DOCUMENTATION FOR DYNAPG SOFTWARE Software Version 1.1 Documentation date: September 6, 1996 Richard A. Snay NOAA/National Geodetic Survey 1315 East-West Highway/Room 8112 Silver Spring, Maryland 20910 tel: 301-713-3205 (ext. 155) fax: 301-713-4327 Internet: rich@dyna.ngs.noaa.gov INTRODUCTION The name, DYNAPG (DYNamic Adjustment Program using a Grid), identifies software for estimating the secular velocity field associated with crustal motion. DYNAPG accepts various geodetic data types including horizontal directions, distances, azimuths, zenith distances, and intersite vectors such as those obtainable with the Global Positioning System (GPS) or with very long baseline interferometry (VLBI). DYNAPG estimates a 3-dimensional velocity vector for each node of a user-specified 2-dimensional grid on a rectangular region that encompasses the geodetic data. The software assumes that the velocity at any other location may be adequately approximated via interpolation. In particular, the software employs bilinear interpolation to transfer crustal deformation information from the geodetic sites to the grid nodes. Also because DYNAPG incorporates the simultaneous-reduction technique described by Bibby (1982), users can apply this software to estimate positional coordinates for geodetic stations as well as various "nuisance" parameters simultaneously along with the velocity vectors. DYNAPG was developed by modifying the DYNAP software (Drew and Snay, 1989) which was in turn developed by modifying the ADJUST software (Milbert and Kass, 1987). The ADJUST software enables users to apply various geodetic data types towards estimating positional coordinates for geodetic marks. ADJUST assumes that these coordinates remain fixed over time. With DYNAP, users may estimate positional coordinates simultaneously with certain parameters that characterize crustal motion. DYNAP, however, severely restricts the spatial variation of the deformation field. DYNAPG was designed to allow for greater spatial variation. In addition to this document, users may find four other documents helpful in understanding the application of DYNAPG to the estimation of secular velocity vectors. In particular, Snay and others (1996) describe the application of DYNAPG to estimate the secular velocity field in the vicinity of the Big Bend of California's San Andreas fault. Drew and Snay (1989) demonstrate some capabilities and limitations of DYNAP with examples using geodetic data for two distinct California regions--the Imperial Valley and the Cascadia subduction zone. The FGCS (1994) specifies formats for the input files called BBOOK and GFILE. (Extensions to these formats are given in Appendix B of this document.) Also, Milbert and Kass (1987) describe the ADJUST software. These four documents may be obtained from the National Geodetic Information Branch, 1315 East-West Highway, Silver Spring, Maryland 20910 (telephone: 301-713-3242). MATHEMATICAL FOUNDATION Let P(n,t), L(n,t), and h(n,t) denote the geodetic latitude (positive north), geodetic longitude (positive east), and ellipsoidal height of the n-th geodetic station at time t. Then the mathematical model encoded into DYNAPG assumes that, for a prespecified reference time to, there are velocities u(L,P), v(L,P), and w(L,P) such that P(n,t) = P(n,to) + u(L(n,to),P(n,to)) * (t - to) L(n,t) = L(n,to) + v(L(n,to),P(n,to)) * (t - to) h(n,t) = h(n,to) + w(L(n,to),P(n,to)) * (t - to). Moreover, DYNAPG assumes that there is a 2-dimensional rectangular grid such that if one knows the 3-dimensional velocity vector (u,v,w) at each node of this grid, then the velocity vector at any other location within the grid may be calculated via bilinear interpolation. In particular, let the n-th station be located in the grid cell whose lower left corner has latitude Pj and longitude Li. Then u(L(n,to),P(n,to)) = [u(Li,Pj)*B*D + u(Li+1,Pj)*A*D + u(Li,Pj+1)*B*C + u(Li+1,Pj+1)*A*C] / [(A + B)*(C + D)]. Here A = L(n,to) - Li B = Li+1 - L(n,to) C = P(n,to) - Pj D = Pj+1 - P(n,to). Similar equations hold for v(L,P) and w(L,P). INPUT Program DYNAPG uses three external files: (1) AFILE, (2) BBOOK, and (3) GFILE. Whereas the files called AFILE and BBOOK are mandatory, the one called GFILE is needed only if there are GPS data to be used. In AFILE (Adjustment File), the user selects from among various options for executing the software. Also in AFILE, the user can input certain types of geodetic information. Format specifications for the AFILE records are given in Appendix A. BBOOK is the file containing horizontal directions and angles, zenith distances, distances, azimuths, and positional information for the geodetic marks (that is, geodetic latitude and longitude, orthometric height, geoid height, and deflections of the vertical). GFILE is the file containing differential GPS vectors and associated information. Format specifications for both the BBOOK records and the GFILE records are given in FGCS (1994). Extensions to these formats are given in Appendix B of this document. OUTPUT DYNAPG directs most of its output to the screen. The user may wish to redirect this screen output to a permanent file. The PP- record of the AFILE controls the content of this output. Among other information, the screen output contains (1) a replication of the input information, (2) a summary of the data for each geodetic station, (3) a list of post-adjustment residuals for all observations, and (4) a list of adjusted positions and derived velocities for all geodetic stations having *80*-records in the BBOOK file. DYNAPG also creates three output files in the user's working directory. The first, called "vfile", is an ASCII file that contains estimated velocities (with associated standard errors) for the grid nodes. The second, called "dfile", is a binary file that contains the covariance matrix for the velocities of the grid nodes. The third file, called "elps", is an ASCII file that containes estimated velocities (with associated standard errors) for all geodetic stations having *80*-records in the BBOOK file. A fortran routine, called DY2MI, uses these three files to create input files for the graphics software package known as MapInfo whereby a person may produce various maps illustrating different aspects of the velocity field or its associated strain field or the associated error statistics. COMPILER INFORMATION DYNAPG is written in standard FORTRAN 77. No special compiler options are needed to compile it. Note however, that standard FORTRAN 77 assumes that INTEGER is equivalent to INTEGER*4. INTEGER*2 is not sufficiently large for all of the integer variables in DYNAPG. The primary storage array is allocated in the main routine. The parameter LENA (defined in a PARAMETER statement in the main routine) is the length of the main storage array in double precision words. This parameter should be increased or decreased as necessary given the restrictions of the computer and/or compiler, and the size of the adjustments. The maximum number of GPS vectors in a group that can be processed by DYNAPG is set by the variable NVECS. The value of NVECS is set by PARAMETER statements in a number of routines. If the maximum number of of GPS vectors in a group needs to be changed, then NVECS must be changed in each routine where it occurs. A special variable, IPRSZ, is used to control printing of special information. If IPRSZ is initialized (in the main routine) to "1", then the amount that remains in the primary storage array is printed several times during the adjustment. The vectors that store the crustal dynamics parameter sets are also printed as they are updated in each iteration. They are not, however, first converted to the output units described above but are printed in their internal units of radians per minute or meters per minute. The matrix manipulations necessary for processing the observation and normal equations are made with a set of routines named "Heart of Gold" (Milbert, 1984). These routines must be compiled and linked with the routines of DYNAPG. If you have any questions about these routines please refer to the "Heart of Gold" documentation. We would greatly appreciate any comments as to form or content of the DYNAPG adjustment program, its output, and the documentation. REFERENCES Bibby, H.M., 1982: Unbiased Estimate of Strain from Triangulation Data Using the Method of Simultaneous Reduction, Tectonophysics 82, 161-174. Drew, A.R., and R.A. Snay, 1989: DYNAP: software for estimating crustal deformation from geodetic data, Tectonophysics, 162, 331-343. Federal Geodetic Control Subcommittee, 1994: Input Formats and Specifications of the National Geodetic Survey Data Base, Volume I. Horizontal Control Data, National Geodetic Information Branch, NOAA, Silver Spring, MD 20910. Milbert, D.G., 1984: Heart of Gold: Computer Routines for Large, Sparse, Least Squares Computations. NOAA Technical Memorandum NOS NGS-39, National Geodetic Information Branch, NOAA, Silver Spring, MD 20910, 28pp. Milbert, D.G., and Kass, W.G., 1987: ADJUST: The Horizontal Observation Adjustment Program. NOAA Technical Memorandum NOS NGS-47, National Geodetic Information Branch, NOAA, Silver Spring, MD 20910, 53pp. Snay, R.A., M.W. Cline, C.R. Philipp, D.D. Jackson, Y. Feng, Z.K. Shen, and M. Lisowski, 1996: Crustal velocity field near the big bend of California's San Andreas fault, J. Geophys. Res., 101, 3173-3185. APPENDIX_A ADJUSTMENT OPTION FILE (AFILE) FOR THE DYNAPG ADJUSTMENT PROGRAM This appendix contains a summary of the adjustment option file (AFILE) records. Changes from the ADJUST options are indicated. In several cases, a certain type of option record must precede another type of option record if they are both in the AFILE. If the option records are ordered alphabetically then all ordering will be correct. The only exception that I regularly make to alphabetic ordering is putting the sometimes very numerous QQ (Accuracy Computation) records at the end. CODE TYPE OF RECORD ----------------------------------------------------------------- ** Comment Record AA Ellipsoid Parameter Record BB Bypass Record CA Constrained Azimuth Record CC Coordinate Constraint Record CD Constrained Distance Record CH Constrained Height Difference Record CZ Constrained Zenith Distance Record EE Default Mean Sea Level Elevation Record GG Default Geoid Height Record GR Grid Record GT D(dX,dY,dZ)/DT record HC Control Point Height Adjustment Record HD Default Height Adjustment Record II Iteration Record LT DL/DT record where L denotes line length MM Adjustment Mode Record PP Print Output Record PV Point velocity record QQ Accuracy Computation Record SM Smoothing Constraint Record SS Auxiliary Parameter Indicator and Constraint Record SV Site velocity record VF Variance Factor Record **_-_Comment_Record: Comment records are ignored by the DYNAPG adjustment program. They ONLY make the option file easier to read. DYNAPG assumes any option file record that starts with two asterisks "**" or two blanks " " is a comment. The comment record is a DYNAPG addition to the ADJUST AFILE options. Columns Descriptions and Defaults Type ------------------------------------------------------------ 01-02 ** or blank 03-80 Anything. Any option record with columns 1 and 2 containing two asterisks or two blanks is treated as a comment and ignored. Totally empty lines are also ignored. AA_-_Ellipsoid_Parameter_Record: The Ellipsoid Parameter record specifies new value(s) for the semi-major axis of the ellip- soid and/or for the square of the eccentricity of the ellip- soid. Columns Descriptions and Defaults Type ------------------------------------------------------------ 01-02 AA 03-12 Semimajor axis, unit of meters default 6378137. meters (real) 13-30 Square of Eccentricity, default 0.0066943800229034156 (real) 31-80 Reserved BB_-_Bypass_Record: The Bypass record directs the adjustment program to ignore certain types of observations. Horizontal directions (*20* and *22* Blue Book records), horizontal angles (*30* and *32* records), zenith distances (*40* and *42* records), distances (both *52* and *54* records), astronomic azimuths (*60* records), and GPS observations (GFILE) may be bypassed. Blue Book observations which are to be ignored are printed (but not given observation numbers) in the BLUE BOOK section of the output. If GPS observations are to be ignored, the GPS observation section of the output is not printed. Bypass records for a particular observation type also cause other AFILE options to be ignored. For example, if the AFILE has a SS record for observation type 52 (mark-to-mark distances) and a BB record with its sixth column non-blank, then this SS record is ignored. For this reason, the BB record MUST precede any DD, SS, TT, and VV records. Columns Descriptions and Defaults Type ------------------------------------------------------------ 01-02 BB 03-03 Non-blank to bypass horizontal directions.(character) 04-04 Non-blank to bypass horizontal angles. (character) 05-05 Non-blank to bypass zenith distances. (character) 06-06 Non-blank to bypass distances. (character) 07-07 Non-blank to bypass astronomic azimuths. (character) 08-08 Non-blank to bypass GPS observations. (character) 09-80 Reserved CA_-_Constrained_Astronomic_Azimuth_Record: A Constrained Azimuth record creates a constraint for the azimuth angle between two stations. The CA records are ignored for one- dimensional adjustments. If CD parameters are being calculated, the constrained azimuths are applied at the reference time. That is, CA records do not give azimuths that are invariant across time. To do that put the same pseudo azimuth observations at two different times in the Blue Book (*60* records). Columns Descriptions and Defaults Type ------------------------------------------------------------ 01-02 CA 03-06 Standpoint Station Serial Number (integer) 07-10 Forepoint Station Serial Number (integer) 11-13 Reserved 14-16 Degrees Azimuth (integer) 17-18 Minutes Azimuth (integer) 19-22 Seconds Azimuth, units of 0.01 arc seconds (integer) 23-27 Azimuth Standard Deviation, units of 0.01 arc seconds, default of 0.01 arc seconds (integer) 28-80 Reserved CC_-_Constrained_Coordinate_Record: A Constrained Coordinate record specifies that a station is a control point, is that its coordinate(s) are constrained or "held fixed" during the adjustment. Only those coordinates which have values given on a CC record are constrained - i.e. there is no default for a constrained coordinate value. Standard deviations for any or all of the constrained coor- dinates on the record can also be specified by the CC record. The units for the standard deviations of the lati- tudes and longitudes as well as for heights are millimeters. The default is 0.1 millimeter for all coordinates. Latitudes and longitudes on CC records are not used for the constraints in one dimensional adjustments and elevations on CC records are not used for constraints in two dimensional adjustments. However, the latitudes and longitudes of CDP reference stations must be be included on a CC record even for one dimensional adjustments. If CD parameters are being calculated, the constrained posi- tions are applied at the reference time. That is, CC records (other than the ones used by TT records) do not give positions that are invariant across time. Columns Descriptions and Defaults Type ------------------------------------------------------------ 01-02 CC 03-10 Reserved 11-14 Station Serial Number (integer) 15-20 Latitude Standard Deviation, units of mm., default 0.1 mm. (integer) 21-26 Longitude Standard Deviation, units of mm., default 0.1 mm. (integer) 27-32 Height Standard Deviation, units of mm., default 0.1 mm. (integer) 33-44 Reserved 45-46 Degrees Latitude (integer) 47-48 Minutes Latitude (integer) 49-55 Seconds Latitude, units of 0.00001 arc sec. (integer) 56-56 Latitude Code N -- POSITIVE NORTH (default) S -- positive South 57-59 Degrees Longitude (integer) 60-61 Minutes Longitude (integer) 62-69 Seconds Longitude, units of 0.00001 arc sec.(integer) 56-56 Longitude Code E -- positive East W -- POSITIVE WEST (default) 70-76 Height, units of millimeters (integer) 77-80 Reserved CD_-_Constrained_Distance_Record: A Constrained Distance record creates a constraint for a mark-to-mark distance between two stations. This record is never ignored. If CD parameters are being calculated, the constrained dis- tances are applied at the reference time. That is, CD records do not give distances that are invariant across time. To do that put the same pseudo distance observations at two different times in the Blue Book (*52* or *54* records). Columns Descriptions and Defaults Type ------------------------------------------------------------ 01-02 CD 03-06 Standpoint Station Serial Number (integer) 07-10 Forepoint Station Serial Number (integer) 11-22 Mark-to-Mark Distance, units of 0.1 mm (integer) 23-27 Distance Standard Deviation, units of 0.1 mm, default of 0.1 mm (integer) 28-80 Reserved CH_-_Constrained_Height_Difference_Record: A Constrained Height Difference record creates a constraint for the difference between the heights of two stations. CH records are ignored in two dimensional adjustments. If the orthometric heights of both endpoints are adjusted, then the height difference given on this record is con- sidered to be orthometric. If the geoid heights of both endpoints are adjusted, then the height difference given on this record is considered to be an undulation difference. Lastly, if the orthometric height of one endpoint is adjusted while the geoid height of the other endpoint is adjusted, then the height difference given on this record is considered to be an ellipsoidal height difference. See the HC record for more information. If CD parameters are being calculated, the constrained heights are applied at the reference time. That is, CH records do not give heights that are invariant across time. Columns Descriptions and Defaults Type ------------------------------------------------------------ 01-02 CH 03-06 Standpoint Station Serial Number (integer) 07-10 Forepoint Station Serial Number (integer) 11-22 Height Difference, units of 0.1 mm (integer) 23-27 Height Standard Deviation, units of 0.1 mm, default of 0.1 mm (integer) 28-80 Reserved CZ_-_Constrained_Zenith_Distance_Record: A Constrained Zenith Distance record creates a constraint for the difference between the zenith angles of the two stations. CZ records are ignored for two dimensional adjustments. If CD parameters are being calculated, the constrained zen- ith distances are applied at the reference time. That is, CZ records do not give azimuths that are invariant across time. To do that put the same pseudo zenith distance obser- vations at two different times in the Blue Book (*40* and *42* records). Columns Descriptions and Defaults Type ------------------------------------------------------------ 01-02 CZ 03-06 Standpoint Station Serial Number (integer) 07-10 Forepoint Station Serial Number (integer) 11-13 Reserved 14-16 Degrees Zenith (integer) 17-18 Minutes Zenith (integer) 19-22 Seconds Zenith, units of 0.01 arc seconds (integer) 23-27 Zenith Standard Deviation, units of 0.01 arc seconds, default of 0.01 arc seconds (integer) 28-80 Reserved EE_-_Default_Mean_See_Level_Elevation_Record: The Default Mean See Level Elevation record specifies a value for the default orthometric height of the adjustment. The default orthometric height is the orthometric height for that is used any station which does not have an elevation in columns 70-75 of its Blue Book *80* (station location) record. Columns Descriptions and Defaults Type ------------------------------------------------------------ 01-02 EE 03-09 Default Orthometric Elevation, units of mm. default 0.0 (integer) 10-80 Reserved GG_-_Default_Geoid_Height_Record: The Default Geoid Height record specifies a value for the default geoid height in the adjustment. The default geoid height is used for all sta- tions which do not have a Blue Book *84* record giving their geoid height. Columns Descriptions and Defaults Type ------------------------------------------------------------ 01-02 GG 03-09 Default geoid height, units of mm., default 0. (integer) 10-80 Reserved GR_-_GRid_Record: The grid record specifies (1) the reference date, (2) the expanse of a grid, and (3) the mesh of this grid. A GR record is required to estimate any velocity vectors as these vectors are associated with the nodes of the grid. Without a GR record, the software will produce estimates for the positional coordinates and the nuisance parameters only. That is, the software will assume that there is no crustal motion. The reference date is the date for which the positional coordinates correspond. The specified grid is to span a rectangular region encompassing the geodetic data. Columns Descriptions and Defaults Type ------------------------------------------------------------ 01-02 GR 03-06 Year of reference (integer) 07-08 Month of reference (integer) 09-10 Day of reference (integer) 11-19 Most western longitude (DDDMMSSss) 20 Sense of most western longitude (E or W) 21-29 Most eastern longitude (DDDMMSSss) 30 Sense of most eastern longitude (E or W) 31-33 Number of grid cells extending from east to west (int.) 34 blank 35-42 Most southern latitude (DDMMSSss) 43 Sense of most southern latitude (N or S) 44 blank 45-52 Most northern latitude (DDMMSSss) 53 Sense of most northern latitude 54-56 Number of grid cells extending from south to north GT_-_D(dX,dY,dZ)/DT record where dX = X2 - X1, dY = Y2 - Y1, and dZ = Z2 - Z1. A GT record introduces three observations: the rates of change of the three components of an intersite vector (dX, dY, dZ). The coordinates of the two sites must be specified in the BBOOK file by *80* records or by *91* records or by one of each. Columns Descriptions and Defaults Type ___________________________________________________________- 01-02 GT 03-06 Station Serial Number of one site 07-10 Station Serial Number of the other site 11-18 D(dX)/DT in m/yr (format = F8.4) 19-26 Standard error for D(dX)/DT in m/yr (F8.4) 27-34 D(dY)/DT in m/yr (F8.4) 35-42 Standard error for D(dY)/DT in m/yr (F8.4) 43-50 D(dZ)/DT in m/yr (F8.4) 51-58 Standard error for D(dZ)/DT in m/yr (F8.4) HC_-_Control_Point_Height_Adjustment_Record: A Control Point Height Adjustment record specified which height (the orthometric or the geoid) is to be adjusted for a control point (constrained location). All control points which do not have a HC record have their height adjusted as specified by the HD record (if there is one), If there is no HD record, then their orthometric height is adjusted. Columns Descriptions and Defaults Type ------------------------------------------------------------ 01-02 HC 03-03 Height Indicator E -- ADJUST ORTHOMETRIC ELEVATIONS (default) G -- Adjust Geoid Height 04-07 Station Serial Number (integer) 08-80 Reserved HD_-_Default_Height_Adjustment_Record: The Default Height Adjustment record specifies which height (the orthometric or the geoid) is to be adjusted for all stations which are not control points (i.e. constrained) and for all control points which do not have HC records. Columns Descriptions and Defaults Type ------------------------------------------------------------ 01-02 HD 03-03 Height Indicator E -- ADJUST ORTHOMETRIC ELEVATIONS (default) G -- Adjust Geoid Height 07-80 Reserved II_-_Iteration_Record: The Iteration record options control the conditions for ceasing to iterate the adjustment solution and what is printed after the iterations are completed. After the adjustment program stops iterating it either aborts or continues to create the final sections of the out- put. If the adjustment program is converging slowly it stops iterating after it has iterated the maximum number of itera- tions specified by this record. An option on this record may be used to control whether the program aborts immedi- ately or continue through the final output sections. Print- ing a (possibly incorrect) output can be very useful for determining why the solution is converging slowly. The absolute value of the maximum allowable normalized resi- dual may be changed with this record. If the absolute values of any of the residuals are larger than the maximum allowable normalized residual, the program aborts. The absolute value of the minimum normalized residual printed with an warning message may also be changed with this record. If on this record the user specifies the value 9999 for the maximum allowable residual, then it is changed to a MUCH larger number. This action effectively eliminates the possibility of large residuals causing the program to abort. Similarly, if the user specifies the value 999 for the minimum absolute normalized residual to be printed with a warning message, then it is also changed to a MUCH larger number - which causes no warning messages for large residu- als. The adjustment program normally does not abort if the vari- ance of unit weight is increasing, that is, if the solution is diverging. This option may be changed with this record. With this option and the maximum iterations options adjust- ment results may still be printed before the program aborts. This can be very useful for determining why the solution is not converging properly. The parameter adjustment is converged if BOTH the RMS change in the station locations AND the minimum number of itera- tions has been reached. If the parameter adjustment is con- verged and if variance factors are to be estimated, then the variance factor(s) are iterated (see the VV record). A minimum number of iterations may be specified with this record. The minimum number of iterations is an addition to the AFILE for the program DYNAPG. It is not in the documen- tation for the ADJUST program. Columns Descriptions and Defaults Type ------------------------------------------------------------ 01-02 II 03-04 Maximum number of iterations, default 5 (integer) 05-08 Maximum allowable normalized residual, units of sigma, default 1200 (integer) 09-11 Minimum printed normalized residual, units of sigma, default 600 (integer) 12-17 RMS station coordinate convergence tolerance, units of millimeters, default 3 (integer) 18-18 Display results for slowly converging or for diverging solution N -- DO NOT DISPLAY (default) Y -- Display 19-20 Minimum number of iterations, default 0 (integer) 21-80 Reserved LT_-_DL/DT record where L denotes line length. An LT record introduces an observation of the rate of change in the line length between two sites whose coordinates are specified in the BBOOK file by *80* records or by *91* records or by one of each. Columns Description and Defaults __________________________________________________________ 01-02 LT 03-06 Station Serial Number of one site 07-10 Station Serial Number of other site 11-18 Rate of change of line length in m/yr (F8.4) 19-26 Standard error of rate in m/yr (F8.4) MM_-_Adjustment_Mode_Record: The Adjustment Mode record controls the mode of adjustment, that is, to not iterate (Mode 0), or what should be done after convergence (Modes 1-3). This record also controls whether or not the standard deviations of the residuals, parameters, and station locations values should be scaled by the a-posteriori standard deviation of unit weight. Regardless of the option indicated on this record, the standard deviations will not be scaled, if the a-posteriori standard deviation of unit weight is less than 0.01. The final two options on this record control whether or not a new Blue Book file should be created containing the adjusted positions. If it is to be created, a name for the file can be specified. There is not room here to explain in detail what each adjustment mode does. A complete explanation is in the ADJUST documentation. Briefly however; Mode 0 - This is a simulation mode. There are no itera- tions. This mode is normally used for network design with simulated data. Mode 1 - The standard deviations of the residuals are assumed to be the same as the standard deviations of the observations. Also, the standard deviations of the parameters and station locations are not calculated. Mode 2 - Same as Mode 1 except standard deviations of param- eters and station locations are calculated. Mode 3 - Same as Mode 2 except standard deviations of resi- duals are calculated. Columns Descriptions and Defaults Type ------------------------------------------------------------ 01-02 MM 03-03 Mode indicator 0 -- Simulation, bypass columns 04-12 of this record 1 -- COMPUTE QUASI-NORMAL RESIDUALS (default) 2 -- Compute quasi-normal residuals and inverse 3 -- Compute normal residuals and inverse 04-04 Post-adjustment indicator N -- DO NOT SCALE STANDARD DEVIATIONS BY A-POSTERIORI STANDARD DEVIATION OF UNIT WEIGHT (default) Y -- Scale standard deviations by a-posteriori standard deviation 05-05 Create Adjusted Data File N -- DO NOT CREATE ADJUSTED DATA FILE (default) Y -- Create adjusted data file 06-12 Adjusted Data File Name, default is "NEWBB" 13-80 Reserved PP_-_Print_Output_Record: The Print Output record contains the options for controlling what is listed in the output and with what format. There are seven options on this record. The first option controls whether all of the Blue Book records are listed on the output or whether only the (non- rejected) observations are listed. Similarly, the second option controls whether all of the GFILE (GPS) records are listed on the output or whether only the (non-rejected) observations are listed. Normally all records should be listed. However, if one or the other of these files is very large and the output needs to be shortened, then these options could be used. The third option controls whether or not the constraints are listed in the output. Again, this option should be used ONLY if there are a very large number of constraints. If the constraints are not listed, then the only place that they appear in the output is in the residual section. Unless you know exactly what will be in the output, the GFILE, BBOOK, and constraints should always be listed when some or all of the residuals are to be not listed. The fourth through the eleventh options control whether or not post-adjustment residuals of observations and con- straints are listed in the output. The forth option con- trols the printing of each residual by its magnitude (divided by its standard deviation). The fifth through the tenth options control the printing of each residual by the observation type and the eleventh controls the printing of the constraint residuals. Note that in Mode=0 post- adjustment residuals are never printed because they can not be calculated (see the MM option record). Normally, all post-adjustment residuals should be printed. Using this option does save some output space as residuals considered too small do not generates output records. However, the blank lines that would have been printed around the small residuals are still printed. This will often result in out- put that only consists of blank lines. It is also probably not a good idea to use these options along with any of the options that control the printing of Blue Book or G-file observations or of constraints as you may not have enough information to determine the cause of problems in the data. The twelfth option is not implemented at present. If it were, it would be used to determine whether or not there should be a section in the output that groups the residuals around the intersection stations. If you do not desire this section to be printed, do not have any intersection stations indicated on the Blue Book *80* Control Point Records (that is do not have a "4" in column 80 of the *80* record). The thirteenth option controls whether or not the shifts in the positions of the stations are to be printed. This shift is the difference between the final adjusted position and the position on the Blue Book *80* records. Not listing the shifts might be used when the station locations on the *80* records are poor. This might occur, for instance, if the station locations are from the NAD27 coordinate system and the constraints are from the NAD83 coordinate system. The fourteenth option controls whether or not the Googe numbers are to be printed. The Googe numbers are always between zero and one. In the adjustment program they deter- mine at which parameters the solution breaks down. A Googe number may be considered a measure of the independence of a parameter. A Googe number of zero indicates that that parameter is completely dependent on the earlier parameters, while a Googe number of one indicates that that parameter is independent of the previous parameters. More information on Googe numbers may be found in the ADJUST documentation. The fifteenth option controls whether or not the observa- tional summary is printed. This option may be used to shor- ten the output when a summary of what types of observation were observed "from" or "to" each station is not of concern. The sixteenth option controls whether or not the adjusted positions are to be printed. This option may be used to shorten the output when the adjusted positions of the sta- tions are not of concern. Columns Descriptions and Defaults Type ------------------------------------------------------------ 01-02 PP 03-03 Echo Blue Book File 0 -- ECHO ALL BLUE BOOK RECORDS (default) 1 -- Echo observations only 2 -- Echo large misclosures only 04-04 Echo G-Format File 0 -- ECHO ALL G-FORMAT RECORDS (default) 1 -- Echo observations only 2 -- Echo large misclosures only 05-05 Display Constraints Y -- DISPLAY CONSTRAINTS (default) N -- Do not display 06-08 Residual Output Tolerance, units of 0.1 sigma 0 -- DISPLAY ALL RESIDUALS (default) 10 -- Display residuals greater than or equal to 1.0 sigma 999 -- Display no residuals (special 100,000,000.) 09-09 Display Direction Residuals Y -- DISPLAY DIRECTION RESIDUALS (default) N -- Do not display 10-10 Display Angle Residuals Y -- DISPLAY ANGLE RESIDUALS (default) N -- Do not display 11-11 Display Zenith Distance Residuals Y -- DISPLAY ZENITH DISTANCE RESIDUALS (default) N -- Do not display 12-12 Display Distance Residuals Y -- DISPLAY DISTANCE RESIDUALS (default) N -- Do not display 13-13 Display Astro-Azimuth Residuals Y -- DISPLAY ASTRO-AZIMUTH RESIDUALS (default) N -- Do not display 14-14 Display GPS Residuals Y -- DISPLAY GPS RESIDUALS (default) N -- Do not display 15-15 Display Constrained Residuals Y -- DISPLAY CONSTRAINED RESIDUALS (default) N -- Do not display 16-16 Display Residuals Grouped Around Intersection Sta. NOTE - This option is not yet implemented Y -- Display Residuals Grouped Around intersection stations N -- DO NOT DISPLAY (default) 17-17 Display Position Shifts Y -- DISPLAY POSITION SHIFTS (default) N -- Do not display 18-18 Display Position Googe Numbers Y -- DISPLAY GOOGE NUMBERS (default) N -- Do not display Googe numbers 19-19 Display Observational Summary Y -- DISPLAY OBSERVATIONAL SUMMARY (default) N -- Do not display Observational Summary 20-20 Display Adjusted Positions Y -- DISPLAY ADJUSTED POSITIONS (default) N -- Do not display 21-80 Reserved PV_-_Point_Velocity_Record: A point velocity record introduces an observation of 3-dimensional velocity at the location whose latitude and longitude are specified. Columns Description and Defaults _____________________________________________ 01-02 PV 04-11 Latitude (DDMMSSss) 12-12 Sense of latitude (N or S) 13-21 Longitude (DDDMMSSss) 22-22 Sense of longitude (E or W) 23-28 Northward velocity (units = 0.01 mm/yr) 29-34 Std. dev. of northward velocity (units = 0.01 mm/yr) 35-40 Eastward velocity (units = 0.01 mm/yr) 41-46 Std. dev. of eastward velocity (units = 0.01 mm/yr) 47-52 Upward velocity (units = 0.01 mm/yr) 53-58 Std. dev. of upward velocity (units = 0.01 mm/yr) QQ_-_Accuracy_Computation_Record: An Accuracy Computation records directs that the adjusted value of the distance, azimuth, and vertical angle between two stations be calcu- lated together with associated statistics indicating the accuracy of these adjusted quantities. These records are ignored if the Mode (from the MM Adjustment Mode record) is 1. The Station Serial Numbers of the two stations are required. If one (or both) are blank or are otherwise incorrect, a warning message is printed but the program does not abort. The distances, azimuths, and vertical angles and their pro- pagated standard deviations are printed in the output sec- tion entitled "LENGTH RELATIVE ACCURACIES". The horizontal shift of the length and the accuracy fraction for the dis- tance between the two stations are also listed. This section is at the end of the adjustment report. Note that the columns for the Station Serial Numbers on the QQ record corresponds to the columns for Station Serial Numbers on the Blue Book observation records. Columns Descriptions and Defaults Type ------------------------------------------------------------ 01-02 QQ 03-10 Reserved 11-14 Station Serial Number, required (integer) 15-50 Reserved 51-54 Station Serial Number, required (integer) 55-80 Reserved SM_-_Smoothing Constraint Record: The SM record enters three constraints for every grid node. The constraints specify that each component of the 3D velocity vector equals the weighted average of the corresponding component of its adjacent grid nodes. Standard errors for these constrains are specified as V/(B*B) where V is in mm/yr and B is in km. B is usually taken as the approximate distance between grid nodes. Columns Descriptions and Defaults __________________________________________________________ 01-02 SM 03-07 V for northward smooting in mm/yr (integer) 08-12 V for eastward smoothing in mm/yr (integer) 13-17 V for vertical smoothing in mm/yr (integer) 18-22 B for weighting base in km (integer) SS_-_Auxiliary_Parameter_Indicator_and_Constraint_Record: An Auxiliary Parameter Indicator and Constraint record creates an auxiliary parameter. Auxiliary parameters are used to estimate (a) scale factors for GPS data; (b) refraction parameters for zenith distances; or (c) scale factors for mark-to-mark distances. In this section, these will all be referred to as scale corrections. Each SS record creates an auxiliary parameter for one obser- vation type. The observations are grouped, for a particular observation type, by a range of dates. There is no default for the observation type. The defaults for the starting and ending dates are set to include all valid observation dates. There may be no overlapping dates on auxiliary parameters for the same observation type. The auxiliary parameters may be constrained or not con- strained to particular values. The initial values for the scale corrections may be specified with this record. If a standard deviation is included on this record, then the scale factor is constrained with that standard deviation. The auxiliary parameters estimate scale corrections for various types of observations from the observations within a given time period. In general, they should be included in an adjustment only after blunders and other non-systematic errors have been removed from the data set. However, if you suspect that the scale for a set of observations is large, the SS record may be included in the AFILE at any time. Columns Descriptions and Defaults Type ------------------------------------------------------------ 01-02 SS 03-04 Observation Type 25 -- GPS (scale) 40 -- Zenith distances (refraction) (also *42*) 52 -- Mark-to-mark distances (scale) (also *54*) 05-08 Start year, default 1801 (integer) 09-10 Start month, default 01 (integer) 11-12 Start day, default 01 (integer) 13-14 Start hour, default 00 (integer) 15-16 Start min., default 00 (integer) 17-17 Time Code, Start Time, default Z 18-21 End year, default 2099 (integer) 22-23 End month, default 12 (integer) 24-25 End day, default 31 (integer) 26-27 End hour, default 23 (integer) 28-29 End min., default 59 (integer) 30-30 Time Code, Start Time, default Z 31-35 Scale Correction, units of .00000001 (.01 ppm), default 0 (integer) 36-40 Standard Deviation of Scale, units of .01 ppm, default 100 (integer) 41-80 Reserved SV_-_Site Velocity Record: An SV record introduces an observation of 3-dimensional velocity at a geodetic site whose latitude and longitude are specified in the BBOOK file by either a *80* record or a *91* record. Columns Descriptions and Defaults __________________________________________________________ 01-02 SV 03-06 Station Serial Number 23-28 Northward velocity (units = 0.01 mm/yr) (integer) 29-34 Standard error of northward velocity (units = 0.01 mm/yr) 35-40 Eastward velocity (units = 0.01 mm/yr) 41-46 Standard error of eastward velocity (units = 0.01 mm/yr) 47-52 Upward velocity (units = 0.01 mm/yr) 53-58 Standard error of upward velocity (units = 0.01 mm/yr) VF_-_Variance_Factor_Record: A variance factor record specifies an observation type and a value. DYNAPG scales the standard error of each observation of the specified type by the square root of the given value. Columns Descriptions and Defaults ___________________________________________________________ 01-02 VF 03-04 Observation type 04 -- GPS X-component 05 -- GPS Y-component 06 -- GPS Z-component 07 -- distance 08 -- azimuth 09 -- zenith distance 10 -- horizontal angle 11 -- horizontal direction 19 -- point velocity (north) 20 -- point velocity (east) 21 -- point velocity (up) 25 -- site velocity (north) 26 -- site velocity (east) 27 -- site velocity (up) 31 -- dL/dt 05-09 Square of scaling factor (Fortran format = F5.2) APPENDIX_B BLUE BOOK DATA FILE FORMAT (BBOOK) FOR THE DYNAPG ADJUSTMENT PROGRAM This appendix is to be used only with the DYNAPG adjustment pro- gram. It is a supplement to the official Blue Book as published by NOAA. A number of records which are a part of the Blue Book format definition are ignored by DYNAPG (or ADJUST). Also, some options on particular records are not recognized. Finally, one record which is not part of the Blue Book format is recognized by DYNAPG (but not by ADJUST). This appendix provides a brief sum- mary. The first table provides a list of Blue Book format data records which are ignored by DYNAPG. Alternate records are suggested, when appropriate. If the information on the record is never used in the DYNAPG adjustment then the table lists it as ignored. Com- ment records are not on this list, but are likewise ignored. Blue Book records which are ignored by DYNAPG, are nontheless printed in the Blue Book section of the output (subject to the options specified on the AFILE "PP" Print Output record). Also note that a "O" in column 6 of the observation records (*20*, *22*, *30*, *32*, *40*, *42*, *52*, *54*, and *60*) will cause that record to be rejected (i.e. treated as a comment). CODE TYPE OF RECORD ALTERNATE ----------------------------------------------------------------- Data_Set_Identification_and_Data_Set_Termination_records: ignored. *10*_and_*11*_Project_Title_records: ignored. *12*_Project_Information_record: ignored. *13*_Geodetic_Datum_and_Ellipsoid_Record: use an AFILE "AA" record. *25*,_*27*,_and_*28*_Global_Positioning_System_(GPS)_Blue_Book Records: ignored. All GPS required data is in the GFILE - See Appendix C. *50*_Taped_Distance_Record: use a *52* record (but see *52* record below). *51*_Unreduced_Distance_Record: use a *52* record (but see *52* record below). *53*_Unreduced_Long_Line_Record: use a *54* record (but see *52* record below). *61*_Geodetic_Azimuth_Record: use a *60* record. *70*_Instrument_Records: ignored. *81*_Control_Point_Record_(UTM/SPC): use a *80* record. *82*_Reference_or_Azimuth_Mark_Record: use a *80* record. *83*_Bench_Mark_Record: use a *80* record. *90*__Fixed_Control_Record: use AFILE "CC" record. The next section of this appendix contains a list of the options which MUST be of a specific type for DYNAPG. The list also con- tains other information about the Blue Book records. The list is NOT complete, only the most common and/or important options and information are given. *20*_and_*22*_Horizontal_Direction_Records: The field set number must differ for each *20* record which has the same stand- point. The observation must be reduced to the station marker (i.e. the height of the instrument must be zero). The external consistency is used rather than the internal consistency for the standard deviation of the observation. If the external consistency columns are blank, then the standard deviation of the horizontal direction observation is 1 arc second. *30*_and_*32*_Horizontal_Angle_Records: Same as for the *20* and *22* records except that the standard deviation of the observation is obtained from the number of blanks in the seconds portion of the horizontal angle value. If columns 69-71 are blank, then the standard deviation is set to 1 minute of arc. If columns 70-71 are blank, then the stan- dard deviation is set to 10 seconds of arc. If only column 71 is blank, then the standard deviation is set to 1 second of arc. *40*_and_*42*_Vertical_Angle_Records: Same as for the *20* and *22* records. Also, the angle must be a zenith distance (i.e. column 72 must contain a "Z") *52*_Reduced_Distance_Records_and_*54*_Reduced_Long_Line_Records: The distances on these records must be mark-to-mark dis- tances (i.e. code "X" ). The external consistency columns contain the standard deviation of a mark-to-mark distance. The external consistency on the *52* record is in units of millimeters (not ppm) and has a default of 0.01 mm. The external consistency on the *54* record is in units of decimeters and has a default of 0.1 mm. The geoid heights and elevations of the standpoint and forepoint are obtained from the *80* and *84* records, not these records. *60*_Laplace_Azimuth_Records: The Laplace azimuth on this record is assumed to be in a geodetic coordinate system. The prime vertical component from this record, not from the *85* Deflection record, is used by DYNAPG. If the azimuth is an astronomic azimuth then the prime vertical component should be zero. The observation must be reduced to the station marker (i.e. the height of the instrument and the height of the target must be zero). The external consistency is used rather than the internal consistency for the standard devia- tion of the observation. If the external consistency columns are blank, then the standard deviation of the hor- izontal direction observation is 1 arc second. *80*_Control_Point_Record: If the elevation columns are blank, then the elevation is taken from the AFILE "EE" Default Mean Sea Level Elevation Record. This record is required for all stations participating in the adjustment. That is, every standpoint or forepoint station serial number which is on a non-rejected data record must have a corresponding *80* record. Initial station locations are obtained from this record. *84*_Geoid_Height_Record: This record is required for all sta- tions whose geoid height is not the same as that given on the AFILE "GG" Default Geoid Height record. *85*_Deflection_Record: The *89* Astronomic Station Location record (which is not a valid Blue Book record) may be used instead of this record for deflection information. The last section of this appendix contains a table of non-Blue Book records which,if they are in the BBOOK file, will be treated by DYNAPG as if they are Blue Book records. An asterisk between the columns and the description indicates required items for that record. CODE TYPE OF RECORD ----------------------------------------------------------------- *89* Astronomic Station Location Record *89*_-_Astronomic_Station_Location_record: The only record currently of this type is an Astronomic Station Location record. It has been given a data code type of *86*. The Astronomic Station Location record can be used instead of the Blue Book record *85* Deflection Record to specify the astronomic values for a control point. (In DYNAPG the deflection information on a *85* record is combined with the geographic station location information on an *80* Control Point record and saved as an astronomic station location.) Columns Descriptions Type ------------------------------------------------------------ 01-06 Sequence number (Integer) 07-10 * *89* 11-14 * Station Serial Number (Integer) 15-44 Reserved 45-46 * Astronomic Latitude, Degrees (Integer) 47-48 * Astronomic Latitude, Minutes (Integer) 49-55 * Astronomic Latitude, Seconds, units of 0.00001 seconds of arc (Integer) 56-56 Direction of Astronomic Latitude, N or S, default N (Character) 57-59 * Astronomic Longitude, Degrees (Integer) 60-61 * Astronomic Longitude, Minutes (Integer) 62-68 * Astronomic Longitude, Seconds, units of 0.00001 seconds of arc (Integer) 56-56 Direction of Astronomic Longitude, E or W, default W (Character) 57-80 Reserved *91* Control_Point_Record for station whose position is not to be adjusted This record is used to specify positional coordinates for geodetic marks involved in SV records, or LT records, or GT records for which no correesponding *80* records exist. An *91* record has the same format as an *80* record.