Feature List

Isatis.neo Mining Edition is a tailored version of Isatis.neo, Geovariances new generation

software solution in geostatistics. Designed for the Mining Industry, the software has all the
features you need for thorough data analysis and quality control, domaining, variographic
analysis, in-situ and recoverable resource estimation, and uncertainty analysis through
simulations.

The software also offers a preconfigured workflow for an optimized way to Mineral
Resource Estimation and simulation, including Ore Control and Reconciliation (formerly
Minestis).

Its innovative and intuitive user interface makes the software quick to learn and easy to
use and puts the full power of geostatistics within your reach.

Version 2020.06
Experience Isatis.neo Mining Edition ease-of-use according to rules you define. Extract unmasked
through a global interface from which all applications are samples into a new data file.
launched. The software ribbon menu makes it easy to find ─ Clear data from possible duplicates. Only one
all available tools. duplicate will be kept. The samples that are kept will
be extracted to a new file or identified through a
1. DATA HANDLING mask.
─ Make use of an advanced scientific calculator
1.1. Data exchange
based on Python syntax to modify or create new
─ Fully integrate and handle both 2D/3D samples variables. Numerous operations are available
and 3D drill hole data, grid, and mesh files. (mathematical, boolean and interval operators,
trigonometric, date and statistical functions, etc.).
─ Differentiate variables according to their types for
proper use: coordinates, Hole ID, length, numerical, ─ Create sub-block models to be used as the
categorical, text, azimuth, dip, depth from, depth to, support of the estimation. Sub-blocks can also be
density, campaign. filled with block values (estimates or simulation
realizations).
─ Define unit classes to make conversion and
calculation between variables easier.

─ Easily import new point, grid, drill hole data

(possibly with deviation survey) through CSV/Excel
files. Load data from Isatis, acQuire, Datamine, or
Vulcan. Variables are loaded with their units and
names. The detection limit of devices is also saved.
─ Load raster geospatial data from GDAL-supported
files (including Z-Map, Arcview, GeoTiff, netCDF
formats).
─ Flatten folded geological structures for a more
─ Import wireframes or surfaces from CAD mesh
robust variography. Either unfolding is made by a 2
files with elevation variable if available (DXF, DWG,
vertical translation of grid nodes (Projection mode)
STL, OBJ, OFF) and Vulcan.
or by a translation normal to a reference level (Bend
─ Import polylines or polygons from Datamine and mode). The reference level can be a surface, or the
GOCAD files, or from vector graphics data files top, bottom or medium level of a block model.
such as Shapefiles or Autocad DXF/DWG files,
─ Compute the minimum distance between data
using a GDAL/OGR library-based utility.
(grid nodes, sample points, borehole samples), and
─ Export point, grid, and drill hole data to a mesh and store results in a variable.
CSV/Excel or Datamine files or to the Isatis
database. Export 3D grids to Vulcan block-models. 1.3. Data analysis
─ Export mesh files (3D solids, wireframes or ─ Quickly compute histograms, standard statistics
surfaces) to CSV or Excel files, Isatis Wireframes (global or by category), quantiles, correlations,
or CAD files (DXF, DWG, STL, OBJ, OFF). and contingency between variables (possibly
─ Export polygons and polylines to CSV/Excel or weighted).
Datamine files, or to vector graphics data files such ─ Compute cross-file statistics and copy data
as Shapefiles or Autocad DXF/DWG files using a between files.
GDAL/OGR library-based utility.
─ Analyze borehole sample lengths and achieve
1.2. Data manipulat ion length regularization (compositing) to improve
estimation quality and avoid bias (support effect
─ Easily check and manage your data. Rename or considered). Compositing can be done by bench or
delete variables, compute statistics, and edit length, considering mesh boundaries (domain or
properties. Combine data. wireframe).
─ Quickly edit and modify values through a user-
friendly spreadsheet.
─ Build masks on samples to ignore some samples
in the further geostatistical analysis. Filter them out

www.geovariances.com  [email protected] +33 1 6074 9090

─ Achieve relevant data declustering based on
statistics to correct possible spatial sampling bias
(due to data clusters).

─ Model the envelope of each class identified

through a previous sample clustering:
• Set the external constraints that will be used to
─ Cap the higher or lower grades in heavy-tailed refine the envelopes: contact points between
distributions to ignore outliers in the estimation domains, control points inside/outside a
process. Test a range of bottom or top-cut domain, surfaces, gradients, and external drifts.
values to identify the most appropriate one for your
• Get the probability to be inside a domain.
orebody. Support your decision-making with a set
of statistics computed before and after capping for • Compute the volume distribution curve and
each test value. domain uncertainty.

─ Prepare your data to block kriging. Compute a ─ Analyse grade variations from either side of
support correction considering or not the domain boundaries using the Contact Analysis
Information Effect. tool. It helps you qualify how domains can be
treated at estimation (i.e., hard/soft boundary,
─ Compute local anisotropy directions (gradient 3
transition domain definition, etc.).
and rotation) relative to a reference surface or two
surfaces at any node of a grid file. They will be used
in kriging with the option Use Local Anisotropies.
2. EXPLORATORY DAT A
AN ALYSIS
─ Remove imperfections in images by running
morphological operations on grids: Isatis.neo Mining Edition offers an integrated and
• Dilation to make objects more visible, fill in powerful application for Exploratory Data Analysis.
small holes or join objects. The whole statistical and geostatistical analysis is
• Erosion to remove small objects or noise or performed from that single tool. Easy to use, you need
detach larger objects. to select a variable in the variable
• Closing to remove small objects while list and drag and drop it into the display window to
preserving larger objects. have it represented in as many statistical charts as you
• Opening to remove small objects from an wish.
image while preserving larger objects.
─ Run connectivity analysis on grids and identify
connected components. The application can be
used to locate permeable pathways inside a
lithostratigraphic unit.

1.4. Domaining
─ Perform sample clustering using a powerful
machine-learning algorithm. Samples are
automatically grouped into homogeneous The strength of this tool is also in its dedicated object
geological or mining domains according to their called the Geostatistical Set which gathers all the
spatial dependency and their degree of similarity inputs that will be used later in the kriging or simulation
(considering various variables and statistics step: variable statistics, experimental variogram,
criteria).

www.geovariances.com  [email protected] +33 1 6074 9090

variogram model, stationarity option and anamorphosis 2.2. Variography
function if needed. This guarantees consistency and
─ Quickly identify and handle anisotropies from the
makes the parameter setting more straightforward
interactive variogram map. The defined
when defining the estimation parameters.
anisotropy is automatically sent to the experimental
variogram parameters.
2.1. Statist ics
─ Get instantaneous univariate or multivariate
statistics (possibly weighted) on the considered
variables:
• Basemap: to locate the sample points.
• Histograms with standard statistics: minimum,
maximum, mean, and standard deviation
values. Assess data global distribution.
• Cross-plot / Box-plot / Swath-plot with
auxiliary variables: for multivariate data ─ Refine experimental variogram lags at small
analysis taking into account the correlation distances to better capture variable spatial
between variables to guarantee estimation behavior.
consistency. Also used to identify different data
─ Choose the variogram curve that you wish to
populations.
adjust: classical variogram, pairwise relative
• Variogram map, univariate or multivariate variogram, covariance, the ratio cross variogram
experimental variogram, variogram cloud, over simple variogram, madogram, or correlogram.
downhole variogram and variogram model.
Characterize the spatial variability of the
variables and highlight outlier samples.
• Theoretical grade-tonnage curves.

─ Define the stationarity option, intrinsic or strict,

that will be applied when kriging (for ordinary or
simple kriging).
─ Handle the spatial variability of your multivariate
model to ensure better estimation coherence with
any other correlated variables.

─ Refine all calculated statistics by interactively ─ Experience automatic variogram fitting and
tuning parameters. instantaneously get a model well-suited to your
input data: automatic modeling of multivariate,
─ Benefit from the dynamic link between the different
multidirectional and multi-structure variograms
representations and thoroughly explore your data:
starting from a combination of basic structures.
when you highlight points in the base map or
histogram bars, they are immediately highlighted in ─ Interactively fit your variogram models and tune
the other representations so that you can check the involved parameters (structure type –
their locations. You may also ignore some samples exponential, spherical, cubic, power, Gaussian,
(erroneous or outlier samples, for instance) and cardinal sine, Stable, Gamma, Bessel-J, Bessel-K
instantaneously check the impact on the variogram – and range and sill for each structure). You may
and further maps. add several structures if the variables show
different behaviors at different scales. Choose
whether to add or not a nugget effect.

www.geovariances.com  [email protected] +33 1 6074 9090

 Or fit the variogram visually using the mouse and • Use Variance of Measurement Error. To filter
the anchors displayed on the variograms. out random errors.
─ Fit variograms on raw data and/or Gaussian • Cokriging or Collocated Cokriging. In
transformed data at once by applying an multivariate cases (with one or several
anamorphosis function. This can improve structure collocated variables).
identification while reducing data variability. • Filtering Model Components. To filter out
Variography can also be performed on residuals. some components of the variogram models, i.e.
Either select a known drift to be subtracted from the nugget effect to remove the noisy component.
raw data or calculate residuals from a multilinear ─ Set a unique or a moving neighborhood. Many
regression of all chosen drift variables. advanced parameters are available in addition to
─ Validate and select the variogram model and the the classical rotation and size of the search ellipsoid
neighborhood that provide the best estimates using and the number of neighbors, to fit the
cross-validation. neighborhood to your data at best and get more
robust results.
─ Quickly compute experimental variograms from
gridded data. Identify and filter-out drift
components from models.

3. ESTIMATION

3.1. Interpolation
─ Quickly interpolate data using various techniques:
inverse distance, nearest neighbor, moving
average, moving median, linear kriging, spline
kriging, linear interpolation for numerical and
rotation variables.
─ Fill blank areas in grids while reproducing the 5
─ Run Kriging Neighborhood Analysis to support
property variability using kriging or simulation.
your neighborhood parameter setting.

3.2. Kriging ─ If needed, define a nested neighborhood to run

multiple passes of the estimates. You may also cap
To facilitate the parameter setting, the different kriging
outliers that might skew the estimations.
options and parameters are grouped into a single
dialog box. ─ Verify the relevance of the neighborhood
parameters and the quality of the estimate by
─ Run point or block kriging, simple or ordinary
displaying various statistics for any target cell: Sum
kriging, universal or external drift kriging, on raw
of Weights, Sum of Positive Weights, Weight of the
or Gaussian-transformed variables, on indicators or
Mean, Variance of Z*, Covariance Between Z and
residuals.
Z*, Correlation Between Z and Z*, Slope of the
Several specific options are available:
Regression Z | Z*, Kriging Efficiency.
• Use Sampling Density Variance. Check this
─ Save kriging results in variables that will be used
option to calculate a volume-like quantity
to control estimate quality: Standard Deviation,
representative of the spatial sampling density.
Variance, Number of Neighbors, Mean Distance to
This volume can then be converted into a risk
Neighbors, Sum of Weights, Sum of Positive
index for mineral resource classification. The
Weights, Lagrange Parameter.
technique requires the definition of a production
volume corresponding to a specific production ─ Display kriging weights in the 3D Viewer.
period and a mean grade. ─ If your dataset is too big, you may parcel it to
• Use Customized Block Size. Kriging is run on perform calculations on subsets of samples. The
irregular grids or sub-block models. Option output grid can be split into as many sub-grids to
available for block kriging only. improve computation performances.
• Use Local Anisotropies. Kriging uses local
variograms and local neighborhoods.

www.geovariances.com  [email protected] +33 1 6074 9090

 3.3. Recoverable resources ─ Select a representative subset of simulations to
characterize the risk attached to a project with the
─ Assess recoverable mineral resources using
Simulation Reduction tool. It allows reducing the
Uniform Conditioning (UC) or Multiple Indicator
initial number of realizations into a few more
Kriging (MIK).
manageable items while preserving the statistics of
─ Localize estimates computed on blocks to SMU. the original simulation set.If your dataset is too big,
─ Derive Grade Tonnage Curves of mineral you may parcel it to perform calculations on
resources from kriging, Uniform Conditioning (UC), subsets of samples. The output grid can be split in
Multiple Indicator Kriging (MIK) or simulation as many sub-grids to improve computation
outputs, or from an anamorphosis function. performances.
─ Produce detailed and geologically sound facies
4. SIMULATIONS models of complex orebody geology using the
Plurigaussian simulations.
To facilitate the parameter setting, the different
simulation options and parameters are grouped into a
5. DISPLAY AND REPORTING
single dialog box.
─ Run conditional or non-conditional, univariate or 5.1. 2D Maps
multivariate simulations (Turning Bands) on points
─ Benefit from usual geographical tools. The display
or blocks, using stationary, intrinsic or non-
is organized in superimposed layers with all the
stationary models, with possible specific options:
attached functionalities: display/mask, add/delete,
• Collocated Cosimulation in multivariate cases edit, move forward/backward, zoom in/out, scale,
(one or several collocated variables). compass grid, etc.) so that you customize the view
• Filtering Model Components. To filter out to your own needs.
some components of the variogram models, i.e., ─ Measure distances between points and angles
nugget effect to remove the noisy component. between consecutive segments of a polygon.
─ Benefit from the Direct Block Simulations, a
unique and powerful algorithm enabling to bypass 6
point scale models to build block scale models
directly. It allows saving considerable time and disk
space.
─ Use a unique or moving neighborhood.

─ Save simulation statistics in variables that will be

used to perform uncertainty analysis: Mean,
Standard Deviation, Variance, Minimum, Maximum,
Quantiles, Confidence Intervals, Mean Above
Threshold, Probability Above Threshold,
Accumulation Above Threshold.

─ Validate simulation outputs by checking ─ Create Polygons to delineate specific areas

consistency between the experimental variograms (facility outlines, yards, etc.) or for delimiting the
calculated on a subset of simulations and the one estimation area. Highlight/mask/edit samples or
from the original data. polygons.
─ Enrich the scene with vector or raster images:
besides the standard image file formats (bmp, gif,
jpg, png, tif), Isatis.neo may load many file-formats
including CAD DXF, GeoTIFF, MapInfo or ESRI
Shapefile.
─ Geolocalize imported images in the project.

www.geovariances.com  [email protected] +33 1 6074 9090

─ Display in a table a chosen set of information 5.4. Report ing
related to selected objects from the 2D map.
Enjoy the efficiency of an integrated word
processor.
Isatis.neo Mining Edition offers all the essential tools
to create your own report as the project is built. You
can send 2D or 3D views, graphics and message
content to the word processor, edit and format texts,
store the report into pdf, odt or html documents, etc.

6. TASK AUTOMATION
Easily record your workflow in a batch file to:
─ keep track of the full process,

─ store your parameter settings to make it possible

─ Drape 2D images onto a surface in the 3D Viewer.
to check them later and modify them if needed,
─ Take advantage of the dynamic link between the
─ rerun the whole project interactively or in an
display window and the database allowing an
automatic way from Isatis.neo or in a command-line
automatic update of the view with new data.
mode outside the software.
─ Connect a GPS and precisely locate any point
on a site: the location is then automatically marked 7. STUDY MAN AGEMENT
out by a symbol on the map.
The connection is made between the software and ─ Switch your project list from icon to list view and
the GPS (with NMEA output) through a specific display general information regarding a project
interface. (e.g., location on disk, size, version).

─ Save 2D views as PNG or SVG images. ─ Attach/detach a project from your list of projects.
Backup/restore a project for future use. 7
5.2. 3DViewer ─ Access the most recent projects in a mouse click.
─ In a single view, gather point data, drillhole data,
surfaces, polygons, faults, wireframes, grids.
─ Customize the display according to your visual
style guidelines. Symbol shape, color, size, color
scales, isolines can be set.
─ Analyse your 3D objects and estimation models
using slicing, clipping, zooming, filtering tools and
database query. Display related information
(position and value).
─ Save 3D views as PNG images.

5.3. Log display

Display
drillhole logs
picking its
symbol on the
map or
selecting them
it in the 3D
view.

www.geovariances.com  [email protected] +33 1 6074 9090

 RESOURCES WORKFLOW

Resources wokflow provides the user with the industry-standard in geostatistics for in-situ and recoverable resource
estimation, ore control and reconciliation, in a preconfigured and easy-to-use workflow. It ensures coherent management
of geological domains for reliable resource estimates and a perfect consistency between long-term and ore control models.

Resources is an integrated and optional module of Isatis.neo Mining Edition.

1. DATA V ALIDATION area from zones and a categorical variable. Each

sample will be assigned to only one domain.
1.1. Grade data validat ion
─ Quickly get the detail of the partition statistics:
─ Define the field limits, value detection limit, and the number of ignored/considered samples per domain,
value for non-sampled core. domain interval characteristics per sample
(continuous or divided interval).
─ Clear data from possible duplicates. All
duplicates can either be removed, or one duplicate ─ Convert each borehole to a 2D sample. The
can be kept. Values below detection limit are also conversion from assay grades to 2D variables takes
automatically processed (assigning them either 0, the form of an accumulation.
the detection limit or half this limit). Sample overlaps
or duplicates are automatically detected. The
process creates a new “Validated database”.

1.2. 3D domain validation and

partit ioning
─ Build coherent 3D partitions of your estimation
area from simple rules (inside/outside a wireframe,
above/below a surface, filter on grid variables). 8
Each sample will be assigned to only one domain.

2. LOCAL/GLOBAL RESOURCE
ESTIMATION
─ Easily perform multivariate or univariate
estimation and choose as many secondary
variables as you wish.

─ Quickly test and compare various estimation

─ Detect possible gaps using morphological scenarios involving SMU size, declustering
operations window size, compositing length, choice of capping
─ Make sure that domains do not overlap. Manage or not, and more. Find out which scenario gives the
areas outside domains by creating a best result.
complementary domain. ─ Consider grades and densities or accumulations
─ Quickly get the volumes of each domain. and densities in the estimation process according
to your deposit.
1.3. 2D domain validation and
partit ioning 2.1. Data analysis
The Resources workflow enables the estimation of ─ Analyze drillhole sample lengths and achieve
deposits with a layer-cake geometry. Assay grades length regularization inside each domain for
are converted to 2D accumulation variables. coherent domain estimation.

─ Build coherent 2D partitions of your estimation

www.geovariances.com  [email protected] +33 1 6074 9090

─ Achieve relevant data declustering based on different graphics and thoroughly explore your data.
statistics. When you highlight points in the 3D viewer or
histogram bars, they are immediately highlighted in
the other graphics so that you can check where they
are. You may also apply a filter or mask samples on
the fly and instantaneously check the impact on the
variogram and further maps.

2.2. Variography
─ Experience automatic variogram fitting and
instantaneously get a model well-suited to your raw
grades and/or Gaussian transformed grades.
─ Interactively fit your variogram model (simple
and cross-variograms in multivariate cases) and
─ Manage outliers that might bias the estimations by adjust the involved parameters (structure type -
capping grade values above or below a given spherical, exponential - range and sill for each
threshold. structure). You may add several structures if the
─ Instantaneously perform structural analysis on grade shows different behaviors at different scales.
both the raw grades and the Gaussian Choose to consider or not a nugget effect.
transformed grades. Results are displayed in Or fit the variogram visually using the mouse and
several views: the anchors displayed on the variograms.

─ Interactively identify and handle anisotropies

from horizontal and vertical variograms through
variogram maps.

• 3D Viewer. Visualize and locate the boreholes

inside the study area.
• Histogram with basic statistics: minimum,
maximum, mean, and standard deviation ─ Apply a support correction to your data
values. Assess the data’s global distribution and considering or not the Information Effect.
check the data transformation quality if
performed (see below). 2.3. Resource est imation
• Experimental variogram (simple and cross- ─ Ordinary or Simple Kriging, Uniform
variograms), variogram cloud, and Conditioning, Local Uniform Conditioning, and
variogram model. Characterize the spatial Simulations are implemented for block estimation
variability of the mineralization and highlight and recoverable resource estimation. Deterministic
outlier samples. methods such as Inverse Distance and Nearest
• Swath Plots to check variable stationarity. Neighbor are also available.

• Cross-plot to analyse the linear correlation ─ A moving neighborhood is used. You may tune
between two variables and identify different the number of samples as well as the size of the
data populations if any. ellipsoid to get more robust results.

─ Refine all calculated statistics/functions by ─ If needed, a nested neighborhood can also be

tuning the involved parameters interactively. used to improve the neighborhood search.

─ Benefit from the dynamic link between these ─ Kriging Neighborhood Analysis helps you

www.geovariances.com  [email protected] +33 1 6074 9090

 choose relevant neighborhood parameters for your variables and get your Recoverable Resources for
data. each class.
─ Verify the relevance of the defined kriging ─ Get quantile maps.
parameters by displaying, for any target cell, the
─ Validate your simulation process by comparing
kriging weights in a 3D view.
the input Gaussian model to the output results.

─ Define panel and SMU sizes and run kriging,

Uniform Conditioning, Localized Uniform
Conditioning and/or Turning Bands simulations
on both supports in one hit.
─ Save all kriging results (e.g., Correlation Z/Z*,
Covariance Z/Z*, Kriging Efficiency, Slope of
─ Benefit from a flexible reporting format. Display
Regression Z/Z*, Weight of the Mean, Number
the outputs of your choice in a single table:
Neighborhood, Mean Distance to Neighbors) as
recoverable tonnage, metal, grade and/or
new variables that can be used to validate your
benefit, computed for one, several or all metal
estimation and classify your resources.
grades, on panels or SMUs, by domain or over the
─ Validate your estimation by comparing estimates whole study area, for each category of resource.
(from inverse distances or nearest neighbor and
kriging) versus data information through swath
plots, statistics, histograms, and 2D/3D 10
visualizations.

─ Easily display Grade Tonnage Curves of results

independently or superimposed.

3. RECOVERABLE RESOURCES
─ Get recoverable resources (both at global and
local scales) for the current partition and/or for
each domain in the partition from kriging, Uniform
Conditioning, Localized Uniform Conditioning,
and/or simulation outputs. At either panel or SMU
support, results can be computed for a list of
cutoffs applied on the main grade or using
categories defined from multivariate rules.

─ Define your categories for resource

classification based on all outputs and created

www.geovariances.com  [email protected] +33 1 6074 9090

4. ORE CONTROL &
RECONCILI ATION
─ Load and validate data in a separate blasthole
database.

─ Always be informed of the next step to come thanks

to the tooltip displayed on each variable.

─ Build an ore-control model from blast hole data

using variogram model and neighborhood
parameters defined for the long-term estimate. This
ensures perfect consistency between the two
models.

─ Set simple rules to define ore categories.

─ Draw mining polygons on each bench and for

each resource category.
─ Export mining polygons to your planning
11
software.
─ Compare long-term and ore control models and
perform reconciliation inside a selection of SMUs.

─ Quantify how the tonnage, the metal quantity

and the mean grade derived from the ore control
model have changed from what was estimated in
the long-term model.

5. PROJECT MANAG EMENT

─ Benefit from a fully automated, auditable and
traceable estimation workflow. Update daily,
weekly or monthly estimates with just one
mouse click as soon as new data are available,
while being able to check and validate what is being
done. If needed, quickly modify the estimation
parameters to adapt them to your data.

www.geovariances.com  [email protected] +33 1 6074 9090