Accurate classification of salt marsh vegetation is crucial for coastal wetland monitoring, but fine-grained species discrimination remains difficult, particularly when only limited training data are available for deep learning approaches. To address this challenge, this paper presents a transfer learning-based framework for classifying salt marsh vegetation using UAV multispectral imagery, focusing on a seven-class taxonomy representative of dominant species and water surfaces. Multispectral orthophotos acquired with a MicaSense Dual-Camera system (10 spectral bands) are combined with five vegetation indices to create rich multi-channel inputs. A classification architecture inspired by heterogeneous transfer learning is developed, where a feature-encoding branch compresses the 15-channel input into three channels before processing through a VGG-16 Convolutional Neural Network (CNN), pre-trained on RGB imagery. By leveraging transfer learning from VGG-16, the proposed model achieves high classification accuracy even with limited training data. Performance is compared with traditional machine learning classifiers, namely Support Vector Machines (SVMs) and Random Forest (RF). Results show that the deep learning approach significantly outperforms SVM and RF, achieving an overall accuracy of 98.4% when jointly using spectral bands and vegetation indices. These findings demonstrate the potential of integrating multispectral UAV data and CNN-based classification to support accurate mapping of heterogeneous salt marsh communities for ecological monitoring and coastal management.
Forest structural heterogeneity is widely recognized as a key driver of biodiversity, yet its effects on different components of understory diversity across spatial scales remain insufficiently understood, particularly when assessed using high-resolution remote sensing techniques. In this study, we examined the relationship between forest structure and understory plant diversity along secondary forest succession using high-density uncrewed laser scanning (ULS) data. We derived a suite of LiDAR-based structural metrics describing canopy height, vertical heterogeneity and canopy openness, and evaluated their association with both taxonomic and functional diversity at alpha- and beta-scales, through linear mixed models and distance-based redundancy analysis. Structural metrics showed a limited explanatory power for local (alpha) diversity, with generally weak relationships across taxonomic and functional indices. In contrast, forest structure consistently explained variation in beta diversity, with canopy openness and vertical heterogeneity emerging as the strongest predictors of community dissimilarity. These findings suggest that forest structure plays a greater role in shaping differences in species composition among stands than in determining within-stand diversity. Overall, our findings highlight the potential of ULS-derived structural metrics to capture spatial patterns of understory diversity, while underscoring the importance of accounting for multiple environmental drivers and scale-dependent processes.