It was better than Christmas. I was handed the much anticipated Kenai Peninsula 2008 LiDAR data. For a GIS (Geographic Information System) analyst, it is like ice cream to a baby, and we were given 500 gigabytes of the stuff. I've distributed portions of the data to geologists and engineers, only to have them show up a few days later wide eyed and wanting more.
So what's the big deal? Light Detection And Ranging (LiDAR) measures the distance to an object using light. By knowing the speed of light, the range of the target is determined by the round trip time of the laser. Much like SoNAR (Sound Navigation And Ranging) uses sound to expose the details of the sea floor, LiDAR uses a laser to measure the elevation across a landscape. Sounds simple, but imagine you are trying to take a measurement every three feet, in an aircraft 6,000 feet above the ground, and traveling over 200 mph. These are some of the many challenges Aerometric of Anchorage faced to meet the contract requirements.
The area surveyed included the Western Kenai Peninsula and the upper Kenai River Watershed. The LiDAR data are used to produce a Digital Elevation Model (DEM) and a Digital Surface Model (DSM). A DEM represents a bare (bald) earth model of the land, as if all the trees and structures were removed. From the DEM, a variety of products can be produced relatively easily such as elevation contours (down to 2 feet), flood plain and tidal surge maps, and wildfire risk assessments. A DSM on the other hand includes all the trees and structures and is useful for aerial navigation, forestry, viewsheds, radio propagation, etc. These products require some heavy computing power to produce, but are relatively straight forward. The raw LiDAR data itself holds a wealth of information; the hard part is filtering out the parts you want.
The first LiDAR applications were in the 1930s using search lights. Lasers have been used since the 1960s. The technology really started to advance in the 1990s with solid state lasers, well developed global positioning system, and advancements in mobile computing capabilities. The first commercial airborne LiDAR systems came out in the 1990s. Unlike terrestrial systems, airborne LiDAR is not only moving but the aircraft is encountering roll, pitch and yaw. To account for the constant change in position and orientation the airborne LiDAR incorporates an inertial navigation system to measure all these factors and correct the data.
In practice, the aircraft flies a series of flight lines over the survey area. The LiDAR unit emits tens of thousands of laser pulses a second in back and forth sweeps. Each individual laser pulse starts as a narrow beam. However, by the time it travels 6000 feet it is about 3 feet wide. As the laser pulse contacts the ground, it is reflected by the terrain and received back at the aircraft as a complex waveform. The waveform is representative of what the laser hits on its way to the ground. For example, if the laser shines on a tree, the wave will have peaks from the tree top, branches, and finally the ground. These peaks are often separated by software into four "returns" (first, second, third, and last). Each of these "returns" has a position (latitude and longitude), and an elevation. Together they are referred to as a point cloud. In general the first return represents the surface with all the trees and structures and is used to create the Digital Surface Model. The last return is used to generate the bare earth or Digital Elevation Model.
With a laser emitting about 33 to 70 thousand cycles a second and each cycle producing up to four returns, for days on end -- that's a lot of data. That is why it took years to produce the final product. To produce a bare earth model, structures must be removed. Much of the processing is automated by custom programming, but a lot is done manually.
Before the LiDAR was delivered, it had to undergo a rigorous quality control process. This was done by the University of Alaska Fairbanks Geophysical Institute.
So how did it all turn out? For the western Peninsula the contract called for a vertical accuracy less than one foot and horizontal accuracy of 3.2 feet. The accuracy assessment calculated the accuracy to be even better. This is an extremely high quality and rich dataset that will provide many benefits to the Kenai Peninsula. A large consortium of Federal, State, and local public and private stakeholders were responsible for seeing this project through (including the EPA, USGS, Kenai Borough, and Kenai Watershed Forum).
The data is publically available through Alaska Mapped and the Statewide Digital Mapping Initiative (http://www.alaskamapped.org/).
Mark Laker is an Ecologist with Kenai National Wildlife Refuge.
Previous Refuge Notebook articles can be viewed on the refuge website, http://kenai.fws.gov/. You can check on new bird arrivals or report your bird sighting on the Kenai National Wildlife Refuge Birding Hotline at 907-262-2300.