Harvard Forest >

Harvard Forest Symposium Abstract 2014

• Title: Terrestrial Lidar Scanning at Harvard Forest with the DWEL and CBL Instruments
• Primary Author: Alan Strahler (Boston University)
• Additional Authors: Crystal Schaaf (University of Massachusetts - Boston)
• Abstract:

  Terrestrial lidar instruments are increasingly being used to survey canopy and subcanopy forest structure and the acquired information is being used to characterize forest ecosystems in terms of their role in biogeochemical cycling, surface energy budgets, and biodiversity. These same terrestrial lidar data, which include higher resolution vegetation structure, understory information, and surface features obscured by forest canopy, are also being used to enhance and validate satellite and airborne forest characterizations. The combined Strahler (BU) and Schaaf (UMB) teams have been testing and evaluating these lidar systems at Harvard Forest. In particular, data from both the full-waveform Dual Wavelength Echidna Lidar (DWEL) and the first/last discrete-return Canopy Biomass Lidar (CBL) were obtained during the 2013 growing season at the Harvard Hardwood site. These were accompanied by TRAC LAI measures, GoPro and DHP hemispherical photos, and field forestry measures as well as leaf, bark, and litter spectra. Additional CBL scans were taken at Pam Templar and James Tang’s soil and nivian zone sites; these were also repeated later after leaf drop).
  
  
  The Dual-Wavelength Echidna lidar (DWEL) emits simultaneous laser pulses at 1064 and 1548 nm, providing an ability to separate returns from foliage from those of woody biomass (trunks, branches) based on the strong absorption of radiation at 1548 nm by leaf water content. Building on the design of the Echidna Validation Instrument (EVI) constructed by the Australian Commonwealth Scientific and Industrial Research Organization (CSIRO), the DWEL uses a rotating mirror scan mechanism, coupled with full digitization of return waveforms, to identify and locate scattering events in an approximately 100 m radius hemisphere space around the scanner. The scans have been shown to provide high quality forest structural parameters (diameter at breast height (DBH), stem count density, basal area, canopy height, leaf area index, foliage profile, and above ground biomass) using the data processing approaches pioneered with the Echidna Validation Instrument (EVI). Multiple scans can be merged to generate three dimensional reconstructions of forest structure.
  
  
  The Canopy Biomass lidar (CBL) is an inexpensive, highly portable, fast-scanning (33 seconds), time-of-flight, terrestrial laser scanning (TLS) instrument, originally conceived by the Katholieke Universiteit Leuven (KUL), refined by the Rochester Institute of Technology (RIT), and further refined by University of Massachusetts, Boston (UMB). The instrument is built around a 905 nm SICK laser with a 0.25° azimuth and zenith scanning resolution. Multiple CBL hemispherical point clouds can also be combined to generate coarser resolution reconstructions of the forest structure. Two CBLs have been constructed by UMB to contribute to ongoing forestry studies, and to explore new avenues of study in coastal systems.
  
  
  The efforts at Harvard Forest have focused on using the hyper-portable CBLs in tandem with the more detailed full-waveform DWEL to achieve extremely high-quality, three-dimensional characterization and validation of forestry systems without having to sacrifice spatial coverage, and therefore representativeness, of the sample. Sampling schema for forestry characterization and validation are often plot-based with a grid overlay, in order to capture a particular square area. The full waveform and higher-resolution DWEL can be used to gather hemispherical information at the key positions of grid patterns, usually towards the center, while the hyper-portable CBLs can rapidly capture the edge information, and limit occlusion, aiding in the completion of larger three-dimensional reconstructions, extending estimates of forestry measures, and performing rapid redeployment to capture disturbance or alteration.

• Research Category: Ecological Informatics and Modelling