What is Sampling?
Sampling is the process of collecting a portion of material from a larger mass in such a way that the sample accurately represents the characteristics of the whole. In gold exploration and mining, this means obtaining a subset of rock, soil, drill cuttings, or stream sediments whose gold content reflects that of the deposit being evaluated. The goal is to obtain reliable data for grade estimation, geological interpretation, and decision-making.
In essence, good sampling is about representativeness—capturing the true average grade and variability of gold within a given volume of rock. Because gold mineralization is often erratic, the challenge lies in designing a sampling approach that minimizes bias and sampling errors.
The Importance of Representativeness
A sample is said to be representative if the gold grade measured in the sample equals, within acceptable limits, the true grade of the material it represents. Representativeness depends on:
-
The sampling method used (e.g., channel, grab, chip, or drill-based)
-
Sample mass and distribution relative to the heterogeneity of the deposit
-
Proper sample preparation and subsampling
-
QA/QC implementation
Failure in any of these aspects can lead to bias, which may systematically overestimate or underestimate the gold grade, leading to serious financial and operational consequences.
Types of Sampling Errors
Understanding error types is essential to designing a sound sampling program. Pierre Gy’s sampling theory remains the cornerstone of modern sampling science and identifies several key sources of error:
-
Fundamental Sampling Error (FSE)
-
Results from the heterogeneity of gold particles within the sample.
-
Highly influenced by particle size, density, and distribution.
-
For gold, FSE is usually high due to coarse particle size and clustering of gold grains.
-
-
Grouping and Segregation Error (GSE)
-
Occurs when particles segregate during handling or transport due to differences in size or density.
-
Common during crushing, splitting, or transportation stages.
-
-
Delimitation and Extraction Error
-
Arises when the physical boundaries of a sample are not clearly defined or properly extracted.
-
For example, taking a biased chip sample along visible mineralization rather than across it.
-
-
Preparation Error
-
Occurs during crushing, splitting, or pulverizing when subsamples do not maintain proportional representation of the original material.
-
-
Analytical Error
-
Stems from limitations in laboratory precision and accuracy, including contamination, incomplete digestion, or calibration issues.
-
-
Human Error
-
Results from inconsistent field practices, poor labeling, or inadequate supervision.
-
Often the most underestimated yet impactful source of bias.
Gold’s Unique Sampling Challenges
Gold presents special difficulties in sampling because of its nugget effect—a term describing the uneven distribution of gold particles in the rock mass. This leads to high local grade variability, meaning that two samples taken just centimeters apart can yield vastly different assay results.
Other challenges include: -
Particle size variation: From micron-sized gold in sulphides to visible nuggets in quartz veins.
-
Density contrast: Gold’s high density (~19.3 g/cm³) causes segregation during handling.
-
Association with specific minerals: Gold may occur as free grains, locked in sulphides, or adsorbed on carbonaceous matter, affecting recovery and assay precision.
These factors make sampling, sample preparation, and QA/QC even more critical in gold projects.
Key Principles for Reliable Sampling
To achieve reliable sampling in gold exploration, adhere to these guiding principles:
-
-
Understand the geology: Know your deposit type, mineralogy, and grade distribution before designing sampling protocols.
-
Select appropriate methods: Match sampling techniques (e.g., RC, diamond, trenching) to deposit scale and exploration stage.
-
Control sample mass and spacing: Larger sample volumes reduce error, but must remain cost-effective.
-
Standardize procedures: Use written protocols and ensure all team members are trained.
-
Implement QA/QC: Regularly insert blanks, standards, and duplicates to track precision and accuracy.
-
Document and review: Maintain detailed sampling logs, photos, and chain-of-custody forms.
No comments:
Post a Comment