School of Forest Resources, University of Maine, Orono, ME 04469, USA.
School of Biology and Ecology, University of Maine, Orono, ME 04469, USA.
Sensors (Basel). 2024 Sep 23;24(18):6129. doi: 10.3390/s24186129.
Canopy foliar traits serve as crucial indicators of plant health and productivity, forming a vital link between plant conditions and ecosystem dynamics. In this study, the use of hyperspectral data and foliar traits for white pine needle damage (WPND) detection was investigated for the first time. Eastern White Pine ( L., EWP), a species of ecological and economic significance in the Northeastern USA, faces a growing threat from WPND. We used field-measured leaf traits and hyperspectral remote sensing data using parametric and non-parametric methods for WPND detection in the green stage. Results indicated that the random forest (RF) model based solely on remotely sensed spectral vegetation indices (SVIs) demonstrated the highest accuracy of nearly 87% and Kappa coefficient (K) of 0.68 for disease classification into asymptomatic and symptomatic classes. The combination of field-measured traits and remote sensing data indicated an overall accuracy of 77% with a Kappa coefficient (K) of 0.46. These findings contribute valuable insights and highlight the potential of both field-derived foliar and remote sensing data for WPND detection in EWP. With an exponential rise in forest pests and pathogens in recent years, remote sensing techniques can prove beneficial for the timely and accurate detection of disease and improved forest management practices.
林冠叶特性可作为植物健康和生产力的重要指标，在植物状况和生态系统动态之间形成至关重要的联系。本研究首次探讨了利用高光谱数据和叶特性来检测白松针叶损伤（WPND）。在北美东北部，东方白松（L.，EWP）是一种具有生态和经济意义的物种，正面临着 WPND 的日益威胁。我们使用实地测量的叶片特性和高光谱遥感数据，采用参数和非参数方法，在绿色阶段检测 WPND。结果表明，仅基于遥感光谱植被指数（SVIs）的随机森林（RF）模型的准确性最高，接近 87%，无症状和有症状两类疾病分类的 Kappa 系数（K）为 0.68。实地测量特性和遥感数据的结合表明整体准确性为 77%，Kappa 系数（K）为 0.46。这些发现提供了有价值的见解，并强调了实地叶特性和遥感数据在 EWP 中检测 WPND 的潜力。近年来，森林害虫和病原体呈指数级增长，遥感技术可用于及时准确地检测疾病并改善森林管理实践。
