Subheadings for this chapter are as follows:
This section describes the conventions used throughout MicroMODEL. MicroMODEL uses standard modeling conventions and terminology as often as possible. Volume I, Section 7, of the User Guide contains a glossary of MicroMODEL terminology. In addition to this information, each program contains on-line help to further assist the user and avoid repeated reference to these documents.
The Project Information File must be created before any drill hole data analysis or modeling can proceed. This file contains the basic information MicroMODEL will need to perform all tasks within the system. If the file is not created, no programs in the system can be accessed.
The model, or local grid, is oriented relative to the global coordinate system as shown in Figure 3.1.
FIGURE 3.1 ORIGIN LOCATION = North:250 East:350 Elevation:9500 Rotation Angle:320 Degrees (Clockwise from North left edge)
The model's size is determined by the number of rows, columns, and levels, and dimensions of each as demonstrated in Figure 3.2. The lower edge and left edge of the model are always relative to the location of the rows and columns as shown in Figure 3.1.
FIGURE 3.2 Width of Model= (Number of Rows) x (Row Dimension) 4 Rows x 30M = 120M Length of Model= (Number of Columns) x (Column Dimension) 3 Columns x 50M = 150M Height of Model= (Number of Levels) x (Level Dimension) 5 Levels x 20M = 100M
A label is any quality or variable that is to be entered or manipulated as drill hole data, or modeled or manipulated as a grade model. Examples of labels are quality or assay data such as Gold, % Fe, oz/ton, milliounce/ton, etc.
MicroMODEL will work with any units the user requires with the exception of length. To achieve proper map scales and volumetrics, the system must know if the modeling will be performed using English (feet) or Metric (meters) units.
Two types of drill hole databases are allowed simultaneously within MicroMODEL, sampled drill holes and composited drill holes.
Sampled drill holes are entered into MicroMODEL as the data is gathered in the field. Generally, sampled drill hole data will contain multiple sample intervals for each bench intercept.
A composited drill hole, as made by the compositing program using the bench compositing option, contains one drill hole interval for each corresponding bench intercept. Compositing prorates the sample intervals by length, such that each bench intercept is represented by only one interval. See Volume II, Compositing, Calculation of Composite Values and Figure 4.1 for a comparison of the same drill hole as it exists in sample format versus bench, drill hole and rock unit composited drill hole databases. If the input drill hole data is already in the composited form, it should still be entered in the Data Entry Module.
All drill holes, both sampled and composited, are specified by collar Northing, Easting, elevation, bearing, plunge, and optional downhole survey records. This information is always entered relative to the global coordinate system (Northings and Eastings) established in the Project Information File.
The collar bearing and survey bearings are specified as degree azimuths clockwise from North to the plan view projection of the drill hole segment. The bearing is sometimes referred to as the dip direction. The plunge is the degree measurement from horizontal to the drill hole segment. Downward plunging drill holes have positive plunge. Therefore, a vertically downward plunging drill hole has a plunge of +90.0.
Downhole survey information can be entered with the drill hole to record the "wandering" of the drill hole as it was drilled. MicroMODEL specifies downhole directional surveys with the distance downhole to the survey, the bearing and the plunge at the survey point. Up to 1000 downhole surveys are allowed per hole (drill hole collar and 999 downhole survey cards). In MicroMODEL, the drill hole is straight until it encounters a downhole survey. At the point of the survey, the hole "bends" to the bearing and plunge at the survey and continues straight until the next survey or end of hole.
Drill hole intervals are specified by the downhole distance to the start (FROM) and finish (TO) of the interval, the rock code, and assay data that apply to the interval. The rock code represents the material contained in the interval and must be an integer from 1 to 9999.
MicroMODEL flags unsampled drill hole intervals with a "-999.99" as label values (assays). The unsampled label values are not included in any statistical or modeling computations. MicroMODEL printouts and plots register blanks for unsampled labels as opposed to "-999.99". To read in an existing database, the ASCII file must use the same label flag for unsampled or unspecified assay data. The only valid way to enter unsampled values interactively is to use the "U" value of the capturing and editing programs (a -999.99 value may be represented internally as -999.98888 and later interpreted by the programs as a valid sampled value.)
Drill holes and drill hole segments are allowed outside of the model area. These drill holes will not appear on most forms of plan view plots, but can be used as valid input data during the statistical analysis and modeling stages.
In MicroMODEL, the topography is modeled as a two-dimensional grid of elevations. Each cell in the topography model is assigned an elevation which applies to the entire cell area. The size of the topography grid is determined by the number of rows and columns in the Project Information File. The topography grid is later used to define the boundary between rock and air.
After presorting, and before MicroMODEL actually assigns values to the topography grid cells, the entire grid is initialized at background cell values of "-999.99". Cells that cannot be estimated according to the given presorting and modeling parameters remain at the unestimated background code, "-999.99". Every attempt should be made by the user to ensure all cells in the topography model are estimated. The cone mining algorithm in the open pit module will not work if there are any unestimated cells in the 2-D topo grid.
Prior to grade modeling, a rock model is constructed which aids in the grade modeling process. In MicroMODEL, any blocks which fall completely above the topography are assigned an integer rock code of 0 (zero) in the rock model. Any blocks which fall below the topography or are cut by the topography are assigned a rock code from 1 to 9999 as defined by the user.
The entire block is considered to be homogeneously filled by the type of rock assigned to the block. MicroMODEL does not perform block partials except during pit design volumetric calculations. Therefore, a block which contains only a small fraction of rock at the topography interface will be displayed throughout the system as a full block. However, as stated earlier, volumetrics will be calculated using block partials where necessary.
The variogram (or semivariogram) programs will always access the data of the current label. If the programs are invoked from the Surface Modeling Module, prepared topography data will be analyzed. Likewise, invoking the Variogram Menu from the Grade Modeling Module enables the user to geostatistically analyze sampled or composited drill hole data for the current label.
The variogram azimuth angle from North is the direction along which data is paired. The tolerance of the variogram is the deviation on either side of the bearing in which data will be paired. A non-directional (global) variogram can be obtained with a tolerance of 90 degrees. MicroMODEL uses a maximum of 25 lag intervals, so the maximum extent of the variogram is (25 x Lag Distance).
In MicroMODEL, a spherical variogram model is specified by its nugget, sill, and range. The nugget is the discontinuity jump at the origin. The model sill (also called the No nugget sill) is related to the total sill and nugget constant by:
Total sill = Nugget + No nugget sill
The range is the distance at which the data shows no correlation. See Figure 3.3 for an illustration of a typical spherical variogram model.
FIGURE 3.3 SPHERICAL VARIOGRAM
For h=distance a=range C=sill Gamma = C * ( 1.5 * (h/a) - 0.5 * (h/a) ** 2 ) ( h< a ) Gamma = C ( h >= a )
For h=distance a=range C=sill Gamma = C * (h/a) ( h < a ) Gamma = C ( h >= a )
For h=distance a=range C=sill Gamma = C * ( 1.0 - e ** (-h/a) )
For h=distance a=range C=sill Gamma = C * ( 1.0 - e ** ( (-h/a) * (h/a) ) )
The point validation presort and modeling programs enable the user to appraise different presorting and/or grade modeling parameters or variogram models. The point validation programs always access data for the current label.
The MicroMODEL grade model(s) are stored as three-dimensional matrices called grids. All three-dimensional grids in a database will have the same dimensions as defined by the number of rows, columns, and levels in the Project Information File. Each grid block contains modeled values relative to the grid as specified by the label (e.g. gold, % Fe). All material contained within a block is considered to be of the grade value assigned to it (i.e. no block partials are performed during modeling).
After presorting, and before MicroMODEL actually assigns values to the grid blocks, the entire grid is initialized at a background value of -999.99. Blocks that cannot be estimated according to the given presorting and modeling parameters remain at the unestimated background code, -999.99.
Ellipses can be used during surface (two-dimensional) modeling. Ellipsoids can also be used during rock and grade (three-dimensional) modeling. When discussing concepts which apply to both ellipses and ellipsoids, this text will use the term "ellipsoid" since an ellipse can be thought of as a two-dimensional ellipsoid.
MicroMODEL utilizes two types of ellipsoids; search ellipsoids and weighting ellipsoids.
A search ellipsoid is used during presorting of data which will subsequently be used for modeling. Points that are located outside of the search ellipsoid are not included in the presort for the particular block, and are ignored during modeling.
A weighting ellipsoid (not necessarily the same as the search ellipsoid) is used during modeling to weight different points according to their relative position in the weighting ellipsoid. Points that are located upon the same elliptical "shell" will be considered as having the same distances to the block center in the modeling calculations. As a consequence, points that occupy smaller elliptical shells usually receive more weight during modeling than points that occupy larger shells.
The shape of either a search or weighting ellipsoid is determined by its relative axis lengths. Generally, the longest axis is the reference axis, called the "primary" axis. The secondary and tertiary axes are perpendicular to the primary axis and to each other. See Figure 3.4 for a typical ellipsoid specification.
FIGURE 3.4 THREE DIMENSIONAL ELLIPSOID USING TILT
The orientations of both search and weighting ellipsoids are specified by their axis rotation angles. The primary rotation angle is the azimuth from North to the plan view projection of the Primary axis regardless of the axis lengths. The secondary rotation is the dip or plunge of the primary axis (+90 is vertical down). Tertiary rotation can be applied in one of two ways. The original convention is that the third angle is applied as a TILT. This is the amount of rotation applied to the ellipsoid while viewing the primary axis as a point. The TILT angle is also consistent with the way MEDSystem (tm) defines rotations. However, the third rotation can also be defined in terms of a RAKE angle. The rake angle is a rotation around the axis that is perpendicular to the plane defined by the first and second rotations. Generally, an ellipsoid can be defined without using the tertiary rotation.
For the RAKE angle rotation, a negative angle is used for the counter-clockwise direction. A posite angle is used for the clockwise direction.
Finally, the size of a search ellipsoid is defined by its search radius or search range. The search radius defines the radius of the primary axis. The secondary and tertiary axis radii of a search ellipsoid are then calculated by default according to the shape of the ellipsoid.
As opposed to a search ellipsoid, a weighting ellipsoid does not have an absolute size since it is only concerned with the relative distances of points that have already been presorted.
During the Open Pit Design Program (OPD), a bench is evaluated relative to a user specified or computer-generated pit polygon. All blocks whose centroids fall within the pit polygon will be completely extracted as shown in Figure 3.5. Partial block volumes are calculated where blocks are cut by the topography.
FIGURE 3.5 BLOCK EXTRACTION WITH PIT POLYGONS
Blocks that are contained within a pit polygon are "mined" by reversing the sign of the rock code within each block. This method prevents a "mined" block from being available for subsequent pit increments. Therefore, if a block contains a positive rock code, it is available to be removed by OPD. If a block contains a negative rock code, OPD will not tabulate the material in the block since it has been previously "mined".
Many MicroMODEL programs refer to "cone surfaces." In general, any 2-D topography grid is considered a "cone surface," even if it wasn't created by the floating cone algorithm.
For example, the program that calculates waste dump volumes will create an updated "cone surface" that reflects a change from the prior topo grid due to the dumping of waste. While this new surface is not technically a cone surface, it is still addressed as such throughout the suite of MicroMODEL programs.
It is also possible to store any mined out OPD pit configuration as a "cone surface." While this is not technically a surface created by the floating cone algorithm, it is still addressed as a "cone surface" by the MicroMODEL programs.
This nomenclature may be confusing to the new user. The thing to remember is that whenever the term "cone surface" is used by a MicroMODEL program, that program is really referring to any 2-D topo grid.