Forest structure and separation of leafy and woody material using terrestrial full waveform lidar: Results from the 2012 NEON Harvard Forest campaign

Background/Question/Methods

In August 2012, the National Ecological Observatory Network (NEON) undertook an initial campaign with its Airborne Observation Platform (AOP) over Harvard Forest, Petersham, MA. The AOP includes an instrument suite with a visible-to-shortwave infrared imaging spectrometer, a full waveform light detection and ranging (Lidar) and a high-resolution digital camera. During the campaign, field measures over two sites at Harvard Forest (hardwood and hemlock plots) were collected by teams from Boston University, University of Massachusetts Boston, and University of Massachusetts Lowell. These same two plots have been scanned by a full waveform terrestrial Lidar, the Echidna^® Validation Instrument (EVI) at periodic intervals since 2007. During the campaign, the Dual Wavelength Echidna^® Lidar (DWEL), a successor instrument to EVI, was also deployed at the hardwood site. The instrument was scanning in an engineering mode and acquired its first scans with a 1548-nm laser. DWEL is designed to utilize two wavelengths – 1064 nm (as with the heritage EVI) and 1548 nm in order to provide better separation of leaves and trunks and hence better measurement of forest structures from ground at the site-level scale. The ground sampling campaign at the two sites also collected LAI measurements with a TRAC instrument, PAI measurements from hemispherical photos, and updated the tree inventory data from earlier years. The ground-based DWEL data and field measurements will serve as part of the validation dataset for processing NEON’s airborne data at Harvard Forest and will also serve to bridge the information provided by the forest observations at the ground plot-level scales with regional and larger scales captured by other airborne and spaceborne platforms.

Results/Conclusions

The image and point cloud from the early DWEL engineering scans, merged with EVI data acquired in 2010, demonstrate how much darker leaves are and more evident trunks are at the SWIR wavelength of 1548 nm than the NIR wavelength of 1064 nm. The terrestrial DWEL and the airborne NEON Lidar data together improve our understanding and characterization of forests at different scales from both below and above the canopy.