Channel sampling in mining means cutting a continuous, even strip of rock across a mineralised zone, at a face, a wall, a trench or an outcrop, and bagging everything that comes out of it as one sample. Chip sampling is its quicker cousin: a line of chips knocked off with a hammer and chisel, touching or spaced apart. Both are used to measure grade and width where rock is exposed, from early exploration trenches to daily grade control in underground mines.

Try it: Channel sample mass and cutting-time calculator →

They are cheap compared with exploration drilling, which is why they are everywhere. They are also easy to do badly. A sampler who takes a little more of the soft, shiny vein and a little less of the hard wall rock produces a higher grade than the rock really carries. Nobody intends it; the rock and the tools make it happen unless the method is designed to stop it.

“In one mine study, 75% of saw-cut channel samples hit their target mass; none of the chip samples did.”

Channel, chip-channel, chip and panel: the four linear sample types

A detailed review of underground gold grade control by Dominy and co-authors, published in the journal Minerals, sets out clear definitions that work just as well at surface. We follow them here, because loose use of “channel” and “chip” is one of the reasons sampling results get misread.

• Chip sample: individual chips collected in a discontinuous fashion and combined into one sample. Effectively a set of specimens.
• Chip-channel sample: a series of continuous, touching chips across the zone, combined into one sample.
• Channel sample: a consistent volume cut across the sample zone, by hand or with a diamond saw.
• Panel sample: large or multiple chip or chip-channel samples spread over an area of the face rather than along a line.

The JORC Code’s Table 1 uses the same vocabulary. Under sampling techniques it asks public reports to describe the nature and quality of sampling, giving “cut channels” and “random chips” as examples, and to explain the measures taken to make samples representative. Under logging it asks whether core, costean and channel samples were photographed. If a company reports trench or face results, those are the lines to read first.

Chip sampling in mining: fast and flexible, low quality

Chip sampling in mining is quick and needs almost no set-up, which is why it survives. Dominy’s review rates chip samples as low quality, prone to high delimitation and extraction errors driven by operator preference and rock hardness, and labour-intensive though relatively fast. The review also notes that chip samples generally carry a positive bias because quartz vein material gets over-sampled. Chips are best treated as indicators: they show where mineralisation is, not reliably how much.

Channel sampling in mining: slower, much better

The same review suggests channels about 5 to 10 cm wide and 2.5 cm deep, giving 3 to 6 kg per metre, and warns that hand-cut channels are hard to keep uniform and often end up closer to chip-channels. Saw-cut channels, made with two parallel diamond-saw cuts and the block between them chiselled out, are generally higher quality. A typical 1 m saw-cut channel with a 2.5 cm by 5 cm profile should yield about 4 kg, and a well-trained two-person team can cut and collect a metre in 10 to 15 minutes. Hand cutting takes 30 minutes or more per metre.

How method choice shows up in the numbers

The clearest evidence comes from a mass-recovery comparison in Dominy’s review. At a North American gold mine, the same crew sampled 2.5 m by 2.5 m faces carrying a 1.0 to 1.3 m quartz vein in altered dolerite, using six different linear methods. Each sample’s dried mass was compared with the mass its target volume should have produced. A sample within 5% of target counted as “on target”.

Under-sampling is not harmless. If the missing mass is mostly hard wall rock that flew off the chisel, the bag is enriched in softer vein material. If it is soft, crumbly ore that washed or fell away, the bag is depleted. Either way, the assay no longer represents the width it is supposed to.

The spread matters as much as the average

The same table gives the range of masses actually collected per metre. Chip samples with a 2.0 kg/m target ranged from 0.4 to 1.8 kg/m. Saw-cut channels with a 3.0 kg/m target ranged from 2.2 to 3.5 kg/m. A tight range means each sample represents a similar volume of rock, which is exactly what a resource geologist needs when averaging grades along a drive or trench.

Evidence from South African gold mines

The review also cites Witwatersrand studies. In one, duplicate saw-cut channels were taken beside earlier chip samples, and the channel samples showed half the nugget effect of the chips. The authors concluded that channel samples at 5 m spacing were equivalent to chip samples at about 2.5 m. Another Witwatersrand study described an “in-stope sample discard process” in which samplers selected better-looking material, producing a positive “waste discard bias”. Better sampling can mean fewer samples, not more.

How to cut a good channel sample, step by step

Most of what makes channel sampling in mining reliable is preparation and discipline rather than equipment. The steps below follow Dominy’s guidance and apply at surface outcrops and trenches as well as underground.

1. Map first. Record the geology of the face, trench wall or outcrop and decide where the zone boundaries are. The review stresses that mapping and sampling are closely linked.
2. Clean the surface with clean water or brushes, so dust and fines from elsewhere do not contaminate the sample.
3. Mark the line across the zone, at right angles to its boundaries, and split it into intervals at geological contacts.
4. Set the orientation to the geology. The channel can run horizontally, vertically or perpendicular to dip, depending on the structure.
5. Cut to a set width and depth. A template of wood or aluminium cut to the channel dimensions helps check delimitation.
6. Catch everything. Use a bucket, a wide receptacle held close to the rock, a tarpaulin or rubber matting below the line.
7. Work from the bottom up when several samples are taken on one face, so falling fragments do not contaminate the lower bags.
8. Bag, tag and photograph each interval, and have the surveyor pick up its position in 3D.

Controlling fly rock and extraction loss

Fly rock is the most common extraction loss in channel sampling in mining. According to the review, chip and channel sampling can lose as much as 20% of the extracted mass as fly rock, typically 5 to 10%. Sharp, high-quality chisels help, because blunt points eject fine fragments. A two-person team, one cutting and one catching, reduces both delimitation and extraction errors. Air picks are fast but tend to produce fly rock and can cut deeper around soft or high-grade material.

Weigh every sample

The simplest quality check for channel sampling in mining is the one the mass-recovery study used: record the dried mass of each sample at the laboratory and compare it with the target from the channel dimensions. The review describes a nominal acceptance band of plus or minus 5%. Samples outside it are flagged as over- or under-sampled. It is not perfect (a sample with loose material fallen in can hit its mass while being wrong), but it catches most careless cutting cheaply.

Planning a channel sampling in mining program

Good individual samples still add up to a weak dataset if the program behind them is not designed. Before the first line is cut, decide what question the samples must answer: is the vein continuous along strike, how wide is the ore zone, how does grade vary across it? The answers set spacing, interval length and method.

Spacing along the drive or trench

There is no universal spacing. Underground, faces are usually sampled after every development round, so spacing follows the advance. In trenches, lines are set by how quickly the geology changes along strike. The Witwatersrand comparison in Dominy’s review offers a useful way to think about it: if better samples carry less random error, you can space them further apart for the same confidence. Spending more per sample on a saw can mean fewer samples overall.

Interval length across the zone

Break samples at geological boundaries so each bag represents one rock type: hangingwall, vein, footwall, as shown in the face photographs in the review. The CIM guidelines make the same point for drill samples: individual samples should not straddle different lithologies or very different styles of mineralisation. Include shoulder samples of barren wall rock on each side so the grade boundary is closed.

Choosing the method for the ground

• Hard, competent rock with a smooth face: a saw-cut channel gives the best result.
• Rough or blocky faces: saw placement is difficult; a careful hand-cut channel with a template may do better.
• Soft, friable zones: watch for fines washing out of saw cuts and for deeper hand cuts; catch everything on a mat.
• Coarse, clustered gold: move to panel or bulk samples, as the Gwynfynydd example shows.
• Reconnaissance at outcrop: rock chips are acceptable if they are reported as selective.

Whatever the method, write it down in a one-page procedure and train every sampler on it. Dominy’s review is frank that sample quality varies from crew to crew and shift to shift, with heat, humidity, location and fatigue all playing a part. A written procedure, a template for channel dimensions and a mass check are the three cheapest controls a channel sampling in mining program can have.

Recording what matters

For each sample, record the location surveyed in 3D, the orientation of the line, the method, the width and depth cut, the rock type and weathering, the sampler’s name, the target and received mass, and a photograph with the sample line and tag visible. JORC Table 1 asks for the orientation of sampling relative to geological structure and for channel photography, so these fields feed directly into any public report.

Channel sampling in mining exploration: trenches and outcrops

Before there is a mine, channel sampling in mining projects happens on outcrops and in trenches. It is often the first quantitative grade and width information on a prospect, and it frequently decides whether a drill rig is booked (our guide to picking an exploration drilling contractor covers that step). The same rules apply as underground, with some extra challenges: weathering, soil cover and the need to expose fresh rock.

Exposing the rock

Where outcrop is poor, geologists strip soil with machinery or dig trenches to bedrock. Ontario’s early exploration permit guide describes mechanised stripping to expose bedrock, followed by pressure washing of the surface, and notes that stripping is often used to uncover zones identified by geophysics. In Ontario, mechanised stripping of more than 100 square metres, or bedrock excavation of more than three cubic metres, within a 200-metre radius needs an exploration permit. Check your own jurisdiction’s thresholds before bringing in an excavator.

Weathered versus fresh rock

Channels cut in weathered rock can give grades that differ from fresh rock at depth, because some elements are leached and others concentrated near surface. Record the weathering state of each interval and do not treat surface channel grades as a substitute for drilling. They are a guide to where to drill and to which intervals deserve close sampling in core (see our guide to diamond drilling costs, underground vs surface).

Rock chip sampling for prospecting

At the prospecting stage, rock chip and grab samples from outcrops and old workings are used to find anomalous areas. Our guide to soil, stream and plant geochemical surveys covers the wider sampling around them. They are selective by nature and should be reported as such. A single high rock-chip assay shows that mineralisation exists there; it says nothing about width or average grade. Our guide to geological mapping methods and scales explains how mapping, rock chips and channels fit together at outcrop.

Underground: faces, walls and grade control

Underground, channel sampling in mining operations is part of the daily cycle. After each development round, a geologist maps and samples the face or walls so the mine knows whether the next round is ore or waste. Dominy’s review notes that face samples are often taken about 1 m (plus or minus 0.5 m) above the drive floor, although this depends on the orebody, and that some operations divide the face into two horizontal halves with a sample from each.

Panels for coarse gold

Where gold is coarse and clustered, even careful channel sampling in mining may miss or over-count particles. The review describes the Gwynfynydd mine in Wales, where dispersed clusters of coarse gold caused a negative bias in chip-channel samples. Faces were divided into quarters and close-chipped to give 25 kg from each of two panels, and the combined 50 kg micro-bulk sample was processed through a gravity unit on surface. Panel samples can exceed 100 kg, so they are also used for metallurgical testing.

Safety at the face

Channel sampling in mining means standing at a rock face. The review lists fly rock as both a sampling error and a safety issue, and notes that saws need care because of rotating parts. Mines apply their own rules on scaling and barring down loose rock before anyone approaches the face, and loose, jointed ground can also drop non-sample material into the bag. Screen or bar down before sampling, and never sample under unsupported ground.

Where satellite targeting helps before the first channel

Channels and trenches are only as useful as their location. A trench cut across the wrong structure is expensive, and in tropical terrain with deep soil and little outcrop, it is easy to cut many before finding the right one. Satellite analysis narrows the ground first.

Our satellite-based mineral detection analyses multispectral and hyperspectral imagery over your licence to flag alteration halos, faults and fractures and deposit-associated patterns, with no ground disturbance. You send coordinates, a KML/KMZ file or a polygon plus the country and target mineral, and we deliver in 5 to 20 business days. That tells your field team where outcrops and trenches are worth channel sampling. For the wider sequence, see our complete guide to mineral exploration.

• 📊 Premium mineral-intelligence report: high-potential zones, prospectivity heatmaps, estimated location and depth ranges and geological interpretation, as PDF plus GIS files.
• Premium+ with TargetMax™ Drilling Intelligence: drilling-angle recommendations and 3D subsurface models of vein structures for the drilling that follows sampling.
• Economics: early exploration from months to days, at up to 80–85% lower early-exploration cost.
• Experience: 100,000+ hectares scanned, 20+ mineral types, 25+ countries.
• ⚠ Limits: satellite targets are exploration targets. Channels, trenches and drilling test them.

See our satellite-driven 3D mineral prospectivity mapping sample, draw your licence on mining.farmonaut.com: Map Your Mining Site, or ask for pricing on the mining query form.

QA/QC for channel and chip programs

The CIM Mineral Exploration Best Practice Guidelines expect quality control on every sampling campaign, not just drilling: external blanks, certified reference materials, duplicates and checks at a second laboratory. In channel sampling in mining, a field duplicate means cutting a second channel beside the first. Duplicate pairs that disagree widely are a sign that the method, not the laboratory, is the weak link.

“Channel samples at 5 m spacing matched chip samples at about 2.5 m in one Witwatersrand study.”

Frequently asked questions

Reviewed September 2026 against Dominy et al. (2018), “Integrating the Theory of Sampling into Underground Mine Grade Control Strategies”, Minerals 8:232, including its mass-recovery comparison and Witwatersrand and Gwynfynydd case material; the JORC Code (2012) Table 1; the CIM Mineral Exploration Best Practice Guidelines (2018); and Ontario’s Early Exploration Permit Activity Information guide.

Study figures come from specific mines and crews and will differ elsewhere; use them as guidance, not as expected values. Satellite targets are exploration targets, not mineral resources.