sgsim

Process Help

SGSIM - sequential gaussian simulation

Process Name	Menu Path	Link to Command Table
SGSIM	Estimate ribbon \| Conditional Simulation \| Sequential Gaussian	Click here

Introduction
Files, Fields and Parameters
Notes
Example
Errors

Introduction

This is a Superprocess and running it may have an effect on other Datamine files in the project. More...

The maximum number of characters in the path name of the project folder should not exceed 256 characters. An error message will be displayed if this path length is exceeded.

The SGSIM process creates a regular grid of conditionally-simulated points using sequential gaussian simulation.

How to use

Multiple realizations (simulations) can be generated in a single run of the process. The main inputs are:

A sample data file (X,Y,Z and grade field), and the corresponding variogram model.
The definition of the output grid of points.

There is a choice of kriging methods:

simple kriging.
ordinary kriging.
simple kriging with a locally varying mean.
kriging with an external drift.
collocated cokriging with one secondary grade.

The outputs are:

A points file (XPT, YPT, ZPT, grade and realization number) containing the simulated points.
A block model file. This contains the same information as the points file, except that each realization is a separate field in the model. It provides an alternative way of viewing the results.
Output file for the transformation table. This contains the grade field GRADE from the IN sample file, and the field TRANDATA, giving the transformed value. The file is sorted by GRADE.
Output statistics table giving statistics for the input sample data, and the output simulated values.

A statistical analysis of the simulated points can then be carried out with process CSMODEL. The CSMODEL process also includes the option to average simulated points over a parent cell. The process is based on the GSLIB code for sequential gaussian simulation (sgsim).

For details of this method refer to the following resource:

GSLIB. Geostatistical Software Library and User's Guide by Clayton V.Deutsch and Andre G.Journel, published by Oxford University Press, Second Edition, 1998. ISBN 0-19-510015-8.

There are some differences between Datamine and GSLIB conventions. However, unless otherwise stated, Datamine conventions are used here. Where required these are converted to GSLIB conventions by the process.

Defining the Grid

The grid of points can be defined in one of three ways:

By specifying an empty prototype model, and parameters XPPPC, YPPPC, ZPPPC. A set of points is created for every parent cell where the number of points per parent cell is defined by the parameters XPPPC, YPPPC, ZPPPC. In effect the parameters define the subcell splitting where each simulated point corresponds to a subcell centre.
By specifying a prototype model containing cells and subcells, and parameters XPPPC, YPPPC, ZPPPC. This is similar to option 1 except that points are only created for those parent cells that contain one or more subcells. A full set of points is always created for each parent cell.
By specifying the parameters XMIN, XSIZE, NX, YMIN, YSIZE, NY, ZMIN, ZSIZE, NZ. Note that XMIN defines the location of the point with the minimum X value ie the centre of the corresponding subcell, not the model origin. XSIZE defines the spacing between grid points and NX defines the total number of grid points. The parameters XPPPC, YPPPC, ZPPPC then define the number of points per parent cell in the output model containing the simulated points. The values of all 12 parameters are stored as implicit fields in the output POINTS file, so that they can be accessed by subsequent processes (eg CSMODEL).

A simulated value is calculated for each grid point. If there is insufficient data within the search volume then an absent data value is assigned.

Input Sample File &IN

The input sample data file includes the coordinate fields X, Y, Z and the grade field GRADE. The GRADE field may contain either the original sample values or the normal score values, depending on parameter TRANTYPE. Normal score values can be calculated using process NSCORE.

The IN file may also include a declustering weights field DCWGT and a secondary field SECFLD1. The declustering weights are used to adjust the sample histogram. The contents of the secondary field depend on the value of parameter KTYPE.

Input Variogram Parameter File &VMODPARM

The input variogram model should be for the normal scores. ie the sample data should have been normalized (process NSCORE) before the variogram was calculated.

The Co and Ci values are normalized automatically by the process ie they are divided by (Co + ΣCi) so that the sill is 1.

The three rotations defining the orientation of the ellipsoid of ranges must be in the order:

Around the Z axis (azimuth rotation)
Around the X axis (dip rotation)
Around the Y axis (plunge rotation)

Therefore, in the input variogram model parameter file (VMODPARM) the values of the VAXIS fields must be:

VAXIS1 = 3 (Z)
VAXIS2 = 1 (X)
VAXIS3 = 2 (Y)

Input Secondary File &SECFILE

The input secondary file is required if the Kriging Type (@KTYPE) = 2, 3 or 4. Its content depends on the value of @KTYPE as follows:

2.    the locally varying mean for Simple Kriging
3.    the drift variable when kriging with an external drift
4.    the secondary grade for Collocated Cokriging

The secondary file must be a block model file with a cell or subcell and data value for each point to be simulated. The secondary field must not include absent data values.

If the grid of simulated points is defined by the &PROTO file and the file does not contain any records then the &SECFILE file should have the same model parameters (XMORIG, XINC, NX, etc) as the &PROTO file and should have subcells as defined by parameters by XPPPC , YPPPC, ZPPPC, at every possible location.

If the &PROTO file contains records then the &SECFILE file must still have the same model parameters and subcells defined by parameters XPPPC, YPPPC, ZPPPC . However the subcells must only cover a 3D rectangular volume as defined by the minimum and maximum parent cell in each of the X, Y and Z directions.

If the grid of simulated points is defined by parameters XMIN , XSIZE , NX etc then the &SECFILE model must include a cell or subcell, and data value, at every grid location. The model parameters do not matter as long as there is a cell or subcell at all grid points.

Search Volume

As for the variogram parameters, the three rotations for the search volume must be in the order around Z, around X, then around Y.

Output Points File POINTS

The output points file contains the simulated points, sorted on XPT, YPT, ZPT. As well as the coordinate fields XPT, YPT, ZPT the file will contain the simulated value GRADE and field SIMNUM giving the simulation (realization) number. This is an integer, starting at 1, with a maximum of NSIM.

The file also contains 12 implicit field whose default values define the regular grid of simulated points, which are stored in terms of the equivalent block model. These fields will be used subsequently by the CSMODEL process for averaging points into block values and creating block models.

Field	Stored	Description
XPT	Yes	X coordinate of simulated point
YPT	Yes	Y coordinate of simulated point
ZPT	Yes	Z coordinate of simulated point
SIMNUM	Yes	Realization number
{grade}	Yes	Simulated value
XPPPC	No	Number of points per parent cell in the X direction
YPPPC	No	Number of points per parent cell in the Y direction
ZPPPC	No	Number of points per parent cell in the Z direction
XMORIG1	No	X origin of model
YMORIG1	No	Y origin of model
ZMORIG1	No	Z origin of model
XINC1	No	Parent cell size in the X direction
YINC1	No	Parent cell size in the Y direction
ZINC1	No	Parent cell size in the Z direction
NX1	No	Number of parent cells in the X direction
NY1	No	Number of parent cells in the Y direction
NZ1	No	Number of parent cells in the Z direction

The XMORIG1, YMORIG1, and ZMORIG1 field values define the origin of the block model. These are different from the XMIN, YMIN, and ZMIN parameters which define the minimum coordinates of a subcell.

Output Model File &MODEL

The output block model file contains the simulated points. This is the same data as the POINTS file, but in block model format. If multiple realizations have been selected (NSIM greater than 1) then the field name for each realization will be SIM1, SIM2, SIM3, etc.

If the grid of simulated points is defined by the PROTO file then the MODEL file will have the same model parameters (XMORIG, XINC, NX, etc); otherwise the model parameters will be based on parameters XMIN, XSIZE, NX, etc. Note that XMIN is the X coordinate of the location of the point with the minimum X value and so is half a cell displaced from the model origin.

Field	Stored	Description
IJK	Yes	IJK index
XC	Yes	X coordinate of centre of parent cell
YC	Yes	Y coordinate of centre of parent cell
ZC	Yes	Z coordinate of centre of parent cell
SIM1	Yes	1st simulated value
SIM2	Yes	2nd simulated value
....	....	....
SIMn	Yes	nth simulated value
XINC	Yes	Parent cell size in the X direction
YINC	Yes	Parent cell size in the Y direction
ZINC	Yes	Parent cell size in the Z direction
XMORIG	No	X origin of model
YMORIG	No	Y origin of model
ZMORIG	No	Z origin of model
NX	No	Number of parent cells in the X direction
NY	No	Number of parent cells in the Y direction
NZ	No	Number of parent cells in the Z direction

Program Limits

Maximum number of:	Value
samples in input IN file	500,000
simulated points in X direction	See "Maximum number of points, below"
simulated points in Y direction	See below
simulated points in Z direction	See below
points in X direction in covariance table	71
points in Y direction in covariance table	71
points in Z direction in covariance table	21
superblock nodes in X direction	21
superblock nodes in Y direction	21
superblock nodes in Z direction	11
previously simulated nodes to use	48
data for one simulation	48
variogram structures	4
characters in full pathname of project file	256
realizations in the output MODEL file	200

Maximum Number of Points

If the 3D grid of points is defined by parameters NX, NY and NZ and NSIM is the number of simulations then the maximum number of simulated points is:

NX*NY*NZ*NSIM <= 2,000,000,000.

If the 3D grid of points is defined by the input prototype model PROTO then the corresponding limit is:

NX*XPPPC * NY*YPPPC * NZ*ZPPPC * NSIM <= 2,000,000,000

where NX is the number of parent cells in X in the PROTO model and XPPPC is the parameter defining the number of points per parent cell in X, etc. Whether or not the PROTO model includes cells does not affect the calculation.

Project Folder Name

The following limitations currently apply to the folder name of the project:

A maximum of 80 characters in the full path name of the project folder e.g.c:\database\consim\project1.
The full path name of the project folder must not include spaces e.g.c:\database\consim\my_project1.

Command Prompt Window

The Command Prompt window is opened during the processing and contains progress information for the run of the sgsim executable. You can left click the mouse in the window to stop it scrolling, and then right click to start it again. The window is closed automatically when the sgsim executable finishes.

System Files

_sgsdbg.txt	Debug output. Contents depend on the value of the DBGLEVEL parameter
_sgslog.txt	Log file. Only useful if there is a problem.
_sgs_*.txt	Temporary system files. These will be deleted if the process terminates cleanly.
_sp*.dm	Temporary Datamine files. These will be deleted if the process terminates cleanly.

All files matching the template _sgs_*.txt and _sp*.dm will be deleted as the process terminates. Therefore you should not use any of these file names.

Files, Fields and Parameters

Input Files

Name	Description	I/O Status	Required	Type
IN	Input sample data file. This must include the coordinate fields X , Y , Z and the grade field GRADE . It may also include the declustering weights field DCWGT and the secondary field SECFLD1 . If an IN	Input	No	Table
VMODPARM	Variogram model parameter file. Each record in this file defines a variogram model type and its parameters.	Input	Yes	Variogram - Model
PROTO	Input prototype model file to define the regular grid for the simulated points. If the PROTO file contains records then simulated points are created for all parent cells that contain at least one subcell. If the PROTO file does not contain any records then simulated points are created for all parent cells. The number of simulated points per parent cell is defined by the parameters XPPPC , YPPPC , ZPPPC . If the file is not specified then the grid is defined by the parameters XMIN , XSIZE , NX etc.	Input	No	Block Model Prototype
REFDIST	Input reference distribution to define required transformation. As well as specifying the file, parameter TRANTYPE must be set to 2. The file must include the field REFGRADE , to define the reference distribution, and may also include the field REFWGT to define declustering weights.	Input	No	Table
SECFILE	Secondary file, required if KTYPE = 2,3 or 4. This contains the locally varying mean ( KTYPE =2), the external drift variable ( KTYPE =3) or the secondary variable for collocated cokriging ( KTYPE =4). This must be a block model file with a cell or subcell (and data value) at each simulated point. - if the grid of simulated points is defined by the PROTO file and the file does not contain any records then the SECFILE file should have the same model parameters (XMORIG, XINC, NX, etc) as the PROTO file and should have subcells as defined by parameters XPPPC , YPPPC , ZPPPC at every possible location. - if the PROTO file contains records then the SECFILE file must still have the same model parameters and subcells defined by parameters XPPPC , YPPPC , ZPPPC . However the subcells need only cover a 3D rectangular volume as defined by the minimum and maximum parent cell in each of the X, Y and Z directions. - if the grid of simulated points is defined by parameters XMIN , XSIZE , NX etc then the SECFILE model must include a cell or subcell, and data value, at every grid location. The model parameters do not matter as long as there is a cell or subcell at all grid points.	Input	No	Block Model

Output Files

Name	I/O Status	Required	Type	Description
POINTS	Output	Yes	Point Data	Output points file containing simulated points. As well as the coordinate fields XPT, YPT, ZPT the file will contain the simulated value GRADE and field SIMNUM giving the simulation (realization) number. This is an integer, starting at 1, with a maximum of NSIM .
MODEL	Output	No	Block Model	Output block model file containing simulated points. This contains the same data as the POINTS file, but in block model format. If multiple realizations have been selected ( NSIM greater than 1) then the field name for each realization will be SIM1, SIM2, SIM3, etc. A maximum of 44 realizations are allowed in the MODEL file in the single precision version of your project, or 200 in the double precision version. If the grid of simulated points is defined by the PROTO file then the MODEL file will have the same model parameters (XMORIG, XINC, NX, etc); otherwise the model parameters will be based on parameters XMIN , XSIZE , NX etc.
TRANDIST	Output	No	Table	Output file for the transformation table. This will contain the grade field GRADE from the IN sample file and the field TRANDATA giving the transformed value. The file will be sorted by GRADE .
STAT_TBL	Output	No	Table	Output statistics table giving statistics for the input sample data and the output simulated values.

Fields

Name	Description	Source	Required	Type	Default
X	X coordinate of the GRADE field in the IN sample file.	IN	Yes	Numeric	X
Y	Y coordinate of the GRADE field in the IN sample file.	IN	Yes	Numeric	Y
Z	Z coordinate of the GRADE field in the IN sample file.	IN	Yes	Numeric	Z
GRADE	Field in the IN sample file defining the grade to be simulated. This may contain either the original sample values or the normal score values, depending on parameter TRANTYPE .	IN	Yes	Numeric	Undefined
DCWGT	Optional declustering weights field in the IN sample file.	IN	No	Numeric	Undefined
SECFLD1	Optional secondary field in the IN sample file. The contents of this field depend on the value of parameter KTYPE .	IN	No	Numeric	Undefined
SECFLD2	Field in secondary file SECFILE . This is required if KTYPE = 2,3 or 4. The field contains the locally varying mean ( KTYPE =2), the external drift variable ( KTYPE =3) or the secondary variable for collocated cokriging ( KTYPE =4) . If @KTYPE=4 *SECFLD2 must be untransformed.	SECFILE	No	Numeric	Undefined
REFGRADE	Reference grade field, defining the reference distribution, in the REFDIST file.	REFDIST	No	Numeric	Undefined
REFWGT	Optional reference weight field, defining declustering weights, in the REFDIST file.	REFDIST	No	Numeric	Undefined

Parameters

Name

Description

Required

Default

Range

Values

MINGRADE

Minimum GRADE value input from the IN sample file. Values less than the minimum are ignored.

Undefined

MAXGRADE

Maximum GRADE value input from the IN sample file. Values greater than the maximum are ignored.

Undefined

TRANTYPE

Transform data option: 0 = no transformation; the field GRADE in the IN sample file is assumed to have a standard normal distribution, and the simulated value GRADE in the output POINTS file will also be untransformed. 1 = transform the GRADE values using the standard normal distribution. 2 = transform the GRADE values using the distribution defined in the input file REFDIST .

0,2

0,1,2

MINSIMGR

Minimum simulated grade value. This is used to restrict values in the back-transformation process.

Undefined

MAXSIMGR

Maximum simulated grade value. This is used to restrict values in the back-transformation process.

999999

Undefined

LOTAIL

Back-transformation method in the lower tail of the distribution to a minimum grade of MINSIMGR . 1 = Linear interpolation. 2 = Power model interpolation. The power used is defined by LOPAR .

1,2

LOPAR

Power parameter used in back-transformation of grades in the lower tail of the distribution to a minimum of MINSIMGR . LOTAIL must be set to 2.

0,+

Undefined

UPTAIL

Back-transformation method in the upper tail of the distribution to a maximum grade of MAXSIMGR . 1 = Linear interpolation. 2 = Power model interpolation. The power used is defined by UPPAR . 4 = Hyperbolic model extrapolation using power parameter defined by UPPAR .

1,4

1,2,4

UPPAR

Power parameter used in back-transformation of grades in the upper tail of the distribution to a maximum of MAXSIMGR . UPTAIL must be set to 2 or 4.

0,+

Undefined

NSIM

Number of realizations to generate. If a MODEL file has been selected then the maximum number of realizations is 44 for the single precision version of Datamine Studio packages or 200 for the double precision version. If a MODEL file has not been selected then there is no limit on the maximum number of realizations.

1,+

Undefined

XPPPC

The number of simulated points per parent cell to be created in the X direction in the output MODEL file. This corresponds to the number of subcells per parent cell in the X direction. The value of parameter XPPPC is also stored as the default value of implicit field XPPPC in the output POINTS file so that it can be accessed by subsequent processes eg CSMODEL.

Undefined

YPPPC

The number of simulated points per parent cell to be created in theY direction in the output MODEL file. This corresponds to the number of subcells per parent cell in the Y direction. The value of parameter YPPPC is also stored as the default value of implicit field YPPPC in the output POINTS file so that it can be accessed by subsequent processes eg CSMODEL.

Undefined

ZPPPC

The number of simulated points per parent cell to be created in the Z direction in the output MODEL file. This corresponds to the number of subcells per parent cell in the Z direction. The value of parameter ZPPPC is also stored as the default value of implicit field ZPPPC in the output POINTS file so that it can be accessed by subsequent processes eg CSMODEL.

Undefined

XMIN

Minimum X coordinate of the regular grid of simulated points created in the output POINTS and MODEL files. If a PROTO file is specified this parameter is ignored.

Undefined

YMIN

Minimum Y coordinate of the regular grid of simulated points created in the output POINTS and MODEL files. If a PROTO file is specified this parameter is ignored.

Undefined

ZMIN

Minimum Z coordinate of the regular grid of simulated points created in the output POINTS and MODEL files. If a PROTO file is specified this parameter is ignored.

Undefined

XSIZE

Spacing between simulated points in the X direction. If a PROTO file is specified this parameter is ignored.

Undefined

YSIZE

Spacing between simulated points in the Y direction. If a PROTO file is specified this parameter is ignored.

Undefined

ZSIZE

Spacing between simulated points in the Z direction. If a PROTO file is specified this parameter is ignored.

Undefined

Number of simulated points in the X direction. If a PROTO file is specified this parameter is ignored.

1,+

Undefined

Number of simulated points in the Y direction. If a PROTO file is specified this parameter is ignored.

1,+

Undefined

Number of simulated points in the Z direction. If a PROTO file is specified this parameter is ignored.

1,+

Undefined

SEED

Random number seed. This should be a large odd integer. If the same seed is used for multiple runs then the same set of simulated grades will be created.

1,+

Undefined

MINDATPT

The minimum number of original data to be used to simulate a grid node. If there are fewer than MINDATPT data points the node is not simulated.

1,+

Undefined

MAXDATPT

The maximum number of original data to be used to simulate a grid node. If there are more than MAXDATPT data points the nearest points are selected.

1,48

Undefined

MAXSIMPT

The maximum number of previously simulated nodes to use for the simulation of another node.

1,48

Undefined

SSTRAT

Search strategy: 0 = data and previously simulated grid nodes are searched separately. Data are searched with a super block search and previously simulated nodes with a spiral search. 1 = data are relocated to grid nodes and a spiral search is used. MINDATPT and MAXDATPT are not then considered.

0,1

MULTGRID

Search strategy for previously simulated nodes: 0 = spiral search if greater than or equal to 1 then MULTGRID defines the number of grids for a multiple grid simulation.

0,9 Value: 0,1,2,3,4,5,6,7,8,9

MAXPEROC

Maximum number of original data points per octant. If set to zero then octant search is not used. If octant search is used then MAXDATPT is ignored.

0,+

Undefined

SDIST1

Search distance in the X direction. This may be a rotated X direction if parameters SANGLE1 , SANGLE2 or SANGLE3 are non zero.

Undefined

SDIST2

Search distance in the Y direction. This may be a rotated Y direction if parameters SANGLE1 , SANGLE2 or SANGLE3 are non zero.

Undefined

SDIST3

Search distance in the Z direction. This may be a rotated Z direction if parameters SANGLE1 , SANGLE2 or SANGLE3 are non zero.

Undefined

SANGLE1

First rotation angle for search ellipsoid. The rotation must be around the Z axis.

-360,360

Undefined

SANGLE2

Second rotation angle for search ellipsoid. The rotation must be around the X axis.

-360,360

Undefined

SANGLE3

Third rotation angle for search ellipsoid. The rotation must be around the Y axis.

-360,360

Undefined

KTYPE

Kriging type: 0 = simple kriging 1 = ordinary kriging 2 = simple kriging with a locally varying mean where the mean is defined by field SECFLD2 in file SECFILE . 3 = kriging with an external drift where the drift is defined by field SECFLD2 in file SECFILE . 4 = collocated cokriging with one secondary grade defined by field SECFLD2 in file SECFILE .

0,4

0,1,2,3,4

VMODNUM

Variogram model number in VMODPARM file.

Undefined

CORCOEFF

Correlation coefficient used for collocated kriging (ie when KTYPE = 4).

Undefined

VARRED

Variance reduction factor used for collocated kriging (ie when KTYPE = 4) .

Undefined

RECTGRID

Rectangular grid flag. This only applies if there are cells in the input PROTO model.

Option	Description
0	Simulated points will only be output where the points lie within a parent cell as defined by the PROTO model.
1	Simulated points will be output at all locations inside the 3D rectangular volume enclosing the PROTO cells. This is essential if the output model file is to be used as the secondary file for a subsequent run of SGSIM with KTYPE=4 for cosimulation.

0,1

DBGLEVEL

Debug level. Controls the amount of debug information written to the debug file _sgsdbg.txt. 0 is the minimum and 3 the maximum.

0,3

0,1,2,3

Notes

No additional notes.

Example

!SGSIM	&IN(sample1),&VMODPARM(vpar1),&POINTS(simpts1),&MODEL(simmod1),
	&TRANDIST(trandis1),X(XPT),Y(YPT),Z(ZPT),GRADE(AU),
	@MINGRADE=0.0,@MAXGRADE=999.0,@TRANTYPE=1.0,@MINSIMGR=0.0,
	@MAXSIMGR=20.0,@LOTAIL=1.0,@LOPAR=1.0,@UPTAIL=1.0,
	@UPPAR=1.0,@NSIM=5.0,@XMIN=8640.0,@YMIN=3360.0,
	@ZMIN=160.0,@XSIZE=5.0,@YSIZE=5.0,@ZSIZE=5.0,@NX=22.0,
	@NY=20.0,@NZ=18.0,@SEED=69069.0,@MINDATPT=1.0,@MAXDATPT=12.0,
	@MAXSIMPT=12.0,@SSTRAT=0.0,@MULTGRID=0.0,@MAXPEROC=0.0,
	@SDIST1=17.0,@SDIST2=43.0,@SDIST3=15.0,@SANGLE1=20.0,
	@SANGLE2=30.0,@SANGLE3=0.0,@KTYPE=0.0,@VMODNUM=1.0,
	@CORCOEFF=0.0,@VARRED=1.0,@DBGLEVEL=0.0

Output Window

SGSIM

Creating simulations ....
       1 file(s) copied.
Processing simulation results ....

2248 grid points not simulated - assigned absent data

FORMAT TIME > 9:20:29

Summary of input data and simulation runs

========================================================
Data      Number  Minimum   Maximum      Mean   Variance
========================================================
Samples      955    0.500    15.100     5.536      6.122
SIM1        5672    0.149    18.039     5.543      6.243
SIM2        5672    0.278    19.678     5.488      6.129
SIM3        5672    0.156    19.407     5.463      6.360
SIM4        5672    0.030    19.562     5.404      6.476
SIM5        5672    0.155    19.053     5.505      5.899
SIM_all    28360    0.030    19.678     5.481      6.221
--------------------------------------------------------

5 simulations. The following files have been created:
   trandis1: transformed distribution file with 955 records
   simpts1: point data file with 39600 records
   simmod1: model file with 7920 records

Command Prompt Window

Simulated points file opened ok

SGSIM Version: 2.000

data file = c:\database\cond_sim\gslib\testdata\_sgs_in.txt
input columns = 1 2 3 4 0
0
trimming limits = 0.0000000E+00 999.0000
transformation flag = 1
transformation file =
c:\database\cond_sim\gslib\testdata\_sgs_trn.txt
consider smoothed distribution (1=yes) = 0
file with smoothed distribution =
c:\database\cond_sim\gslib\testdata\_sgs_dis.txt
columns = 1 2
data limits (tails) = 0.0000000E+00 20.00000
lower tail = 1 1.000000
upper tail = 1 1.000000
debugging level = 0
debugging file = c:\database\cond_sim\gslib\testdata\_sgsdbg.txt
output file (not used)
c:\database\cond_sim\gslib\testdata\_sgs_out.txt
number of realizations = 5
X grid specification = 22 8640.000 5.000000
Y grid specification = 20 3360.000 5.000000
Z grid specification = 18 160.0000 5.000000
random number seed = 69069
min and max data = 1 12
maximum previous nodes = 12
two-part search flag = 0
multiple grid search flag = 0 0
number of octants = 0
search radii = 17.00000 43.00000 15.00000
search anisotropy angles = 20.00000 -30.00000 0.0000000E+00
kriging type = 0
secondary model file =
c:\database\cond_sim\gslib\testdata\_sgs_sec.txt
column in secondary model file = 1
nst, c0 = 2 0.0000000E+00
it,cc,ang[1,2,3]; 1 0.5737710 20.00000 -30.00000
0.0000000E+00
a1 a2 a3: 12.00000 7.000000 11.00000
it,cc,ang[1,2,3]; 1 0.4262300 20.00000 -30.00000
0.0000000E+00
a1 a2 a3: 43.00000 17.00000 15.00000

WARNING the sill of your variogram is not 1.0!
the sill = 1.000001

Setting up transformation table
Reading input data

Data for SGSIM: Number of acceptable data = 955
Number trimmed = 0
Weighted Average = 5.5359
Weighted Variance = 6.1222

Setting up rotation matrices for variogram and search
Setting up super block search strategy

Working on realization number 1
Realization 1 of 5. Node 792 of 7920
Realization 1 of 5. Node 1584 of 7920
Realization 1 of 5. Node 2376 of 7920
Realization 1 of 5. Node 3168 of 7920
Realization 1 of 5. Node 3960 of 7920
Realization 1 of 5. Node 4752 of 7920
Realization 1 of 5. Node 5544 of 7920
Realization 1 of 5. Node 6336 of 7920
Realization 1 of 5. Node 7128 of 7920
Realization 1 of 5. Node 7920 of 7920

Back transforming for realization 1

Realization 1: number = 5672
mean = 0.0028 (close to 0.0?)
variance = 1.0159 (close to gammabar(V,V)? approx. 1.0)

Working on realization number 2
Realization 2 of 5. Node 792 of 7920
Realization 2 of 5. Node 1584 of 7920
Realization 2 of 5. Node 2376 of 7920
Realization 2 of 5. Node 3168 of 7920
Realization 2 of 5. Node 3960 of 7920
Realization 2 of 5. Node 4752 of 7920
Realization 2 of 5. Node 5544 of 7920
Realization 2 of 5. Node 6336 of 7920
Realization 2 of 5. Node 7128 of 7920
Realization 2 of 5. Node 7920 of 7920

Back transforming for realization 2

Realization 2: number = 5672
mean = -0.0192 (close to 0.0?)
variance = 0.9966 (close to gammabar(V,V)? approx. 1.0)

Working on realization number 3
Realization 3 of 5. Node 792 of 7920
Realization 3 of 5. Node 1584 of 7920
Realization 3 of 5. Node 2376 of 7920
Realization 3 of 5. Node 3168 of 7920
Realization 3 of 5. Node 3960 of 7920
Realization 3 of 5. Node 4752 of 7920
Realization 3 of 5. Node 5544 of 7920
Realization 3 of 5. Node 6336 of 7920
Realization 3 of 5. Node 7128 of 7920
Realization 3 of 5. Node 7920 of 7920

Back transforming for realization 3

Realization 3: number = 5672
mean = -0.0348 (close to 0.0?)
variance = 1.0435 (close to gammabar(V,V)? approx. 1.0)

Working on realization number 4
etc

Error and Warning Messages

Message	Description	Solution

ERROR: nn characters in Project path name. Maximum = 48	The number of characters in the path name of the project folder exceeds 48 characters.	Reduce the path name to 48 or less characters in length.

The command SGSIM is based on the GSLIB program SGSIM.

Copyright (C) 1996,  The Board of Trustees of the Leland Stanford
Junior University.  All rights reserved.

The programs in GSLIB are distributed in the hope that they will be
useful, but WITHOUT ANY WARRANTY.  No author or distributor accepts
responsibility to anyone for the consequences of using them or for
whether they serve any particular purpose or work at all, unless he
says so in writing.  Everyone is granted permission to copy, modify
and redistribute the programs in GSLIB, but only under the condition
that this notice and the above copyright notice remain intact.