GIS 4048 - Module 2 - Forestry and LiDAR

In this module, we explored the use of LiDAR data for forestry analysis, focusing on the Big Meadows area of Shenandoah National Park, Virginia. We were taught how to decompress .las files, create Digital Elevation Models (DEM) and Digital Surface Models (DSM), and calculate tree height, canopy density, and biomass estimations from LiDAR data. 

The first map I created displays tree heights across Big Meadows, with a color ramp transitioning from blue to yellow to show increasing height. 

 

Tree Height Map w/ Histogram

 

This map mostly helps with visualizing the spatial variability of the forest structure. The map also has a histogram that shows the distribution of tree heights, with a mean height of approximately 54.4 feet, a median of 56 feet, and a standard deviation of about 20.5 feet. Most trees in the study area ranged between 30 to 70 feet in height, with very few exceeding 150 feet. This would seem to be the trend with what we would be expected from a mature, mixed species forest common in the Appalachian region. 

The second map I produced represents canopy density, calculated by comparing vegetation returns to total LiDAR returns.

 

Canopy Density Map

Areas with darker green indicate higher canopy density, while lighter greens and whites correspond to clearings, roads, or areas with sparse vegetation. This map would be most valuable to foresters as it provides insights into forest health, stand density, and areas of potential concern for habitat or forest management activities. The density data could be used to guide selective thinning, monitor forest regeneration, or evaluate habitat quality for wildlife species dependent on specific canopy conditions.

The third map is a LiDAR scene combined with a DSM overlay, which allowed for a three-dimensional visualization of the landscape’s surface that includes the tree canopies and man-made structures. 

 

DSM Map

This map shows the area's vertical complexity and reveals features that would be difficult to interpret from a 2D map. 

The single most important aspect I encountered in this lab was the challenge of organizing the numerous datasets generated through each processing step (creating well over 10). Without organization, it would have been easy to lose track of files or repeat steps unnecessarily which I almost did. Processing these large LiDAR datasets caused ArcGIS Pro to lag significantly, especially when running intensive operations like raster calculations or rendering in 3D which has helped me think about saving during my projects much more.