What Is LiDAR and How Is It Used in Surveying?

Lidar stands for light detection and ranging.

It is a pulse-and-return measurement method: the instrument emits laser pulses and records the reflections the surfaces send back.

Each return is recorded as one measured point in the cloud.

In surveying, the raw product is a point cloud.

Caltrans' Surveys Manual — one state DOT's specification — describes it as the raw product of a laser scan, in which every point has X, Y, Z and return intensity, plus RGB values when imagery is overlaid.

The same manual is explicit about where the accuracy comes from: each point's error is the accumulated error of the scan control plus the individual measurement.

In survey work, the scan is tied to survey control — known coordinates.

For pavement elevation surveys, Caltrans' mobile scanning spec requires final values to be checked against independently measured validation points, so the positions in the cloud can be verified against something other than the scan itself.

What the pulses reach shapes what the cloud contains.

USGS notes that lidar point clouds record buildings, vegetation and the ground, which is why a single collection can feed several products — a bare-earth elevation model among them.

The instrument itself is a laser scanner.

O*NET's occupation list has no 'LiDAR technician' title — the closest match is Remote Sensing Technicians (19-4099.03), the occupation whose task list appears later in this guide.

What stays constant across titles is the pipeline — capture a scene as a point cloud, tie it to survey control, then classify and model the points into a deliverable.

Each measurement is a pulse and a return.

The scanner emits laser pulses, surfaces reflect part of that light back to the instrument, and the returns are recorded as points — every one carrying X, Y, Z and return intensity.

How good each of those measurements is depends on the surface, the geometry and the range, and the written specifications spell out each factor.

The intensity of the return is the energy that comes back, and surfaces return it differently.

Caltrans' Surveys Manual notes that new black pavement returns less energy than older grey asphalt, which limits the distance the scanner can measure.

The same manual adds that laser measurements made perpendicular to a surface are more accurate than those at a large angle of incidence, because the beam elongates and produces distance errors.

Range is only half the measurement; the other half is where the sensor itself sits.

On a tripod, that position comes from survey control, and separate set-ups are tied together afterwards.

ASPRS describes registration as combining static scans from several tripod set-ups into one point cloud — by aligning common features in multiple scans, or by registering to known coordinates of features visible in the scans.

Moving platforms carry their own positioning, which the next section covers.

Two limits shape what a scan can deliver.

Caltrans notes that laser-scan accuracy decreases as scan range increases, so its mobile scanning datasets exclude point-cloud data beyond the system's useful range.

And the sensor has to hold its adjustment: Caltrans tells scanner operators to calibrate daily per the manufacturer's procedure, and notes that scanners calibrate at start-up, with wind and vibration interfering with that process.

Finally, the measurement chain is checked, not assumed.

Caltrans' mobile scanning spec says GNSS vertical accuracy limits mean mobile lidar alone will not meet its engineering survey standards for pavement elevation surveys — additional control points in the scan area are required, and final values are checked against independently measured validation points.

The sensor idea is the same on every platform.

What changes is what carries it — and that sets the coverage, the speed and the positioning method a crew has to manage.

Terrestrial (tripod) scanning.

The scanner stays static and the crew moves it between set-ups.

ASPRS describes the workflow: static scans from several tripod set-ups are combined into one point cloud, either by aligning common features captured in multiple scans or by registering to the known coordinates of features visible in them.

Caltrans classes the work by accuracy tier.

Type A is the engineering-accuracy tier — its list includes architectural and historical preservation surveys, structures and bridge clearance, deformation monitoring and as-built surveys — while earthwork surveys of stockpiles, borrow pits and landslides fall into lower-accuracy Type B.

SLAM adds a positioning option that needs no GNSS at all.

ASPRS explains simultaneous localization and mapping as a sensor mapping its surroundings in 3D while locating itself in that space — positioning from targets and landmarks instead of GNSS, which is what lets it work indoors, underground and in confined spaces.

ASPRS notes lidar is the main sensor in the SLAM type commonly used for mapping, while camera-based visual SLAM is more familiar from consumer devices such as robotic vacuums.

With enough identifiable targets, ASPRS puts SLAM positional accuracy at a level similar to a survey-grade PPK GNSS solution — generally 5 cm or better for mapping purposes.

The limits are real: SLAM needs feature-rich surroundings, long empty corridors or continuous canopy cause problems, it is less efficient than GNSS over large outdoor areas, and localization errors can accumulate over time.

Mobile mapping (vehicle-mounted).

Caltrans' Surveys Manual defines mobile terrestrial laser scanning (MTLS) as lidar combined with GNSS, a distance measuring instrument (DMI) and an inertial measurement unit (IMU), producing georeferenced point clouds and imagery from a moving vehicle.

Its platform list includes SUVs, pickup trucks, hi-rail vehicles and boats.

ASPRS describes a mobile mapping system more generally, as a vehicle-mounted sensor array and computer whose most common sensors are integrated GNSS receivers and IMUs, lidar and cameras.

Because the platform is moving, positioning is what limits the accuracy — the control-point and validation-point rule in the previous section is Caltrans' answer to that limit for pavement elevation surveys.

Caltrans also directs MTLS crews to laser eye-safety rules: OSHA's 29 CFR 1926.54, the ASTM E2641-09 standard and manufacturers' recommendations, including never viewing laser beams through magnifying optics and considering the eye safety of the traveling public.

Airborne lidar.

Aircraft collect terrain elevation, and public programs write the requirements down.

The USGS Lidar Base Specification — LBS 2025 revision A, released June 10, 2025 — governs collections for the USGS 3D Elevation Program (3DEP), where the minimum acceptable quality level is QL2.

The specification's quality levels trade density against accuracy.

Aggregate nominal pulse density runs from at least 8.0 pulses per square meter for QL0 and QL1, to at least 2.0 for QL2 and at least 0.5 for QL3.

The absolute non-vegetated vertical accuracy targets are ≤0.050 m for QL0, ≤0.100 m for QL1 and QL2, and ≤0.200 m for QL3.

QL0 and QL1 share the same pulse density — QL0's required vertical accuracy is twice as good, and USGS notes its checkpoint surveys may need static, redundant GPS rather than RTK.

Drone (UAS) lidar.

Drones carry remote sensing equipment over sites that are difficult to access or need high-resolution capture — O*NET's 2026 task list for remote sensing technicians includes operating remote sensing equipment on drones for exactly that purpose.

On the positioning layer, ASPRS' UAS addendum describes what properly configured drone GNSS achieves: PPK frequently reaches 1–2 cm horizontal and 2–4 cm vertical accuracy, while properly set-up RTK may reach typical accuracies of 5–10 cm.

The flight side of this work — pilots, flight ops and the photogrammetry-heavy jobs — belongs to the drone surveyor role.

A lidar project ends in the office, and every deliverable starts from the point cloud the sensor produced.

O*NET's technology skills file lists laser-scanning and point-cloud software — Cyclone for data conversion and Terrasolid TerraScan — among the tools of survey and mapping occupations; for remote sensing technicians it also lists ESRI ArcGIS, Autodesk AutoCAD and Bentley MicroStation.

Classified point clouds.

The classes technicians sort a cloud into come from ASPRS' LAS specification.

LAS 1.4, revision R16 (August 2025), defines standard point classes including 2 Ground, 3 Low Vegetation, 4 Medium Vegetation, 5 High Vegetation, 6 Building, 7 Low Point (Noise) and 9 Water.

For point formats 6 through 10, ASPRS adds classes such as 10 Rail, 11 Road Surface, 13–14 Wires, 15 Transmission Tower, 17 Bridge Deck, 18 High Noise, 19 Overhead Structure, 20 Ignored Ground, 21 Snow and 22 Temporal Exclusion.

Classification is what turns a cloud of returns into separable ground, vegetation and structure layers.

Elevation models and surfaces.

From the ground class, the office derives terrain products.

USGS defines a digital elevation model (DEM) as a representation of the bare-earth surface, excluding trees, buildings and other surface objects, and notes a lidar point cloud is the raw material from which DEMs are one of many products.

ASPRS defines a photogrammetric digital surface model (DSM) as a "first visible" product — the features first visible from the air, such as trees and buildings — a different surface than the DEM's bare earth.

Autodesk's Civil 3D help, one maker's documentation, shows the processing step: Civil 3D can build a TIN surface directly from ReCap point cloud files, selecting the areas wanted and filtering out non-ground points, with its surface wizard offering Planar Average, Kriging Interpolation or no-filter options.

Volumes.

Earthwork and stockpile quantities come from differencing surfaces.

In Civil 3D, a volume surface is the difference or composite between two surfaces — this is how earthwork and stockpile volumes are computed from two surveys of the same site.

ASPRS adds a warning the estimating side needs to hear: an accurate point cloud does not guarantee accurate volumes, because point density that is too low for rough terrain produces inaccurate volume estimates regardless of software.

Documented accuracy.

Deliverable accuracy is assessed against a written standard, not asserted.

The current edition of ASPRS' Positional Accuracy Standards for Digital Geospatial Data is Edition 2, Version 2 (2024).

It dropped the old 95% confidence reporting in favor of RMSE.

It also relaxed the old rule that ground control be four times and checkpoints three times more accurate than the final product, and requires checkpoint survey error to be included in the computation.

A compliant product accuracy assessment uses a minimum of 30 checkpoints and a maximum of 120 on large projects.

For lidar, only Non-Vegetated Vertical Accuracy passes or fails — Vegetated Vertical Accuracy is still evaluated and reported.

The crew role that runs this technology is the LiDAR technician — the scanning specialist on a survey team, from tripod set-ups to point-cloud processing.

The duties, job types, training, software and pay live in the lidar technician career guide; this page covers the technology itself.

The work splits between field capture and office processing, and the federal occupation data reflects both halves.

O*NET's task list for remote sensing technicians (19-4099.03) includes calibrating data collection equipment, monitoring raw data quality during collection and making equipment corrections, correcting raw data for errors such as skew or atmospheric variation, and collecting ground verification data with GPS receivers.

Its work-context survey shows the desk lean: 88 percent of remote sensing technicians said they work indoors in an environmentally controlled setting every day, and 66 percent said they never work outdoors exposed to all weather.

On the outlook side, BLS folds lidar integration into cartographers and photogrammetrists: it says these workers integrate land surveys and LiDAR using GIS and CAD.

That occupation is projected to grow 7 percent from 2025 to 2035 — much faster than average, per the BLS Employment Projections.

The specification-writing side shows who the employers are.

Caltrans' Surveys Manual — the source behind several examples in this guide — is one state DOT's scanning and mobile mapping spec, and federal program work runs on written quality levels, like 3DEP's QL2 minimum.

Drone capture adds another employer type, overlapping the drone surveyor role described above.

When you are ready to look, the openings are collected on the lidar technician jobs page.