A levee is only as good as the ground surface it was designed against.
That sounds obvious until you look at how most levee work actually gets surveyed. Someone walks a line with a rod and collects shots every fifty or hundred feet, then a surface gets interpolated between them. On open, gently sloping ground that works fine. In a timbered bottom with old sloughs, blown-out ditches, relic field ditches, and twenty years of sediment, the ground between your shots is doing things your shots never captured. You design a levee against a surface that isn't there, and you find out during construction.
LiDAR solves a specific piece of that problem, and it's worth being precise about which piece.
A camera can only see the top of the canopy. Photogrammetry builds a surface from overlapping images, and if the images show treetops, the surface is treetops. In an open field photogrammetry is excellent and often cheaper. Over a levee corridor running through hardwood bottom, it produces a beautiful model of the forest and nothing about the dirt.
A LiDAR sensor fires many pulses per square meter. Most of them hit vegetation. Some of them find gaps in the canopy and return off the actual ground. Those ground returns get classified out and used to build a bare-earth surface underneath the trees.
That's the whole trick, and it's why LiDAR is the right tool for levee corridors, wooded drainage, timbered impoundments, and anywhere the design surface is under cover. It is the same drone LiDAR workflow we fly on every mapping job — applied to a corridor instead of a section.
Existing terrain, at real density. Instead of cross sections every hundred feet, you get a continuous bare-earth model across the entire corridor and its drainage. Old ditches, swales, and low spots show up because they were measured, not interpolated.
Levee alignment and crown elevation set against actual ground. You can run the alignment across the real surface and see where the fill gets deep, where you're borrowing from, and where a small shift in the line saves significant dirt.
Cut and fill volumes before anyone moves dirt. This is where the survey usually pays for itself outright. Earthwork is the dominant cost on most levee jobs, and a volume calculated off a dense bare-earth surface is a number you can bid against and hold a contractor to. A volume estimated off sparse cross sections is a number you find out about later.
As-built verification after construction. Fly it again when the work is done. Compare built surface to design surface. You get actual constructed elevation along the full length of the levee, actual volume moved, and documentation of both — rather than spot-checking a few stations and assuming the rest.
Change detection over time. Levees settle. Fly a corridor annually and you have a measured record of settlement and erosion rather than an impression of it.
If you're building an impoundment levee for ducks — the job our flooded timber and waterfowl page covers in detail — the terrain model does something the levee itself can't tell you: it tells you exactly what acreage floods at each pool elevation.
Set the surface, pick an elevation, and you can see the flooded footprint and the depth distribution across it before you build anything. Which means you can size the levee to the flood you actually want — rather than building it, flooding it, and discovering you got four inches over ground you meant to have eighteen inches on, or that a third of the water went somewhere useless.
That's the same bare-earth model, used a different way. If you're doing both — and most people building a duck levee are — it's one flight.
Whatever your engineer works in. Classified LAS or LAZ point clouds, DWG or DXF contours and breaklines, GeoTIFF rasters, shapefiles, and PDF exhibits are all standard.
Everything is processed in-house rather than subcontracted, so the deliverable gets built to your specification instead of pulled off a fixed template. If your engineer needs a specific contour interval, a particular coordinate system, or breaklines pulled a certain way, that's a conversation, not a change order.
Vertical accuracy depends on flight parameters, ground control, and vegetation density. On open ground with proper ground control, survey-grade vertical accuracy is achievable.
Under heavy canopy it degrades, because fewer pulses reach the ground and the bare-earth surface is built from a thinner sample. Dense evergreen cover is the hardest case. That's a physical limit, not a settings problem.
We'll tell you up front what accuracy your specific site and budget will support, rather than quoting a number that doesn't hold once we're over your timber. If a project needs tighter control than the conditions allow, the answer is more ground control and denser flight lines — and we'd rather price that honestly than deliver a surface your engineer can't rely on.
Note: drone LiDAR produces the terrain data an engineer or licensed surveyor designs from. Where a project requires a boundary survey, a plat, or a sealed document, that's work for a licensed land surveyor. We provide the data; we're clear about where our deliverable ends and theirs begins.
Tell us the corridor, the timber, and what your engineer needs — we will tell you what accuracy the site will support and what it costs.