| Three-dimensional reconstructions of forest stands, assembled from multiple scans of the ground-based full-waveform lidar, EchidnaRTM Validation Instrument (EVI), provide a new pathway to estimate biophysical parameters of forest structure, as well as a novel ecological application visualizing bat flight tracks through a reconstructed forest. I reconstruct six 50 m by 50 m forest stands of varying canopy density and species from the Sierra Nevada National Forest, California, and the Harvard Forest, Petersham, Massachusetts, using lidar data acquired in 2008 and 2009. I use commercial software tools to display, manipulate, and produce "fly-through" visualizations of our forest stands. The procedure processes each returned lidar pulse to identify one or multiple "hits"; converts associated peak power to apparent reflectance; classifies hits as ground hits or non-ground hits of either trunk/branch or foliage using apparent reflectance; locates hits in Cartesian coordinates; stores hits as points in a point cloud; and registers five or more overlapping scans into a single point cloud reconstruction. Applying visualization tools, I estimate forest structural parameters that include tree diameter at breast height (DBH), tree height, crown diameter, crown height, and foliage area volume density (FAVD), that agree with field measurements and by airborne lidar data. I generate a fine-resolution digital terrain model (DTM) and canopy height model (CHM) at each site, to measure individual DBH and tree height more accurately. In an ecological application, I reconstruct three-dimensional bat flight trajectories using imaging data from thermal infrared cameras at a bat roosting and maternity site in Petersham, and co-register them to a three-dimensional forest reconstruction built from nine scans at the site. Patterns of flight trajectories during first 5 seconds of emergence from the roosting barn show how the bats chose different flight routes to forage along the edge of a forest and into the understory. |