Introduction

Coastal redwoods, the world's largest coniferous tree species, are a central part of Northern California's ecology and economy. These magnificent trees, found solely along the Pacific Coast, provide vital habitats for a variety of species, including the threatened Northern Spotted Owl. Yet, they face challenges from several disturbances, notably bark stripping by black bears. This behavior can compromise tree health and growth, leading to notable economic implications for timber production. This study explores a novel remote sensing technique to early detect and map the damage inflicted by bears, offering a potential solution to mitigate the adverse effects on redwood timber stands.

Methods

The research leveraged high-resolution hyperspectral imagery from Unmanned Aerial Vehicles (UAVs) to capture detailed spectral signatures of redwood trees. These signatures helped distinguish between healthy trees, those recently attacked by bears, and those with old damage. The study utilized advanced machine learning models to analyze the imagery, focusing on identifying specific spectral features indicative of bear damage. This approach aimed to provide a non-invasive, accurate, and efficient method for monitoring redwood health and assessing the spatial patterns of bear bark stripping.

Results

The study achieved some success in distinguishing healthy trees from those with old bear damage. However, it faced challenges in identifying trees recently attacked by bears due to the subtle spectral changes not adequately captured within the study's timeframe. Despite these limitations, the research uncovered potential spectral bands and indices significant for detecting tree health variations, hinting at the possibility of improving early detection methods with further assessment.

Discussion

This investigation highlights the unique resilience of redwood trees to bear bark stripping, contrasting with the uniform damage patterns observed in other species affected by pests like bark beetles. Although the study faced challenges in early detection of recent damage, it emphasized the potential of UAV-based hyperspectral imaging in forest health monitoring. By refining data collection and analysis methods, there's hope for developing more precise tools to combat and mitigate the impacts of bear bark stripping on redwood forests. The findings underscore the importance of continuous innovation and research in preserving these critical ecosystems for future generations.