Bidding good-bye to winter and saying hello to spring!
In the winter term, Tristan Goodbody and Joe Rakofsky successfully defended their theses in January and March 2019 respectively, this concludes Q11 and Q12 of theme 2. Tristan, now a postdoc, will be focusing on Q1 of theme 1. Congratulations to both of you!
Sam Herniman grew up between the small countries of England and Northern California. Sam fled, at the age of 18, to the countryside of Wales to complete an undergraduate degree in ecology at Bangor University. While in university, he took a year out to research the anthropogenic influences on tropical orchids in China. This project used Landsat-8 images to create environmental niche models for two Chinese orchids. After graduating, Sam spent a short time researching dragonflies in Ugandan caldera lakes before starting a forestry internship in eastern Oregon for the US Bureau of Land Management and the Chicago Botanic Garden. This work in the forest industry led him to seek out education in forestry and remote sensing – a sector the AWARE project easily satisfies.
In January 2018, Sam moved to Vancouver, BC to become a master’s student to work on AWARE question 18 project in the Integrated Remote Sensing Studio under Dr. Nicholas Coops. Question 18 involves creating habitat suitability models for birds in Newfoundland. It has already been established that forest structure information is essential for accurate models of forest birds. Subsequently, Sam is exploring ways to interpolate structural information across Newfoundland, where wall-to-wall airborne laser scanning data are not available.
In his free time, Sam enjoys road trips, camping and playing fetch with his robotic vacuum.
Q7: Using 3-dimensional point clouds to improve characterizations of tree stems across scales in boreal mixedwood forest stands
Chris Mulverhill is a PhD student working under the supervision of Nicholas Coops (UBC). His research project (Question 7 of theme 2) investigates the use of 3D remote sensing technologies to estimate stem volume and diameter in a boreal mixedwood forest near Slave Lake, Alberta.
Part of Chris’s work uses ground-based images to construct 3D point clouds, called Digital Terrestrial Photogrammetry, or DTP. DTP and associated techniques have been the focus of recent research, in response to enhanced data needs of higher resolutions at lower costs. These needs (and areas of study for DTP) are, for example, estimates of tree diameter at breast height (DBH), volume, and taper. However, previous studies using DTP have relied on either the acquisition of up to a few hundred images of individual trees, thereby raising issues of time or data storage requirements in operational capacities. Additionally, most studies have focused on relatively homogeneous stands or a single primary species, and, as a result, an additional analysis of point cloud accuracy across species and size gradients is needed to understand the utility of DTP in irregular stands such as those present in boreal mixedwood forests.
Chris’s recent work has evaluated the use of DTP based on a small set of photos (12 per tree) for estimating tree volume and diameter on a set of 15 trees from 6 different species. Images were acquired using two spherical cameras mounted on a telescoping pole, taking images at each of three heights, approximately 2, 3, and 5 m above the ground, and at each of two locations around the tree (Figure 1). The mount ensured the two cameras remained at a fixed distance apart, which allowed this distance to be input as a scale bar during point cloud processing. Point clouds were processed using an automated workflow in Agisoft Photoscan, and the resulting point clouds were analyzed in Computree, a collaborative and open-source software to derive detailed tree-level estimates from ground-based point clouds. Results of DBH estimation, shown in Figure 2, indicate the general success of this technique in DBH estimation for the sample trees. Through this work, Chris and his collaborators hope to demonstrate the possibility of using DTP to characterize the trees present in complex and changing habitats such as boreal mixedwood forests.
Figure 1: The setup of image acquisition and coded target placement (left), and the use of the cameras and telescoping pole during July 2018 fieldwork (right).
Figure 2: Estimation accuracy of the DBH for sample trees (n = 15).
AWARE projects in general are going very well on all sites. In Alberta, work is continuing on the use of terrestrial photogrammetry for stem characterisation and growth. Over in Ontario, work has picked-up with the acquisition of the single photon LiDAR over the Romeo Mallette Forest and analysis of species at the Petawawa research forest. With the use of new plot data collected from Black Brook, testing and improving the species identification software continues. UBC is working on processing the SPL data and deriving an initial stratification for a new plot network and developing an EFI using the data.
There is great interest in using LiDAR to improve mapping in the Newfoundland and Labrador region as well as in the three growth and yield questions in Q20.
The level of involvement of our industrial partners has increased and expected to continue to improve. This gives us a high chance of clawing back the cash that NSERC withheld in 2018.
AWARE e-lecture series to bring the project results of AWARE to the larger public are coming in the fall of 2019. The online courses will be hosted at the Canadian Institute of Forestry (CIF-IFC)’s online school. Six to eight classes are being planned to run between September to October of 2019. A rolling AWARE lecture series is also in the plan.