Validation is integral part of the geological modelling process to ensure a robust model.
With Leapfrog Geo and Central you can create fully integrated workflows for your team. Add data or revise your interpretations at any stage, and these changes flow downstream, with no additional configuration (scripting/programming) required, for rapid updating, validation and efficient decision making.
Validation is integral part of the geological modelling process to ensure a robust model. With Leapfrog Geo and Central you can create fully integrated workflows for your team. Add data or revise your interpretations at any stage, and these changes flow downstream, with no additional configuration (scripting/programming) required, for rapid updating, validation and efficient decision making.
Join Seequent’s Jillian James (Project Geologist) to discuss:
- Validating your drillhole database
- Visual validation issues
- Using Leapfrog’s statistical tools to validate your model
- Using Query Filters to test areas of your model
- Upload to Central for Peer review and final validation
Project Geologist- Seequent
<v Jillian>Hi, everyone.</v>
Thank you for taking the time to join me today
for this Tip Tuesday on validating your Leapfrog Geo model.
My name is Jillian James,
and I’m a project geologist here in Seequent’s Perth Office.
Seequent has a range of software
for use throughout the exploration and mining value chain.
Today, the focus will be on the functionality
in Leapfrog Geo with a brief introduction
to Seequent Central.
The key aims of this session
are to give you some tips and tricks
on how to validate your Leapfrog Geo model
and make validation an integral part of the modeling process
rather than just an afterthought.
In this session, we’ll cover:
how to check the data feeding into your model is validated,
tips on section by section validation,
how you can use Leapfrog Geo statistical tools
to validate your model,
how you can use query filters to test areas of your model,
for example, if you want to test where the model was barren,
but the grades are above a certain value,
where the model was mineralized,
but the grades are below a certain value,
and where the geology log does not match the model geology.
Finally, we will look at how to upload your project
to Central for peer review and final validation.
The most important step in validating any geological model
is to check that the data
feeding into the model is validated.
Remember, rubbish in equals rubbish out.
Things you should check include looking to see
that all the data is included in your project,
including any unmineralized holes,
understanding the impact that missing
and null values may have on your model,
and how the replacement rules are being implemented.
Are there any extreme outliers, highs or lows,
in your dataset?
Does the data look to be in the correct spatial location
relative to things such as development and topography?
And are there any duplicate points or overlapping intervals?
Leapfrog Geo automatically identifies and flags
common data errors when data tables are imported.
Until errors are corrected,
the rows that contain those errors are excluded.
For example, if the ID for a color
contains an invalid coordinate,
that color will not be displayed in the scene
as its location is not known.
This will also affect any downstream processing
and it will ignore any data associated with this drillhole.
We will now move into the software
to look at this in more detail.
This session will not cover building domains from scratch.
So throughout the demonstration,
I am going to make use of a project already created
using the saved scenes to show the functionality.
I’m going to use a dataset from a real underground goldmine
that has since been mined out.
Data locations and names have been changed
to keep the project anonymous.
Although I am demonstrating this functionality
with respect to our underground narrow vein gold system,
these validation tools can be useful
in almost any deposit style or geological setting.
If you need help with your individual project,
please don’t hesitate to contact
our friendly support team to assist you.
I will begin with validating the drillhole database.
Over in my project tree,
you can see I have a diamond drillhole database here
and I’ve imported all these various table types
to build my database.
The red exclamation marks here on my assay, geology,
and vein codes indicate that there is possibly an error.
The orange exclamation mark on my survey
indicates there is a warning.
The blue downward pointing arrow
indicates that the data has been imported correctly
and there is no errors to fix.
To fix an error, I right-click on one of these tables
that has an error and select fix errors.
This will open up a new tab
and it will show me where these errors are occurring.
This assay table actually has no errors,
but it has some invalid values handling.
If we look at this AU_PPM one,
we can see that some rules have been applied
to the non-positive values.
So by default, it is going to replace them with 0.01.
However, other rules have been applied
for other values with the general rule of thumb
usually replacing them with half the detection limit.
Leapfrog Geo identifies a range of common errors
and all errors identified are grouped by type.
You can click on any error
and expand it to find out more information.
Some errors will involve you
going back to the original database and fixing them
and some errors can be ignored within Leapfrog Geo.
Once the errors in the database are fixed and validated,
you can be sure your model will have the most up-to-date
accurate data feeding into it.
Next, I can check that my drillhole data
that I have imported is in the correct location
to some of my features such as my topography
and development drives.
I can rotate in the 3D scene
to check for any error in the survey data
and ensure my drillholes are where I expect them to be.
Using the slicer tool,
I can then look at validating my model
on a section by section basis.
Here, I’m looking down on a Z section at the 852 RL.
This shows that my mineralization model
is following these development drives.
If I step through a few more sections
using the greater than or less than keys on my keyboard
on a set step width,
I can see the model changes at other RLs
and how this matches my development.
Alternatively, I can do this process
on a vertical cross-section.
Here, I am looking north through a slice of my model
with development drives and drillholes.
If I stepped through my model using the greater than
or less than keys on my keyboard,
you can visually validate that your model
is matching those development drives
and following any drillhole logging or assays.
This method is one of the more traditional ways
to validate a model and it can be quite time-consuming.
Leapfrog also has some other unique tools
that help you identify areas of your model
that may require further investigation.
The first tool I will show you
is the correlation statistical tool.
This shows you the correlation between the model geology
and the log geology.
To do this, you need to evaluate your mineralized model
and your geological model onto your drillholes.
You do this by right-clicking on your drillhole database
and selecting new evaluation table.
I have already completed this process
and if I bring this evaluation of my mineralized domains
into my scene and have a look,
you can see it shows me
where my mineralized domains are on my drillholes.
Once I have a evaluation table set up,
I then can come to their evaluation in my project tree,
right-click and select statistics.
This will bring up a correlation statistics table
showing me my drilling lithology, my drilling length,
my matching length, matching percent,
non-matching length, and non-matching percent of my model.
In this example,
I’ve shown this for the mineralized vein system,
but this can also be used to test an entire geological model
to determine where more refinement is needed.
Alternatively, you can create a merged table
of your evaluated domains with your original assay data,
and then look at the statistics
to help validate your domains.
You do this up here in the drillhole database,
right-click on your drillholes and select new merged table.
Again, I’ve already created this
where I’ve merged my vein lithologies for my domain
with my gold grades.
I think can right-click on this table and select statistics.
You then can choose which type of statistics
you want to look at.
Firstly, I’ll look at this table of statistics
and this will open up a new tab showing you your domains
with the mean grade, standard deviation, variance,
and quantile data for each of your domain codes.
You can also do this for a box plot.
Again, right-click on your merged table, select statistics,
choose the box plot option
where you can view each of your domains as a box plot
to get a visual indication of where any outliers are
within your domains
and the main grades of each of these domains.
I’ve applied a log scale to this chart
to help see this better
as the status here is very negatively skewed.
I can also have a look at the histogram of a single domain.
This will allow me to examine the data distribution
within the domain and check it’s forming a population
that meets the assumptions of stationarity.
This is where the mean and variance
do not change across the domain
and will enable it to be used in downstream estimation
or numeric modeling.
To do this, I come to the merged table, open it up,
and come to the numeric data of interest,
right-click and select statistics.
This will open up a new tab
showing us the histogram of the data.
Here, I can apply a query filter.
I’ve built one just for this 4,400 domain
so I’m only looking at the data within that domain.
I can then detach this
and have a look at where some of the anomalous grades
are occurring in my scene view.
I will hide some of these estimation domains
and demonstrate how you do this.
So if I feel anything above this area was anomalous,
I could highlight and then I could have a look in my scene
to see if that is actually sporadic
or is it forming another domain population.
In relation to this 4,400,
if I turn that on and turn down the translucency,
I can see that this is pretty sporadic across my domain.
I can reattach that tab
and then I can also look at the histogram of the log
or the cumulative histogram.
Once you have a merged table
of your evaluated domains with grade,
you can then utilize query filters
to test things in your model.
Here in my scene, I have applied a query filter
to show me where there are bearing grades
included within my mineralized model.
I’ve created this query filter
by right-clicking on my merged table
and selecting new query filter.
I’ve previously built this so open this one up for you
and you can see the syntax applied to build that query.
We also have the easy build query function
where you can come in and apply your query.
Here, I’ve selected where my AU_PPM
is less than or equal to 0.3
and my estimation domains
are within these four domains shown.
We also have the advanced query filter builder
that you can utilize for your project.
Once I’ve built my query filter,
I can drag and drop that in the scene
and that query will be applied.
You can also apply the query filter
over here in your properties.
If I zoom in and look around my project,
I can now see areas that I need to go and investigate
to understand why these barren grades have been included
within my mineralized model.
Next, I can use a similar process
to test where I have mineralized grade
that has not been included in my mineralized model.
Here in the scene,
you can see I have applied this query filter
and I’ve built it by right-clicking again,
selecting new query filter.
And if I come into my prebuilt query filter, open that up,
you can see I’ve done the reverse of our first one
where I’ve said if it’s greater than or equal to 0.3
and it’s not in our domains,
then it’s considered mineralized outside of the domains.
This is then being applied in my scene.
And in this example,
you can see there’s quite a lot outside my model.
That is because there is actually quite a few
unmodeled domains in this project.
You can also use query filters to test areas of your model
to see where geology code logged
does not match geology modeled.
To do this, I’ve created an evaluation of my host rock model
and merged that with my geological lithology table.
I’ve then built a query filter here
to test where my end site is outside my model.
If I have a look at my scene,
I can rotate and zoom in on some of these areas,
click on a hole and I can go and investigate
why that is outside that geology unit.
If I did want to examine a hole in more detail,
I can use the drillhole correlation tool
here under my drillhole database
to examine in further detail.
I right-click on this and select new drillhole set,
which brings me out to a new drillhole set tab.
I then select the pit color function,
which will take me back to my scene view.
I then choose my select drillhole button
and choose the hole I’m interested.
I then can click save
and then it would appear here in my drillhole set layer
where I can bring across my host rock
and my original logging.
And then I can come in and identify
why I have these different areas
within my model versus my logging.
The final validation step for any project
is always to have your project peer reviewed
by one of your colleagues.
To do this, we are going to use Seequent Central.
I will show you this briefly.
However, we have another webinar on this
that was completed earlier this year
by one of my colleagues,
so please feel free to check out this video recording
on the Seequent website.
To upload your project,
you will need to be connected to the Central server.
You can do so by connecting up here
in your right-hand corner.
Once you are connected, you select publish.
You select which objects you would like to appear
in your online web browser.
Set your project workflow stage.
I’m going to choose peer review for this one.
Add in any revision notes for your colleagues
and then select next to start the publish and upload.
Once your project has finished publishing,
it will say publish successful here in the dropdown menu,
and it will be available for your colleagues to download
if they have the right permissions and access to Central.
It is also now available via the web browser
for non-Leapfrog Geo users such as managers
to review via Web-Verse and make comments.
To do this, you can select go to portal
and this will take you to the appropriate web browser
where your project is stored.
As you can see, it’s got the revision note
saying please review this project.
It’s got the stage of peer review
and it’s got the user who uploaded that project.
You can then select the project
to view within the web browser,
come to your geological model,
open up your domains and have a look at these
without needing access to a Leapfrog Geo license.
Comments can then be made by clicking on this comment button
and selecting add comments.
These can be geo tagged on your domain
and you can tag your colleagues here
so they get an instant notification
on any area to go and look at in further detail.
That brings me to the conclusion of today’s webinar.
If you require any further information,
please refer to the Seequent website,
which has many blogs, webinar recordings, and videos
to help you with your learning.
We also have a great series of free online courses
to help up-skill which can be found
under your MySeequent account.
If you do need any additional help,
please reach out to your local support team
who will be willing to assist.