A Guide To The Parker

Challenge
DISCUSSING THE PURPOSE, RESULTS AND CHALLENGES OF THE PARKER
CHALLENGE AS WELL AS WHAT IT MEANS FOR RESOURCE DISCLOSURE
 Introduction:
1-9

6 Problems With Mineral Estimation and Why The

Parker Challenge Is Necesary:
10 - 12

Results of The Parker Challenge:

13 - 16

Three Main Takeaways From The Parker

Challenge:
17 - 19

Issues With The Parker Challenge To Be Improved

In 2024:
20 - 23

What This Means For Resource Disclosure:

24 - 25
Introduction
 Given the same data set, will geo-
statisticians, geoscientists and geologists
achieve the same estimation? What is
the variability?

The challenge, which has never been done

before, generated some interesting results
that miners should be aware of, particularly in
the resource disclosure space.

In this article, we will run through The Parker

Challenge and the findings that were drawn
from it, as well as the risks that were exposed
for resource disclosure and how you can
minimise your exposure.

1
 Dr Harry Parker
The late Dr Harry Parker (1946-2019)
contributed a considerable amount to
standardising reconciliation nomenclature, as
well as working on the CRIRSCO reporting code.

The Parker Challenge culminates the work Dr

Harry Parker achieved in his life to create
something that he would have been equally
fascinated and perturbed by.

2
 The Parker Challenge
host – Scott Dunham
With over three decades of experience in the Resource
sector, Scott Dunham, the founder of SD2, is now dedicating
himself to imparting his knowledge and expertise to the
industry.

Scott possesses a deep passion for disseminating insights

and perspectives, which is exactly what The Parker
Challenge sets out to accomplish.

Committed to addressing the fundamental factors behind

the intricate challenges faced by the industry on a daily and
yearly basis, Scott searches to make the industry more
efficient.
3
 What Is The Parker Challenge?
The Parker Challenge called on mineral resource estimators to create a classified model from the same data set as with
the results to be presented at the 2023 AusIMM Mineral Resource Estimation Conference. The task they were given was:

“Develop a classified and reportable mineral resource estimate from the supplied
data. The estimate will be used to inform long-term mine planning and investment
decisions.”

The aim was to quantify the “between person variance” or “pattern noise” in mineral resource estimation between
geologists and mineral resource estimation techniques. The estimation results from each participant were
amalgamated and compared to work out just how large human difference can be despite being given the same
data.

6
 What Is The Parker Challenge?
Presenter, Scott Dunham, stated that:

“One of the things that plagues the resource estimation discipline is that we all work independently,
working on individual deposits to produce our estimates. We all understand that how the estimate
performs is really difficult to determine. And one of the things that’s never been measured is that if
you gave that same data, that same information, to a wide group of people and asked them to
come up with their estimate, how different would each one be?” (Dunham, S. 2023)

The range was calculated on three differences:

·Difference in approaches
·The range of outcomes
·Difference in classification decisions

The findings were presented at the conference and highlighted just how different the
estimates can be between different individuals, despite being given the same data – since
the participants haven’t used a one-source-of-truth platform to find the same result together,
the results should be interesting.

Sponsored by Rio Tinto, the winner of The Parker Challenge received $55,000.

Before we get into the results of the challenge, let’s get into some definitions.
7
 Glossary of Definitions
When is something reasonably inferred?
An inferred mineral resource refers to a section of a mineral resource where the estimated
quantity, grade, or quality is determined based on limited geological evidence and
sampling. The available geological evidence is enough to suggest the presence of
geological and grade or quality continuity, but it doesn’t provide conclusive verification.

How precise is an indicated resource?

An indicated mineral resource is a more precise category of resource estimation
compared to an inferred mineral resource. It represents a higher level of confidence in
the estimated quantity, grade, or quality of the resource. This confidence is based on a
greater amount of geological evidence and sampling allowing for a more reliable
assessment of the continuity of the geological features and the grade or quality of the
deposit.

While uncertainties still exist, an indicated mineral resource provides a more reliable
foundation for decision making and planning in mineral exploration and development
projects.

8
 Glossary of Definitions
What is “pattern noise”? What is a measured resource?
Pattern noise, or between person variance, is A measured mineral resource is the most
the difference measured between each precise and reliable category of resource
person’s estimation. The larger the pattern estimation. It represents the highest level of
noise that’s revealed, the bigger the problem confidence in the estimated quantity, grade,
that is discovered from The Parker Challenge. or quality of a mineral deposit. This
confidence is based on extensive geological
Factors that contribute to pattern noise evidence, sampling, and data analysis,
include: which provide a robust understanding of the
deposit’s geological characteristics,
· How many decisions do you make when continuity, and grade or quality distribution.
estimating a mineral resource.
· Which decisions are more critical, and A measured mineral resource undergoes
which are trivial. rigorous and detailed assessment, including
· The real range of reasonable outcomes? comprehensive drilling programs and
thorough geological modeling. This category
The burning question for The Parker of resource estimation is considered highly
Challenge, what drives the noise in the reliable for project evaluation, mine planning
system? and feasibility studies.

9
 6 Problems With Mineral
Estimation and Why The Parker
Challenge Is Necessary
When there’s no target for someone to hit, it’s often difficult to judge what the
best result will be.
 Data Limitations
Insufficient or incomplete data can hinder accurate resource estimation, not that it’s a
problem for The Parker Challenge seeing everyone is given the same data. Limited
access to geological information, incomplete sampling, or inadequate data quality
can introduce uncertainties and biases into the estimation process. Additionally, gaps
in the data can be subject to bias if information is extrapolated.

Geological Complexity
Mineral deposits often exhibit complex geological characteristics, such as irregular
shapes, variable grade distributions, and intricate mineralisation patterns.
Understanding and modeling these complexities accurately pose challenges during
resource estimation, and not everyone uses one project resource estimation template.

10
 Personality Bias
Everyone is subject to personal bias – it’s human nature. Estimations can be changed
based on the bandwagon effect, where we’re swayed by other people’s opinions and
the group that we identify with. Additionally, the “first mover” advantage can sway our
opinion as we tend to agree with whoever speaks first.

Uncertainty and Risk

Resource estimation involves dealing with inherent uncertainties and risks. Uncertainties
can arise from limited data, geological interpretations, sampling errors, or assumptions
made during the estimation process. Managing and quantifying these uncertainties is
crucial for decision-making and project evaluation.

11
 Subjectivity and Interpretation
Estimation involves making interpretations and assumptions based on available data
and geological knowledge – there isn’t one standardised resource estimation
procedure. Subjectivity in interpreting geological features, selecting estimation
methods, or applying cutoff grades can introduce bias and affect the reliability of
resource estimates. Additionally, the different software used can draw different results.

Regulatory Compliance
Mineral resource estimation needs to adhere to specific regulatory guidelines and
industry standards. Ensuring compliance with these standards while addressing
technical challenges can be demanding. Good governance over data and models is
critical.

12
Results of The Parker Challenge
Overall, the Parker Challenge was a success and there were some interesting
findings.
 The Participants
The majority of participants were geologists, which was desired, however there were
more academics than expected. The breakdown of participants is shown below.

13
 Difference In Resource Classification
When it comes to resource classifications, the difference between participants was less than expected,
at least for some classifications. As you can see from the graph below, indicated resources were quite
similar across the board, while inferred resources had a lot of variability.

Funnily enough, the more experience

someone had the less likely they were to
classify a resource as measured, while the less
experience someone had, the more likely they
are to classify an inferred resource.

Additionally, resource estimates were more

similar when the resource classification was
higher, which is surprising? Probably not. It
makes more sense that when a resource is
more certain, there would be less guess work,
and with less guess work, comes less personal
bias and as such less variability.

14
 Variation In
Volume
When it comes to the difference in
volume estimated amongst
participants, variation was large.
Even with the two top and two
bottom results removed from the
data, the range is still near on
10,000Kt Cu metal (measured and
indicated). That’s a difference of
-66% to +91% of metal in the
measured and indicated estimates.

15
 There's Even Variance In
Time Used For The
Challenge
Sometimes a quick estimate is just as good as a
large effort – overthinking can have you chasing
after shadows which won’t move you forward.
Sometimes you should embrace clarity and trust
your instincts because quite often you can find the
best decisions come from simplicity. However, that
isn’t to say you shouldn’t be thorough. Here’s the
results that were yielded along with the time that
each participant spent on their estimate. The key
takeaway from this is that a longer estimate might
make you succumb to your own biases, so
sometimes the best thing to do is keep it simple, and
not overthink your estimates.

16
Three Main Takeaways From
The Parker Challenge
 Classification and Estimation Are
Very Different
The magnitude of variation caused by different geological domains and classification is significantly
higher compared to the variation in grade estimation on a global scale. Surprisingly, many
individuals tend to solely concentrate on the grade estimation and geostatistics aspects, even within
the classification process.

Estimation and classification are distinct processes that require different skills, knowledge, and
experience. It involves making the best possible forecast for a particular purpose, aiming to provide
an accurate assessment.

On the other hand, classification focuses on understanding and evaluating the associated risks.

The JORC Code, and other codes, as a guideline for reporting mineral resources, should ideally
reflect these fundamental differences by providing clear guidance and considerations for both
estimation and classification methodologies.

17
 Mineral Resource Estimation
Software Does Not Replace Expertise
Mineral resource estimation software has a significant impact on various aspects of our work. The
workflows and training provided by vendors often promote specific approaches and can hinder
critical thinking by neglecting to address unknown or unexpected situations.

In order to avoid fundamental errors, machine learning and artificial intelligence (ML/AI)
approaches in resource estimation benefit from the guidance and expertise of experienced
practitioners in the field. This requirement is not unique to ML/AI; it is also applicable to industry-
standard kriging practices. Both methodologies rely on the involvement of skilled professionals to
ensure accurate and reliable resource estimations.

18
 Evidence Based Geology Is
Important
It is essential to ensure that geology is valid, verified, and plausible, meaning it should be supported
by evidence and align with accepted geological principles and knowledge.

The winning entry adopted an integrated approach, seamlessly connecting various stages from
data review to final classification. The result was a cohesive "story" that flowed smoothly and
coherently throughout the process.

There exists a considerable amount of variability among individuals in resource estimation. This
variability arises from interactions between volume, tonnes, and geological domains, as well as the
chosen estimation method and approach. Multiple sources contribute to parametric uncertainty,
further complicating the process. Experience plays a significant role in finding the optimal balance,
commonly referred to as the "sweet spot," where expertise and knowledge align to achieve reliable
and accurate estimations.

19
 Issues With The Parker
Challenge To Be Improved On
In 2024
Even though there were some great take-aways from The Parker Challenge,
and the challenge was a success as a whole, that doesn’t mean there weren’t
problems that can be fixed for next time.
 Significant Dropout Rate
The first problem that occurred was that
there was a significant dropout rate. From
the 306 downloads, only 29 submitted
entries – that’s hardly 10%. Some of the
reasons for this might include:

It was the first time running the

challenge.
It was running over Christmas and New
Year’s so the timing was poor.
It wasn’t exactly what people
expected.
It probably asked too much from
participants (multi-element, multi-
domains, reports etc).

20
 The Challenge Was Subject To Too
Much Variability
If the challenge was to be run again, it
could include isolating some of the key
steps to better understand how much
variability comes from each step. For
example, if everyone started with the same
geological ‘skeleton’, there would be more
clarity about the impact of downstream
processes on the variability between
estimates. Another alternative would be for
participants to be instructed to use the
same estimation method.

21
 Classification Process Itself

The classification process itself is a key step with

significant room for the practitioner to exercise
personal judgement. It would be interesting for the
next challenge to isolate the classification process
by providing participants with the same estimated
model to classify. This approach would enable
transparency of the variability associated with
classification alone.

22
 There Weren't Enough Industry
Experts Involved
Lastly, there were many academics in the
challenge, which is good, but it would be even
more useful if more highly experienced industry
experts were involved. Industry experts would
not have found the $55k prize money worth a
serious attempt and so many probably had
better things to do with their time. Arguably, if
the prize pool was increased, highly
experienced industry experts would find the
challenge more worthy of their time and would
be more likely to participate – which means
there would be more experienced judgement
involved, better reflecting the true state of
competence in the industry.

23
What This Means For Resource
Disclosure
 Governance Is Key
So then, what does this mean for resource disclosure? The Parker Challenge has
expressly shown that governance is key. As we can see, when it comes to project
resource estimation methods and results, it is very easy to arrive at different opinions.

If your team is not on the same page about your resource disclosure and reconciliation,
it can be very easy to land yourself in hot water and waste a significant amount of time
and money.

You can limit variability in your team by limiting the chance for people to have different
estimations which can be done with a resource disclosure and mine reconciliation
platform to assist with your governance.

24
 A Resource Governance and
Disclosure Platform
If you’re looking for the right Resource Disclosure and Mine Reconciliation governance
solutions, you can be sure to remove subjectivity over your mineral estimates and public
reporting with K2fly. K2fly allows you to coordinate your reporting process on a single
standardised platform so you will never have to experience the difficulty of trying to find
or follow up information from your team – you will already know exactly where to find it.

K2fly’s solution let’s you:

Maintain your social license by limiting mistakes and protect your reputation.
Automate your disclosure process to decrease time wasted as well was human-made
errors.
Give yourself a gold star for governance with the highest standard in platform
readability and reliability.
Remove animosity with disclosure by having all of your data in one place.

K2fly’s proprietary software services some of the largest tier one miners in the world,
helping them achieve their ESG goals through governance platforms that provide a
companywide single source of truth. Spanning over 500 sites in more than 62 countries,
we’re sure our solution will be able to service your needs.

25
 Resource Governance solutions
for net positive impact

Get in contact with our team and find out how

we can solve the issues you’re having today.