Summary

The purpose of the trip was to attend the International Workshop on Three-dimensional analysis of Forest Structure and Terrain using LiDAR technology, held at the Pacific Forestry Centre, Victoria, BC, Canada.  This trip was funded by the EOC travel committee and the FWPRDC grant for the Canopy Lidar Initiative.  I presented a poster paper titled ‘LIDAR methods for canopy structure measurement in Australian Forests’ which reported work undertaken in the Bago-Maragle forest to compare results of conventional forest measurement with airborne and ground-based laser systems and hemispherical photography.  A full report of this work will be prepared for the workshop proceedings that will be published in the Canadian Journal of Remote Sensing later this year

The workshop was attended by leading researchers using LIDAR data and developing systems to study vegetation as well as forestry researchers who use commercial systems.  Commercial operators and instrument manufacturers were also represented. 

Keynote speakers reported on current use of both research-based and commercial lidar instruments to measure vegetation.

Wayne Szameitat (Optech) reported on currently available systems and future directions.  One major new development is an airborne instrument that will digitize and record the full waveform return for a subset of laser shots.  This is a big step forward for the more complex canopy profile measurements, but is limited by the small footprint of the laser shot.  He also reported on ILRIS-3D, a ground-based scanning system.  This generated some interest and discussion from those interested in reconstructing scenes for ‘fly-through’ visualization.  This system is similar in capability to the Cyrax and I-Site systems.  None of these scans a full hemisphere and they are designed mainly for scanning solid objects (eg buildings).

There was some discussion of the use of the recorded intensity of lidar returns that is available from most airborne systems.  Martin Flood (Airborne-1) reported that most of his clients requested intensity in their delivered data, but none knew what to do with it.  Some researchers reported unsuccessful attempts to correlate lidar intensity with other coincident data (eg hyperspectral imagery).  This raised issues of BRDF as the lidar data is by nature a hot-spot (shadow-free) reflection.  There was a request to manufacturers that work be done on calibrating the intensity of returns.

David Harding and Michael Lefsky reported on the use of large footprint digitizing lidar systems (SLICER and LVIS) for the extraction of canopy profile information.  While some SLICER data is freely available, these systems are not available for commercial use, so were of limited interest to many of those present.  Harding raised some important issues associated with discrete-return lidar systems that can lead to biased results.  In particular, leading edge detection will always over-estimate the height of the ground.  He also highlighted that dead-time in multiple return systems means that when there is low vegetation, the ground return may be missed even if the laser penetrates to the ground.

The majority of presentations showed the use of commercial airborne systems to map the number and height of trees.  Most were interested in individual tree identification but some reported difficulty in identifying the lidar-measured trees when visiting the field site.  Tree numbers, crown size and height were reported with varying degrees of accuracy.  Zhenkui Ma (Weyerhaeuser) reported operational use of LiDAR to map tree height and crown size and research underway to calculate volume from these results.  A common theme among foresters was that the LiDAR data is most valuable for providing an accurate terrain model to enable them to plan routes for roads and to assess risk of landslide.

Outcomes

Attending this workshop was a valuable experience for me to see where the use of LiDAR for practical forest measurement is up to.  It was also important to present our work and raise the profile of the CSIRO Canopy Lidar Initiative in this forum.  I was able to talk with several of the key participants and there was some interest in the poster. 

• The group from Virginia Tech, lead by Randolph Wynne, were extremely interested in ECHIDNAÔ as they have recently purchased a Cyrax system and plan to use it in association with airborne data.  They requested a copy of the poster and we will keep in touch to see how their system performs. 
• Neil Eggleston of EarthData Technologies also requested a copy of the poster.  He did some work on a ground-based system in the early 1990’s but found that it was‘ahead of its time’.
• The work reported in the poster will be written up as a paper for the workshop proceedings to be published by the Canadian Journal of Remote Sensing.