• LOGIN
    Login with username and password
Repository logo

BORIS Portal

Bern Open Repository and Information System

  • Publications
  • Theses
  • Research Data
  • Projects
  • Organizations
  • Researchers
  • More
  • Collections
  • Statistics
  • LOGIN
    Login with username and password
Repository logo
Unibern.ch
  1. Home
  2. Publications
  3. Tracking Floating Wood During a Flood: New Insights From Drone Surveys and Machine Learning

Tracking Floating Wood During a Flood: New Insights From Drone Surveys and Machine Learning

Details
Files
DOI
10.48620/97230
Publisher DOI
10.1029/2024JF008193
Abstract
Instream large wood (LW) plays a vital role in river morphology and ecology, but its transport can pose risks to infrastructure during floods. Monitoring LW transport during flood events remains limited due to technical and logistical limitations. This study employs drone-based video monitoring and machine learning to analyze LW dynamics during an experimental flood in the Spöl River, Swiss Alps. Using three drones covering a 200-m stretch, we created a high-resolution data set of over 560 pieces and 36,000 wood detections (including individual pieces captured in multiple frames). Convolutional neural networks (CNNs) detected and tracked LW, enabling detailed analysis of trajectories, rotation, and velocity, complemented with flow field characteristics (i.e., surface velocity) derived from Large-Scale Particle Image Velocimetry (LSPIV). Results showed that LW transport was concentrated in high-velocity flow paths and influenced by wood piece dimensions. Longer, thinner pieces moved faster, while thicker pieces faced greater resistance. Flow convergence aligned wood pieces with flow direction, reducing rotation, especially for larger pieces. Although wood piece rotation increased with flow velocity, it plateaued at the highest velocities. Large pieces, while fewer, represented 65% of the total transported volume, emphasizing their role in LW dynamics. By leveraging unmanned aerial vehicles (UAVs) and convolutional neural networks (CNNs), this study offers new insights into interactions between flow conditions, wood size, and transport behavior. Our findings contribute to the understanding of LW dynamics in flood conditions and provide valuable information that can enhance flood risk assessment, support early warning systems, and inform sustainable river management strategies.
Date Issued
2025-11-06
Publication Type
Article
Subject(s)
500 Science > 550 Earth sciences & geology
Subjects
large wood (LW)
•
drone monitoring
•
machine learning
•
river dynamics
•
flood risk assessment
•
wood transport analysis
Language(s)
en
Author(s)
Aarnink, J.  
Institute of Geography  
Fornari, A.
Rouge, F.
Ceriotti, G.
Ruiz-Villanueva, V.  
Institute of Geography, Geomorphology  
Institute of Geography  
Additional Credits
Institute of Geography  
Institute of Geography, Geomorphology  
Journal
Journal of Geophysical Research: Earth Surface
Publisher
American Geophysical Union
ISSN
2169-9003
2169-9011
Access(Rights)
open.access
Show full item
BORIS Portal
Bern Open Repository and Information System
Build: 0eaa7c [ 7.08. 11:06]
Explore
  • Projects
  • Funding
  • Publications
  • Research Data
  • Organizations
  • Researchers
  • Audiovisual Material
  • Software & other digital items
  • Events
More
  • About BORIS Portal
  • BORIS Portal & Open Science
  • Send Feedback
  • Cookie settings
  • Service Policy
Follow us on
  • Mastodon
  • YouTube
  • LinkedIn
UniBe logo
Repository logo COAR Notify