GIS 5100 Module 2 - LiDAR and Forestry

Module 2 of Applications in GIS was an extremely different GIS application than the crime analysis study completed in Module 1; Module 2 focused on forestry analysis on a LiDAR dataset. LiDAR [Light Detection and Ranging] is a remote sensing technology where thousands, if not millions, of laser pulses are emitted from a sensor, bounce off the surfaces below, and the amount of time it takes for the pulse to return back to the sensor is measured to determine the distance between the landscape and the sensor. Once the pulse returns, the X, Y, and Z data is collected, and the entire dataset is compiled into a LiDAR point cloud data file. This compiled dataset can display all features that are present on the landscape. The LiDAR point cloud raster shown below is from a swath of forestry in the state of Virginia. As illustrated in the legend, areas of higher elevation are displayed by warm colors of the spectrum and transition to the cool colors of the spectrum as elevation decreases. 

After the LiDAR point cloud dataset has been imported and processed accordingly, the continuous raster dataset can be converted into a two-dimensional multi-point dataset. Once this has been accomplished, geoprocessing tools can be ran to transform this vector dataset into Digital Surface and Digital Elevation Models [DSM and DEM, respectively]. A Digital Surface Model is an elevation model that captures both the environment's natural and artificial features. Conversely, a Digital Elevation Model is a two-dimensional raster output representation of a continuous surface. As LiDAR sensors can have many returns from the same laser pulse, the processing software can distinguish between returns that are representing characteristics protruding from the landscape from those that are actually hitting the terrain. Once this entire dataset has been processed, a DEM can be derived, creating an accurate representation of the terrain that lies below any protrusions from the Earth's surface; this Digital Elevation Model is also known as a Bare Earth Model. The DEM of the Virginian forest swath can be seen below. The same color scheme used for the LiDAR point cloud dataset was applied to the DEM; this was an intentional choice to show the topographical similarities between the two datasets.

After the Digital Elevation Models and Digital Surface Models have been generated, further geoprocessing functions can be employed to conduct further analysis on the study area. Geospatial analysis examples could include vegetation height [map below, left] and canopy density within the study area [map below, right]. The information gathered by these data outputs can easily be displayed in a graphical form, such as the Distribution of Vegetation Heights histogram, displayed in the lower left-hand corner of the map below. 

Some noteworthy characteristics of this Virginian Forestry Analysis, is the majority of vegetation heights present [within study area], are within the range of 40 to 80 feet, with the statistical mean at approximately 54 feet. Also, a direct correlation between vegetation height and canopy density can be discerned upon initial observations of the map.

This module was very informative on how continuous, raster-based datasets can be used to gather further information on land classification and other characteristics of the natural environment.