Background and Research Approach

Recent years have seen escalating wildfire risk in the historically fire-resilient Cross Timbers region of Oklahoma, largely driven by landscape fragmentation and the encroachment of fire-intolerant tree species such as eastern redcedar. Recognizing limitations in broad-scale fire risk models for this highly parcelled landscape, our international research team set out to develop a novel, high-precision risk assessment tool. This tool leverages advanced digital forestry to enable robust modeling of wildfire risk at fine spatial scales, ultimately supporting conservation efforts and management strategies.

Methods

Our multi-institutional group (Georg-August-University of Göttingen, TU Berlin, Oklahoma State University, and Kent State University) initiated a month-long field campaign within a privately owned, unmanaged forest stand in Stillwater. The methodology combines field data, terrestrial laser scanning, and unmanned aerial vehicle (UAV)-derived imagery for modeling three-dimensional biomass distribution and moisture. Vegetation was systematically mapped and analyzed: we innovated a probabilistic selection protocol to identify typical eastern redcedars representing key height-perimeter classes, using drone imagery to segment crowns and guide field navigation. Selected trees were scanned on the ground level to obtain dense three-dimensional point clouds. Branch and stem sampling protocols were developed to quantify heartwood-to-sapwood ratios, biomass, and moisture content at various locations within each tree.

Achievements and Challenges

A major challenge encountered was the extreme density and limited accessibility of the forest, which made identifying and physically reaching suitable trees a logistical hurdle. We had to manually clear paths, which limited the number of trees accessible within the timeframe. However, these difficulties inspired a methodological innovation: high-resolution UAV imagery was used in real time to select and localize representative individuals, a significant and scalable advance for digital forestry in dense environments. Sensitive equipment occasionally malfunctioned due to environmental conditions, but field adaptations and close collaboration with OSU partners allowed us to resolve these technical issues. Despite time constraints, all core objectives were met, although the sample size was somewhat reduced. A highlight of the mission was an experimental side trip to a recently-burnt area where we acquired unique laser scanner data capturing post-fire forest structure, providing a comparative dimension to our core dataset.

Outputs and Outlook

The project results in an integrated dataset: terrestrial laser scan point clouds, UAV-based orthophotos, and detailed physiological measurements from stem/branch discs. These data will underpin forthcoming method-oriented and application-focused publications on tree-level biomass and moisture distribution, and high-resolution fire modeling. Additionally, the field campaign markedly strengthened our collaborative ties with Oklahoma State University, paving the way for future joint research. Ongoing analysis will further refine these outputs, advancing science-based wildfire resilience management in the Cross Timbers and beyond.

• Innovative Tree Selection with Drone Imagery:Faced with challenging field conditions, the team pioneered a new protocol for selecting representative trees using real-time high-resolution UAV imagery—enabling efficient, unbiased sampling even in dense and hard-to-navigate forests.
• Acquisition of Unique Post-Fire Data: An impromptu research trip to a wildfire-impacted area allowed the team to collect valuable terrestrial laser scans of burnt forest structure, providing essential data to assess changes in fuel distribution and fire resilience.
• Strengthened International Collaboration The intensive fieldwork deepened institutional partnerships, especially with Oklahoma State University, laying a strong foundation for future joint research projects and advancing international cooperation in wildfire resilience science.