Photodegradation is considered as a universal contributing factor to litter decomposition and carbon (C) cycling within the Earth’s biomes. Identifying how solar radiation modifies the molecular structure of litter is essential to understand the mechanism controlling its decomposition and reaction to shifts in climatic conditions and land-use. In this study, we performed a spectral-attenuation experiment following litter decomposition in an understory and gap of a temperate deciduous forest. We found that short-wavelength visible light, especially blue light, was the main factor driving variation in litter molecular structure of Fagus crenata Blume, Quercus crispula Blume, Acer carpinifolium Siebold & Zuccarini and Betula platyphylla Sukaczev, explaining respectively 56.5%, 19.4%, 66.3%, and 16.7% of variation in its chemical composition. However, the variation also depended on canopy openness: Only in the forest gap was lignin aromatic C negatively associated with C-oxygen (C–O) bonding in polysaccharides receiving treatments containing blue light of the full spectrum of solar radiation. Regardless of species, the decomposition index of litter that explained changes in mass and lignin loss was driven by the relative content of C–O stretching in polysaccharides and lignin aromatic C. The results suggest that the availability of readily degradable polysaccharides produced by the reduction in lignin aromatic C most plausibly explains the rate of litter photodegradation. Photo-products of photodegradation might augment the C pool destabilized by the input of readily degradable organic compounds (i.e., polysaccharides).

To better understand the effects of ground-level ozone (O₃) on nutrients and stoichiometry in different plant organs, urban tree species Celtis sinensis, Cyclocarya paliurus, Quercus acutissima, and Quercus nuttallii were subjected to a constant exposure to charcoal-filtered air (CF), nonfiltered air (NF), or NF + 40, 60, or 80 nmol O₃ mol^–1 (NF40, NF60, and NF80) starting early in the summer of the growing season. At the end of summer, net CO₂ assimilation rate (A), stomatal conductance (g _s), leaf mass per area (LMA), and/or leaf greenness (SPAD) either were not significantly affected by elevated O₃ or were even higher in some cases during the summer compared with the CF or NF controls. LMA was significantly lower in autumn only after the highest O₃ exposures. Compared to NF, NF40 caused a large increase in g _s across species in late summer and more K and Mn in stems. At the end of the growing season, nutrient status and stoichiometric ratios in different organs were variously altered under O₃ stress; many changes were large and often species-specific. Across O₃ treatments, LMA was primarily associated with C and Mg levels in leaves and Ca levels in leaves and stems. NF40 enriched K, P, Fe, and Mn in stems, relative to NF, and NF60 enhanced Ca in leaves relative to CF and NF40. Moreover, NF resulted in a higher Ca/Mg ratio in leaves of Q. acutissima only, relative to the other O₃ regimes. Interestingly, across species, O₃ stress led to different nutrient modifications in different organs (stems + branches vs leaves). Thus, ambient and/or elevated O₃ exposures can alter the dynamics and distribution of nutrients and disrupt stoichiometry in different organs in a species-specific manner. Changes in stoichiometry reflect an important defense mechanism in plants under O₃, and O₃ pollution adds more risk to ecological stoichiometries in urban areas.

Tree trunk instance segmentation is crucial for under-canopy unmanned aerial vehicles (UAVs) to autonomously extract standing tree stem attributes. Using cameras as sensors makes these UAVs compact and lightweight, facilitating safe and flexible navigation in dense forests. However, their limited onboard computational power makes real-time, image-based tree trunk segmentation challenging, emphasizing the urgent need for lightweight and efficient segmentation models. In this study, we present RT-Trunk, a model specifically designed for real-time tree trunk instance segmentation in complex forest environments. To ensure real-time performance, we selected SparseInst as the base framework. We incorporated ConvNeXt-T as the backbone to enhance feature extraction for tree trunks, thereby improving segmentation accuracy. We further integrate the lightweight convolutional block attention module (CBAM), enabling the model to focus on tree trunk features while suppressing irrelevant information, which leads to additional gains in segmentation accuracy. To enable RT-Trunk to operate effectively under diverse complex forest environments, we constructed a comprehensive dataset for training and testing by combining self-collected data with multiple public datasets covering different locations, seasons, weather conditions, tree species, and levels of forest clutter. Compared with the other tree trunk segmentation methods, the RT-Trunk method achieved an average precision of 91.4% and the fastest inference speed of 32.9 frames per second. Overall, the proposed RT-Trunk provides superior trunk segmentation performance that balances speed and accuracy, making it a promising solution for supporting under-canopy UAVs in the autonomous extraction of standing tree stem attributes. The code for this work is available at https://github.com/NEFU-CVRG/RT-Trunk.