Vertical Structure Complexity Assessment of Tropical Forests from a Portable LiDAR System
Amanda
N
Cooper, University of Central Florida, acooper@pegasus.cc.ucf.edu
(Presenting)
John
F
Weishampel, University of Central Florida, jweisham@mail.ucf.edu
Jason
Drake, US Forest Service, jasondrake@fs.fed.us
David
Clark, La Selva Biological Station, dbclark@sloth.ots.ac.cr
Geoffrey
Parker, Smithsonian Environmental Research Center, parkerg@si.edu
Vertical structure is an important physical attribute of a forest, influencing the microclimate, biogeochemical cycling, and biodiversity. Tropical forests have a highly complex structure that is altered by both natural and anthropogenic disturbances. Such disturbances could permanently affect the abundance of diversity these forest support. Current field methods for quantifying vertical structure include field-based forest survey methods which utilize indicator values such as stem density and dbh (diameter at breast height) and labor intensive optical point quadrate methods that maybe inconsistently interpreted. LiDAR (Light Detection And Ranging) remote sensing provides a method for surveying forest structure that is repeatable and less exhaustive for researchers. Traditionally, LiDAR is collected via satellite (e.g. GLAS) or airborne (e.g. LVIS, EAARL) platforms. Satellite-based LiDAR is still lacking at moderate resolutions and airborne LiDAR has only been collected in a few broad-scale studies because of the costs of data collection. This research focuses on the use of a portable LiDAR system for tropical forest survey. Our system, SYCLPS (Structure Yielding Canopy LiDAR Portable System), utilizes a first return, upward facing LiDAR (Reigl LD90-3100VHS-FLP) to provide distributional information of the canopy components. Surveys at the La Selva Biological Station, Costa Rica in July 2005 demonstrate that SYCLPS is a useful tool for defining canopy vertical structure. SYCLPS data were able to highlight differences in canopy organization between primary and secondary forests at La Selva. Further work will use SYCLPS to develop pseudo-waveforms to mimic large-footprint sensors and extend the use of SYCLPS into forest management applications.