How LiDAR Mining Survey Works

TL;DR: A LiDAR mining survey works by firing hundreds of thousands of laser pulses per second from an aircraft- or ground-mounted sensor, timing each return to build a georeferenced 3D point cloud of the pit, stockpiles or plant. Tied to GDA2020/MGA2020 control and AHD heights, that point cloud becomes the surface, volume and as-built data your mine plan, reconciliation and compliance reporting depend on — captured in hours rather than days.

Key takeaways

• LiDAR measures distance by timing laser pulses (time-of-flight); a UAV sensor such as a DJI Zenmuse L2 or a Riegl-class payload fires 200,000–500,000 pulses per second to generate millions of measured points across a pit in a single flight.
• Survey-grade accuracy comes from control, not the laser alone — RTK/PPK positioning plus GDA2020/MGA2020 ground control and AHD levelling deliver 20–50 mm absolute accuracy on bare earth; check-shot validation against total-station marks is what makes the data defensible.
• LiDAR penetrates vegetation and captures last-return ground points, so it outperforms photogrammetry on partly grassed waste dumps, rehabilitation areas and batters where a camera only sees the canopy.
• Commercial UAV LiDAR in Australia is flown under CASA Part 101 by a RePL-licensed pilot under an ReOC, typically within 120 m AGL and visual line of sight, with site-specific approvals on active operations.
• A drone LiDAR pit-and-stockpile survey usually runs AUD $2,500–$10,000+ depending on area, control density and deliverables — a fraction of the cost and downtime of walking a total station over the same ground.

What LiDAR actually measures

LiDAR — Light Detection and Ranging — is an active sensor. Where a camera passively records reflected sunlight, a LiDAR sensor emits its own near-infrared laser pulses and measures the time each pulse takes to travel to a surface and back. Because the speed of light is known, that time-of-flight converts directly into a precise distance, or range.

The sensor sweeps a rotating or oscillating mirror so the beam covers a wide swath beneath the aircraft. Combine the range to each point with the exact position and orientation of the sensor at the instant of firing, and you get an XYZ coordinate for every return. Do that several hundred thousand times a second and you build a dense, measured 3D model of the ground — the point cloud.

Two characteristics make LiDAR particularly well suited to mining. First, it captures multiple returns from a single pulse: a beam clipping the edge of low scrub on a rehabilitation area returns one pulse from the vegetation and a later pulse from the bare earth beneath. Classification software keeps the last (ground) return, giving you a true bare-earth surface that photogrammetry cannot reconstruct under cover. Second, LiDAR is largely independent of lighting — it works in flat overcast, low sun and even at night, which matters on a 24/7 operation where survey windows are dictated by blast and haulage schedules, not daylight.

How a LiDAR mining survey works, step by step

A field-to-finish LiDAR survey on an Australian mine site follows a consistent sequence. Understanding it helps you scope the work and judge whether a deliverable is trustworthy.

Step 1: Establish survey control (GDA2020 / AHD)

Before a single pulse is fired, the survey is anchored to the national datum. Ground control points and check points are coordinated in GDA2020, projected into the relevant MGA2020 zone (for example Zone 50 across most of the Pilbara, Zone 51 through Kalgoorlie, Zone 55/56 through the Bowen Basin and Hunter Valley), with heights on the Australian Height Datum (AHD). Control is observed with a GNSS base or CORS connection and verified with a Leica or Trimble total station. Without this step, a point cloud is internally consistent but floating — useless for reconciliation against a mine plan held in the same coordinate system.

Step 2: Plan the flight and exclusion zones

The pilot designs flight lines for the target swath overlap (typically 30–50%), altitude and ground speed needed to hit the required point density. On an active pit this is also a safety and airspace exercise: deconfliction with the operation's spotters and dispatch, exclusion zones around the digging face, and any CASA approvals for operations near a controlled aerodrome or above standard limits.

Step 3: Capture the point cloud

The UAV flies the programmed mission while the LiDAR sensor and its onboard GNSS/IMU log ranges and trajectory continuously. A single 30–40 minute flight can blanket a large pit or a stockpile yard with tens of millions of points. For underground voids, confined plant or steep highwalls, a terrestrial scanner (a Leica RTC360 or FARO Focus, for instance) or a mobile/handheld SLAM unit fills the gaps the drone cannot see.

Step 4: Process the trajectory and georeference

Back in the office, the raw laser ranges are fused with post-processed kinematic (PPK) trajectory data and tied to the ground control. This produces a georeferenced point cloud in MGA2020/AHD. Strip-to-strip alignment is checked, and the cloud is validated against the independent check points held back in Step 1.

Step 5: Classify, model and deliver

The point cloud is classified — ground, vegetation, infrastructure, noise — and thinned into a digital terrain model (DTM) or triangulated surface (TIN). From there the deliverables flow: end-of-month surfaces, stockpile and pit volumes, contours, cross-sections, haul-road and batter checks, and rehabilitation comparisons against prior epochs.

LiDAR vs photogrammetry on a mine site

Drone photogrammetry and drone LiDAR are often discussed together, but they answer different questions. Photogrammetry reconstructs a surface from overlapping photos and excels at bare, well-textured ground — open pit floors, fresh stockpiles, clean earthworks — where it delivers excellent visual orthomosaics and 1–3 cm accuracy at low cost. LiDAR earns its premium where photogrammetry struggles.

In practice many ISS surveys are hybrid: LiDAR for the vegetated rehabilitation cells and a photogrammetry pass for the orthomosaic and the clean active pit, both tied to the same control so they merge cleanly.

Accuracy: where it comes from and how to verify it

A common misconception is that buying a more expensive laser buys accuracy. The sensor sets the precision of each range, but the absolute accuracy of the survey — how well the cloud sits in MGA2020 and AHD — is governed by control and trajectory quality.

Three things determine whether a LiDAR mining survey is fit for reconciliation:

• Ground control density and distribution. Well-distributed GDA2020 control surveyed with a total station or GNSS gives the processing software the constraints it needs to remove systematic tilt and drift from the cloud.
• RTK/PPK positioning of the aircraft. Logging the sensor's trajectory against a local base or CORS, then post-processing it, is what pulls absolute accuracy into the tens-of-millimetres range.
• Independent check shots. The deliverable should be validated against marks that were not used in processing. A statement such as "RMS vertical residual of 28 mm against 12 independent check points" is the difference between a defensible volume and a guess.

For stockpile reconciliation — where a 1% error on a 200,000-tonne pad is 2,000 tonnes of unaccounted product — that verification is not a formality. It is the whole point.

Where LiDAR mining surveys are used

The same workflow underpins a broad set of mining tasks across Australian operations, from Pilbara iron ore through Goldfields operations around Kalgoorlie and Leonora to Bowen Basin and Hunter Valley coal:

• Open-pit and waste-dump surfaces for end-of-month survey, design conformance and movement tracking.
• Stockpile volumetrics on ROM pads and product yards for monthly reconciliation and inventory.
• Rehabilitation and closure monitoring, where vegetation penetration makes LiDAR the standard for proving bare-earth landform against an approved closure design.
• Tailings storage facilities (TSF), capturing freeboard, beach slopes and embankment crests for dam-safety reporting.
• Haul roads and batters, checking gradient, crossfall, width and bench geometry against design for both productivity and geotechnical safety.
• Conveyor corridors, ROM bins and plant areas, often combined with terrestrial scanning for as-built and clash detection during shutdowns.

Regulatory and safety context in Australia

Commercial drone LiDAR is flown under the Civil Aviation Safety Authority's CASA Part 101. The operator must hold a Remote Pilot Licence (RePL) and operate under a Remotely Piloted Aircraft Operator's Certificate (ReOC). Standard conditions cap operations at 120 m above ground level, within visual line of sight and clear of people; flying above those limits, at night or near a controlled aerodrome requires specific CASA approval. Most large operations also impose their own RPAS procedures, ground-spotter requirements and dispatch deconfliction on top of the CASA baseline.

Beyond aviation, the survey itself sits within the operation's site safety system — JSAs, exclusion zones around active digging, and integration with blast and traffic-management plans. The flip side is that LiDAR removes the surveyor from the highwall, the batter and the live haul road entirely, which is one of the strongest arguments for it on a geotechnically active pit.

Cost considerations

A drone LiDAR pit-and-stockpile survey in Australia typically runs AUD $2,500–$10,000+. The spread is driven by a handful of factors:

Against the alternative — walking a total station over a pit and dumps across multiple days of crew time and lost access — LiDAR is usually the cheaper way to get more data with less site exposure.

Frequently asked questions

How accurate is a LiDAR mining survey?

With good GDA2020 ground control and PPK positioning, UAV LiDAR delivers roughly 20–50 mm absolute accuracy on bare earth. Terrestrial scanners such as the Leica RTC360 reach a few millimetres on hard surfaces at close range. The figure that matters is the residual against independent check points — always ask for it.

Is LiDAR better than drone photogrammetry for mining?

It depends on the surface. LiDAR is better where ground is partly vegetated — rehabilitation cells, waste dumps, batters — because it captures the bare earth beneath cover. Photogrammetry is more cost-effective and gives a photoreal orthomosaic on clean pit floors and fresh stockpiles. Many sites use both.

What coordinate system and datum is used?

Australian mining surveys are delivered in GDA2020, projected to the local MGA2020 zone, with heights on AHD — the same system your mine plan, designs and statutory reporting already use, so the data drops straight in.

Do you need a licence to fly LiDAR drones on a mine?

Yes. Commercial operations require a CASA Part 101 RePL-licensed pilot working under an ReOC, plus compliance with the site's own RPAS and safety procedures. Operations near controlled airspace or beyond standard limits need additional CASA approval.

How long does a LiDAR survey take?

Field capture for a typical pit and stockpile yard is often a single day, sometimes a few hours of flying. Processed and verified deliverables — surfaces, volumes and a report — usually follow within a few business days, faster for routine monthly reconciliation on an established control network.

Talk to ISS about your LiDAR mining survey

If you need defensible pit surfaces, stockpile reconciliation, rehabilitation monitoring or tailings reporting captured quickly and safely, Industrial Spatial Solutions plans, flies and processes LiDAR surveys tied to GDA2020/MGA2020 and AHD, with CASA Part 101 compliance and check-shot validation built in. Call 0407 057 015 to scope your site and request a fixed-price quote for your next survey.