Towards operational tree-level timber quality and volume estimation using UAV multi-sensor remote sensing
Bibliographic Data
| ID | 22240233 |
|---|---|
| Authors | Paul Eisenschink (0009-0003-3847-6223, Ludwig-Maximilians-Universität München, corresponding author), Thomas Knoke (0000-0003-0535-5946, Technical University of Munich), Wolfgang Obermeier (Ludwig-Maximilians-Universität München), Lukas Lehnert (0000-0002-5229-2282, Ludwig-Maximilians-Universität München) |
| Year | 2026 |
| Volume | 26 |
| Pages | 101375 |
| Publication date | 2026-07-01 |
| Peer Reviewed | Yes |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | Trees Forests and People (JOURNAL) |
| Journal identifiers | ISSN: 2666-7193 • E-ISSN: 2666-7193 |
| Publisher | Elsevier BV (PUBLISHER) |
| DOI | 10.1016/j.tfp.2026.101375 |
| OpenAlex | W7167721561 |
| Language | EN |
| References cited | 54 |
Forests provide numerous ecosystem services, including carbon sequestration, water purification, and serving recreational needs. However, many forest owners regard their forest primarily as a capital asset, intended to provide timber. Adapted forest management is required to ensure sustained provision of ecosystem services even under fastening climate change, which can strongly benefit from high-quality data. To facilitate that, some manual techniques exist, but the collection of data using those techniques often results in imprecise results and time-intensive when entire forest areas are to be investigated. We present a new framework based on unmanned aerial vehicles (UAV) remote sensing, combining LiDAR and multispectral imagery, applied to a Bavarian forest owned by Ludwig-Maximilians-Universität München. This framework combines tree-level species information obtained via a machine learning approach that integrates multispectral data with tree properties, such as stem diameter and tree height from LiDAR data. Additionally, we present novel methods for assessing stem straightness from LiDAR data, that are validated using photography-based reference data: Distance From Line (DFL) and Continuous Segment Assessment (CSA). The DFL method provides a single straightness index per tree stem, allowing the analysis of stand differences in stem straightness, whereas the CSA quantifies the length and number of straight stem segments. This combined framework enables the detection of growth-related differences, wood volume, and stem straightness with good accuracy within our study sites and allows for the identification of trees based on specific wood requirements. Integrating structural and spectral UAV data supports data-driven forest management decisions, such as balancing timber production with ecological sustainability
Data quality · Forest ecology · Forest Inventory · Forest management · Identification (biology) · Lidar · Logging · Multispectral image · Tree (set theory) · Plant Surface Properties and Treatments · Remote Sensing and LiDAR Applications · Remote Sensing in Agriculture
Textural Features for Image Classification
A soil-adjusted vegetation index (Savi)
Forest biodiversity, ecosystem functioning and the provision of ecosystem services
Feature Selection with the Boruta Package
Restoring natural forests is the best way to remove atmospheric carbon
Red and photographic infrared linear combinations for monitoring vegetation
Costs of forest carbon sequestration in the presence of climate change impacts
European private forest owner typologies
Forest Management and Biodiversity Conservation Based on Natural Ecosystem Dynamics in Northern Europe
| Citation velocity | historical |
|---|---|
| Highly cited | No |