LiDAR for Metal Detecting: How to Find Untouched Colonial Sites Before You Swing
LiDAR for Metal Detecting: How to Find Untouched Colonial Sites Before You Swing
If you've been detecting for any amount of time, you've probably noticed something at club meetings: certain people consistently pull colonial-era finds while everyone else is digging clad and pulltabs at the local park. They're not luckier than you. They're doing better research.
And increasingly, that research involves LiDAR.
The Detectorist's Edge
LiDAR has become something of an open secret in the metal detecting community. The experienced hunters who use it don't always advertise the fact — and for good reason. When you can see cellar holes, abandoned roads, and farmstead clusters that are invisible on the ground and in aerial photography, you have a significant advantage in finding productive, unhunted sites.
The technology works by firing laser pulses from aircraft and measuring the returns. Pulses that hit tree canopy are filtered out, leaving only ground-level data. The result is a "bare earth" terrain model that reveals every bump, depression, ridge, and linear feature on the landscape — even under dense forest cover.
For detectorists, this means you can identify sites from your computer that would take days or weeks of bushwhacking to find on foot. And more importantly, you can identify sites that nobody else knows about.
What to Look For in LiDAR Data
Once you learn to read LiDAR imagery, certain features become unmistakable:
Cellar holes are the holy grail. They appear as rectangular or square depressions, often 15 to 30 feet across, sometimes with visible stone-lined walls. A cellar hole means a dwelling stood there, and where people lived, they lost things. The yard around a cellar hole — especially the area near where the front door would have been — is prime detecting ground. Look for associated features nearby: a privy (small circular depression offset from the main structure), a well (circular depression, often stone-lined), and outbuilding foundations.
Old roads show up as sunken linear paths connecting clusters of features. These are goldmines in their own right — people traveled these roads daily for decades, dropping coins, buttons, buckles, and personal items along the way. Where an old road intersects another road, or passes close to a dwelling site, is especially productive.
Stone walls indicate former agricultural boundaries. While the walls themselves aren't detecting targets, they tell you where the fields were — and field edges, gates, and corners where walls meet are places where people congregated, rested, and dropped things.
Scatter patterns are subtler. A cluster of small mounds in a cleared area might indicate a dump site. A flat terrace cut into a hillside could be a building platform for a structure that didn't use a stone foundation — and these sites are often completely untouched because there's nothing visible on the surface to attract attention.
Why Raw LiDAR Data Is Hard to Work With
Here's the catch: while LiDAR data is publicly available through the USGS for much of the country, actually using it requires specialized skills. The raw data comes as LAZ files — compressed point clouds containing millions of data points. You need GIS software to open them, process them into usable terrain models, and generate the hillshade and slope visualizations that make features visible.
Most detectorists who use LiDAR rely on state-level viewers where they exist — Connecticut's LiDAR viewer, New Hampshire's Stone Wall Mapper, and similar tools. These are useful but limited: you can browse the terrain but you can't get a detailed, annotated analysis of a specific property. You're looking at the raw imagery and interpreting it yourself, which takes practice and still misses subtle features.
What GroundSight Does Differently
We built GroundSight specifically to bridge this gap. You give us a property address, draw the boundary on a map, and we run the LiDAR data through a multi-model AI analysis pipeline that's been trained to identify the kinds of features that matter to detectorists and property researchers.
The output is a detailed PDF report with annotated maps showing every identified feature — classified by type (foundation, wall, road, depression, earthwork), rated by confidence level, and placed in historical context. We process the terrain data into multiple visualization types (hillshade from different angles, slope analysis, elevation models) because different features show up better under different lighting conditions.
It's not a substitute for doing your own research — you still need to check historical maps, verify land ownership, get permission, and ultimately put boots on the ground. But it gives you a massive head start. Instead of wandering through the woods hoping to stumble on something, you arrive at a site knowing exactly where the cellar holes are, where the old road runs, and where the most promising detecting ground is likely to be.
A Note on Ethics and Permissions
This should go without saying, but: always get landowner permission before detecting on private property. If you discover something that appears to be a Native American artifact or burial site, stop immediately — these are protected under federal and state law (NAGPRA and ARPA). And if you find something historically significant, consider reporting it to your local historical society or State Historic Preservation Office. The detecting community's reputation depends on all of us doing this right.
Stop Hunting Blind
The difference between a productive hunt and a frustrating one usually comes down to research. LiDAR is the most powerful research tool available to modern detectorists — it shows you what's really on the ground before you ever leave your house.