                                      
                 ---------------                                               
                 I   DECIMATE  I      <---- decimation to hourly samples       
                 ---------------                                               
                                                                               
      It is in any case recommended to store the original data files  to  be   
      edited on a backup medium, e.g. magneto-optical disk, before  starting   
      the graphical editing with PREGRED.                                      
                                                                               
      PREGRED may be executed by entering                                      
                                                                               
                  <pregred projectname.suffix>                                 
                                                                               
      If you enter <pregred -h>, you will  see  a help  screen  showing  the 
      command line options. If you have a high resolution  graphics  adapter 
      and screen, you may want to work  with  a  high  resolution  image  by 
      entering
 
        <pregred projectname.suffix -r800>  for  800 x  600 pixel resolution,
        <pregred projectname.suffix -r1024> for 1024 x  768 pixel resolution.
        <pregred projectname.suffix -r1280> for 1280 x 1024 pixel resolution.
  
      PREGRED overwrites the original file with the edited data,  but  saves   
      the original data in  a  backup  file.  If  you  edit  e.g.  the  file   
      "sg102tes.tid", the original data are  saved  in file  "sg102tes.#ti", 
      whereas the edited  data  are  stored  in  file  "sg102tes.tid"  after 
      finishing the execution of program PREGRED. You should  never input  a 
      backup file to PREGRED, because in this case it may be impossible  for
      PREGRED to save the backup file. If you have made a mistake during the
      editing with PREGRED, you should finish the execution of PREGRED, copy
      the backup file to a file having the original name,  and  restart  the 
      editing procedure.                                 
                                                                               
      PREGRED reports the modification which have been made at the data file   
      in a report file with name "filename".PRO. As an example, we  give  in   
      directory "\eterna33\1mindat" the report file "sg102tes.pro" which was
      established by PREGRED when editing file "sg102tes.cor".
                                                                               
      For the execution of PREGRED, there must  exist  a  control  parameter   
      file with name "projectname.ini" and a calibration parameter file with
      name "projectname.cal" in the directory, from which PREGRED  has  been
      executed and in which the data file  to  be  edited  is  stored.  This
      control parameter file defines the  control  parameters  necessary  to 
      execute PREGRED. The structure of the control parameter file  used  by 
      PREGRED is identical to the structure of the control  parameter  files
      used in the ETERNA 3.40 package by other programs (e.g. PREDICT).  For
      the execution of PREGRED, the following control parameters have to  be   
      defined:                                                                 
                                                                               
      SAMPLERATE= data sample interval in s (col. 12...26, free format). The   
                  program PREGRED is designed for 60  s  raw  data  sampling   
                  interval, but may also be used for another sampling inter-   
                  val (between 1 s and 3600 s). Recommended sample rates are
                  1 s, 2 s, 3 s, 5 s, 10 s, 20 s, 30 s,  1 min,  5  min,  10  
                  min, 20 min, 30 min and 60 min.                           
      NUMRAWCHAN= number  of  raw  data  channels  (before  calibration  and
                  detiding). This  parameter  is  valid only for files  with
                  suffix "raw".
      NUMBCHANEL= number of final channels (after detiding)  for  all  files
                  where the suffix is not equal to "raw".  
      HPERPAGE=   displayed hours per page on the original (unzoomed) screen   
                  (col. 12...26, free format).                                 
      PGUPPGDN=   time shift between successive windows in hours (col. 12...   
                  26, free format).                                            
      RANGETCx=   vertical range for channel x in units of channel  1,  when
                  editing calibrated files with suffix CAL, TID, COR or  FIL
                  (col. 12...26, free format). x denotes the channel  number
                  and is restricted to 1...9.                                  
      ASSOC_LTO1= parameter for copying modifications made with  PREGRED  in   
                  the last channel (detide signal) to channel 1 (col. 12 ...
                  26, free format, 1 = yes). If you interpolate gaps in  the
                  last channel and ASSOC_LTO1 is  set  to  1,  the  data  in 
                  channel is computed by:                                    
                                                                            
                  ch(1) = ch(last) + sum over [channel(i)*METCORFACT(i)], 
                                                              i=2,last-1  
 
      RANGRAWx=   vertical range for channel x in Volt, when editing an  un-   
                  calibrated  file  with  suffix  RAW  (col. 12 ... 26, free   
                  format).                                                     
      INTEPOCx=   Number of samples to be averaged before and  after  a  gap
                  for gap interpolation in channel x  (col. 12 ... 26,  free 
                  format). By default, 10 samples will be  used.  x  denotes 
                  the channel number and is restricted to 1...9.
      MAXGAP=     Maximum number of hours of a gap  to  be  interpolated  by 
                  PREGRED.  Gaps exceeding MAXGAP cannot be interpolated.
 
      The calibration parameter file contains a file header and a a file 
      body. The file header contains alphanumeric information to describe
      the file contents, and ends with C********* in cols. 1...10.
 
      The first record of the file body contains
      in cols.  1..11:  RAWCHAUNIT=
      in cols. 12..14:  blanks
      in cols. 15..24:  unit of raw data in channel 1 (CHARACTER*10)
      in cols. 25..34:  unit of raw data in channel 2 (CHARACTER*10)
      in cols. 35..44:  unit of raw data in channel 3 (CHARACTER*10)
      in cols. 45..54:  unit of raw data in channel 4 (CHARACTER*10) ...
 
      The second record of the file body contains
      in cols.  1..11:  CALCHAUNIT=
      in cols. 12..14:  blanks
      in cols. 15..24:  unit of calibrated data in channel 1 (CHARACTER*10)
      in cols. 25..34:  unit of calibrated data in channel 2 (CHARACTER*10)
      in cols. 35..44:  unit of calibrated data in channel 3 (CHARACTER*10)
      in cols. 45..54:  unit of calibrated data in channel 4 (CHARACTER*10) ...
 
      The third record of the file body contains
      in cols. 1...11: METCORFACT= 
      in cols. 12..14: blanks
      in cols. 15..24: correction factor for channel 1 (has to be 1.0)
      in cols. 25..34: correction factor for channel 2
      in cols. 35..44: correction factor for channel 3  
      in cols. 45..54: correction factor for channel 4 .... 
      
      The calibration parameter file to be used with PREGRED is identical to
      the calibration  parameter  file  for  programs  DETIDE  and  DESPIKE.
      PREGRED reads however only the chanel  units  and  correction  factors
      for the meteorological channels from this file.
   
      We give below a copy of  file  "sg102tes.cal"  provided  in  directory
      "\eterna33\1mindat":
  
      File              : sg02tes.cal
      Status            : 1996.09.02
      Contents          : calibration parameters for data file sg02tes.raw
                          cryogenic gravimeter SG102 and barometer BMG3 
                          at BFO Schiltach.
 
      Sequence is: 
      raw data:        unit       unit
      detided data:    unit       unit       unit    unit
      correction:      fact.1     fact.2    fact.3  
      calib.param:    linear     linear
                      quadratic  quadratic
                      cubic      cubic
                      time lag   time lag
      C**************************************************************
      RAWCHAUNIT=       Volt        Volt
      CALCHAUNIT=       nm/s**2      hPa    nm/s**2   nm/s**2
      METCORFACT=        1.0000   -3.2000    1.0000 
      19940602 120000 3439.7000  -23.0424
      19940602           0.0000    0.0000
      19940602           0.0000    0.0000
      19940602          57.6000   21.0300
      19940622 120000 3439.7000  -23.0424
      19940622           0.0000    0.0000
      19940622           0.0000    0.0000
      19940622          57.6000   21.0300
      99999999
 
      While reading data from the input file, PREGRED  performs  a  sequence   
      test that checks for gaps in the data file. In case of gaps or  faults   
      in the date and time of the observation,  the  user  is  requested  to   
      decide either to terminate  the  execution  of  PREGRED  or  to  allow   
      PREGRED to fill the gaps with flags. Please  have  in  mind  that  the   
      samples should be given at full minutes if the sample interval is e.g.   
      60 s, because otherwise PREGRED will start to correct the time of  the   
      observation.                                                             
                                                                               
      After having performed the sequence check, PREGRED shows the  data  of   
      the first page on the graphical screen. Using function key 1 shows the   
      on-line help screen of program PREGRED, which explains the  operations   
      allowed with PREGRED.                                                    
                                                                               
      We give in the following the help screen of PREGRED version 3.40:        
                                                                               
        PgUp      = save gaps, display prev. page                             
        Ctrl+PgUp = save gaps, display first page
        PgDn      = save gaps, display next page
        Ctrl+PgUp = save gaps, dosplay last page                               
        ESC    = save gaps and EXIT program                                    
        a,A    = switch active channel RED-WHITE                               
        Arrows = move cursor, move slow use Ctrl-key additional                
        g,G    = start gap                                                     
        e,E    = end   gap                                                     
        f,F    = fill gaps for this channel by linear interpolation.
        c,C    = save gaps, display next channel as active                     
        d,D    = display both channels <-> only the active channel             
        u,U    = UNDO gaps for active day                                      
                                                                               
        s,S    = save gaps and define shift for active channel                 
                 to actual cursor position                                     
        r,R    = save gaps, refresh screen                                     
        z,Z    = save gaps, zoom in both axis                                  
        o,O    = save gaps, zoom out in both axis                              
        T      = save gaps, zoom in Time-axis                                  
        t      = save gaps, zoom out in Time-axis                              
        V      = save gaps, zoom in value axis                                 
        v      = save gaps, zoom out in value axis                             
        1      = save gaps, zoomfactor 1 for both axis                         
        i,I    = switch ranges from INI-File - calculated ranges               
                                                                               
        Continue PREGRED - Hit RETURN                                          
                                                                               
      Description of the on-line commands:                                     
      -----------------------------------                                      
                                                                               
      ESC   : By hitting the "ESC" key, the program PREGRED  terminates  the   
              operation. Eventual modifications which have been made to  the   
              data and are not yet saved are now saved in the data file. The
              program checks and lists remaining gaps in the data.       
      PgUp  : Switch screen to previous page (if it exists).  If there  have   
              been made any modifications to the data on the  current  page,   
              these modifications are now stored  permanently  in  the  data   
              file. The modifications cannot be  undone  after  hitting  the   
              "PgUp" key.                                                      
      Ctrl+PgUp: Switch to first screen. If there have been made any changes
              in the data on the current page, these modifications  are  now
              stored permanently in the data file. The modifications  cannot
              be undone after hitting the "Ctrl+PgUp" key.
      PgDn  : Switch screen to next page (if it exists). If there have  been   
              made any modifications to the data on the current page,  these   
              modifications are now stored permanently in the data file. The   
              modifications cannot be undone after hitting the "PgDn" key.     
      Ctrl+PgDn: Switch to last screen. If there have been made any  changes
              in the data on the current page, these modifications  are  now
              stored permanently in the data file. The modifications  cannot
              be undone after hitting the "Ctrl+PgDn" key.
      Pos1(Home): Move cursor to left border of the screen.
      End:    Move cursor to the right border of the screen.
      d,D   : There are usually displayed  two  channels  simultaneously  on   
              the page  by  PREGRED.  The  decision  of  PREGRED  which  two   
              channels (of the up to four channels available) are  displayed   
              depends on the  suffix  of  the  file  which  is  edited:  for   
              suffixes RAW  and  CAL  (and  suffixes  unknown  to  PREGRED),   
              channel 1 is initially the active channel plotted in  red  and   
              channel 2 is initially the inactive channel plotted in  white.   
              For suffixes TID, COR, FIL and DEC, channel 4 is initially the   
              active channel plotted in red and channel 2 is  initially  the   
              inactive channel plotted  in  white.  Which  data  channel  is   
              currently plotted as red or white channel is displayed at  the   
              left side of the page. By hitting key "d" or "D",  the  active   
              channel is displayed only and the inactive channel is  deleted   
              from the page. This may help the editing  in  case  of  larger   
              disturbances or crossings of red and white traces. The "d" and   
              "D" command is valid for all pages of the session.               
      a,A   : There are usually two channels plotted simultaneously  on  the   
              page, one in white and one in red. One  of  both  channels  is   
              called the "active channel", which means that modifications to   
              the data (deleting data of destepping data) are made  to  this   
              channel only. Which one of the channels is  currently  active,   
              is displayed in the lower right part of  the  page.  For  each   
              session, initially the red channel is active. If you  hit  the   
              "a" or "A" key, the active channel is switched  to  the  white   
              channel, which was previously inactive. If you hit  again  the   
              "a" or "A" key, the  active channel is again the red  channel.   
              The "a" and "A" command is valid for all pages of the session.   
     c,C    : The current active channel is replaced by the  next  available   
              channel with higher number of the current file to  be  edited.   
              If the active channel is 1 and you hit the "c" or "C" key, the   
              channel 2 will be plotted as active channel (except if channel   
              2 is already  plotted  as  inactive  channel;  in  this  case,   
              channel 3 will be displayed). If the active channel is already   
              the last channel available for the file and you hit the "c" or   
              "C" key, channel 1 will be displayed as active channel.          
     Arrows : The cursor keys may be used to  change  the  position  of  the   
              cursor (white cross, being positioned initially in the  center   
              of the page). By using simultaneously the "Cntrl" key  and the   
              cursor keys, you have a low speed motion with  minimum  incre-   
              ment of 1 pixel resp. the sample interval. The date  and  time   
              corresponding to the current cursor position are shown in  the   
              upper left part of the page; the numerical value of the active   
              channel corresponding to the current cursor position is  shown   
              at the upper right part of the page.                             
      z,Z   : The trace of the active channel of the page (plotted  in  red)   
              is zoomed around the current cursor position by a factor of  2   
              in both vertical  and  horizontal  directions  (in  fact  both   
              channels are zoomed in horizontal direction, but  in  vertical   
              direction the active channel is zoomed only).  If  you  hit  n   
              times the "z" or "Z" key, the active channel is  zoomed  by  a   
              factor of 2**n. The "z" and  "Z"  command  is  valid  for  the   
              current page only.                                               
      T     : The trace of both channels of the page is  zoomed  around  the   
              current cursor position by a factor of 2 in horizontal  direc-   
              tion only. If you hit n times  the  "T"  key,  the  screen  is   
              zoomed by a factor of 2**n in horizontal  direction.  The  "T"   
              command is valid for the current page only.                      
      t     : The trace of both channels of the page is de-zoomed around the   
              current cursor position by a factor of 2 in horizontal  direc-   
              tion only (this works only, if the page  has  previously  been   
              zoomed in horizontal direction by hitting "z","Z" or "T" key).   
              If you hit n times the "t" key, the page  is  de-zoomed  by  a   
              factor of 2**n in horizontal direction.                          
      V     : The trace of the active channel of the page is  zoomed  around   
              the current cursor position by  a  factor  of  2  in  vertical   
              direction only. If you hit n times the "V" key, the  trace  of   
              the active channel is zoomed by a factor of 2**n  in  vertical   
              direction. The "V" command is valid for the current page only.   
      v     : The trace of the active channel  of  the  page  is   de-zoomed   
              around the current  cursor  position  by  a  factor  of  2  in   
              vertical direction only (this works  only,  if  the  page  has   
              previously been zoomed in vertical direction by  hitting  "z",   
              "Z" or "V" key). If you hit n times the "v" key, the  page  is   
              de-zoomed by a factor of 2**n in vertical direction.             
      o,O   : The trace of the active  channel  of  the  page  is  de-zoomed   
              around the current cursor position by a factor of  2  in  both   
              vertical and horizontal directions (this works  only,  if  the   
              page has previously been zoomed). If you hit n times  the  "o"   
              or the "O" key, the trace of the active channel  is  de-zoomed   
              by a factor of 2**n.                                             
      1     : The page is plotted  with  the  initial  ranges;  any  zooming   
              which may have been performed is removed.                        
      g,G   : Start of a gap at the current (horizontal) cursor position. To   
              flag (delete) disturbed data, you move the cursor in  horizon-   
              tal direction to the start of the gap you want to create,  and   
              you hit the "g" or "G" key if the cursor  is  at  the  correct   
              position.                                                        
      e,E   : End of a gap at the current (horizontal) cursor  position.  To   
              flag (delet) disturbed data, you move the cursor in horizontal   
              direction to the end of the gap you want to  create,  and  you   
              hit the "e" or "E"  key  if  the  cursor  is  at  its  correct   
              position. These keys work only if the start  of  the  gap  has   
              already been defined on the same screen!                         
      f,F   : Linear interpolation of the current gap in the active channel.
              This key works only if the cursor is positioned inside  a  gap
              (in time direction). The last n samples before the gap and the
              first n samples after the gap are  used  to  linearily  inter-
              polate the gap, if the length of the gap is less  than  MAXGAP
              hours. The number of samples to be averaged can be defined  by
              parameter INTEPOCx. By default, INTEPOCx is 10.  
      u,U   : A gap which has been created on  the  current  screen  may  be   
              undone (if the data of the current screen have  not  yet  been   
              saved) by hitting the "u" or "U" key.                            
      s,S   : Step correction. To correct the data of the active channel for   
              the complete data file, move the cursor to the position at the   
              right hand side of a step,  where  the  trace  of  the  active   
              channel before the step would be located  if  the  step  would   
              not exist. Before using the "s" or "S" keys, it is recommended   
              to zoom the screen around the step in vertical direction.        
                                                                               
                                   step                                        
                                    -----------------------                    
                                                                               
              ----------------------  x= cursor position                       
                                                                               
              Hit the "s" or "S" key, and the program PREGRED will show  you   
              how the step will be removed, if you agree with  the  proposed   
              shift (the program asks you to confirm the step removal opera-   
              tion by entering either "y" for yes or "n" for no).              
                                                                               
                                    --                                         
                                       <------ spike to be deleted             
              ----------------------  x--------------------                    
                                                                               
              It is usually necessary, to delete a small part  of  the  data   
              around the step after its removal by using  the  "g"  and  "e"   
              keys, see above.                                                 
                                                                               
      Error messages of program PREGRED:
      ----------------------------------
 
      Message:                   Description / Action
 
      Wrong samplingrate xxx     Check the parameter SAMPLERATE= in the 
      - consult manual           INI-File. The Samplingrate has to be between 
                                 1 and 3600 and 3600 must be divisible by 
                                 Samplingrate without a rest.
 
      Can't find font oem10.fon  Copy the font-files into your working-
                                 directory or define a directory in your 
                                 AUTOEXEC.BAT which contains the font-files.
                                 Use SET MSCFONTS=<directory> e.g.
                                 SET MSCFONTS=C:/eterna33/bin
                                 !! DON'T use a backslash "\" as seperator !!
 
      channel numbers not        If you use a command-line-option to choose 
      allowed                    the channels to be displayed, the channel
      and                        numbers have to be different and they have
      channels have to be        to be between 1 and the number of channels
      different                  in the data-file
 
      Unknown Parameter -x       You used an unknown command-line-parameter
 
      File xxx not found         PREGRED could not find your input-file
 
      Can't open xxx-File        PREGRED could not open a file for writing.
                                 May be your filesystem is full.
 
      Wrong RANGETC-Number at    The number of Channels is limited to 11.
      INI-File                   Check your parameters in the INI-File
 
      Wrong RANGRAW-Number at    The number of Channels in RAW-Files is 
      INI-File                   limited to 9. Check your Parameters in the 
                                 INI-File.
 
      Wrong INTEPOC-Number at    The number of Channels is limited to 11. 
      INI-File                   Check your Parameters in the INI-File
 
      AIRPCORFAC, METCORFACx     Delete the AIRPCORFAC and METCORFACx 
      are not supported any more parameters in the INI-File and define them
      METCORFACT has to be       in the CAL-File
      defined in the CAL-File
 
      No INI-File found....      Generate a INI-File or copy the existing
                                 INI-File to the working directory
 
      No CAL-File found....      Generate a CAL-File or copy the exsiting
                                 CAL-File to the working directory
 
      CAL-File uncomplete        One of the following parameters is NOT
      - Program aborted          defined in the CAL-File:
                                 METCORFACT, RAWCHAUNIT, CALCHAUNIT
                                 Define the missing Parameter.
 
      Number of Chanels in INI-  In the datafile, the number of offsets
      File different from num-   from the number of channels, described 
      ber of first offsets       in the INI-File (NUMBCHANEL or NUMRAWCHAN)
 
      Max. xx channels allowed   For RAW-Files, only 9 channels are allowed
                                 and for other files the limit is 11 channels
 
      ASSOC_LTO1 ignored in      In RAW-Files, the last channel can't be 
      RAW-Files                  associated to channel no. 1
 
 
      Data error                 No data found in the file
 
      First Data-Line contains   The first line of data contains less
      not enough channels        channels than described in the INI-File
 
      Found Data-Line containing A single data record contains less channels
      not enough channels        than described in the INI-File
 
      Wrong time data detected.  The detected time doesn't fit to the 
      Please select action       sampling interval
      Correct time by program
      (=1) abort Program (=2)
 
      At <time> additional       Offsets with code 77777 may not occur in
      Offsets in Rawfile         a RAW-file and are not processed
      ignored
 
      Not enough memory          Take care to have at least 500kB of free
                                 memory in the lower memory section or 
                                 decrease the parameter HPERPAGE in the 
                                 INI-File
 
      Can't interpolate data     PREGRED only interpolates data if there is
      No gap at actual cursor-   a gap at the current cursor position. Move 
      position - Hit RETRUN      the cursor to a gap or delete some data 
                                 before gap interpolation.
 
      This is already the FIRST  You tried to go forward/backward at the 
      Page                       end/start of the datafile
        or
      This is already the LAST
      Page
 
      xxx gaps detected -        After terminating the program, PREGRED 
      see PRO-File               detected xxx gaps in the data, described
                                 in the PRO-File
 
      No data at the END of gap  You tried to interpolate a gap, but there
                                 are no data at the end of the gap
 
      No data at the BEGIN of    You tried to interpolate a gap, but there
      gap                        are no data at the beginning of the gap
 
      Gap exceeds maximum length You tried to interpolate a gap with a 
      of xxx hours - See Para-   length exceeding the MAXGAP-Parameter
      meter MAXGAP in INI-File   in the INI-file
 
      Gap in Meteorological Data You tried to interpolate a gap in the 
      Can't calculate Channel 1  channel with parameter ASSOC_LTO1 = 1. 
      - run DETIDE first         PREGRED can't calculate the data for 
      - run DETIDE first         channel 1 if there are gaps in any
                                 meteorological data at the same time. 
                                 Interpolate the meteo-channels first and
                                 run DETIDE-Program.
 
      Gap (99999.999) at act.    You tried to correct an offset but at the
      Position - Can't calcu-    current cursorposition are no data. Move 
      late an offset             the cursor to a position containing data 
                                 and repeat the step-correction.
 
      Gap detected at <time>     PREGRED detected a gap of xxx records. 
      xxx records missing -      Select the action, which PREGRED shall 
      Fill gap with flags(=1)    perform.
      abort Program (=2)
 
      You decided to fill        Be shure to fill such a large gap with
      a large gap of more than   flags.
      10 hours
 
      Section 13: Description of program DECIMATE                   
      --------------------------------------------                   
                                                                               
      Program DECIMATE can be used  to  decimate  data  stored  in  standard 
      ETERNA format after numerical lowpass filtering to a lower  resampling
      interval. If data have e.g. been recorded with 10 s sampling interval,
      program DECIMATE can be used to numerically lowpass  filter  the  data 
      and to resample the data at 60 s resampling interval. After data  pre-
      processing with DETIDE and DESPIKE, the preprocessed  data  can  again
      be decimated to hourly interval using DECIMATE. Program DECIMATE  uses
      FIR zero phase numerical lowpass filters; the apropriate filter can be
      chosen by the user from a variety of available filters.  
      
      Section 13.1: Decimation of data
      --------------------------------
 
      For program DETIDE, there are in principle allowed sampling intervals
      between 1 and 3600 s. However, we have provided numerical lowpass 
      filters for the following sampling / resampling intervals only:
 
      filter name       sampling / resampling interval      enclosure
 
      n01s1m01.nlf          1 s /     1 min                   009
      n02s1m01.nlf          2 s /     1 min                   010
      n03s1m01.nlf          3 s /     1 min                   011
      n05s1m01.nlf          5 s /     1 min                   012
      n10s1m01.nlf         10 s /     1 min                   013
      n20s1m01.nlf         20 s /     1 min                   014 
      n30s1m01.nlf         30 s /     1 min                   015
                                          
      n02s5m02.nlf          2 s /     5 min                   016
      n03s5m02.nlf          3 s /     5 min                   017
      n05s5m02.nlf          5 s /     5 min                   018
      n10s5m02.nlf         10 s /     5 min                   019
      n20s5m02.nlf         20 s /     5 min                   020
      n30s5m02.nlf         30 s /     5 min                   021
      n60s5m02.nlf         60 s /     5 min                   022
 
      n2h1m001.nlf        1 min /     5 min                   023
      n14h5m01.nlf        5 min /    60 min                   024 
 
      The filter gain of these lowpass  filters can be plotted with  program 
      PLOTFILT.
 
      Program DECIMATE reads the project name "pn" from file  "project"  and
      the necessary control parameters from file "pn".ini from the directory
      from which DECIMATE is called. The data to be  decimated  have  to  be 
      provided in the same directory; the name of the input data file has to
      be defined in the control parameter file "pn".ini. The decimated  data
      are written in standard ETERNA format to a file in the same directory.
      The name of the data output file has to  be  defined  in  the  control 
      parameter file "pn".ini. The length of the data files to be  decimated
      with DECIMATE is restricted to 10000 days = 30 years.  The  length  of 
      the numerical filters to be used with DECIMATE is restricted  to  4501
      coefficients.
 
      Section 13.2: Description of the project file "project"
      -------------------------------------------------------
 
      The ASCII project name file "project"  for  program DECIMATE  contains
      one single variable only, the project name  which  is  abbreviated  as
      "pn" in this manual. The project  name  is  an  alphanumerical  string 
      consisting of 8 characters at maximum. The project name is  read  from 
      the first 8 characters of  the  first  record  of  project  name  file 
      "project". The project name is used to define  the  control  parameter 
      file name "pn".ini for program DECIMATE.
   
      Section 13.3: Description of control parameter file "pn".ini    
      ------------------------------------------------------------    
                                                                               
      The control parameter file "pn".ini (where "pn" stands for the project  
      name defined in file "project")  defines  several  control  parameters 
      necessary for the operation of program DECIMATE. The control parameter
      file has an identical structure to the control parameter files used by
      programs RECTIDE, PREDICT, DETIDE, DESPIKE, PREGRED and ANALYZE. Some
      of the variables are identical to  those  used  by  programs  RECTIDE, 
      DETIDE, DESPIKE, DECIMATE, PREGRED and ANALYZE.  Thus  it  is  usually  
      possible to use the same control parameter file  for  data  recording, 
      data preprocessing and data analysis.                      
                                                                               
      All control parameter statements must begin with the control parameter   
      name (column 1...10) which ends with an equal sign. The control parame-  
      ter names must be spelled exactly with uppercase  letters  as  defined   
      below. Control parameters  with  unknown  names  will  be  ignored  by   
      program DESPIKE. The control parameter variables must be  input  after   
      the equal sign following the control parameter name, with  appropriate   
      format. The control parameter statements  must  not  have  a  specific   
      sequence. A record starting with # in column 1 will be ignored and may   
      be used to comment the control parameters.  All  characters  following   
      # in a record will be ignored and may be used to comment  the  control   
      parameters.                                                              
                                                                               
      The following control parameters are valid for program DECIMATE: 
                                                                       
      DECFILNAME= lowpass filter file name (CHARACTER*12). Valid lowpass
                  filter file names are         
 
                  n01s1m01.nlf   n02s1m01.nlf   n03s1m01.nlf  n05s1m01.nlf
                  n10s1m01.nlf   n20s1m01.nlf   n30s1m01.nlf  n02s5m02.nlf 
                  n03s5m02.nlf   n05s5m02.nlf   n10s5m02.nlf  n20s5m02.nlf
                  n30s5m02.nlf   n60s5m02.nlf   n2h1m001.nlf  n14h5m01.nlf
 
      INPFILNAME= input data file name (CHARACTER*12).                 
      OUTFILNAME= output data file name (CHARACETR*12).                
      SAMPLERATE= sampling interval in s.                              
      NUMBCHANEL= number of data channels to be decimated. NUMBCHANEL is 
                  restricted to 9.             
      DECIMATION= integer decimation factor. The decimation factor is
                  restriced by the filter cutoff period. 
                                                                       
 
      Section 14: Description of program DETIDE                  
      -----------------------------------------                  
                                                                               
      The program DETIDE is a Fortran 90 program for  the  preprocessing  of   
      high rate earth tide observations (1 s ... 30 min sampling  interval).
      DETIDE was a part of the larger program PRETERNA  developed  by  H.-G. 
      Wenzel in 1991, and is described in Wenzel (1994). Program DETIDE uses
      the ETERNA standard format  for  the  storage  and  exchange  of  high 
      resolution and high rate earth tide data described in  Wenzel  (1995).   
      The output file of program DETIDE may directly be used to  input  into   
      the graphical editor PREGRED or the despiking program DESPIKE  or  the
      decimation  program  DECIMATE  or  the  earth  tide  analysis  program 
      ANALYZE.
                      
      The earth tide preprocessing programs DETIDE,  DESPIKE,  DECIMATE  and
      PREGRED allow the preprocessing of high rate earth tide data. The data
      preprocessing is carried out using a remove-restore technique (similar
      to the remove-restore technique known from gravity  field  transforma-
      tion). We at first remove all known signals (i.e. computed model tides
      and influences of meteorological parameters (e.g. air  pressure).  The   
      remaining residual signal (including the earth tide sensor's drift) is
      then cleaned (destepped, despiked and degapped), and the known signals
      are subsequently added back to the cleaned residual signal. Errors  in 
      the model earth tide signal and model meteorological parameter contri-
      bution  will  by  this  procedure  be  cancelled  to  a  large  extend
      (naturally except for the gaps, where we use the model earth tide  and
      model meteorological parameter contribution  to  interpolate the  data
      during the gap). The corrected samples at the original sampling inter-
      val may finally be numerically  filtered  and  decimated  to  a  lower 
      sampling interval. The data preprocessing should  be  carried  out  in 
      different steps, and one should check  the  result  of  the  different  
      steps by plotting the data using programm PLOTDATA or PREGRED. 
 
      There exist several different procedures for the preprocessing of one
      data set; the ETERNA 3.40 package should be flexible enough to  allow
      to establish very special data preprocessing preocdures. 
      
      The experience of several groups with  PRETERNA  has  shown,  that  an 
      optimal data preprocessing my be the following procedure:
 
      - Decimate the raw data for  e.g. one  month  from  e.g. 1 s  sampling 
        interval to 1 min sampling interval using DECIMATE,
      - Check the 1 min observations  with  the  graphical  editor  PREGRED,
        especially for the meteorological channels. Remove spikes and  steps
        and interpolate gaps in the meteorological channels with PREGRED.
      - Calibrate and detide the earth tide data with DETIDE,
      - Check the detided earth tide data with the graphical editor PREGRED.
        Parts of the data with large distortions e.g. due  to  earth  quakes
        should be eliminated (make a gap). Slow steps and parts with several
        steps in short time should be removed.
      - Despike the earth tide data with program DESPIKE.
      - Check the  despiked  earth  tide  data  with  the  graphical  editor 
        PREGRED.
      - Decimate the despiked earth tide data with DECIMATE  to  e.g.  5 min
        resampling interval.
      - Collect several 5 min data sets and merge these data together,
      - Decimate the merged data set to e.g. 1 hours resamling interval,
      - Perform an analysis of the 1 hour data set using program ANALYZE.
   
      There may be used with program DETIDE:                                
                                                                               
      - as observations:                                                       
         tidal potential, gravity tides, tilt tides, vertical displacements,   
         horizontal displacements,  vertical strain tides, horizontal strain   
         tides, areal strain tides, shear strain tides, volume strain tides,   
         and oceanic tides.                                                    
                                                                               
      - high rate earth tide data for up to 100 days, containing up to      
        500 gaps at maximum (with a maximum length of 30 000 samples),
                                                                               
      - up to 85 tidal wave groups,                                           
                                                                               
      - up to 6 additional meteorological parameter,                           
                                                                               
      - 7 different tidal potential catalogues  (Doodson  1921, Cartwright -   
        Tayler-Edden-Doodson 1973, Buellesfeld 1985,  Tamura 1987,  Xi 1989,
        Roosbeek 1996 and Hartmann and Wenzel 1995),    
                                                                               
      There  have  been  made  significant  improvements  and  changes  over   
      previous program versions. Program DETIDE has been extensively  tested
      using model data and observed data sets  for  gravity  only.  However, 
      many routines are identical in DETIDE, PREDICT and ANALYZE,  and  have   
      been tested with gravity, tilt, strain, and ocean tides. 
                                                                               
      The structure of the data files is identical for programs DECIMATE,
      DETIDE, DESPIKE, PREGRED and PLOTDATA.
                                                                               
      The project file PROJECT is described in section B5.6 and the  control   
      parameter file "project name".INI is described in section B5.7.          
 
     Section 14.1: Calibration of data
     ---------------------------------
                                                                               
      The name of the formatted data input file to program DETIDE has to  be   
      defined in the control parameter file "pn".ini. When reading the  data   
      from the data input file, the sequence of the data is checked. No  gap
      in the sequence of the data is allowed; if there  exist  gaps  in  the
      sequence of the data, these gaps have to be  filled with flagged  data
      which means that a sample value of 99999.999 has to be put in the data 
      input file at the proper  positions.  The  flagging  of  gaps  may  be 
      carried out using an alphanumerical standard editor, which is  a  very
      tedious task for longer gaps. We recommend to use the graphical editor
      PREGRED to check for gaps and to flagg gaps.                          
                                                                               
      We use 3rd degree polynomials with time  dependent  coefficients  (see
      below) to calibrate the earth tide signal from the raw data unit (e.g. 
      Volt) to the component dependent earth tide signal unit                 
                                                                               
         m**2/s**2          for tidal potential,                               
         nm/s**2            for tidal gravity,                                 
         mas                for tidal tilt,                                    
         mm                 for tidal vertical displacement,                   
         mm                 for tidal horizontal displacement,                 
         nstr = 10**-9 str  for tidal vertical strain,                         
         nstr = 10**-9 str  for tidal horizontal strain,                       
         nstr = 10**-9 str  for tidal areal strain,                            
         nstr = 10**-9 str  for tidal shear strain,                            
         nstr = 10**-9 str  for tidal volume strain, and                       
         mm                 for ocean tides.                                   
                                                                               
      The meteorological signals have to be calibrated  from  the  raw  data
      unit to the proper unit of the meteorological parameter  (e.g. hPascal
      for air pressure, deg Celsius for temperature),  where  a  3rd  degree 
      calibration polynomial (with time variable parameters, see below)  may
      be used. If there is known a linear  calibration  factor  only  for  a 
      specific meteorological sensor, set the 2nd and 3rd degree  polynomial
      calibration coefficients simply to zero.  The  calibrated  earth  tide 
      observation will be stored as channel 1 and the calibrated  meteorolo-
      gical observations will be stored as channels 2..NRAW in the formatted
      data output file of program DETIDE.                          
                                                                               
      Section 14.2: Computation of model tides and residual signal             
      ------------------------------------------------------------             
                                                                               
      The calibrated data are used for this computation  step.  DETIDE  will 
      compute model earth tides for the specified earth  tide  component  at 
      the specified sampling interbal, delayed for the instrumental time lag
      of the earth tide sensor (read from the calibration  parameter  file). 
      These model earth tides are stored as channel  NRAW+1  in  the  output  
      output data file "pn".TID of computation step 3. The model earth tides   
      are computed using one of seven available tidal potential  catalogues,   
      with observed amplitude factors and phase leads defined in the control   
      parameter file "pn".INI. As tidal potential catalogue,                   
                                                                               
      - the Doodson (1921) tidal potential catalogue with 379 waves,           
      - the Cartwright-Tayler-Edden-Doodson (1973) tidal potential catalogue   
        with 505 waves,                                                        
      - the Buellesfeld (1985) tidal potential catalogue with 656 waves        
      - the Tamura (1987) tidal potential catalogue with 1200 waves,          
      - the Xi (1989) tidal potential catalogue with 2934 waves,
      - the Roosbeek (1996) tidal potential catalogue with 6499 waves,
      - the Hartmann and Wenzel (1995) tidal potential catalogue with
        12935 waves
                                                                              
      may be selected. For the contribution of meteorological parameters,
      we use a linear regression model with regression parameters defined 
      in the calibration data file. The residual signal is computed by      
                                                                            
        ch(last) = ch(1) - sum over [chanel(i)*METCORFACT(i)], i=2,NRAW+1     
 
      and stored as last channel in the output data file of program DETIDE.
                                                                               
      Section 14.3: Description of the project file "project"
      -------------------------------------------------------
 
      The ASCII project name  file "project"  for  program  DETIDE  contains
      one single variable only, the project name  which  is  abbreviated  as
      "pn" in this manual. The project  name  is  an  alphanumerical  string 
      consisting of 8 characters at maximum. The project name is  read  from 
      the first 8 characters of  the  first  record  of  project  name  file 
      "project". The project name is used to define  the  control  parameter 
      file name "pn".ini for program DETIDE.
   
      Section 14.4: Description of control parameter file "pn".ini    
      ------------------------------------------------------------    
                                                                               
      The control parameter file "pn".ini (where "pn" stands for the project  
      name defined in file "project")  defines  several  control  parameters 
      necessary for the operation of program DETIDE.  The  control parameter
      file has an identical structure to the control parameter files used by
      programs  RECTIDE,  PREDICT,  DECIMATE,  DESPIKE, PREGRED and ANALYZE.
      Some of the variables are identical to those used by programs RECTIDE,
      PREDICT, DECIMATE, DESPIKE, PREGRED and ANALYZE.  Thus it  is  usually  
      possible to use the same control parameter file  for  data  recording, 
      data preprocessing and data analysis.                      
                                                                               
      All control parameter statements must begin with the control parameter   
      name (column 1...10) which ends with an equal sign. The control parame-  
      ter names must be spelled exactly with uppercase  letters  as  defined   
      below. Control parameters  with  unknown  names  will  be  ignored  by   
      program DETIDE. The control parameter variables must be  input  after   
      the equal sign following the control parameter name, with  appropriate   
      format. The control parameter statements  must  not  have  a  specific   
      sequence. A record starting with # in column 1 will be ignored and may   
      be used to comment the control parameters.  All  characters  following   
      # in a record will be ignored and may be used to comment  the  control   
      parameters.                                                              
                                                                               
      The following control parameters are valid for program DECIMATE: 
 
      SENSORNAME= name of the earth tide sensor (CHARACTER*10)
      INPFILNAME= name of the raw data input file (CHARACTER*12)
      CALFILNAME= name of the calibration data file (CHARACTER*12)
      OUTFILNAME= name of the output data file (CHARACTER*12)
      NUMRAWCHAN= number of raw data channels (INTEGER)
      SAMPLERATE= Data sample interval in s (col. 12 ... 26, integer).         
      STATLATITU= ellipsoidal latitude of the station in degree referring to   
                  WGS84 reference system (col.12...26).                        
      STATLONITU= ellipsoidal longitude of the station in  degree  referring   
                  to WGS84 reference system (col.12...26).                     
      STATELEVAT= ellipsoidal height of the station in meter referring to      
                  WGS84 reference system (col.12...26).                        
      STATGRAVIT= gravity of the station in m/s**2, necessary for tidal tilt   
                  only (col. 12...26,  real  with  decimal  point).  If  the   
                  gravity is unknown, use a value of less than 1.0  and  the   
                  program will  compute  and  subsequently  use  the  normal   
                  gravity value referring to GRS80 reference system.           
      STATAZIMUT= azimuth of the  instrument  in  degree  decimal,  reckoned   
                  clockwise from north (col. 12 ... 26,  real  with  decimal   
                  point). This parameter is used for tidal tilt,  horizontal   
                  displacement and horizontal strain only.                     
      TIDALCOMPO= Earth tide component (col. 12 ... 26, integer).              
                  = -1 for tidal potential in m**2/s**2.                       
                  =  0 for tidal gravity in nm/s**2.                           
                  =  1 for tidal tilt in mas, at azimuth STATAZIMUT.           
                  =  2 for tidal vertical displacement in mm.                  
                  =  3 for tidal horizontal displacement in mm at              
                       azimuth STATAZIMUT.                                     
                  =  4 for tidal vertical   strain in 10**-9 = nstr.           
                  =  5 for tidal horizontal strain in 10**-9 = nstr,           
                       at azimuth STATAZIMUT.                                  
                  =  6 for tidal areal  strain in 10**-9 = nstr.               
                  =  7 for tidal shear  strain in 10**-9 = nstr.               
                  =  8 for tidal volume strain in 10**-9 = nstr.               
                  =  9 for ocean tides in mm.                                  
                  The computed model  tides  will  be  given  in  the  units   
                  defined above.                                               
      TIDALPOTEN= parameter for the tidal potential catalog to be used.        
                  1 = Doodson (1921) tidal potential catalog,                  
                  2 = Cartwright-Tayler-Edden (1973) tidal potential catalog   
                  3 = Buellesfeld (1985) tidal potential catalog,            
                  4 = Tamura (1987) tidal potential catalog,
                  5 = Xi (1989) tidal potential catalog,
                  6 = Roosbeek (1996) tidal potential catalog,
                  7 = Hartmann and Wenzel (1995) tidal potential catalog.
                  As default, TIDALPOTEN= 7 is used.                   
      AMTRUNCATE= Amplitude threshold for the tidal potential  catalogue  in
                  m**2/s**2. Only tidal waves with amplitudes exceeding  the
                  amplitude threshold are used  for  the  computation.  This 
                  reduces the execution time, but also the accuracy  of  the
                  computed tidal signales. For mean latitudes, the  relation
                  between amplitude threshold and gravity tide accuracy   is
                  for the Hartmann and Wenzel (1995) tidal potential catalog
 
                           threshold              rms error             
                                                 [nm/s**2]          
 
                             1.D-01                88.40              
                             1.D-02                14.40              
                             1.D-03                 2.25              
                             1.D-04                 0.44              
                             1.D-05                 0.068             
                             1.D-06                 0.011             
                             1.D-07                 0.002             
                             1.D-08                 0.001             
                             1.D-09                 0.001              
                             1.D-10                 0.001
      RIGIDEARTH= Parameter for using a rigid earth model for  the  adjusted   
                  earth  tide  parameters (col. 12...26, integer,  1 = yes).
                  For program DETIDE, parameter RIGIDEARTH should, always be
                  1.0000 except for the processing of benchmark series.
      TIDALPARAM= wave group selection for a priori model tide  computation.   
                  The number of wavegroups is restricted to  85.  The  tidal   
                  parameters may be taken from an  earth  tide  analysis  of   
                  another data set in the same station,  or  from  a  neigh-   
                  bouring station, or from synthetic gravity tide parameters   
                  (Timmen and Wenzel 1994, 1995), or from a guess only.        
                  Col.12..21: DFRA#=  start frequency of the wave  group  in   
                                      cpd (real).                              
                  Col.22..31: DFRE#=  end   frequency of the wave  group  in   
                                      cpd (real).                              
                  Col.32..41: DGAM0#= amplitude factor  of  the  wave  group   
                                      (real).                                  
                  Col.42..51: DPHI0#= phase lead of the wave  group  in  deg   
                                      (real).                                  
                                                                               
      The wave group selection for model tide  computation  should  be  made   
      according to the wave group selection for the analysis of  Earth  tide   
      data (see the program manual for ANALYZE); the list given below  is  a
      proposal only). The frequencies are given in cycles per day (cpd).      
                                                                               
      group        > 1 month            > 6 months          > 1 year           
                   from      to         from      to        from      to       
                  [cpd]   [cpd]        [cpd]   [cpd]       [cpd]   [cpd]       
                                                                               
      SA             -       -            -       -       0.001379  0.004107   
      SSA            -       -      0.004108  0.020884    0.004108  0.020884   
      MM      0.020885  0.054747    0.020885  0.054747    0.020885  0.054747
      MF      0.054748  0.091348    0.054748  0.091348    0.054748  0.091348
      MTM     0.091349  0.501369    0.091349  0.501369    0.091349  0.501369   
      Q1      0.501370  0.911390    0.501370  0.911390    0.501370  0.911390   
      O1      0.911391  0.947991    0.911391  0.947991    0.911391  0.947991   
      M1      0.947992  0.981854    0.947992  0.981854    0.947992  0.981854
      P1             -       -      0.981855  0.998631    0.981855  0.998631   
      S1             -       -            -       -       0.998632  1.001369   
      K1      0.981855  1.023622    0.998632  1.023622    1.001370  1.004107   
      PSI1           -       -            -       -       1.004108  1.006845   
      PHI1           -       -            -       -       1.006846  1.023622   
      J1      1.023623  1.057485    1.023623  1.057485    1.023623  1.057485   
      OO1     1.057486  1.470243    1.057486  1.470243    1.057486  1.470243   
      2N2     1.470244  1.880264    1.470244  1.880264    1.470244  1.880264   
      N2      1.880265  1.914128    1.880265  1.914128    1.880265  1.914128   
      M2      1.914129  1.950419    1.914129  1.950419    1.914129  1.950419   
      L2      1.950420  1.984282    1.950420  1.984282    1.950420  1.984282   
      S2      1.984283  2.451943    1.984283  2.002736    1.984283  2.002736  
      K2             -       -      2.002737  2.451943    2.002737  2.451943   
      M3M6    2.451944  7.000000    2.451944  7.000000    2.451944  7.000000
                                                                               
      Section 14.5: Description of the calibration parameter file
      -----------------------------------------------------------
 
      The calibration parameter file contains a file header and a file body. 
      The file header contains alphanumeric information to describe the file
      contents, and ends with C********* in cols. 1...10.
 
      The first record of the file body contains
      in cols.  1..11:  RAWCHAUNIT=
      in cols. 12..14:  blanks
      in cols. 15..24:  unit of raw data in channel 1 (CHARACTER*10)
      in cols. 25..34:  unit of raw data in channel 2 (CHARACTER*10)
      in cols. 35..44:  unit of raw data in channel 3 (CHARACTER*10)
      in cols. 45..54:  unit of raw data in channel 4 (CHARACTER*10) ...
 
      The second record of the file body contains
      in cols.  1..11:  CALCHAUNIT=
      in cols. 12..14:  blanks
      in cols. 15..24:  unit of calibrated data in channel 1 (CHARACTER*10)
      in cols. 25..34:  unit of calibrated data in channel 2 (CHARACTER*10)
      in cols. 35..44:  unit of calibrated data in channel 3 (CHARACTER*10)
      in cols. 45..54:  unit of calibrated data in channel 4 (CHARACTER*10) ...
 
      The third record of the file body contains
      in cols. 1...11: METCORFACT= 
      in cols. 12..14: blanks
      in cols. 15..24: correction factor for channel 1 (has to be 1.0)
      in cols. 25..34: correction factor for channel 2
      in cols. 35..44: correction factor for channel 3  
      in cols. 45..54: correction factor for channel 4 .... 
      
      Following the third record, there has to be given a table of 
      time depnendent 3rd degree calibration coefficients for all raw
      data channels. For each calibration epoch, you have to enter
      four records. 
      First calibration table record:
      in cols. 1... 8: date of the calibration in yyyymmdd (year,month,day).
      in cols.10...15: time of the calibration in hhmmss (hour,min,sec).
      in cols.16...25: 1st degree calibration coefficient for channel 1.
      in cols.26...35: 1st degree calibration coefficient for channel 2.
      in cols.36...45: 1st degree calibration coefficient for channel 3.
      in cols.46...55: 1st degree calibration coefficient for channel 4.
      Second calibration table record:
      in cols. 1... 8: date of the calibration in yyyymmdd (year,month,day).
      in cols.10...15: time of the calibration in hhmmss (hour,min,sec).
      in cols.16...25: 2nd degree calibration coefficient for channel 1.
      in cols.26...35: 2nd degree calibration coefficient for channel 2.
      in cols.36...45: 2nd degree calibration coefficient for channel 3.
      in cols.46...55: 2nd degree calibration coefficient for channel 4.
      Third calibration table record:
      in cols. 1... 8: date of the calibration in yyyymmdd (year,month,day).
      in cols.10...15: time of the calibration in hhmmss (hour,min,sec).
      in cols.16...25: 3rd degree calibration coefficient for channel 1.
      in cols.26...35: 3rd degree calibration coefficient for channel 2.
      in cols.36...45: 3rd degree calibration coefficient for channel 3.
      in cols.46...55: 3rd degree calibration coefficient for channel 4.
      Fourth calibration table record:
      in cols. 1... 8: date of the calibration in yyyymmdd (year,month,day).
      in cols.10...15: time of the calibration in hhmmss (hour,min,sec).
      in cols.16...25: time lag in sec of earth tide sensor.
      in cols.26...35: time lag in sec of meteorological sensor 1.
      in cols.36...45: time lag in sec of meteorological sensor 2.
      in cols.46...55: time lag in sec of meteorological sensor 3.
 
      After the first calibration table epoch, there might follow up to  299
      other calibration epochs. 
 
      Program DETIDE will linearily interpolate the  calibration  parameters
      and time lags from the table. If you dont know one of the  calibration
      parameters, it might make sense to simply put the  parameter  to  zero 
      (e.g. if your sensor is perfectly linear, you can set the 2nd  and 3rd
      degree coeffcients to zero). If the calibration parameters are  stable 
      with time, you can use one calibration epoch only. If  an  observation
      is taken before the date of the first calibration, DETIDE will use the
      calibration parameters of the first  calibration  epoch  to  calibrate 
      this observation. If an observation is taken after the  last  calibra-
      tion epoch, DETIDE will use the calibration  parameters  of  the  last
      epoch to calibrate this observation. 
 
      The calibration parameter file to be used with DETIDE is identical  to
      the calibration  parameter  file  for  programs PREGRED  and  DESPIKE.
   
      We give below a copy of  file  "sg102tes.cal"  provided  in  directory
      "\eterna33\1mindat":
  
      File              : sg02tes.cal
      Status            : 1996.09.02
      Contents          : calibration parameters for data file sg02tes.raw
                          cryogenic gravimeter SG102 and barometer BMG3 
                          at BFO Schiltach.
 
      Sequence is: 
      raw data:        unit       unit
      detided data:    unit       unit       unit    unit
      correction:      fact.1     fact.2    fact.3  
      calib.param:    linear     linear
                      quadratic  quadratic
                      cubic      cubic
                      time lag   time lag
      C**************************************************************
      RAWCHAUNIT=       Volt        Volt
      CALCHAUNIT=       nm/s**2      hPa    nm/s**2   nm/s**2
      METCORFACT=        1.0000   -3.2000    1.0000 
      19940602 120000 3439.7000  -23.0424
      19940602           0.0000    0.0000
      19940602           0.0000    0.0000
      19940602          57.6000   21.0300
      19940622 120000 3439.7000  -23.0424
      19940622           0.0000    0.0000
      19940622           0.0000    0.0000
      19940622          57.6000   21.0300
      99999999
 
      Section 15: Description of program DESPIKE                         
      ------------------------------------------                         
                                                                               
      The program DESPIKE is a Fortran 90 program for the  preprocessing  of   
      high rate earth tide observations (1 s ... 30 min sampling  interval).
      DESPIKE was a part of the larger program PRETERNA developed  by  H.-G. 
      Wenzel in 1991, and is described in  Wenzel  (1994).  Program  DESPIKE 
      uses the ETERNA standard format for the storage and exchange  of  high 
      resolution and high rate earth tide data described in  Wenzel  (1995).   
      The output file of program DESPIKE may directly be used to input  into   
      the graphical editor PREGRED or the plotting program PLOTDATA  or  the
      decimation  program  DECIMATE  or  the  earth  tide  analysis  program 
      ANALYZE.
                      
      The earth tide preprocessing programs DETIDE,  DESPIKE,  DECIMATE  and
      PREGRED allow the preprocessing of high rate earth tide data. The data
      preprocessing is carried out using a remove-restore technique (similar
      to the remove-restore technique known from gravity  field  transforma-
      tion). We at first remove all known signals (i.e. computed model tides
      and influences of meteorological parameters (e.g. air  pressure).  The   
      remaining residual signal (including the earth tide sensor's drift) is
      then cleaned (destepped, despiked and degapped), and the known signals
      are subsequently added back to the cleaned residual signal. Errors  in 
      the model earth tide signal and model meteorological parameter contri-
      bution  will  by  this  procedure  be  cancelled  to  a  large  extend
      (naturally except for the gaps, where we use the model earth tide  and
      model meteorological parameter contribution  to  interpolate the  data
      during the gap). The corrected samples at the original sampling inter-
      val may finally be numerically  filtered  and  decimated  to  a  lower 
      sampling interval. The data preprocessing should  be  carried  out  in 
      different steps, and one should check  the  result  of  the  different  
      steps by plotting the data using programm PLOTDATA or PREGRED. 
 
      The project name for program DESPIKE is read  from  the  project  file 
      "project", which has to be stored in the directory from which  DESPIKE
      is executed. The control parameters for  DESPIKE  are  read  from  the
      control parameter file  "pn".ini,  which  has  to  be  stored  in  the 
      directory from which DESPIKE is executed. 
 
      Section 15.1: Destepping, despiking and degapping
      -------------------------------------------------
 
      DESPIKE tries at first to destep the data (i.e. to find steps  and  to
      remove these steps). Half hourly averages of the residual  signal  are 
      computed together with their  standard  deviations.  With  these  half
      hourly averages, an iterative search for steps is carried out. We  use 
      a 1st degree least squares polynomial fitted to the half  hourly  mean
      values over 10 hours, with two different zero degree coefficients (one
      for the first half of the samples and one for the second half  of  the
      samples) being adjusted. The standard deviations of  the  half  hourly
      averages estimated from the averaging are used to derive  weights  for
      the step  adjustment. The  difference  between  the  two  zero  degree
      coefficients is the estimated step. A step is assumed to be found,  if
      it exceeds three times its estimated standard  deviation,  and  if  it
      exceeds the threshold STEPDETLIM  read  from  control  parameter  file
      "pn".INI (we don't want to correct very small steps in  case  of  very
      small noise). Within one iteration  loop  for  the  step  search,  the 
      data window for the fit is moved over the recorded time span with half
      an hour shift between successive adjustments. For the iteration  loop,
      the step with the maximum signal to noise ratio is assumed to be found 
      and subsequently corrected at the half hourly averages and the  sample 
      data. The iteration is continued until no more steps are found.         
                                                                               
      <-    half hourly averages           ->                                  
                                                                               
      x x x x x x x x x x o o o o o o o o o o        1st fit                   
        x x x x x x x x x x o o o o o o o o o o      2nd fit                   
          x x x x x x x x x x o o o o o o o o o o    3rd fit                   
            x x x x x x x x x x o o o o o o o o o o  4th fit                   
                                                                               
      It should be noted that the automatic destepping  procedure  described   
      above can generally detect and correct single large  steps  only.  For
      the correction of small steps or of several large steps,  the  use  of
      the graphical editor PREGRED is recommended.             
                                                                               
      After destepping, program DESPIKE is starting the despiking loop.  The
      despiking is carried out in an iterative search  for  and  removal  of   
      spikes using the original samples of the residual signal from the last 
      input data channel. A 2nd degree polynomial is fitted to the  residual 
      signal over 120 min time span,  and  the  maximum  residual  inside  a 
      central window of +/- 30 min around the central time  of  the  fit  is 
      tested against a  threshold.  This  threshold  is  determined  by  the 
      despiking limit SPIKDETLIM read from control parameter file  "pn".INI, 
      if it exceeds three times the rms of fit of  the  previous  iteration. 
      The maximum residual inside the central window exceeding the threshold
      is then corrected at the sample data, and the fit is repeated with the
      time span for the fit shifted by 15 min.   
                                                                               
      The threshold is initially set to 64 times  the  parameter  SPIKDETLIM   
      read from control parameter file "pn".INI, and then subsequently divi-   
      ded by two for each despiking loop.  Within  a  despiking  loop,  this   
      threshold remains constant. Within each despiking loop, the time  span   
      for the fit is shifted over the whole record of 1 min samples.  If  we   
      for instance input as despiking threshold SPIKDETLIM= 2 nm/s**2 for  a   
      gravimeter record, the  first  despiking  loop  uses  128  nm/s**2  as   
      threshold and removes spikes exceeding 128 nm/s**2 only. In the second   
      despiking loop, we use 64 nm/s**2 as threshold and  we  remove  spikes   
      exceeding 64 nm/s**2 only. The threshold is divided by  two  for  each   
      despiking loop, until it is below the parameter SPIKDETLIM and below 3   
      times the rms of fit of the previous despiking  loop.  This  procedure   
      enables an automatic adaption of the despiking limit to the  noise  of   
      the record for normal recording periods. But  for  periods  containing   
      large disturbances (e.g. earthquakes), these  disturbances  would  not   
      allow the despiking threshold to become normal. Therefore the input of   
      a hard despiking limit enables the  despiking  of  periods  containing   
      large disturbances. Another good procedure is to  delete  the  gravity   
      data during large disturbances, and to interpolate the artificial  gap   
      later on during the degapping.                                           
                                                                               
      Shifting of the despiking time span:                                     
                                                                               
               central window                                                  
      --------xxxxxxxxxxxxxxxx--------                           1st fit       
          --------xxxxxxxxxxxxxxxx--------                       2nd fit       
              --------xxxxxxxxxxxxxxxx--------                   3rd fit       
                  --------xxxxxxxxxxxxxxxx--------               4th fit       
                      --------xxxxxxxxxxxxxxxx--------           5th fit       
                          --------xxxxxxxxxxxxxxxx--------       6th fit       
                                                                               
                                                                               
      After despiking, DESPIKE looks for gaps  in  the  data  and  tries  to 
      interpolate data in the gaps. For  the  degapping,  DESPIKE  looks  at 
      first for gaps in the meteorological channels record and  subsequently
      interpolate the gaps in these channels by a 1st degree polynomial. For
      the gaps in the residual signal, we use a  least  squares  1st  degree 
      polynomial, where samples over 10 min are used on both  sides  of  the 
      gap. The number of gaps is restricted to 500, and the  maximum  length
      of a gap is restricted to 30 000 samples. Thus, the maximum gap length
      depneds on the sampling interval.                    
                                                                               
      Section 15.2: Description of the project file "project"
      -------------------------------------------------------
 
      The ASCII project name file "project"  for  program  DESPIKE  contains
      one single variable only, the project name  which  is  abbreviated  as
      "pn" in this manual. The project  name  is  an  alphanumerical  string 
      consisting of 8 characters at maximum. The project name is  read  from 
      the first 8 characters of  the  first  record  of  project  name  file 
      "project". The project name is used to define  the  control  parameter 
      file name "pn".ini for program DESPIKE.  
 
      Section 15.3: Description of control parameter file "pn".ini    
      ------------------------------------------------------------    
                                                                               
      The control parameter file "pn".ini (where "pn" stands for the project  
      name defined in file "project")  defines  several  control  parameters 
      necessary for the operation of program DESPIKE. This control parameter
      file has an identical structure to the control parameter files used by
      programs RECTIDE, PREDICT, DECIMATE, DETIDE, PREGRED and ANALYZE. Some
      of the variables are identical to  those  used  by  programs  RECTIDE, 
      DETIDE, DESPIKE, DECIMATE, PREGRED and ANALYZE.  Thus  it  is  usually  
      possible to use the same control parameter file  for  data  recording, 
      data preprocessing and data analysis.                      
                                                                               
      All control parameter statements must begin with the control parameter   
      name (column 1...10) which ends with an equal sign. The control parame-  
      ter names must be spelled exactly with uppercase  letters  as  defined   
      below. Control parameters  with  unknown  names  will  be  ignored  by   
      program DESPIKE. The control parameter variables must be  input  after   
      the equal sign following the control parameter name, with  appropriate   
      format. The control parameter statements  must  not  have  a  specific   
      sequence. A record starting with # in column 1 will be ignored and may   
      be used to comment the control parameters.  All  characters  following   
      # in a record will be ignored and may be used to comment  the  control   
      parameters.                                                              
                                                                               
      The following are valid control parameters for program DESPIKE:
 
      NUMBCHANEL= integer number of data channels, including the model
                  tide channel and the residual signal channel (col. 12..
                  21).
      INPFILNAME= input  data file name (col.12..23, 12 characters).
      OUTFILNAME= output data file name (col.12..23, 12 characters).
      CALFILNAME= calibration data file name (col.12..23, 12 characters).
      SENSORNAME= alphanumeric name of the earth tide sensor (col 12 ... 21,  
                  left bounded). SENSORNAME is restricted to 10  characters.  
      SAMPLERATE= input data sample interval in s (col. 12 ... 26, integer).   
                  The program DESPIKE is designed for 60 s sampling interval
                  but may also be used for  other  sampling  intervals.  The 
                  sampling interval should however not exceed 30 min. Please
                  note that  the  samples  have  to  be  given  e.g. at full
                  minutes if the sample interval is 60 s.    
      STEPDETLIM= Step detection threshold in earth tide units (col. 22  ...    
                  26, real with decimal point). There will be removed  steps    
                  exceeding STEPDETLIM and three times their estimated stan-    
                  dard deviation.   
      SPKIDETLIM= Despiking threshold in earth tide  units  (col. 22 ... 26,   
                  real with decimal point).              
                                                                               
      Section 15.4: Examples for data preprocessing
      ---------------------------------------------
 
      In directory "\eterna33\1mindat" we give  several  example  files  for 
      data preprocessing for the  three  projects "kal299tt", "sg102tes" and
      "bfhw9502". 
 
      Project "kal299tt" contains  gravity  data  recorded  with  gravimeter 
      LCR-G299 at station Karlrsuhe using 5 s sampling interval with program
      RECTIDE and online decimation to 1 min data. The raw  data are  stored  
      in file "kal299tt.raw". In the raw data file, several large spikes and
      one large step has been introduced in order to  test  data  processing
      software. The detided data using program DETIDE  (see  encl.  025  and
      026) are stored  in  file  "kal299tt.tid".  The  despiked  data  using 
      program  DESPIKE  (see encl.  027  and  028)   are   stored   in  file 
      "kal299tt.cor". The data decimated to 5 min resampling interval  using
      program DECIMATE are stored in file "kal299tt.fil" (see encl. 029  and
      030). The data decimated to 1 h sampling interval using a second  time
      program DECIMATE (but with a different filter, naturally)  are  stored
      in file "kal299tt.dec".
  
      Project "sg102tes" contains  gravity  data  recorded  with  gravimeter 
      GWR-SG102 (courtesy of Institut fuer Angewandte Geodaesie,  Frankfurt)
      at BFO Schiltach in 1994 using 5 s sampling interval.  The  data  were
      numerically filtered and decimated to 1 min resampling interval  after
      recording: these 1 min data are stored  in  file  "sg102tes.raw".  The
      detided data using program DETIDE are stored  in  file  "sg102tes.tid"
      (see encl. 031, 32 amd 33). Please note  a  very  small  step  of  1.4
      nm/s**2 in encl. 033, which could not be detected with  the  automatic
      despiking program DESPIKE, but  has been (together  with  other  steps
      and spikes) edited with program PREGRED (encl. 034). The output  data 
      of the automatic despiking of the manually edited  file "sg102tes.cor"
      using program DESPIKE is stored in file "sg102tes.des". The  despiking 
      of disturbances during a large  earthquake  (see  encl. 035  and  036) 
      shows good supression of short periodic noise, but reveals also a step
      and long periodic disturbances  (perhaps  due  to  clipping  of  large
      voltages in the  data  aquisition).  These  longperiodic  disturbances 
      have to be removed with program PREGRED (see encl. 037) before decima-
      tion ofthe data. The data decimated to 5 min resampling interval  (see
      encl. 038 and 039) are  stored
      in file "sg102tes.fil", the data decimated to 1 h resampling  interval
      are stored in file "sg102tes.dec".
 
      Project  "bfhw9502"  contains  sythetic  gravity  data  computed  with 
      program PREDICT  at  1  min  sampling  interval.  The  gravity  signal 
      includes gravity tides using the  Hartmann  and  Wenzel  (1995)  tidal 
      potential catalogue, pole tides,  LOD  tides,  plus  three  additional 
      channels (sine waves with periods 5.0 h, 5.5 h  and  7.0 h).  We  have 
      built into the data one step, several spikes  and  several  gaps.  The 
      synthetic  gravity   signal  (see   encl.   042)  is  stored  in  file 
      "bfhw9502.rrr" and the detided gravity  signal  using  program  DETIDE 
      (see encl. 043) is stored in file "bfhw9502.tid". The despiked detided
      data after application of program DESPIKE (see encl. 044) is stored in 
      file "bfhw9502.cor". The data decimated to 5 min  resampling  interval
      (encl. 046, please have in mind the  quantization  interval  of  0.001 
      nm/s**2  =  0.1  ngal)  using  program  DECIMATE  are  stored in  file 
      "bfhw9502.fil", and the data  decimated to  hourly  sampling  interval
      using a second time program DECIMATE are stored in file "bfhw9502.dec". 
 
      Section 16: Description of program PREPLOT                               
      ------------------------------------------                               
                                                                               
      The program PREPLOT has been written in Microsoft QuickBasic  4.0  and   
      may be used to plot the data of individual channels of  selected  data   
      files on the screen of the pc. There might be  visualized  raw,  cali-   
      brated, detided or preprocessed 1 min data. The executable file 
      "preplot.exe" stored in directory "\eterna33\bin" may be executed by
      entering
 
                 <preplot>
 
      from any directory. The program may be (modified and)  interpreted  by
      by calling the Microsoft QuickBasic interpreter available under MS-DOS   
      4.0 upwards by entering                                            
                                                                               
                  <qbasic preplot>                                             
                                                                               
      from keyboard. The program asks for the input of  some  parameters  to   
      be entered into the program via keyboard. The data of  one  individual   
      channel is displayed in white color, whereas the numerically differen-   
      tiated data of the channel is displayed in red on the same screen. The   
      initial time window is 1 day, but the screen may be zoomed  in  either   
      horizontal or vertical direction. The screen may be dumped to a matrix   
      printer (e.g. EPSON LQ850) by initializing the MS-DOS routine GRAPHICS   
      in the start file AUTOEXEC.BAT and hitting keys                          
                                                                               
             SHIFT and PRNT                                                    
                                                                               
      from keyboard.
                                                             
      Section 17: Description of program ANALYZE
      ------------------------------------------
 
      The program ANALYZE is a Fortran 90 program for the analysis of  earth
      tide observations. It follows a combination of the ideas of  Chojnicki
      (1973), Schueller (1976) and Wenzel (1976, 1977, 1994). ANALYZE is the
      successor of the former program ETERNA.                                  
                                                                              
      There have been made significant improvements and changes compared  to   
      previous versions of ANALYZE. The program has been tested  extensively
      using model data and observed  data  sets  for  tidal  gravity,  tilt, 
      vertical strain, horizontal strain, volume strain, and ocean tides.
                                                                              
      The program allows the adjustment of  tidal parameters, meteorological  
      regression parameters and pole tide  regression  parameters,  and  the   
      computation of a Fourier amplitude spectrum of the residuals. To  care   
      for instrumental drifts, either drift elimination by highpass  filter-   
      ing or drift modelization by Tschebyscheff  polynomials  is  possible.   
      The stacking of instruments  with different precision and instrumental   
      characteristics at the same station is possible by applying individual   
      instrumental time lag corrections, scale factors and weights for indi-   
      vidual observation blocks. For the least squares adjustment  of  tidal   
      parameters, the unity and Hann window  may  be  applied  as  weighting   
      function (e.g. Schueller 1976).  For  the  tidal  waves  of  different   
      degree and order, a priori amplitude factors from an approximation  to   
      the Wahr-Dehant-Zschau model (Dehant, V. 1987) for an elliptical, uni-   
      formly rotating oceanless Earth with liquid  outer  core  and  viscous   
      mantle may be used to represent the different response of the  elastic   
      Earth to tidal waves of different degree and  order  (but  nearly  the   
      same frequency) within one wave group. For these  a  priori  amplitude   
      factors, the nearly diurnal free wobble resonance in the diurnal  band   
      is taken into account from Wahr's 1981  model.  The  printed  adjusted   
      tidal parameters are scaled  by  the  a  priori  body  tide  amplitude   
      factors of the main wave in the group, just as they  would  have  been   
      obtained referring to a rigid Earth model.                               
                                                                               
      When using unfiltered observations, tidal  parameters  from  long-  to   
      shortperiodic waves may be adjusted.  Observations  of  meteorological 
      parameters (e.g. air pressure) may be input, which will be used to fit   
      a  simple  linear  regression  model  to  the  earth tide observation.   
      Additionally, Tschebyscheff  polynomial  bias  parameters  of  varying   
      degree for the individual data blocks may  be  adjusted  in  order  to   
      approximate the drift. The Tschebyscheff polynomials have been chosen,   
      because  they  are  orthogonal  with respect to normalized time (range   
      +/- 1)  and  have  much  better  numerical  properties  than  ordinary   
      polynomials. The  degree of bias polynomials is not restricted, except   
      from  the  total  number  of  unknowns.  The number of polynomial bias   
      parameters  per  individual  block  should  not exceed the blocklength   
      divided  by the longest period of the tidal wave, for which parameters   
      will be adjusted. For a  block  with  N  observations,  numbered  from  
      J=1...N, the adjusted drift for observation no. J may be computed from
      the adjusted Tschebyscheff polynomial bias parameters DBIAS(1...NBIAS)
      (listed in the print file, containing the parameters  for  degree 0...
      NBIAS-1) by                                                
                                                                               
            DTN=(DBLE(J-1)-DBLE(N-1)*0.5D0)/(DBLE(N-1)*0.5D0)                  
            DAK(1)=1.D0                                                        
            DAK(2)=DTN                                                         
            DRIFT=DBIAS(1)*DAK(1)+DBIAS(2)*DAK(2)                              
            DO 4560 I=3,NBIAS                                                  
            DAK(I)=2.D0*DTN*DAK(I-1)-DAK(I-2)                                  
     4560   DRIFT=DRIFT+DBIAS(I)*DAK(I)                                        
                                                                                
      Although the adjustment of Tschebyscheff  polynomials  has been tested   
      with simulated data up to degree 100, it is  recommended  to  use  low   
      degree polynomials only, especially when using the Hann window for the   
      least squares adjustment. In any case, the drift approximation  should   
      be controlled by e.g. plotting the residuals after adjustment.           
                                                                               
      When using highpass filtered observations, the earth tide observations   
      are  splitted  into a  low  frequency part (below about 0.8 cycles per   
      day) and a  high  frequency  part  (above  about 0.8  cycles  per day)   
      by  numerical  filtering. The least  squares adjustment of tidal para-   
      meters for diurnal to sixthdiurnal  wave  groups  is  carried  out  by   
      using the highpass filtered earth tide observations.  Observations  of   
      meteorological   parameters  (e.g.  air  pressure)  may  be  input  as
      additional channels, which will (after highpass  filtering  equivalent
      to the earth tide  observations)  be  used  to  fit  a  simple  linear 
      regression model to the earth  tide  observations.  For  the  adjusted
      tidal parameters, the frequency transfer function of the used highpass
      filter is corrected.    
                                                                               
      The highpass filtering is in fact done in the program  by  application 
      of  a  symmetrical  numerical  lowpass  (or  bandreject)  filter,  and 
      subtracting   the  lowpass  filtered  observation  from  the  original
      observation. This is equivalent to highpass (or  bandpass)  filtering.
      There are supplied different symmetrical numerical  non-recursive  FIR
      (finite impulse response) filters  for  different  sampling  intervals 
      (see below), which  -  due to their exact symmetry - do  not  generate
      any phase shift. The quality of the filters may be estimated from  the
      transfer function of the filters, which can be plotted  using  program
      PLOTFILT. 
                                                                               
      In contrast  to  all  previous  tidal  analysis  programs,  which were 
      restricted to hourly sampling interval, we allow for  program  ANALYZE
      for artificial smapling intervals,  e.g.  1 min ... 1  hour.  The  re- 
      commended sampling intervals for ANALYZE simply depend  on the numeri-
      cal filters which are available in the current version, and are 1 min,
      5 min, 10 min, 20 min, 30 min and 1 hour.
 
      The following numerical filters are available for ANALYZE:
  
      n1h1h001.nlf      lowpass for ANALYZE, 1 h sampl. (old filter no. 1) 
      n1h1h002.nlf      lowpass for ANALYZE, 1 h sampl. (old filter no. 2)
      n1h1h003.nlf      lowpass for ANALYZE, 1 h sampl. (old filter no. 3)
      n1h1h004.nlf      lowpass for ANALYZE, 1 h sampl. (old filter no. 4)
      n1h1h005.nlf      lowpass for ANALYZE, 1 h sampl. (old filter no. 5)
      n1h1h006.nlf      lowpass for ANALYZE, 1 h sampl. (old filter no. 6)
      n1h1h007.nlf      lowpass for ANALYZE, 1 h sampl. (old filter no. 7)
      n1h1h008.nlf      lowpass for ANALYZE, 1 h sampl. (old filter no. 8)
      n05m05m2.nlf      lowpass for ANALYZE,  5 min sampling interval
      n10m10m2.nlf      lowpass for ANALYZE, 10 min sampling interval
      n20m20m2.nlf      lowpass for ANALYZE, 20 min sampling interval
      n30m30m2.nlf      lowpass for ANALYZE, 30 min sampling interval
      n60m60m2.nlf      lowpass for ANALYZE, 60 min sampling interval
 
      In general, the tidal  parameters adjusted from highpass filtered data    
      do not change significantly when using different  filters;  but  there    
      exist small changes  at  the  residuals  and  the  estimated  standard    
      deviations of the tidal parameters. 
                                                                                 
      The error  estimation for the adjusted tidal parameters is done in two    
      different ways with ANALYZE:                                               
                                                                                 
      - by least squares  adjustment procedure,  which  generally gives  too    
        optimistic estimates due to the neglection of error correlations,       
                                                                                 
      - by Fourier amplitude spectrum of the residuals  of  the  adjustment,    
        using average spectral amplitudes over individual tidal  bands (e.g.    
        Wenzel 1976, 1977). Please have in mind, that the  Fourier amplitude    
        spectrum at frequencies below 0.8 cpd (the filter cut-off frequency)    
        is of no meaning when using the highpass filter option!                 
                                                                                 
      The error estimation by Fourier amplitude spectrum of the residuals is    
      known to give more realistic error estimations  (Wenzel  1976,  1977),    
      especially when adjusting simultaneously long- and shortperiodic tidal    
      parameters. The error estimation procedure uses average noise amplitu-    
      des from the computed residual spectrum in 0.1  deg/h  =  0.00666  cpd    
      interval :                                                                
                                                                                
       0.1... 2.9 deg/h = 0.007...0.193 cpd for longperiodic waves,             
      12.0...17.9 deg/h = 0.800...1.193 cpd for diurnal tidal waves,            
      26.0...31.9 deg/h = 1.733...2.127 cpd for semidiurnal waves,              
      42.0...47.9 deg/h = 2.800...3.193 cpd for terdiurnal  waves,              
      57.0...62.9 deg/h = 3.800...4.193 cpd for quaterdiurnal waves.            
                                                                                
      For longperiodic waves, we use the frequency dependent noise model        
                                                                                 
                             DMSE0                                              
                 noise(f) =  -----                                              
                               f                                                
                                                                                
      with f = frequency in cycles per day. The constant DMSE0 is computed      
      by taking the average of                                                  
                                                                                
                          noise(f)*f                                            
                                                                                
      between 0.1 ... 2.9 deg/h = 0.007 ... 0.193 cpd.                          
                                                                                
      The standard deviations computed by the least squares adjustment are      
      scaled (up or down) by the relation                                       
                                                                                 
                           noise(f)                                             
                 scale =   -------- ,                                           
                              wn                                                
                                                                                
      where wn is the estimated Fourier amplitude  of  white  noise  in  the    
      frequency band 0.01...6.0 cpd.                                            
                                                                                
      There has been built in a quick-look option,  which  carries  out  the    
      adjustment of parameters and error estimation by least squares method,    
      but computes  neither  residuals nor the spectrum of residuals. In any    
      case, this option should  only  be  used for a quick-look of the data,    
      and the final  analysis  should  be made using  the  full  computation    
      option (QUICKLOOKA=0).                                                    
                                                                                
      There may be used with ANALYZE                                           
                                                                               
      - up to 500 projects, whith the names of the projects to be defined in
        file "project".
 
      - as observations:                                                        
        tidal potential, gravity tides, tilt tides, vertical displacements,    
        horizontal displacements,  vertical strain tides, horizontal strain    
        tides, areal strain tides, shear strain tides, volume strain tides,    
        and oceanic tides.                                                     
                                                                               
      - unlimited number of hourly earth tide observations within one  block   
        (i.e. without gap),                                                    
                                                                               
      - up to 300 blocks of data (i.e. data spans without interruption),       
                                                                               
      - up to 175 unkown parameters,                                           
                                                                               
      - up to 85 tidal wave groups (i.e. up to 170 tidal parameters),          
                                                                               
      - up to 6 additional meteorological parameters,                          
                                                                               
      - either no highpass filtering or highpass filtering,                    
                                                                               
      - different numerical filters of different length and quality,
        there can even be used numerical filters computed by the user,         
                                                                               
      - 7 different tidal  potential  catalogs  (Doodson  1921,  Cartwright-   
        Tayler-Edden-Doodson 1973, Buellesfeld 1985, Tamura  1987,  Xi 1989,
        Roosbeek 1996 and Hartmann and Wenzel 1995 tidal potential catalog).
                                                                               
      Section 17.1: Description of project name file "project"                 
      --------------------------------------------------------                 
                                                                               
      The ASCII project name file "project" contains the project  names  for
      program ANALYZE. There might be defined one single  project  only,  or
      up to 500 projects within one "project" file. In  this  case,  ANALYZE
      performs subsequently the earth tide analyses  for  all  the  projects
      within one batch run. The project name is abbreviated as "pn" in  this
      manual. The project name is an alphanumerical string consisting  of  8 
      characters at maximum. The project names are read  from  the  first  8
      characters of the non-empty records of project  name  file  "project". 
      The project names are used in ANALYZE to define the control  parameter
      file names and the names of data input and output files.                 
                                                                               
      The program ANALYZE uses the following file name convention:      
                                                                               
      - control parameter file:           "project name".ini                   
      - hourly data input file:           "project name".dat                   
      - print file:                       "project name".prn                   
      - hourly residuals of adjustemnt:   "project name".res                   
      - spectrum of residuals:            "project name".far                   
      - adjusted parameters:              "project name".par
      - normal equation system file:      "project name".new                   
 
      Section 17.2: Description of the control parameter files "pn".INI       
      -----------------------------------------------------------------       
 
      For program ANALYZE, there might be defined a group of common  control
      parameters for all  analyses  in  file  "default.ini"  containing  the  
      common default control parameters, and there might be defined  indivi-
      dual control parameters for each project  in  the  individual  control
      parameter file "pn".ini. The control parameter files "pn".ini are used   
      by programs RECTIDE, DECIMATE, DETIDE, DESPIKE, PREGRED,  PREDICT  and
      ANALYZE. Because these programs use partly the same control parameters
      but are written  in  different  languages,  we  use  a  common  simple
      structure to input the control parameters to the individual  programs.   
      A control parameter file may consist of control parameter records  and   
      and of comment records. A comment  record  starts  with  symbol  #  in   
      column 1, and gives just an explanation. A  control  parameter  record   
      starts with the name of the control parameter in columm 1, which  must    
      be written in capital letters and must be followed by an  equal  sign.    
      The  length  of  the  control  parameter  name  is  restricted  to  10    
      characters excluding the equal sign. After the control parameter name,    
      the value of the control parameter(s) have to be given.  There  is   a    
      format specified for the control parameters, and a real parameter  has    
      naturally to be input with a point (e.g.  3.1415962)  and  an  integer    
      parameter has to be input without a point. If there is given a control    
      parameter which is not valid  for  the  used  program  but  valid  for    
      another program, the control parameter record is simply  neglected  by    
      the used program. The sequence of the control parameter records is  in    
      general of no importance  with the exception of the TEXTHEADER control    
      parameters. The latter have to be input in their proper sequence.         
                                                                                
      The following control parameters are valid for program ANALYZE:
 
                                                                               
      TEXTHEADER= Alphanumeric text, which will  be  printed  as  header  at   
                  the print file of ANALYZE. The number of header records is
                  records is restricted to 10.                                 
      STATLATITU= Ellipsoidal latitude of the  station  in  degree  decimal,  
                  positive north of the equator, referring to World Geodetic   
                  System 1984 (col. 12...26, real with decimal point).         
      STATLONITU= Ellipsoidal longitude of the station  in  degree  decimal,   
                  positive east of Greenwhich, referring to  World  Geodetic   
                  System 1984 (col. 12...26, real with decimal point).         
      STATELEVAT= Ellipsoidal height of the station in  meter  referring  to   
                  World  Geodetic  System  1984  (col.  12...26,  real  with   
                  decimal point).                                              
      STATGRAVIT= Gravity of the station in m/s**2, necessary for tilt only.   
                  (col. 12...26, real with decimal point). If the gravity is   
                  unknown, use a value of less than 1. and the program  will   
                  compute and subsequently use the normal gravity value.       
      STATAZIMUT= Azimuth of the  instrument  in  degree  decimal,  reckoned   
                  clockwise from north (col. 12 ... 26,  real  with  decimal   
                  point). This parameter is used for tilt,  horizontal  dis-   
                  placement and horizontal strain only.                        
      TIDALCOMPO= Earth tide component (col. 12 ... 26, integer).               
                  = -1 for tidal potential in m**2/s**2.                        
                  =  0 for tidal gravity in nm/s**2.                            
                  =  1 for tidal tilt in mas, at azimuth STATAZIMUT.            
                  =  2 for tidal vertical displacement in mm.                   
                  =  3 for tidal horizontal displacement in mm at               
                       azimuth STATAZIMUT.                                      
                  =  4 for tidal vertical   strain in 10**-9 = nstr.            
                  =  5 for tidal horizontal strain in 10**-9 = nstr,            
                       at azimuth STATAZIMUT.                                   
                  =  6 for tidal areal  strain in 10**-9 = nstr.                
                  =  7 for tidal shear  strain in 10**-9 = nstr.                
                  =  8 for tidal volume strain in 10**-9 = nstr.                
                  =  9 for ocean tides in mm.                                   
                  The earth tide observations must be  input  in  the  units    
                  given above!!!                                                
      TIDALPOTEN= parameter for the tidal potential catalog to be used.        
                  1 = Doodson (1921) tidal potential catalog,                  
                  2 = Cartwright-Tayler-Edden (1973) tidal potential catalog   
                  3 = Buellesfeld (1985) tidal potential catalog,            
                  4 = Tamura (1987) tidal potential catalog,
                  5 = Xi (1989) tidal potential catalog,
                  6 = Roosbeek (1996) tidal potential catalog,
                  7 = Hartmann and Wenzel (1995) tidal potential catalog.
                  As default, TIDALPOTEN= 7 is used.                   
      PRINTDEVEL= Parameter for printout of the tidal waves for the specific    
                  component (col. 12...26, integer).                            
                  = 1 : tidal waves will be printed.                            
      SEARDATLIM= Parameter for testing the  observations  on  gross  errors    
                  (col. 12...26, real with decimal point).                      
                  For positive SEARDATLIM, the observations will  be  tested    
                  on gross errors using a filter of 23 hours length  (Wenzel    
                  1976). SEARDATLIM is the threshold for testing the  obser-    
                  vations on gross errors in units of the  observations.  If    
                  any computed test value exceeds  SEARDATLIM,  the  program    
                  will list the first 20 test values  exceeding  SEARDATLIM.    
                  It is however not recommended to use this method of  gross    
                  error detection, because  in  the  case  of  multi-channel    
                  input, the influence of the  additional  channels  on  the    
                  earth tide signal is not  taken  into  account  with  this    
                  method of gross error detection.                              
      NUMHIGPASS= Parameter deciding on numerical highpass filtering of  the
                  data before analysis (col. 12 ... 26, integer, 1  =  yes).
                  We do highpass filtering by subtracting the lowpass filte-   
                  red signal from the observation. This is because  we gene-   
                  rate less algebraic correlation than  by  direct  highpass   
                  filtering. We use  symmterical  filters  with  zero  phase   
                  shift only.                    
                                                                                
                  = 0: no filtering will be applied. In this case, bias         
                       parameters may be adjusted for each block of data.       
                  = 1: highpass filtering will be applied. In this case,
                       the input of a valid name for the high pass filter
                       file (see below) is necessary.
                                                                                 
      NUMFILNAME= alphanumeric name of the highpass filter file, CHARACTER*12.
                  The following numerical filters are available for ANALYZE
                  (in directory \eterna33\commdat): 
 
                  file name     coeff. lenght   sampling     gain
                                                intvl.
                  n1h1h001.nlf    37    37 h    1 h          encl. 046
                  n1h1h002.nlf    51    51 h    1 h          encl. 047
                  n1h1h003.nlf   143   143 h    1 h          encl. 048
                  n1h1h004.nlf   239   239 h    1 h          encl. 049
                  n1h1h005.nlf    49    49 h    1 h          encl. 050
                  n1h1h006.nlf    49    49 h    1 h          encl. 051
                  n1h1h007.nlf   145   145 h    1 h          encl. 052
                  n1h1h008.nlf   241   241 h    1 h          encl. 053
                  n05m05m2.nlf  2001   167 h     5 min       encl. 054
                  n10m10m2.nlf  1001   167 h    10 min       encl. 055
                  n20m20m2.nlf   501   167 h    20 min       encl. 056
                  n30m30m2.nlf   335   167 h    30 min       encl. 057
                  n60m60m2.nlf   167   167 h    60 min       encl. 058
                  
                  The first eight filters are the "old" filters already
                  used in previous versions of the program. The mnemotic
                  of the filter names is "n1h1h00x" is old filter no. x.
                  We recommend to use one of the last 5 filters, because 
                  they all have approximately the same filter gain (which
                  is partly much better than the gain of the first eight
                  filters).   
 
      PRINTOBSER= Parameter for writing the original observations  to  print   
                  file "pn".PRN (col. 12...26, integer, 1 = yes).              
      PRINTLFOBS= Parameter for writing the lowpass filtered observations to   
                  print file "pn".PRN (col. 12...26, integer, 1 = yes).        
      RIGIDEARTH= Parameter for using a rigid earth model for the adjustment   
                  of earth tide parameters (col. 12...26, integer, 1 = yes).   
                  A priori amplitude factors 1.0000 are used for  all  tidal   
                  waves, independent of their degree and order. RIGIDEARTH =   
                  1 should  be  used  for  analyzing  computed  model  tides   
                  referring to a rigid Earth model (e.g.  benchmark  series)   
                  only. For analyzing observed tides, RIGIDEARTH=0 should be   
                  used. In the latter case, a  priori  body  tide  amplitude   
                  factors for the tidal waves of different degree and  order   
                  from an approximation to the Wahr-Dehant earth model  will   
                  be used, including Wahr's model of the nearly diurnal free   
                  wobble resonance.                                            
      HANNWINDOW= Parameter for applying a Hann-window  (see Schueller 1976)   
                  to the weights of the observations  before  least  squares   
                  adjustment (col. 12...26, integer, 1 = yes).  This  option   
                  may be  used  to  reduce  leakage  effects.  However,  the   
                  application of the Hann-window has been found to be dange-   
                  rous if bias parameters are adjusted together  with  tidal   
                  parameters in case of NUMHIGPASS = 0 (no highpass  filter-   
                  ing).                                                        
      QUICKLOOKA= Parameter for carrying out a quick look  adjustment  (col.   
                  12...26, integer, 1 = yes). For QUICKLOOKA = 1, the  para-   
                  meters will be adjusted, but  neither  residuals  nor  the   
                  Fourier spectrum of residuals  will  be  computed.  Please   
                  have in mind that the error estimation  is  in  this  case   
                  purely from least squares adjustment techniques neglecting   
                  correlations and  thus  generally  being  too  optimistic.   
                  The quick-look mode needs about 30% of the execution  time   
                  for a complete analysis only.                                
      POLETIDCOR= Correction for pole tide  (gravity  only)  with  amplitude   
                  factor 1.164 and no phase lead  (col. 12 ... 26,  integer,   
                  1 = yes). This option is useful  only  if  the  NUMHIGPASS   
                  parameter is set to zero (no highpass filtering). To  com-   
                  pute the pole tide, we use pole coordinates given in  file   
                  PUTC7294.DAT for each day at 0  UTC  between  January  1st   
                  1972 and October 30th 1994. The pole coordinates are named   
                  EOPC04 and taken from  the  International  Earth  Rotation   
                  Service IERS. We quadratically interpolate the daily  pole   
                  coordinates to derive the pole tide at each sample.          
      POLETIDREG= Adjustment of a linear regression  model  between  gravity   
                  observations and gravity pole tides computed  for  a rigid   
                  Earth  model  (col. 12 ... 26,  integer, 1  =  yes).  This   
                  option is useful only if the NUMHIGPASS parameter  is  set   
                  to zero (no highpass filtering). To  compute  the  gravity   
                  pole   tide,   we  use  pole  coordinates  given  in  file   
                  PUTC7294.DAT for each day at 0  UTC  between  January  1st   
                  1972 and October 30th 1994. The pole coordinates are named   
                  EOPC04 and  taken from the  International  Earth  Rotation   
                  Service IERS. We quadratically interpolate the daily  pole   
                  coordinates to derive the pole tide and its  time  deriva-   
                  tive at each sample. A regression parameter  in  amplitude   
                  and time lag is adjusted.  For  the  error  estimation  of   
                  these pole tide regression parameters,  we  use  the  same   
                  frequency dependent noise model as is used for  the  error   
                  estimation of long periodic tidal parameters.                
      LODTIDECOR= LOD correction of gravity (gravity variations due to vari-   
                  ation in length of day = LOD) will be applied (col. 12 ...   
                  26, integer, 1 = yes). This option is useful only  if  the   
                  NUMHIGPASS parameter is set to zero (no  highpass  filter-   
                  ing). To compute the LOD-tide, we use the differences DTAI   
                  = TAI minus UT1 (International Atomic Time minus Universal   
                  Time 1) given in file PUTC7294.DAT for each day at  0  UTC   
                  between January 1st 1972 and October 30th 1994. The DTAI's   
                  are taken from the International  Earth  Rotation  Service   
                  IERS. We  quadratically  interpolate  the  daily  DTAI  to   
                  compute the length of day at each sample. The LOD  correc-   
                  tion of gravity is computed in routine ETPOLC by             
                                                                               
                                        2                2                     
                               2 * omega * dLOD * a * cos(psi)                 
                  dg = 1.164 * ------------------------------- * 1.D9          
                                           86400                               
                                                                               
                  with dg   = LOD correction of gravity in  nm/s**2,  to  be   
                              subtracted from observed gravity,                
                       omega= 7.292115D-5 = rotation speed of the Earth,       
                       dLOD = d(DTAI)/dt  = time derivative of the length      
                                            of day,                            
                       a    = 6378137 m   = major semi axis of the Earth,      
                       psi  = geocentric latitude.                             
                                                                               
                  The LOD correction of gravity is however relatively  small   
                  and may be significant for very precise observations only.   
      STORENEQSY= parameter for storage of the normal equation system.         
                = 1: normal equation system will be stored on file "pn".NEQ.   
                  The normal equations may be used to  stack  data  sets  of   
                  different instruments at different stations.                 
      WAVEGROUPI= wave group definition for the adjustment of tidal  parame-   
                  ters. The number of wavegroups is restricted to 85.  There   
                  have to be input the five parameters  DFRA,  DFRE,  DFTFD,   
                  DFTFP and CNSY for each wave group in the same record.       
                  col. 12...21: DFRA = frequency of first tidal wave in  the   
                                       wave group, in cpd (real).              
                  col. 22...31: DFRE = frequency of last tidal wave  in  the   
                                       wave group, in cpd (real).              
                  col. 32...41: DFTFD= instrumental gain for the wave group,   
                                       (real), should be approx. 1.            
                  col. 42...51: DFTFP= instrumental phase lag for  the  wave   
                                       group in degree decimal  (real).  For   
                                       instruments  with  linear  phase  lag   
                                       with respect to frequency, it is  re-   
                                       commended to input  the  instrumental   
                                       time lag in secs in the block  header   
                                       record  in  file  "pn".DAT.  In  this   
                                       case, DFTFD should be  set  to  1.000   
                                       and DFTFP should be set to 0.000.       
                  col. 52...60: CNSY = alphanumeric name of the  wave  group   
                                       (8 characters max.).                    
                                                                               
      Standard  wave  groups for analysis of earth tide data (the list given   
      below is a proposal only). The  frequencies  are  given  in cycles per   
      day (cpd).                                                               
                                                                               
      group        > 1 month          > 6 months        > 1 year               
                   from      to       from      to      from      to           
                  [cpd]   [cpd]      [cpd]   [cpd]     [cpd]   [cpd]           
                                                                               
      SA             -       -          -       -     0.001379  0.004107       
      SSA            -       -    0.004108  0.020884  0.004108  0.020884       
      MM      0.020885  0.054747  0.020885  0.054747  0.020885  0.054747       
      MF      0.054748  0.091348  0.054748  0.091348  0.054748  0.091348       
      MTM     0.091349  0.501369  0.091349  0.501369  0.091349  0.501369       
      Q1      0.501370  0.911390  0.501370  0.911390  0.501370  0.911390       
      O1      0.911391  0.947991  0.911391  0.947991  0.911391  0.947991       
      M1      0.947992  0.981854  0.947992  0.981854  0.947992  0.981854       
      P1             -       -    0.981855  0.998631  0.981855  0.998631       
      S1             -       -          -       -     0.998632  1.001369       
      K1      0.981855  1.023622  0.998632  1.023622  1.001370  1.004107       
      PSI1           -       -          -       -     1.004108  1.006845       
      PHI1           -       -          -       -     1.006846  1.023622       
      J1      1.023623  1.057485  1.023623  1.057485  1.023623  1.057485       
      OO1     1.057486  1.470243  1.057486  1.470243  1.057486  1.470243       
      2N2     1.470244  1.880264  1.470244  1.880264  1.470244  1.880264       
      N2      1.880265  1.914128  1.880265  1.914128  1.880265  1.914128       
      M2      1.914129  1.950419  1.914129  1.950419  1.914129  1.950419     
      L2      1.950420  1.984282  1.950420  1.984282  1.950420  1.984282       
      S2      1.984283  2.451943  1.984283  2.002736  1.984283  2.002736       
      K2             -       -    2.002737  2.451943  2.002737  2.451943       
      M3M6    2.451944  7.000000  2.451944  7.000000  2.451944  7.000000      
                                                                               
      METEOPARAM= regression to meteorological parameter will  be  adjusted.   
                  The number of meteorological parameters is  restricted  to   
                  8. There have  to  be  input  the  four  parameters  IREG,   
                  DMECOR, CFY1 and CFY2 for each meteorological  channel  in   
                  the same record.                                             
                  col. 12...21: IREG   = number  of  meteorological  channel   
                                         (integer).                            
                  col. 22...31: DMECOR = a   priori   regression   parameter   
                                         (real). DMECOR is currently unused.   
                  col. 32...41: CFY1   = alphanumerical description  of  the   
                                         meteorological parameter.             
                  col. 42...51: CFY2   = alphanumerical description  of  the   
                                         unit in  which  the  meteorological   
                                         parameter is given.                   
                                                                               
 
      We give in the following a listing of the file HAL29901.INI, which  is  
      provided in directory "\eterna33\hourdat":   
                                                                               
      # This file HAL29901.INI status 1996.09.05 containing control parameters
      # for program package ETERNA 3.40
 
      # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
      # ! NOTE: The datalines have to start with their names.       !
      # !       An additional comment may follow after the values,  !
      # !       delimited by a whitespace                           !
      # ! Values of 0 or less causes PREGRED to calculate the       !
      # ! range(s) automatically resp. to use default values        !
      # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
 
      # a commentline starts with an '#', it may appear at any position
      # in this file. Empty lines may appear too
 
      TEXTHEADER=Gravimetric Earth tide station Hannover no.709 Germany
      TEXTHEADER=Institut fuer Erdmessung, Universitaet Hannover
      TEXTHEADER=52.387N   9.712E  H50M  P2M  D180KM    VERTICALCOMPONENT
      TEXTHEADER=GRAVIMETER LACOSTE-ROMBERG NR.G299     DR.BREIN IFAG FRANKFURT
      TEXTHEADER=ZERO METHOD INSTRUMENT WITH ELECTROMAGNETIC FEEDBACK
      TEXTHEADER=1973.05.29 - 1973.07.31      63.5 DAYS
      TEXTHEADER=INSTALLATION       DR.BREIN,SCHREINER FRANKFURT
      TEXTHEADER=MAINTENANCE        H.G.WENZEL HANNOVER
      TEXTHEADER=CALIBRATED TO IGSN71
      TEXTHEADER=INSTRUMENTAL LAG CORRECTED FOR 0.59 DEG O1 AND 1.23 DEG M2
 
      SENSORNAME=LCR-G299-BR     #earth tide sensor name
      SAMPLERATE=   3600         #sampling interval in seconds 
      STATLATITU=     52.387     #stations latitude  in degree
      STATLONITU=      9.712     #stations longitude in degree
      STATELEVAT=     50.000     #stations elevation in meter
      STATGRAVIT=      0.        #stations gravity in m/s**2
      STATAZIMUT=      0.        #stations azimuth in degree from north
      TIDALCOMPO=      0         #tidal component, see manual
      AMTRUNCATE=     5.D-5      #truncation threshold for tidal waves
      STEPDETLIM=      5.        #DESPIKE limit for step detection  (nm/s**2) 
      SPIKDETLIM=      2.        #DESPIKE limit for spike detection (nm/s**2)
 
      TIDALPARAM=  0.000000  0.600000   1.15000    0.0000 long #tidal param.
      TIDALPARAM=  0.600001  0.910000   1.14972   -0.7206 Q1   #tidal param.
      TIDALPARAM=  0.910001  0.949000   1.15182    0.1099 O1   #tidal param.
      TIDALPARAM=  0.949001  0.980000   1.11863    1.0176 M1   #tidal param.
      TIDALPARAM=  0.980001  1.012000   1.13894    0.1373 K1   #tidal param.
      TIDALPARAM=  1.012001  1.050000   1.14967   -0.3844 J1   #tidal param.
      TIDALPARAM=  1.050001  1.500000   1.13796    2.1804 OO1  #tidal param.
      TIDALPARAM=  1.500001  1.875000   1.15147    1.8050 2N2  #tidal param.
      TIDALPARAM=  1.875001  1.910000   1.16998    2.0400 N2   #tidal param.
      TIDALPARAM=  1.910001  1.950000   1.18803    1.6049 M2   #tidal param.
      TIDALPARAM=  1.950001  1.985000   1.21930    1.7852 L2   #tidal param.
      TIDALPARAM=  1.985001  2.500000   1.19502    0.1603 S2   #tidal param.
      TIDALPARAM=  2.500001  7.000000   1.25291   -5.7544 M3M6 #tidal param.
 
      PRINTDEVEL=      0      #ANALYZE print param. for tidal development (1=yes)
      SEARDATLIM=     -1.     #ANALYZE search for data error threshold 
      NUMHIGPASS=      1      #ANALYZE highpass filtering = 1 
      NUMFILNAME=n1h1h002.nlf #ANALYZE Pertzev (1959) highpass filter
      PRINTOBSER=      0      #ANALYZE print parameter for observations (1=yes)
      PRINTLFOBS=      0      #ANALYZE print parameter fo lowpass filtered obs.
      RIGIDEARTH=      0      #ANALYZE parameter for rigid earth model (1=yes)
      HANNWINDOW=      0      #ANALYZE parameter for Hann-window (1=yes)
      QUICKLOOKA=      0      #ANALYZE parameter for quick look analysis (1=yes)
 
      WAVEGROUPI=  0.501370  0.911390   1.00000     0.000 Q1  
      WAVEGROUPI=  0.911391  0.947991   1.00000     0.000 O1  
      WAVEGROUPI=  0.947992  0.981854   1.00000     0.000 M1   
      WAVEGROUPI=  0.981855  1.023622   1.00000     0.000 K1   
      WAVEGROUPI=  1.023623  1.057485   1.00000     0.000 J1       
      WAVEGROUPI=  1.057486  1.470243   1.00000     0.000 OO1     
      WAVEGROUPI=  1.470244  1.880264   1.00000     0.000 2N2     
      WAVEGROUPI=  1.880265  1.914128   1.00000     0.000 N2      
      WAVEGROUPI=  1.914129  1.950419   1.00000     0.000 M2     
      WAVEGROUPI=  1.950420  1.984282   1.00000     0.000 L2      
      WAVEGROUPI=  1.984283  2.451943   1.00000     0.000 S2      
      WAVEGROUPI=  2.451944  7.000000   1.00000     0.000 M3M6    
 
      METEOPARAM=         1      3.20airpress. hPa        #ANALYZE meteorol.
 
      HPERPAGE=24        #PREGRED displayed hours per page
 
      PGUPPGDN=12        #New hours per Pageup / Pagedown
 
      RANGETC1=8000      #PREGRED range for Channel 1 in nm/s^2
      RANGETC2=30        #PREGRED range for Channel 2 in hPascal
      RANGETC3=2500      #PREGRED range for Channel 3 in nm/s^2
      RANGETC4=250       #PREGRED range for Channel 4 in nm/s^2
 
      # ranges for RAW-Files
 
      RANGRAW1=2         #PREGRED range for Channel 1 in Volts
      RANGRAW2=2         #PREGRED range for Channel 2 in Volts
 
      # End of file HAL29901.INI
 
      Section 17.3: Description of the data input files "pn".dat              
      ----------------------------------------------------------               
                                                                               
      The first  part of the data  file may contain a  general  alphanumeric   
      file header of 80 columns per record, which is used as  a  description   
      of the data given in the file only. The header is not used by any pro-   
      gram but simply read and eventually written to the print file. In any    
      case, the file header must finish with                                   
                                                                               
      C********* starting at the first column.                                 
 
                                                                               
      Now the  first data block of earth tide observations follows, starting   
      with a header record and an offset record.                             
                                                                               
      Header record FORMAT (A10,5X,2F10.4,F10.3,I10):                      
      -----------------------------------------------                      
                                                                               
      CINSTR  = Alphanumeric name of the instrument (cols.1..10).              
      DCAL    = Calibration  factor for the  earth tide observations of this   
                data block, which will be multiplied to the  hourly  samples   
                in order to get proper calibration. (cols. 16...25).
      DSAPR   = A priori  standard deviation  of the earth tide observations   
                of this data block in proper unit. The parameter  DSAPR will   
                be used to give the observations of the current data block a   
                proper statistical weight for the least squares  adjustment.   
                If DSAPR is unknown, use 1.0 as standard deviation. 
                (cols. 26...35).    
      DTLAG   = Instrumental time lag of the instrument  in  sec  (including   
                filter, recording system). The input of a time lag per block   
                makes  sense  only  for  instruments  with  linear phase lag   
                versus frequency and should be  preferred  in  those  cases.   
                (cols. 36...45).      
      NBIAS   = Number of  Tschebyscheff  polynomial  bias  parameters to be   
                adjusted for the current data block. This information is re-   
                quired for parameter NUMHIGPASS = 0 only. The number of bias   
                parameters should not exceed the length of  the  data  block   
                divided  by  the  longest  period  of the  tidal waves to be   
                adjusted. Otherwise, ETERNA will print a warning.  If  NBIAS
                is set to a negative value, this block will not be used  for
                the analysis, independent of the parameter NUMHIGHPASS.
                (cols.46..55).              
                                                                               
      Offset record FORMAT I8,7X,6F10.3):                                 
      -----------------------------------                                  
                                                                               
      IDUMMY       = 77777777 dummy parameter (cols. 1...7).
      DOFFS(1...9) = Offset values (cols.  16...25,  26...35,...).  DOFFS(1)
                     will be added to all following earth tide  observations
                     (after having calibrated the earth  tide observations). 
                     DOFFS(2) will be added to  all  following  observations 
                     of the first meteorological parameter, and so on.   
                                                                               
      Data records FORMAT = I8,1X,I6,8F10.3:                                
      -------------------------------------                                
                                                                               
      IDAT    =    Date of the observations of  the  record (cols. 1 ... 8).   
                   IDAT = 19891214 means year = 1989, month = 12, day = 14.   
      ITIM    =    Time of the observations of the record in  UTC  (cols. 10
                   ...15). ITIM = 130000 means  13  hours,  00  minutes,  00 
                   seconds.        
      DCIN(1) =    Earth tide sample (cols. 16...25). The earth tide  sample
                   DCIN(1) will be scaled with DCAL  before  being  used  in
                   program ANALYZE.                                
      DCIN(2..8) = meteorological samples (cols. 26...35, 36...45, 46...55,
                   ....).                                         
                                                                               
      The end of a data block is defined by 999999999 as  date  IDAT.  After   
      the end of a data block, a new data block may follow  starting with  a   
      block header record and an offset record, or the  end  of  earth  tide   
      data flagged by 888888888 as date IDAT. We list in  the  folllowing as   
      example for input data a part of file BFL24903.DAT  (1  meteorological   
      parameter):                                                              
                                                                               
      L249               1.0000     0.977   139.400         0   BLOCK 1        
      77777777            0.000     0.000                                      
      19881208      0  -672.700    16.130                                      
      19881208  10000  -342.500    16.210                                      
      19881208  20000    79.600    16.610                                      
      19881208  30000   523.100    16.440                                      
      19881208  40000   912.100    16.970                                      
      19881208  50000  1198.800    17.150                                      
      19881208  60000  1355.300    17.450                                      
      19881208  70000  1391.400    17.720                                      
      19881208  80000  1343.700    17.770                                      
      19881208  90000  1258.700    18.490                                      
      19881208 100000  1191.300    18.930                                      
      19881208 110000  1179.700    18.210                                      
      ....                                                                     
      99999999                                                                 
      88888888                                                                 
                                                                               
 
      Section 17.4: Description of the residual output files "pn".res          
      ---------------------------------------------------------------          
                                                                               
      The ASCII output data files with suffix "res" have the same  structure   
      and use the same standard format as the data input files  with  suffix   
      "dat". If the data input file has NC channels, the corresponding resi-   
      dual output file has NC + n channels, where n = 1 in case of POLTIDREG   
      = 0 and n  =  3 in case of POLTIDREG = 1. The first NC channels of the   
      residual output file contain highpass filtered observations in case of   
      NUMHIGPASS greater zero. In case of POLTIDREG = 1, the channel  NC + 1   
      contains the computed gravity pole tide referring  to  a  rigid  earth   
      model and channel NC + 2 contains the time derivative of  the  gravity   
      pole tide in nm/s**2 per day. The  residuals  of  the  adjustment  are   
      always stored in the last channel. We list in the folllowing  a  short   
      example for a residual file, taken from file BFL24903.RES (1  meteoro-   
      logical parameter):                                                      
                                                                               
      L249               1.0000    0.8500   139.400         0RESIDUALS         
      77777777            0.000     0.000     0.000                            
      19881211      0 -1647.100     0.427    -0.519                            
      19881211  10000 -1635.102     0.177    -0.367  <----- residual           
      19881211  20000 -1432.571    -0.124    -0.172                            
      19881211  30000 -1077.091    -0.432    -0.111                            
      19881211  40000  -631.130    -0.690    -0.207                            
      19881211  50000  -168.266    -0.835    -0.345                            
      19881211  60000   242.645    -0.833    -0.431                            
      19881211  70000   551.564    -0.687    -0.476                            
      19881211  80000   736.371    -0.453    -0.516                            
      19881211  90000   804.971    -0.239    -0.489                            
      19881211 100000   790.014    -0.165    -0.270                            
      19881211 110000   737.602    -0.283     0.130                            
      19881211 120000   692.827    -0.526     0.500                            
      19881211 130000   685.977    -0.738     0.652                            
      19881211 140000   723.432    -0.778     0.691                            
      .......                                                                  
                                                                               
      Section 17.5: Description of the residual amplitude spectrum file        
                    "pn".far                                                   
      -----------------------------------------------------------------        
                                                                               
      The ASCII output data files with suffix  "far"  contain  the  Fourier    
      amplitude spectrum of residuals to be plotted with  program  PLOTSPEC.   
      It starts with the file header of the corresponding  data  input  file   
      with suffix DAT. The file header is finished with                        
                                                                               
      C*********   starting at the first column.                               
                                                                               
      The following records contain in each record the frequency in cpd and
      the Fourier amplitude in the unit of the tidal component.
                                                                               
      Section 17.6: Description of examples
      -------------------------------------
 
      In  directory  "\eterna33\hourdat"  we  have  given  files  for  seven
      different examples (see also encl 059... 093). The individual examples
      are also described in the header of their data file (with suffix dat).
  
      example   station        data type           length       epoch
 
      hal29901  Hannover       observed gravity      63.5 d     1973
      bfl24903  BFO Schiltach  observed gravity     121.0 d     1991
      bfd00801  BFO Schiltach  observed strain      110.0 d     1974
      bfet1907  BFO Schiltach  observed gravity     286.0 d     1991
      bhtt4003  Bad Homburg    observed gravity    1004.5 d     1982
      bfhw9501  BFO Schiltach  synthetic gravity   2731.0 d     1990
      bfde403f  BFO Schiltach  benchmark gravity   2922.0 d     1990
 
      Example hal29901 contains gravity tide observations with electromagne-
      tic feedback gravimeter CLR-G299 (transformed by Dr. R.  Brein,  IfAG)
      at station Hannover in 1973, 63.5  days,  analog  recording,  together 
      with air pressure observations. Plots to example  hal29901  are  given 
      in encl. 059...063.
  
      Example bfl24903 contains gravity tide observations with SWR  electro-
      static feedback gravimeter LCR-G249F at station BFO Schiltach in 1988-
      1989, 121 days, digitally recorded, together with air pressure  obser-
      vations. Plots for this example are given in encl. 064...068. 
 
      Example bfd00801 contains horizontal strain tides from 10 m invar wire
      strainmeter D008 at station BFO Schiltach azimuth 2 deg from north  to
      east, at position 190, analog recording.
      Reference:  Beavan, J.,  R.  Bilham,  D.  Emter  and  G. King  (1979): 
      Observations of strain enhancement across a fissure.  In:  Bonatz,  M.
      (editor):  Beitraege  zur  Erdgezeitenforschung   des   Arbeitskreises
      Geodaesie/Geophysik der Bundesrepublik Deutschland,  Deutsche  Geodae-
      tische Kommission, Reihe B, Heft  Nr.  231,  47 - 58,  Muenchen  1979.
      Plots for this example are given in encl. 069...073.
 
      Example bfet1907 contains  gravity  tide  observations  with  LaCoste-
      Romberg earth tide gravimeter ET19 with Larson electrostatic  feedback
      at station BFO Schiltach in 1991, 286 days, together with air pressure
      observations. The data have been digitally recorded with 5 s  sampling
      interval and 21 bit resolution, and decimated to hourly samples  after
      preprocessing with program PRETERNA. Plots for this example are  given
      in encl. 074...078.   
 
      Example bhtt4003 contains  gravity tide  observations  with  supercon-
      ducting gravimeter GWR-TT40 at station Bad Homburg  1981-1984,  1004.5
      days, together with air pressure  observations  (Richter  1987).  This 
      data set is included in the ETERNA package with kind permission of Dr.
      B.  Richter, IfAG  Frankfurt. The  analysis of  this data set has been 
      carried out without highpass filtering by adjusting  longperiodic  and
      and shortperiodic tidal parameters, pole  tide  parameters  and  drift
      parameters. Please have in mind that the noise level is generally much
      lower when applying highpass filtering.  Plots  to  this  example  are 
      given in encl. 079...083.   
 
      Examnple bfhw9501  contains  synthetic (i.e. computed)  gravity  tides 
      from Hartmann  and  Wenzel  (1995)  tidal  potential  catalogue  using 
      program PREDICT for station BFO Schiltach, 1987...1994, 2731 days. The
      analysis of this  data  set  with  program  ANALYZE  demonstrates  the 
      consistency between programs PREDICT and ANALYZE  (standard  deviation
      of residuals 0.000 nm/s**2), even when solving for longperiodic  tidal
      parameters. Plots to this examples are given in encl. 084...088.  
 
      Example bfde403f contains a gravity  tide  benchmark  series  computed
      from DE403 ephemerides (Wenzel 1996) for station BFO  Schiltach,  1987
      - 1994, 2922 days. The analysis of this benchmark series with  program
      ANALYZE when using the Hartmann  and  Wenzel  (1995)  tidal  potential
      catalogue shows the  very  small  errors  of  the Hartmann and  Wenzel 
      (1995) tidal potential catalogue of 0.001 nm/s**2 rms. You can  verify
      yourself that the other tidal potential catalogues  have  much  larger
      errors. Plots to this example are given in encl. 089...093.
 
      Section 17.7: Execution time for the example files                   
      --------------------------------------------------                   
                                                                               
      The cpu execution time of ANALYZE mainly depends on the number of data   
      to be processed,  the  tidal  potential to be used  and  the number of   
      parameters to be adjusted.  The  execution time has been measured  for   
      different data sets provided on the  CD-rom  using different cpu's.  
                                                                               
      operating system:     MS-DOS      MS-DOS      MS-DOS               
      processor:            486DX2      PENTIUM     PENTIUM               
      speed:                66 MHz       90 MHz     100 MHz               
                                                                              
      sample file   days                                                       
                                                                               
      hal29901       63.5    23.29 s           s      4.84 s           
      bfl24903      121.0    78.59 s           s     15.49 s                  
      bfd00801      110.0    54.05 s           s     11.20 s
      bfet1907      286.0   261.61 s           s     47.07 s
      bhtt4003     1004.5   479.45 s           s     83.00 s 
      bfhw9501     2731.0 13915.54 s           s   2384.81 s
      bfde403f     2922.0 14850.26 s           s   2531.41 s
                                                                               
      Section 18: Description of program TRANS                                
      ----------------------------------------                                
                                                                               
      The program TRANS transforms old ETERNA version 2.3 data  input  files   
      (international format and 6 sample per record format) into the  ETERNA   
      standard format (Wenzel 1995). The program TRANS has been  written  in   
      Fortran 90 and has been compiled using the Lahey Fortran 90 compiler.
      After entering        
                                                                               
            <trans>                                                            
                                                                               
      the program asks for the data file to be  transformed. The  data  file   
      must be an  ETERNA  2.3  data  input  file,  as  given  e.g.  in  file   
      "trans.dat". The program generates the control  parameter  file  under   
      filename "trans.ini" and the ETERNA 3.40 data input  file  under  name   
      "trans.out". A protocol file is available under name "trans.prn".        
                                                                               
      There is no transformation program for ETERNA 3.0 data input files  to   
      ETERNA 3.21 data input files, because we believe that  this  transfor-   
      mation can easily be done using a standard alphanumeric editor.          
 
      Section 19: Description of program PLOTDATA
      -------------------------------------------
 
      The program PLOTDATA performs a plot of  data  files  containing  time
      dependent equally spaced data in standard ETERNA format  (e.g.  Wenzel
      1995). The plot can either be visualized on the screen of  the  pc  or 
      printed on paper (after establishing  file  "plotdata.hpg"  containing 
      plot commands in HPGL-plotter language or file "plotdata.lsr" contai-
      ning plotcommands for HP-Laser printer). The  program  is  written  in 
      Fortran 90 but contains calls to non-standard Fortran 90 Calcomp  plot
      routines, which have been compiled using the Lahey Graphoria  package.   
      Program PLOTDATA reads the  necessary  control  parameters  from  file
      "plotdata.ini". PLOTDATA can be called from any directory, but expects
      the driver files in directory \eterna33\bin.  
                                                                               
      The following control parameters are valid for program PLOTDATA: 
                                                                       
      DATFILNAME= name of file containing the data to be plotted 
                  (CHARACTER*12).      
      PLOT-TITEL= titel of the plot (CHARACTER*64).                    
      X-PLOTSIZE= size of inner frame in x-direction in inch (2.52 cm).
      Y-PLOTSIZE= size of inner frame in y-direction in inch (2.52 cm).
      Y-AXISUNIT= y-axis unit (CHARACTER*15).                          
      Y-PMINIMUM= y-minimum of the plot, in world y-coordinates.
      Y-PMAXIMUM= y-maximum of the plot, in world y-coordinates.
      Y-TICKDIST= y-tick difference, in world y-coordinates.
      X-GRIDLINE= plots grid lines in x-direction at y-axis ticks.     
      Y-GRIDLINE= plots grid lines in y-direction at date/time ticks.  
      PLOTCHANEL= plot channel number (INTEGER).
      PLOTCOLOUR= plot colour number (INTEGER, 0...15).
                  0 = black.               1 = dark blue.
                  2 = green.               3 = light blue.
                  4 = red.                 5 = dark magenta.
                  6 = brown.               7 = light grey.
                  8 = dark grey.           9 = bright dark blue.
                 10 = bright light green. 11 = bright light blue.
                 12 = bright red.         13 = bright magenta.
                 14 = bright yellow.      15 = bright white.
      STARDATEIS= start date of the plot in yyyymmdd (year, month, day).
                  E.g. 19870101 is January 1st, 1987.                  
      STARTIMEIS= start time of the plot in hhmmss (hours, minutes, seconds).
                  E.g. 131214 is 13 hours 12 minutes 14 seconds.
      STOPDATEIS= stop date of the plot in yyyymmdd.                    
      STOPTIMEIS= stop time of the plot in hhmmss.                      
      TICKDATEIS= x-tick difference in yyyymmdd.                       
      TICKTIMEIS= x-tick difference in hhmmss.                         
      NUMDIFFERN= numerically differentiation of the signal. For NUMDIFFERN=
                  1, the differences of subsequent samples will be plotted.
                                                                        
      For an example of control parameters for program  PLOTDATA,  see  file
      "plotdata.ini" in directory \eterna33\hourdat.
 
      On  option, program  PLOTDATA  establishes  the  file  "plotdata.hpg",
      which can be used with batch file PLOTRAUS.BAT (contained in directory
      \eterna33\bin) to print the plot on paper. PLOTRAUS can  be  used  for
      different printers (see below).                          
                                                                       
      Section 20: Description of program PLOTFILT
      -------------------------------------------
 
      The program PLOTFILT performs a plot of the gain of  a  numerical  FIR
      zero phase filter, as e.g. used  in  program  ANALYZE.  The  plot  can 
      either be visualized on the screen of  the  pc  or  printed  on  paper 
      (after establishing file "plotfilt.hpg" containing the  plot  commands 
      in HPGL-plotter language of file "plotfilt.lsr" containing plotcommans
      for HP-Laser printer). The program is written in Fortran 90  but  con- 
      tains calls to non-standard Fortran 90 Calcomp  plot  routines,  which
      have been compiled using the Lahey Graphoria package. Program PLOTFILT
      reads the  necessary  control  parameters  from  file  "plotfilt.ini". 
      PLOTFILT can be executed from any  directory, but expects  the  driver 
      driver files on directory \eterna33\bin.  
 
      The following control parameters are valid for program PLOTFILT: 
                                                                       
      DATFILNAME= name of data file containing the numerical filter
                  (CHARACTER*12).   
      PLOT-TITEL= titel of the plot (CHARACTER*64).                    
      INVERSEFIL= plots the gain of the inverse filter if INVERSEFIL=1.
      X-FREQUNIT= frequency unit for the x-axis, in cps.                
      X-PLOTSIZE= plot size (inner frame) in x-direction in inch.      
      Y-PLOTSIZE= plot size (inner frame) in y-direction in inch.      
      X-PMINIMUM= x-minimum of the plot in world x-coordinates.        
      X-PMAXIMUM= x-maximum of the plot in world x-coordinates.        
      X-TICKDIST= x-tick distance in world x-coordinates.              
      X-AXISUNIT= x-axis unit (CHARACTER*15).                          
      Y-PMINIMUM= y-minimum of the plot in world y-coordinates.        
      Y-PMAXIMUM= y-maximum of the plot in world y-coordinates.        
      Y-TICKDIST= y-tick distance in world y-coordinates.              
      Y-AXISUNIT= y-axis unit (CHARACTER*15).                          
      X-GRIDLINE= plots horizontal gridlines if X-GRIDLINE=1.          
      Y-GRIDLINE= plots vertical   gridlines if Y-GRIDLINE=1.          
                                                                       
      For an example of control parameters for program  PLOTFILT,  see  file
      "plotfil.ini" in directory \eterna33\commdat.
 
      On  option, program  PLOTFILT  establishes  the  file  "plotfilt.hpg",
      which can be used with batch file PLOTRAUS.BAT (contained in directory
      \eterna33\bin) to print the plot on paper. PLOTRAUS can  be  used  for
      different printers (see below).                          
                                                                       
      Section 21: Description of program PLOTHIST
      -------------------------------------------
 
      Program PLOTHIST computes and plots a histogram of samples  stored  in
      standard ETERNA format. E.g., a histogram  of  residuals  after  least
      squares adjustment with program ANALYZE can be plotted. The  plot  can 
      either be visualized on the screen of  the  pc  or  printed  on  paper 
      (after establishing file "plothist.hpg" containing the  plot  commands 
      in HPGL-plotter language or file "plothist.lsr"  containing  the  plot 
      commands for an HP-Laser printer). The program is written  in  Fortran
      90 but contains calls to non-standard Fortran 90 Calcomp plot routines,
      which have been compiled using the Lahey  Graphoria  package.  Program 
      PLOTHIST reads necessary control parameters from file  "plothist.ini". 
      PLOTFILT can be executed from any  directory, but expects  the  driver 
      driver files on directory \eterna33\bin.  
 
      The following control parameters are valid for program PLOTHIST:
 
      The following control parameters are valid for PLOTHIST:         
                                                                       
      DATFILNAME= name of file containing the sample data (CHARACTER*12). 
      PLOT-TITEL= title of the plot (CHARACTER*64).                    
      X-PLOTSIZE= size of inner frame in x-direction in inch (2.52 cm).
      Y-PLOTSIZE= size of inner frame in y-direction in inch (2.52 cm).
      X-AXISUNIT= x-axis unit (CHARACTER*15).                          
      X-PMINIMUM= x-minimum of the histogram in units of the samples.  
      X-PMAXIMUM= x-maximum of the histogram in units of the samples.  
      X-TICKDIST= x-tick distance in units of the samples.             
      PLOTCHANEL= channel numer in which samples are stored.           
      PLOTCOLOUR= plot colour number (INTEGER, 0...15).
                  0 = black.               1 = dark blue.
                  2 = green.               3 = light blue.
                  4 = red.                 5 = dark magenta.
                  6 = brown.               7 = light grey.
                  8 = dark grey.           9 = bright dark blue.
                 10 = bright light green. 11 = bright light blue.
                 12 = bright red.         13 = bright magenta.
                 14 = bright yellow.      15 = bright white.
      Y-PMAXIMUM= y-maximum of the plot in %.                          
      Y-TICKDIST= y-tick distance in %.                                
                                                                       
      For an example of control parameters for program  PLOTHIST,  see  file
      "plothist.ini" in directory \eterna33\hourdat.
 
      On  option, program  PLOTHIST  establishes  the  file  "plothist.hpg",
      which can be used with batch file PLOTRAUS.BAT (contained in directory
      \eterna33\bin) to print the plot on paper. PLOTRAUS can  be  used  for
      different printers (see below).                          
                                                                       
      Section 22: Description of program PLOTSPEC
      -------------------------------------------
                                                                               
      Program PLOTSPEC performs a plot of the Fourier amplitude spectrum  of   
      residuals after least squares  adjustment, as stored in an output file
      of program ANALYZE with suffix "far". The plot can either be  visuali-
      zed on the screen of the pc or printed on  paper  (after  establishing 
      file "plotspec.hpg"  containing  the  plot  commands  in  HPGL-plotter
      language or file "plotspec.lsr" containing the plot  commands  for HP-
      laser printer). The program is written  in  Fortran  90  but  contains 
      calls to non-standard Fortran 90 Calcomp  plot  routines,  which  have
      been compiled using the  Lahey  Graphoria  package.  Program  PLOTSPEC 
      reads the  necessary  control  parameters  from  file  "plotspec.ini". 
      PLOTSPEC can be executed from any  directory,  but  expects the driver
      files in directory \eterna33\bin.  
 
      The following control parameters are valid for program PLOTSPEC:
 
      DATFILNAME= name of the data file containing the spectrum to be
                  plotted (CHARACETR*12).       
      PLOT-TITEL= titel of the plot (CHARACETR*64).                    
      X-PLOTSIZE= size of inner frame in x-direction in inch (2.52 cm).
      Y-PLOTSIZE= size of inner frame in y-direction in inch (2.52 cm).
      X-AXISUNIT= x-axis unit (CHARACTER*15).                          
      Y-AXISUNIT= y-axis unit (CHARACTER*15).                          
      Y-PLOTLOGS= 0 for linear, 1 for logarithmic y-axis.              
      X-PMINIMUM= x-minimum of the plot, in world x-coordinates.
      X-PMAXIMUM= x-maximum of the plot, in world x-coordinates.
      X-TICKDIST= x-tick distance, in world x-coordinates.
      Y-PMINIMUM= y-minimum of the plot, in world y-coordinates.
      Y-PMAXIMUM= y-maximum of the plot, in world y-coordinates.
      Y-TICKDIST= y-tick distance, in world y-coordinates.
      X-GRIDLINE= plot of vertical gridlines    if X-GRIDLINE=1.
      Y-GRIDLINE= plot of horizontal gridlines  if Y-GRIDLINE=1.
                                                                       
      For an example of control parameters for program  PLOTSPEC,  see  file
      "plotspec.ini" in directory \eterna33\hourdat.
 
      On  option, program  PLOTSPEC  establishes  the  file  "plotspec.hpg",
      which can be used with batch file PLOTRAUS.BAT (contained in directory
      \eterna33\bin) to print the plot on paper. PLOTRAUS can  be  used  for
      different printers (see below).                          
                                                                       
      Section 23: Description of program PRINTGL
      ------------------------------------------
 
      Program PLOTRAUS is a batch file in  directory  "\eterna33\bin"  which 
      calls program PRINTGL from Ravitz Software Inc. to make a  preview  or
      a hardcopy from a HPGL-plotfile. You should edit  file  "plotraus.bat"
      in order to set the correct options for your printer.
 
      The call of PRINTGL is
  
      PRINTGL [/P]plotfile [/option1] [/option2] ...
  
      Valid options are
 
      /A[O][x,y]  puts plot center or x,y (inches) at print center or 
                  Origin 
      /Cc..       up to 8 pen colors of B C G K M R W Y or shade mixed colors.
      /D[+]f      destination printer port|file|device, +append 
 
      /Fc[-|+[b]|*][!][~][x,y]  format, -lo|+hi|*max, !comprs, ~no clr, 
                                x,y dpi  
      The following printers may be chosen:
 
       J=Canon BJ130                   G=GEM IMG ~ 
       Q=HP QJ                         6=IBM X24 +*
       9=NEC 24 +*                     C=CGA
       7=Canon BJ                      8=HP-GL     
       4=IBM LP                        5=X24 alt1 +
       S=PostScript                    E=EGA screen
       D=Canon LBP                     L=HP LJ 
       1=IBM Pro +                     X=X24 alt2 +
       O=Tosh 24 +*                    M=EGAmono
       N=Epson 9 +*                    I=HP LJ3   
       2=IBM QW2                       K=Kodak C4 
       Z=ZS PCX                        H=HGC
       T=Epson 24 +*                   P=HP PJ 
       3=IBM QW3                       A=MT MT92C 
       V=VGA +b                 
 
       Other options of PRINTGL are displayed on the screen by entering
       <printgl>
 
      Section 24: Description of program LOAD89
      -----------------------------------------
 
      Program LOAD89 has  kindly  been  provided  by  Dr.  O.  Francis  from 
      International Center for Earth Tides  (ICET)  Bruxelles.  It  computes
      ocean tide loading for a  different  ocean  tide  models  provided  in
      directory "\eterna33\oceload". Unfortunetely, we were unable to  write
      a detailed program manual for program LOAD89 with this version of  the
      ETERNA package; but we have included two  Fortran  source  code  files 
      (load89.f for a Sun workstation, load89.for for a pc using  the  Lahey
      compiler). 
      
      Section 25: Description of program WPAREX
      -----------------------------------------
 
      Program WPAREX interpolates synthetic  gravity  tide  parameters  from
      the 1 by 1 degree grid computed  by  Timmen  and  Wenzel (1994)  using
      Schwiderski (1980) ocean tide  model. The  coordinates  (latitude  and
      longitude) of the stations to be interpolated have  to  be  stored  in
      file "wparex.dat" before execution of program WPAREX, and  the  inter-
      polated gravity tide parameters will be stored  in file   "wparex.prn"
      after execution of program WPAREX. Program WPAREX can only be executed
      from directory "\eterna33\oceload". The print file of  program  WPAREX
      for the example input file is given in encl. 094. 
  
      end end end end end end end end end end end end end end end end end end




34


