1 of 11

Comparing Drone and Traditional Airborne Lidar at Gallon Jug, Belize: Assessing Visibility, Resolution, and Terrain Modeling for Archaeology

Chris Ploetz PhD Student, University of Texas Department of Geography and the Environment

2 of 11

What is Lidar and Why does it type matter in archaeology?

Lidar (Light Detection and Ranging) allows archaeologists to “see through” vegetation by using laser pulses to generate 3D models of the landscape.�*But not all LiDAR is created equal.
Different platforms—drones vs. airplanes—and sensors yield different results.

This project compares two LiDAR datasets over the same Maya site using two sensors:

Drone: DJI Zenmuse L1
Airborne: Optech Titan (collected by NCALM)

3 of 11

Same site. �Different sensors.

Site: Gallon Jug
Zones: Forest vs. Pasture
Sensors:

Optech Titan
DJI Zenmuse L1

Metrics derived:

DSM (Digital Surface Model)
DEM (Digital Elevation Model)
CHM (Canopy Height Model)
DoD (DEM of Difference)
% ground points and point density

Pasture

Forest

4 of 11

Workflow

5 of 11

Total Points:�Drone ~ 428 million points�NCALM ~ 28 Million points

Zone	Sensor	Total Points	Ground Points	% Ground Returns	Area (m²)	Point Density (pts/m²)
Forest	Drone	282,436,728	4,902,202	1.74%	605,124	466.76
Forest	NCALM	18,206,985	1,206,480	6.63%	605,124	30.09
Pasture	Drone	145,777,208	1,452,650	1.00%	437,470	333.20
Pasture	NCALM	9,509,008	575,202	6.05%	437,470	21.73

Return Number

Classification

High resolution ≠ High Ground Visibility

6 of 11

Drone

NCALM

7 of 11

Vegetation Structure

Drone

NCALM

8 of 11

DEM and Hillshade Comparison

Drone

NCALM

9 of 11

DEM Differencing (DoD Raster)

Red Represents a Positive Shift
Blue represents a Negative shift

This figure shows a raster of the elevation difference between the drone and airborne DEMs. The red areas indicate where the drone returned higher elevation values

Blue areas show where the drone picked up lower terrain

What's really interesting is how spatially patterned these differences are. In the upper section, dominated by forest, we see a lot more red and blue variation—evidence of the drone's sensitivity to small vegetation differences.

In contrast, the bottom portion, which is pasture, shows much more uniform agreement between the sensors.

This supports my conclusion: while drone LiDAR captures more data, that richness can introduce noise in dense forest.

But it also provides an opportunity—it reveals fine-scale forest structure that could be highly valuable for ecological studies.

10 of 11

Conclusion

Drone Lidar = higher point density and richer vegetation structure
NCALM Lidar = fewer points overall but better ground classification
Sensor performance depends on target: vegetation structure vs. bare-earth terrain

11 of 11

Thanks to…

Dr. Brett Houk

Dr. Amy Thompson

Mark Willis

Alex Marden

NCALM

Libraries Scholar lab