Forest Remote Sensing Days

Disentangling dispersion from mean, reveals true heterogeneity-diversity relationships

Foundational theory on how habitat heterogeneity impacts biodiversity is hindered by mean-biased heterogeneity measures. By deriving a mean-independent measure of heterogeneity, we reveal monotonic positive biodiversity-relationships from formerly inconsistent hump-shaped observations. The mean-independent measure of heterogeneity, which disentangles statistical dispersion from mean for beta and gamma distributed variables, is provided as a solution for monitoring and understanding the environment’s impact on biodiversity.

Prof. Rubén Valbuena, Forest remote sensing, Dept. Forest resource management, Swedish University of Agricultural Sciences

Wednesday 5 November, 9:30-10:10 PO Bäckströms sal (Aula)

Beyond Standard Workflows: Accessing additional water body information from full-waveform LiDAR bathymetry data

The presentation focuses on novel approaches to analyze full-waveform LiDAR bathymetry data. Standard methods often reach their limits, especially with regard to the evaluable water depth and the derived information content. The presentation provides an introduction to novel processing methods that address these challenges and provide additional water body information.

Dr.-Ing. Katja Richter, TUD Dresden University of Technology, Institute of Photogrammetry and Remote Sensing, Germany

Wednesday 5 November, 10:30-11:10, PO Bäckströms sal (Aula)

Forests in 3D: terrestrial point clouds for forestry. Two examples: 3DFin and 3DFoS

An overview of how terrestrial point clouds enable detailed 3D analysis of forests, featuring two open-source tools that operate at plot level: 3DFin for automatic measurement and 3DFoS for semantic segmentation.

Prof. Carlos Cabo Gómez, Oviedo School of Mining, Energy, and Materials Engineering, University of Oviedo, Spain

Wednesday 5 November, 13:00-13:40, PO Bäckströms sal (Aula)

Towards remote sensing-based environmental forest inventories: Experiences from Norway

Environmental information is essential for supporting sustainable forest management. Remotely sensed data offers opportunities to provide such information objectively. The presentation will give an overview of the experiences and ongoing work in Norway related to the development of remote sensing-based environmental forest inventories.

Assoc. Prof. Hans Ole Ørka, Dept. Forestry and Renewable Energy Section, Norwegian University of Life Sciences, Ås, Norway

Wednesday 5 November, 15:00-15:40, PO Bäckströms sal (Aula)

Expanding data availability for tree-level remote sensing-based forest inventories

This doctoral thesis investigates how complementary data sources can expand the availability of tree-level information for forest inventories based on remote sensing. These include close-range laser scanning, harvester production reports, and synthetic point clouds. The four studies demonstrate methods for deriving stem attributes, training airborne laser scanning-based models, automating annotation, and simulating stem defects. Results show that such data can provide ground-truth information of sufficient quality and scale to support modelling.

PhD candidate Raul de Paula Pires, Forest remote sensing, Dept. Forest resource management, Swedish University of Agricultural Sciences

Thursday 6 November, 9:00-10:00, break, and second part of the dissertation, PO Bäckströms sal (Aula)

New insights into the structure of old trees from terrestrial lidar

Old trees are of great interest culturally and ecologically, but we currently know relatively little about how and why great age is achieved. I will discuss some of the interests around old trees and how measuring their structure can help provide new insights into how these trees respond to environment and how we can perhaps understand, protect and manage them better.

Prof. Remote Sensing, Mathias Disney, University College London, England

Friday 7 November, 10:00-10:40, PO Bäckströms sal (Aula)