Every LiDAR quote quotes a point density. Most of them quote it as a single number without explaining what it's measuring or why. The density that's exactly right for a bare civil site is wildly insufficient for canopy work and wasteful for engineering inspection. Specifying it correctly saves real money.
If you've compared LiDAR quotes side-by-side and seen 50 pt/m² on one and 200 pt/m² on another for what looks like the same project, you've met the point-density question. Both numbers can be correct. Both can be appropriate. Both can also be wrong for the actual deliverable the project needs.
This article is the working version. What density actually measures, the per-deliverable targets, the diminishing-returns curve, and how to specify density properly in a brief so the quotes you receive are genuinely comparable.
Point density is reported as points per square metre — written variously as pt/m², ppm² or PPM. The metric measures how many captured LiDAR returns land in each square metre of the project area on average.
The headline number hides three distinctions that matter:
1. All-returns density vs ground-classified density. A LiDAR pulse can produce multiple returns. "All-returns density" counts every return from every pulse; "ground-classified density" counts only the returns the algorithm has classified as ground. Under canopy, the ratio can be 10:1 or worse — 200 pt/m² all-returns might be only 8-15 pt/m² ground-classified, and the DTM is built from the ground-classified subset.
A quote citing "200 pt/m²" without specifying which is implicitly quoting the larger number. The relevant number for DTM accuracy is usually the smaller one.
2. Density on bare ground vs density through canopy. On bare ground, every pulse becomes a ground return. Through canopy, only a fraction of pulses reach the ground. The achieved ground-classified density beneath canopy is a fraction of the nominal capture density — typically 5-30%, depending on density of cover.
3. Average density vs achieved density everywhere. The headline density is an average across the project. Boundaries, overlap zones and shadow areas may be substantially below average. The QA point-density map (see the QA report article) shows the distribution that the average hides.
Four flight parameters set the achieved density:
The combination of these four determines the achieved capture density. Doubling any one of them (lower altitude, slower speed, higher overlap, higher pulse rate) roughly doubles capture density and roughly doubles airtime.
The honest scoping numbers for typical engineering work:
| Deliverable | Ground-class. target | Notes | | --------------------------------- | -------------------- | ------------------------------ | | Planning-grade DTM (bare/sparse) | 8 — 20 pt/m² | Rural feasibility, ELVIS-comparable | | Engineering-grade DTM, bare | 30 — 80 pt/m² | Standard civil, road, mining | | Engineering-grade DTM, mod. canopy| 80 — 200 pt/m² | Suburban with treelines | | Engineering-grade DTM, dense canopy| 200 — 500 pt/m² | Bushland, riparian, forest | | Contour generation at 0.25 m | 60 — 150 pt/m² | Standard civil drawings | | Contour generation at 0.1 m | 200 — 400 pt/m² | Detailed design | | CAD linework extraction | 100 — 250 pt/m² | Edges, kerbs, formation | | Vegetation / fuel classification | 80 — 150 pt/m² | Canopy + structural metrics | | Powerline catenary modelling | 200 — 600 pt/m² * | On the conductors themselves | | Asset / structural inspection | 500 — 2000+ pt/m² | Member-level dimensioning | | BIM-ready building capture | 1000 — 5000+ pt/m² | Object classification depth |
*Powerline density is on the conductors specifically; the surrounding terrain is typically captured at standard corridor density.
The pattern: bare ground is easy, canopy needs 3-5× the density, contours scale with interval, asset inspection needs an order of magnitude more than terrain work.
Where projects most commonly under-specify is DTM under canopy. Quoting "60 pt/m²" sounds reasonable until the bushland section of the project produces a DTM with metre-scale holes because only 4-8 pt/m² actually reached the ground.
Density and accuracy don't scale linearly. Doubling density from 80 to 160 pt/m² doesn't double the DTM accuracy — it improves it noticeably on canopy and edge cases, marginally on bare ground.
The rough relationship for DTM accuracy as a function of ground-classified density:
The implication: a 300 pt/m² spec on a bare civil site is paying for capture you don't get useful accuracy from. A 60 pt/m² spec on a forested site is paying for capture that doesn't reach the ground at the spec'd density.
Match the spec to where the project actually lives, not to the sensor specification.
Capture cost roughly tracks total points captured — area × density. A 50 ha site at 80 pt/m² is 4 million points per hectare or 200 million points total; the same site at 200 pt/m² is 500 million points. Processing time and storage scale with point count; airtime scales (less directly, but meaningfully) with density.
The practical cost differential for doubling density on a typical engineering capture:
Aggregate cost impact: typically 30 — 60% more for double density. Worth it when the project genuinely needs the extra density (canopy DTM, asset inspection); wasteful when it doesn't (bare civil work).
Three patterns worth correcting:
"Higher density is always better." Higher density helps where the data is sparse (canopy, thin features, very small detail). On bare ground past ~100 pt/m², extra density doesn't improve the engineering output and adds processing cost.
"Sensor spec equals achieved density." Sensor pulse-rate specifications describe what the sensor can do; achieved density depends on flight parameters too. A 2 MHz sensor flown at 200 m AGL at 60 km/h produces wildly different ground-density than the same sensor at 80 m AGL at 30 km/h.
"Density and accuracy are the same metric." Different metrics measuring different things. A high-density cloud with poor control is positionally wrong (densely so). A low-density cloud with rigorous control is positionally accurate but sparsely sampled. Both matter; neither substitutes for the other.
The clean density specification at scoping covers three things:
A clean scoping note reads:
"Capture target: 100 pt/m² ground-classified on bare areas, 250 pt/m² ground-classified beneath moderate canopy. Minimum acceptable: no cell below 40 pt/m² ground-classified. QA density map at 1 m grid expected in delivery."
That sentence eliminates 90% of the post-delivery density disputes — quotes against it are genuinely comparable, and the QA pack has a clear acceptance criterion.
Point density is a scoping lever, not a quality badge. Higher isn't always better; it's better in some places and wasteful in others.
Match the spec to the deliverable and the cover. Bare-ground DTMs at 30 — 80 pt/m² ground-classified; canopy DTMs at 200 — 500; contours scale with interval; asset inspection needs an order of magnitude more.
Specify both the target density AND the surface conditions it applies to AND a minimum acceptable. Quotes against a clear spec are comparable; quotes against a single-number spec are not.
Tell us the deliverable, the cover conditions and the accuracy spec. We'll size the density correctly — high enough where the project needs it, no higher — and quote the airtime that actually delivers it.
How density fits into the four levers that move project price — alongside site size, vegetation and accuracy.
Why under-canopy ground-classified density is so much less than all-returns density, and what the classifier does with what it gets.