Dr. Douglas Morton, NASA Goddard Space Flight Center
Dr. Douglas Morton is an Earth scientist at NASA's Goddard Space Flight Center and an Adjunct Professor at the University of Maryland. He leads an interdisciplinary lab at NASA to conduct large-scale ecological research using data from NASA's satellites and airborne platforms, ecosystem models, and fieldwork. Dr. Morton's work focuses on tropical forests, fires, and food production. He has worked in Brazil for the past 15 years, with an emphasis on agricultural frontiers in the Amazon and Cerrado biomes and dynamics of deforestation, forest degradation, and agricultural management following forest conversion. Dr. Morton contributes to the Global Fire Emissions Database (GFED, www.globalfiredata.org), a collaborative effort to characterize the impact of global burned area and carbon emissions from fire activity on the Earth system. Dr. Morton is also actively engaged in international efforts to Reduce Emissions from Deforestation and forest Degradation (REDD+), and serves as a technical advisor to SilvaCarbon, a US-Government initiative to build capacity in tropical forest countries to monitor and manage their forest resources.
Dr. Morton will speak on the 'Long-term Carbon Consequences of Amazon Forest Degradation'.
Dr. Thomas Fox, Vice President of Research and Sustainability with Rayonier, Inc.
Tom Fox is a silviculturist with expertise in forest fertilization and tree nutrition, silviculture of Appalachian hardwoods, intensive silviculture of southern pine plantation; forest soils, rhizosphere chemistry, environmental sustainability of intensive forest management, and ecophysiology of managed forest stands. He is a former Honorable Garland Gray Professor of Forestry and Professor of Forest Soils and Silviculture at Virginia Tech, the former Co-Director of the Forest Productivity Cooperative, and the former Site Director of the NSF Center for Advanced Forestry Systems.
Dr. Fox will speak on the use of lidar and remote sensing as forest management tools.
Dr. Joanne C. White, Canadian Forest Service
Joanne White is a Research Scientist with the Canadian Forest Service (Pacific Forestry Centre, Victoria, BC). Joanne has worked in the fields of remote sensing and GIS, in a forestry context, for more than 20 years, and has been employed by private, provincial, and federal forest agencies. Specializing in remote sensing applications for forest inventory and monitoring, primarily with optical and lidar data, Joanne has co-authored more than 100 peer-reviewed scientific publications.
Title: Digital aerial photogrammetry versus airborne laser scanning
data: exploring trade-offs for forest management
The capacity to acquire information characterizing three-dimensional (3D) forest vertical structure has transformed forest inventories around the globe. Airborne laser scanning (ALS) data have been the primary source for this 3D information; however, in an effort to reduce costs, there is increasing interest in the use of 3D information generated from high spatial resolution digital aerial photogrammetry (DAP). Several studies have explored the capacity of DAP for estimating a basic suite of forest inventory attributes such as height, basal area, and volume, across a range of forest environments, and have reported similar accuracies for estimates derived from DAP and ALS (using an area-based approach). However, comparing outcomes across studies is confounded by the use of different camera systems, acquisition parameters, and image matching algorithms. Moreover, what is the current state of knowledge concerning large-area applications of DAP data for forest inventory? While ALS data have demonstrated capacity for characterizing forest structure for a broad range of ecosystem goods and services such as habitat and riparian applications, the utility of DAP data for such applications is less well understood. In a recent study, we have compared the relative performance of ALS and DAP data for detecting and mapping canopy gaps in a coastal forest environment in British Columbia, Canada. Results of this gap-detection work will be presented and considered in the broader context of information needs for forest management.
Dr. Mathais Disney, University College London
Mat is a Reader in the Geography Department at University College London, and his current research interests are very much aimed at trying to quantify and understand the structure and function of trees using lidar, particularly terrestrial laser scanning (TLS). Mat arrived there via a slightly circuitous route, with an undergrad in Physics, Masters in Remote Sensing and a PhD in 3D radiative transfer modelling of vegetation. Mat developed his interests to cover a range of areas including remote sensing of the land surface, 3D measurement and modelling methods, the carbon cycle and climate and the role of trees and forests. Mat has built a team over the last few years who have focused on using TLS in new ways, particularly in tropical forests, but also more unusual temperate and urban environments, to measure trees in unprecedented detail. These data are providing fascinating new insights into forest biomass, tree size and and allometry, and fundamental scaling relationships. Mat collaborates with groups across the globe, including NASA and ESA colleagues developing new space-based instruments for observing tropical forests.
Title: Weighing trees with lasers: TLS for biomass from tropical forests to city churchyards
Terrestrial laser scanning (TLS) has been used for some time for forestry applications, but exploitation of the rich 3D TLS point cloud data for ecological applications is relatively new. TLS has the potential to provide new insights into tree structure and function across diverse environments which are difficult to measure in any other way. I will show how we can use TLS to 'weigh' trees with unprecedented accuracy and avoiding incredibly costly (or even impossible) destructive harvesting. I will describe new TLS work across a range of domains, from tropical to urban, which is providing new insights, opportunities and developments for estimating tree and forest structure, particularly above ground biomass (AGB), but also tree size and shape. I will show how TLS can address the limitations of current (near universal) approaches to estimating AGB based on empirical allometric scaling equations (ASEs), particularly the lack of harvest data across tree size and shape distributions, which underpin all large-scale estimates of AGB. I will discuss new model developments that enable quantitative 3D models of tree structure to be generated from TLS data, which can drastically improve estimates of structure, AGB, and allometry as well as provide information on a range of fundamental tree properties which are near-impossible to measure any other way. I will show how TLS work on AGB is being used to help calibrate and validate tropical AGB estimates from forthcoming ESA and NASA space missions. I will describe new developments in TLS measurement and analysis allowing us to capture very accurate 3D point clouds in environments which are often hard to measure, particularly tropical forests, but also urban environments. Recent work has shown that in London for example, small pockets of large trees in squares and cemeteries can have biomass per unit area equivalent to or greater than tropical forests. I will outline some of the remaining challenges for the wider adoption of TLS methods for measuring very detailed tree structure across domains. Lastly, I will touch on some of the fascinating new opportunities that TLS data are opening up - not just scientific, but also capturing the wider public imagination through 3D animation, VR/AR tools and simply by providing beautiful and unusual views of what we think of as familiar objects.
Dr. Günther Bronner, Umweltdata GmbH, Wolfsgraben, Austria
Günther Bronner has been implementing forest remote sensing approaches for practical forest inventory and management planning services for 30 years. In the 1990s, he designed the GIS of Austrian State Forests, then changed to a private forest service company in 2000. Since 2006 his team is steadily improving forest inventory and mapping methods based on ALS data. Currently, he is involved in several research projects focusing on the development of drone-based forest inventory methods.
Title: Forest as Point Cloud: Launching a novel approach of forest representation
For generations, forest scientists and practitioners used to describe a forest by segmenting it into stand polygons adding site and stand attributes. New sensors and remote sensing methods produce a multitude of detailed, accurate and up to date information about forest ecosystems, and most forest site & stand properties change rather smoothly than abruptly along coordinate space. Laser scanning can deliver features and coordinates of single trees, and as a consequence, traditional polygon- or raster-based systems of forest representation come to their limits. During our research for an all-purpose data format, which can describe big forest areas in a high level of detail and enable fast data processing as well as the implementation of user-friendly management tools, we ended up in successfully 'misapplying' the LAS-format for this purpose. Grids with variable mesh size in the range of 1 to 10 meters are supported as well as datasets with single tree properties. LAS 1.4, allowing up to 256 extra-bytes for freely defined attributes, could perfectly meet our requirements. Some of Martin Isenburgs open source LAS-tools components were adapted and LAZ-compression was used to dramatically reduce the data volume. The presentation covers the theoretical background of this approach as well as numerous possibilities of application in forest science and practical daily forest business. The entire workflow of data acquisition and data consolidation is illustrated; automated data update strategies are described. Examples from forest inventory, management planning, growth simulation and new sustainability monitoring approaches illustrate the huge potential of Forest Point Clouds as a perfectly adaptable data format for representing all kinds of forests in high spatial resolution. At the users front end it enables flexible forest management planning and decision support. As open source platform for exchange of data and methods, a Forest Point Cloud Initiative is meant to serve the community joining scientists, service providers and the industry.