Search

Accurate individual tree crown delineation (ITCD), which can be used to estimate various forest parameters such as biomass, stem density, and carbon storage, stands as an essential component of precision forestry. Currently, raster data such as the canopy height model derived from airborne light detection and ranging (LiDAR) data have been widely used in large-scale ITCD. However, the accuracy of current existing algorithms is limited due to the influence of understory vegetation and variations in tree crown geometry (e.g., the delineated crown boundaries consistently extend beyond their actual boundaries). In this study, we achieved more accurate crown delineation results based on an expansion process. First, the initial crown boundaries were extracted through watershed segmentation. Then, a "from the inside out" expansion process was guided by a novel gradient feature to obtain accurate crown delineation results across different forest conditions. Results show that our method produced much better performance (~75% matched on average) than other commonly used methods across all test forest plots. The erroneous situation of "match but over-grow" is significantly reduced, regardless of forest conditions. Compared to other methods, our method demonstrates a notable increase in the precisely matched rate across different plot types, with an average increase of 25% in broadleaf plots, 18% in coniferous plots, 23% in mixed plots, 15% in high-density plots, and 32% in medium-density plots, without increasing over- and under- segmentation errors. Our method demonstrates potential applicability across various forest conditions, facilitating future large-scale ITCD tasks and precision forestry applications. [ABSTRACT FROM AUTHOR]

Natural secondary forests play a crucial role in global ecological security, climate change mitigation, and biodiversity conservation. However, accurately delineating individual tree crowns and identifying tree species in dense natural secondary forests remains a challenge. This study combines deep learning with traditional image segmentation methods to improve individual tree crown detection and species classification. The approach utilizes hyperspectral, unmanned aerial vehicle laser scanning data, and ground survey data from Maoershan Forest Farm in Heilongjiang Province, China. The study consists of two main processes: (1) combining semantic segmentation algorithms (U-Net and Deeplab V3 Plus) with watershed transform (WTS) for tree crown detection (U-WTS and D-WTS algorithms); (2) resampling the original images to different pixel densities (16 × 16, 32 × 32, and 64 × 64 pixels) and inputting them into five 3D-CNN models (ResNet10, ResNet18, ResNet34, ResNet50, VGG16). For tree species classification, the MSFB combined with the CNN models were used. The results show that the U-WTS algorithm achieved a recall of 0.809, precision of 0.885, and an F-score of 0.845. ResNet18 with a pixel density of 64 × 64 pixels achieved the highest overall accuracy (OA) of 0.916, an improvement of 0.049 over the original images. After incorporating MSFB, the OA improved by approximately 0.04 across all models, with only a 6% increase in model parameters. Notably, the floating-point operations (FLOPs) of ResNet18 + MSFB were only one-eighth of those of ResNet18 with 64 × 64 pixels, while achieving similar accuracy (OA: 0.912 vs. 0.916). This framework offers a scalable solution for large-scale tree species distribution mapping and forest resource inventories. [ABSTRACT FROM AUTHOR]

The circumpolar Taiga–Tundra Ecotone significantly influences the feedback mechanism of global climate change. Achieving large-scale individual tree crown (ITC) extraction in the transition zone is crucial for estimating vegetation biomass in the transition zone and studying plants' response to climate change. This study employed aerial images and airborne LiDAR data covering several typical transitional zone regions in northern Finland to explore the ITC delineation method based on deep learning. First, this study developed an improved multi-scale ITC delineation method to enable the semi-automatic assembly of the ITC sample collection. This approach led to the creation of an individual tree dataset containing over 20,000 trees in the transitional zone. Then, this study explored the ITC delineation method using the Mask R-CNN model. The accuracies of the Mask R-CNN model were compared with two traditional ITC delineation methods: the improved multi-scale ITC delineation method and the local maxima clustering method based on point cloud distribution. For trees with a height greater than 1.3 m, the Mask R-CNN model achieved an overall recall rate ( A r ) of 96.60%. Compared to the two conventional ITC delineation methods, the A r of Mask R-CNN showed an increase of 1.99 and 5.52 points in percentage, respectively, indicating that the Mask R-CNN model can significantly improve the accuracy of ITC delineation. These results highlight the potential of Mask R-CNN in extracting low trees with relatively small crowns in transitional zones using high-resolution aerial imagery and low-density airborne point cloud data for the first time. [ABSTRACT FROM AUTHOR]

Neuro-symbolic models combine deep learning and symbolic reasoning to produce better-performing hybrids. Not only do neuro-symbolic models perform better, but they also deal better with data scarcity, enable the direct incorporation of high-level domain knowledge, and are more explainable. However, these benefits come at the cost of increased complexity, which may deter the uninitiated from using these models. In this work, we present a framework to simplify the creation of neuro-symbolic models for tree crown delineation and tree species classification via the use of object-oriented programming and hyperparameter tuning algorithms. We show that models created using our framework outperform their non-neuro-symbolic counterparts by as much as two F1 points for crown delineation and three F1 points for species classification. Furthermore, our use of hyperparameter tuning algorithms allows users to experiment with multiple formulations of domain knowledge without the burden of manual tuning. [ABSTRACT FROM AUTHOR]

Single-tree segmentation can greatly contribute to the extraction of forest structure parameters. The LiDAR point cloud from unmanned aerial vehicle (UAV) was a commonly used data source for the single tree segmentation. However, the point cloud data was large, and the processing was complex. This study aimed to accurately segment the single-tree crown from the point cloud data using deep learning. The main purpose was to improve the processing of point cloud data and the accuracy of single tree segmentation in complex stands. Point cloud rasterization was combined with deep learning to segment the canopy of a single tree. Firstly, the D2000 UAV (Feima Robotics company) platform equipped with the LiDAR sensor (DLiDAR2000) was used to acquire the point cloud data of a mixed coniferous and broadleaf forest. Lidar360 software was then employed for point cloud denoising, ground point classification, and point cloud normalization preprocessing. Subsequently, the top-down rasterization of the sample plot cloud was performed to calculate the maximum height, maximum intensity, and density information per unit rasterized area of the point cloud. The RGB channels were then mapped corresponding to the rasterized image pixels, in order to make the tree canopy clearer in the rasterized image. Secondly, according to the Mask RCNN model within the Detectron2 framework, the number of layers and iterations of different backbone networks were compared to select a backbone network that provided superior segmentation performance. Thirdly, the Global Context Network (GC) and Attention Mechanism modules were integrated into the ResNet network. A comparison was made with the simultaneous introduction of the GC Net and Attention Mechanism modules to enhance the segmentation accuracy of the Mask R-CNN model. To validate the practicality of the improved Mask R-CNN model, its segmentation accuracy was compared with that of similar deep learning networks (U-Net and DeepLabv3+). Finally, the tree crown masks that were segmented by the improved Mask R-CNN were used to segment the point cloud of individual tree crowns. The segmentation of the test plot was compared and evaluated using the watershed, K-means, and the improved Mask R-CNN. Among the three backbone networks of Mask R-CNN, the R50-FPN-3X network saved some training time and computational resources, compared with the R101- FPN-3X network. An average accuracy of 76.72% was achieved to increase by 1.01 percentage points higher than that of the R50-FPN-1X network. In the R50-FPN-3X backbone model, after introducing the Squeeze-and-Excitation (SE), Coordinate Attention (CA), and Convolutional Block Attention Module (CBAM) mechanisms, the average accuracy increased by 1.41, 2.14, and 4.65 percentage points compared to the original model, respectively. The GC Net module was integrated into the ResNet network, and the model accuracy was 80.35%, which was an improvement of 3.63 percentage points over the original model. While both CBAM and GC Net attention mechanisms achieved the highest accuracy of 82.91%, thus increasing by 1.54 percentage points and 2.56 percentage points, compared with the two modules alone. The improved Mask R-CNN model achieved an average accuracy that was 7.27 percentage points higher than the U-Net model and 4.62 percentage points higher than the DeepLabv3+ model. The improved Mask R-CNN outperformed the Watershed and K-means algorithms in single-tree crown point cloud segmentation, indicating the highest recall, precision, and F-score of 81.19%, 78.85%, and 80.00%, respectively. The point cloud segmentation with the improved Mask R-CNN network demonstrated the robustness of the tree crown segmentation in mixed coniferous and broadleaf forests. Point cloud data processing was also integrated with deep learning models. The accuracy of single-tree crown segmentation was enhanced significantly, thereby providing reliable foundational data and technical references to assess the forest resource, biomass, and carbon stock. [ABSTRACT FROM AUTHOR]

Tree detection and canopy area measurement are important and difficult tasks in forest inventory, which are important for understanding forest stand structure. This study utilized remotely piloted aircraft (RPA) aerial photography technology to collect remote sensing images of forests in Xiong County, China, creating a dataset comprising 1200 images of six tree species. Based on this dataset, the paper proposes an optimized model, Att-Mask R-CNN, for canopy detection and segmentation. Att-Mask R-CNN outperforms the original models (Mask R-CNN and MS R-CNN) by achieving 65.29% mean average precision for detection, 80.44% mean intersection over union for segmentation, and 90.67% overall recognition rate for the six tree species. In addition, a pixel statistics method based on segmentation masks is introduced for estimating the vertical projected area of individual tree crowns, and comparisons between the measured and predicted vertical projected area of the crowns of six tree species (100 trees of each class) show an overall goodness-of-fit R2 of 85% and a relative root-mean-square error rRMSE of 12.81%. By using remote sensing images from RPAs and optimizing existing deep learning models, the detection and segmentation of individual tree canopies can be achieved, resulting in a more accurate understanding of forest structure, which provides scientific support for forest management and resource monitoring. [ABSTRACT FROM AUTHOR]

The effect of the parent tree on seedling recruitment has been studied in various research studies. The Janzen–Connell (JC) hypothesis states that the closer the seedlings are to the source tree, the greater the risk of mortality and/or impact from pathogens and herbivores. Despite the extensive existing literature, there are not many studies that evaluate the influence of crown area, as well as the effects on leaf asymmetry, an important measure of biotic and abiotic stress. (1) This study evaluates the effect of distance from the parent tree and the crown's area of influence on mortality, growth, and leaf asymmetry of Manilkara zapota seedlings, as well as insect herbivory and damage from leaf pathogens in a Mexican neotropical forest. (2) We selected 10 reproductive adult trees (Diameter at breast height, DBH ~ 10–25 cm) and established four 10 m × 1 m transects around each tree in four directions (north, south, east, and west). Each transect produced 10 quadrants of 1 m², and the quadrant where the shadow of the parent tree extended was marked as either under crown or crown-free. All M. zapota seedlings were counted in each quadrant. For one seedling in each quadrant, we recorded height, leaf asymmetry (LA), insect herbivory, and damage from leaf pathogens. Herbivory by insects, damage from leaf pathogens, and LA were only measured on the newest leaves. Mortality was determined after 9 months per quadrant, as well as light availability (photosynthetic photon flux density), temperature, and relative humidity. (3) We found that mortality and relative growth rate (RGRHeight) increased near and under the parent tree. Furthermore, LA decreased at greater distances from the parent tree and only outside the crown's influence. Additionally, LA had a strong positive influence on damage caused by insect herbivory and leaf pathogens, impacting both more strongly under the crown. A high dependency of leaf pathogens on damage from insect herbivory was also recorded. Finally, the most frequent type of herbivory was that caused by chewing insects. (4) To our knowledge, we present one of the few studies that has addressed the JC hypothesis, considering not only the distance from the parent tree and seedling density but also the influence of the crown on the performance of M. zapota seedlings. Studies that consider the influence of the microenvironment are of fundamental importance for a comprehensive understanding of the JC hypothesis. [ABSTRACT FROM AUTHOR]

For the accurate and automatic mapping of forest stands based on very-high-resolution satellite imagery and digital orthophotos, precise object detection at the individual tree level is necessary. Currently, supervised deep learning models are primarily applied for this task. To train a reliable model, it is crucial to have an accurate tree crown annotation dataset. The current method of generating these training datasets still relies on manual annotation and labeling. Because of the intricate contours of tree crowns, vegetation density in natural forests and the insufficient ground sampling distance of the imagery, manually generated annotations are error-prone. It is unlikely that the manually delineated tree crowns represent the true conditions on the ground. If these error-prone annotations are used as training data for deep learning models, this may lead to inaccurate mapping results for the models. This study critically validates manual tree crown annotations on two study sites: a forest-like plantation on a cemetery and a natural city forest. The validation is based on tree reference data in the form of an official tree register and tree segments extracted from UAV laser scanning (ULS) data for the quality assessment of a training dataset. The validation results reveal that the manual annotations detect only 37% of the tree crowns in the forest-like plantation area and 10% of the tree crowns in the natural forest correctly. Furthermore, it is frequent for multiple trees to be interpreted in the annotation as a single tree at both study sites. [ABSTRACT FROM AUTHOR]

The frequency of forest disturbances has increased in recent years, provoking widespread defoliation, crown dieback and tree mortality. The ICP Forests monitoring network offers a unique platform for observing the impacts on forests of heatwaves, droughts and other extreme climatic events, as well as the trends of defoliation and mortality. The Italian ICP Forests Level I network consists of 261 permanent plots where tree crown defoliation and damage symptoms are assessed visually each year by well-trained crews of the Corpo Forestale dello Stato (2001-2016) and the Carabinieri Forestale from 2017 onward. This paper aims to assess the main tree species' responses, in terms of defoliation and mortality, to severe climatic events. The results are discussed in relation to species-specific physiological behaviour and bioclimatic regions. A significant trend toward increasing defoliation and mortality has been observed since 2010 in both conifers and broadleaves. Conifers (especially Picea abies), which are largely diffuse in the Alpine regions, have suffered from bark beetle outbreaks due to severe windstorms (such as Vaia in 2018) and recurrent dry years. In the temperate regions, characterised by deciduous broadleaved trees, the most relevant defoliation events coincided with the driest and hottest years, with low relative humidity (2012, 2017 and 2021-2022), only partially recovering in the subsequent years. Among them, Fagus sylvatica and Quercus cerris, along with increased defoliation, showed symptoms caused by fungi of the genus Biscogniauxia, causal agents of "charcoal canker", in less favourable site conditions. Quercus pubescens was the most resilient species, able to restore its crown after defoliation. The Mediterranean forests, with evergreen broadleaved species, showed no significant trends but were impacted at the most drought-prone coastal sites. The findings evidenced that the current ICP Forests network in Italy represents a fundamental infrastructure for monitoring impacts and trends connected to climate change and species-specific responses. A local intensification of the grid would help to capture under-represented species or ecological conditions. [ABSTRACT FROM AUTHOR]

Single-tree segmentation on multispectral UAV images shows significant potential for effective forest management such as automating forest inventories or detecting damage and diseases when using an additional classifier. We propose an automated workflow for segmentation on high-resolution data and provide our trained models in a Toolbox for ArcGIS Pro on our GitHub repository for other researchers. The database used for this study consists of multispectral UAV data (RGB, NIR and red edge bands) of a forest area in Germany consisting of a mix of tree species consisting of five deciduous trees and three conifer tree species in the matured closed canopy stage at approximately 90 years. Information of NIR and Red Edge bands are evaluated for tree segmentation using different vegetation indices (VIs) in comparison to only using RGB information. We trained Faster R-CNN, Mask R-CNN, TensorMask and SAM in several experiments and evaluated model performance on different data combinations. All models with the exception of SAM show good performance on our test data with the Faster R-CNN model trained on the red and green bands and the Normalized Difference Red Edge Index (NDRE) achieving best results with an F1-Score of 83.5% and an Intersection over Union of 65.3% on highly detailed labels. All models are provided in our TreeSeg toolbox and allow the user to apply the pre-trained models on new data. [ABSTRACT FROM AUTHOR]

As the main area for photosynthesis in trees, the canopy absorbs a large amount of carbon dioxide and plays an irreplaceable role in regulating the carbon cycle in the atmosphere and mitigating climate change. Therefore, monitoring the growth of the canopy is crucial. However, traditional field investigation methods are often limited by time-consuming and labor-intensive methods, as well as limitations in coverage, which may result in incomplete and inaccurate assessments. In response to the challenges encountered in the application of tree crown segmentation algorithms, such as adhesion between individual tree crowns and insufficient generalization ability of the algorithm, this study proposes an improved algorithm based on Mask R-CNN (Mask Region-based Convolutional Neural Network), which identifies irregular edges of tree crowns in RGB images obtained from drones. Firstly, it optimizes the backbone network by improving it to ResNeXt and embedding the SENet (Squeeze-and-Excitation Networks) module to enhance the model's feature extraction capability. Secondly, the BiFPN-CBAM module is introduced to enable the model to learn and utilize features more effectively. Finally, it optimizes the mask loss function to the Boundary-Dice loss function to further improve the tree crown segmentation effect. In this study, TCSNet also incorporated the concept of panoptic segmentation, achieving the coherent and consistent segmentation of tree crowns throughout the entire scene through fine tree crown boundary recognition and integration. TCSNet was tested on three datasets with different geographical environments and forest types, namely artificial forests, natural forests, and urban forests, with artificial forests performing the best. Compared with the original algorithm, on the artificial forest dataset, the precision increased by 6.6%, the recall rate increased by 1.8%, and the F1-score increased by 4.2%, highlighting its potential and robustness in tree detection and segmentation. [ABSTRACT FROM AUTHOR]

Urban tree classification enables informed decision-making processes in urban planning and management. This paper introduces a novel data reformation method, pseudo tree crown (PTC), which enhances the feature difference in the input layer and results in the improvement of the accuracy and efficiency of urban tree classification by utilizing artificial intelligence (AI) techniques. The study involved a comparative analysis of the performance of various machine learning (ML) classifiers. The results revealed a significant enhancement in classification accuracy, with an improvement exceeding 10% observed when high spatial resolution imagery captured by an unmanned aerial vehicle (UAV) was utilized. Furthermore, the study found an impressive average classification accuracy of 93% achieved by a classifier built on the PyTorch framework, with ResNet50 leveraged as its convolutional neural network layer. These findings underscore the potential of AI-driven approaches in advancing urban tree classification methodologies for enhanced urban planning and management practices. [ABSTRACT FROM AUTHOR]

Fractal dimension (D) can be used to characterise temporal changes of crown architecture of individual trees. Our goal in this study was to analyse seasonal changes in tree crown fractal dimension of two species of deciduous oaks (Quercus castanea and Q. obtusata) coexisting in a natural forest in central Mexico using low cost sampling, and relate these changes to morphological attributes and environmental variables. Every two months, from May 2017 to September 2018, for each oak species, we photographed fixed portions of the crowns of individual trees, measured their trunk diameters, and obtained average temperature and accumulated precipitation data recorded for the sampling date. From the obtained images, we calculated D values by the semivariogram method using three different variability estimators (square increment, isotropic, and transect variation). We identified a positive correlation between D and temperature, and a negative correlation between temperature and crown cover. The fractal dimension (D) of crowns of two deciduous oak species changes according to the tree's phenological stage. D values varied through time in relation to tree crown phenological variation, but not with crown cover dimension. We propose a model of annual D value fluctuation in deciduous trees, characterised by two high complexity peaks and two low complexity valleys, corresponding to the effects on crown cover of annual periods of leaf abscission and development. [ABSTRACT FROM AUTHOR]

Understanding the spatial heterogeneity of forest structure is crucial for comprehending ecosystem dynamics and promoting sustainable forest management. Unmanned aerial vehicle (UAV) LiDAR technology provides a promising method to capture detailed three-dimensional (3D) information about forest canopies, aiding in management and silvicultural practices. This study investigates the heterogeneity of forest structure in broadleaf forests using UAV LiDAR data, with a particular focus on tree crown features and their different information content compared to diameters. We explored a non-conventionally used method that emphasizes crown competition by employing a nearest neighbor selection technique based on metrics derived from UAV point cloud profiles at the tree level, rather than traditional DBH (diameter at breast height) spatial arrangement. About 300 vegetation elements within 10 plots collected in a managed Beech forest were used as reference data. We demonstrate that crown-based approaches, which are feasible with UAV LiDAR data at a reasonable cost and time, significantly enhances the understanding of forest heterogeneity, adding new information content for managers. Our findings underscore the utility of UAV LiDAR in characterizing the complexity and variability of forest structure at high resolution, offering valuable insights for carbon accounting and sustainable forest management. [ABSTRACT FROM AUTHOR]

Trees respond to global change in myriad ways, many of which may be linked to adaptations relating to tree crown architecture. However, there is a paucity of theory capable of predicting the adaptive importance and dynamics of crown architecture, most likely because of the difficulties involved in measuring the three‐dimensional arrangement and orientation of tree leaves within individual crowns. Here, we describe a theory of tree crown economics, and use measurements from new lidar (light detection and ranging) instruments, UAVs (unoccupied aerial vehicles), and time‐lapse camera imagery to identify support for two predictions of the theory, that (1) a light competition versus water use economic trade‐off drives covariance among three tree crown functional traits (mean leaf angle, crown density, and crown rugosity), and (2) crown traits can drive spatial and temporal variability in near‐infrared spectral reflectance and related ecosystem functions. Tree crown economic theory can complement leaf economic theory in helping ecologists map and model forest ecosystem responses to global change. [ABSTRACT FROM AUTHOR]

The surveying of forestry resources has recently shifted toward precision and real-time monitoring. This study utilized the BlendMask algorithm for accurately outlining tree crowns and introduced a Bayesian neural network to create a model linking individual tree crown size with diameter at breast height (DBH). BlendMask accurately outlines tree crown shapes and contours, outperforming traditional watershed algorithms in segmentation accuracy while preserving edge details across different scales. Subsequently, the Bayesian neural network constructs a model predicting DBH from the measured crown area, providing essential data for managing forest resources and conducting biodiversity research. Evaluation metrics like precision rate, recall rate, F1-score, and mAP index comprehensively assess the method's performance regarding tree density. BlendMask demonstrated higher accuracy at 0.893 compared to the traditional watershed algorithm's 0.721 accuracy based on experimental results. Importantly, BlendMask effectively handles over-segmentation problems while preserving edge details across different scales. Moreover, adjusting parameters during execution allows for flexibility in achieving diverse image segmentation effects. This study addresses image segmentation challenges and builds a model linking crown area to DBH using the BlendMask algorithm and a Bayesian neural network. The average discrepancies between calculated and measured DBH for Ginkgo biloba, Pinus tabuliformis, and Populus nigra varitalica were 0.15 cm, 0.29 cm, and 0.49cm, respectively, all within the acceptable forestry error margin of 1 cm. BlendMask, besides its effectiveness in crown segmentation, proves useful for various vegetation classification tasks like broad-leaved forests, coniferous forests, and grasslands. With abundant training data and ongoing parameter adjustments, BlendMask attains improved classification accuracy. This new approach shows great potential for real-world use, offering crucial data for managing forest resources, biodiversity research, and related fields, aiding decision-making processes. [ABSTRACT FROM AUTHOR]

Forest stands are developing in changeable climate conditions that influence stand health and consequently assortment quality. Narrow-leaved ash is strongly affected by dieback because of new fungal diseases. The main aim of this study was to determine the quantity and quality of produced wood assortments in dieback-affected narrow-leaved ash stands. Based on the study results, the average tree value increased with tree diameter and partially decreased with tree crown defoliation degree. The healthy (crown defoliated up to 25%) and 3A (crown defoliated from 61 to 80%) trees had significantly higher average tree values (EUR/m3) compared to the significantly defoliated 3B trees (crown defoliated from 81 to 99%) and dead trees (100% defoliated crown). The influence of stand age and share of narrow-leaved ash in stand volume were confirmed as factors influencing the average tree value. Wood chips quality remained the same regardless of tree crown defoliation degree. Based on the significance influence of the tree crown defoliation degree on the average tree value, current assortment tables should be expanded in order to achieve more accurate expected values. [ABSTRACT FROM AUTHOR]

In urban forestry, managing trees is crucial for sustainable urban environments, especially in the context of climate change and the urban heat island effect. This research explores the complex dynamics of tree crown geometry development by asking the question: how do surrounding objects, such as nearby trees, buildings, and other urban structures, affect the shape of tree crowns? It aims to uncover how competition for light and space influences tree crown development in competitive urban environments. Our study employs machine learning models on six main species in Munich, using the measured data from the LiDAR scans, with the Hist Gradient Boosting Regressor (HGBR) emerging as the most promising performer across various metrics. Notably, the evaluation of 13 models reveals the HGBR's consistent ranking as the best or second-best across all tree crown dimensions assessed, with R 2 values reaching 0.83 for the tree height model and 0.7 on average for crown radiuses in eight directions. Employing SHapley Additive exPlanations (SHAP) values elucidate factors influencing model predictions, emphasising the significant impact of adjacent trees and buildings. After evaluating the models to include additional tree species in Munich, the models show strong predictive capabilities for some additional species. Despite the studies' limitations - the models are only valid for selected species, and there are constraints in predicting tree crown start height - our findings contribute valuable insights for urban forestry management and planning. [Display omitted] • Machine learning method helps the tree crown development prediction in cities. • The tree-to-tree light competition effect on crown geometry is predictable and measurable by machine learning. • Among local environmental factors, neighbouring trees have the greatest impact on the geometry of a tree's crown shape. [ABSTRACT FROM AUTHOR]

Urban street greening is usually used as the main measure to mitigate vehicle emissions and improve air quality. However, recent studies have shown that the planting of trees in street canyons may easily lead to the obstruction of internal air flow and the accumulation of pollutants. In order to further reveal the influence of tree crown characteristics on the distribution of the particulate pollutants in street canyon, the urban microclimate numerical simulation software ENVI-met was used to simulate and analyze the influence of the four tree crown profiles on the flow field and PM 10 concentration distribution in street canyon under three wind directions of parallel, oblique and vertical inlet winds downward. The results showed that there was a blocking effect on the air flow in the street canyon of four street tree crowns regardless of the inlet wind downward. The leaf area density(LAD) distribution of tree canopy was an important factor influencing the blocking effect of tree crown. In general, the canopy height with the highest LAD had the most significant decrease on wind speed. There was a certain difference in the influence of tree crown on the PM 10 dispersion of pedestrian height on both sides under different inlet winds. The order of PM 10 accumulation under parallel and oblique wind direction was: conical crown, spherical crown, oval crown, obovate crown, while under vertical wind direction was: obovate crown, spherical crown, oval crown, conical crown. In the future, in the process of tree planting and management in street canyons, it should be combined with the wind direction to select appropriate trees or prune tree crowns to reduce the risk of pollution exposure for pedestrians. The results of the study provide a reference for the planting and management of street trees. [ABSTRACT FROM AUTHOR]

Predicting forest recovery at landscape scales will aid forest restoration efforts. The first step in successful forest recovery is tree recruitment. Forecasts of tree recruit abundance, derived from the landscape‐scale distribution of seed sources (i.e., adult trees), could assist efforts to identify sites with high potential for natural regeneration. However, previous work revealed wide variation in the effect of seed sources on seedling abundance, from positive to no effect. We quantified the relationship between adult tree seed sources and tree recruits and predicted where natural recruitment would occur in a fragmented, tropical, agricultural landscape. We integrated species‐specific tree crown maps generated from hyperspectral imagery and property ownership data with field data on the spatial distribution of tree recruits from five species. We then developed hierarchical Bayesian models to predict landscape‐scale recruit abundance. Our models revealed that species‐specific maps of tree crowns improved recruit abundance predictions. Conspecific crown area had a much stronger impact on recruitment abundance (8.00% increase in recruit abundance when conspecific tree density increases from zero to one tree; 95% credible interval (CI): 0.80% to 11.57%) than heterospecific crown area (0.03% increase with the addition of a single heterospecific tree, 95% CI: −0.60% to 0.68%). Individual property ownership was also an important predictor of recruit abundance: The best performing model had varying effects of conspecific and heterospecific crown area on recruit abundance, depending on individual property ownership. We demonstrate how novel remote sensing approaches and cadastral data can be used to generate high‐resolution and landscape‐level maps of tree recruit abundance. Spatial models parameterized with field, cadastral, and remote sensing data are poised to assist decision support for forest landscape restoration. [ABSTRACT FROM AUTHOR]

Rapid and accurate tree-crown detection is significant to forestry management and precision forestry. In the past few decades, the development and maturity of remote sensing technology has created more convenience for tree-crown detection and planting management. However, the variability of the data source leads to significant differences between feature distributions, bringing great challenges for traditional deep-learning-based methods on cross-regional detection. Moreover, compared with other tasks, tree-crown detection has the problems of a poor abundance of objects, an overwhelming number of easy samples and the existence of a quantity of impervious background similar to the tree crown, which make it difficult for the classifier to learn discriminative features. To solve these problems, we apply domain adaptation (DA) to tree-crown detection and propose a DA cascade tree-crown detection framework with multiple region proposal networks, dubbed CAS-DA, realizing cross-regional tree-crown detection and counting from multiple-source remote sensing images. The essence of the multiple region proposal networks in CAS-DA is obtaining the multilevel features and enhancing deeper label classifiers gradually by filtering simple samples of source domain at an early stage. Then, the cascade structure is integrated with a DA object detector and the end-to-end training is realized through the proposed cascade loss function. Moreover, a filtering strategy based on the planting rules of tree crowns is designed and applied to filter wrongly detected trees by CAS-DA. We verify the effectiveness of our method in two different domain shift scenarios, including adaptation between satellite and drone images and cross-satellite adaptation. The results show that, compared to the existing DA methods, our method achieves the best average F1-score in all adaptions. It is also found that the performance between satellite and drone images is significantly worse than that between different satellite images, with average F1-scores of 68.95% and 88.83%, respectively. Nevertheless, there is an improvement of 11.88%~40.00% in the former, which is greater than 0.50%~5.02% in the latter. The above results prove that in tree-crown detection, it is more effective for the DA detector to improve the detection performance on the source domain than to diminish the domain shift alone, especially when a large domain shift exists. [ABSTRACT FROM AUTHOR]

The shape and area of the crown of each tree are among the most influential parameters for identifying and controlling the processes of photosynthesis, respiration, transpiration and its management. In such a way that various physiographic functions, such as carbon dioxide absorption, light energy absorption, oxygen release and transpiration, which are vital for the growth and development of the tree, are done in the crown. In this research, the RGB image of the UAV with a spatial resolution of 2 cm was resampled to three pixel sizes of 10, 30 and 50 cm. Then, each image was classified separately by SVM, ANN and MLC algorithms, which are all part of Ensemble. In the next step, each of the obtained crowns was compared with the digitization of the same crown, and based on the area of the crown obtained from each classification and normalization method, the weight was obtained specifically for the same crown. Finally, by using the weighted majority voting method, classifications were fusioned at the decision level. The results showed that the ANN method gives better results in all pixel sizes compared to MLC and SVM. Also, the combination of different classification methods with the weighted majority voting method based on the weight assigned to the same crown based on each classification method has significantly increased the classification accuracy of the tree crown in all the sizes of the analyzed pixels. [ABSTRACT FROM AUTHOR]

Tree crown diameter (CD) values, relating to the rate of material exchange between the forest and the atmosphere, can be used to evaluate forest biomass and carbon stock. To map tree CD values using meter-level optical remote sensing images, we propose a novel method that interprets the relationships between the spectral reflectance of pixels and the CD. The approach employs the spectral reflectance of pixels in the tree crown to express the diversity of inclination angles of leaves based on the radiative transfer model and the spatial heterogeneity of these pixels. Then, simulated and acquired GF-2 images are applied to verify the relationships between spatial heterogeneity and the tree CD. Meanwhile, filter-based and object-based methods are also employed to extract three types of variables (spectral features, texture features, and spatial heterogeneity). Finally, the tree CD values are mapped by four models (random forest (RF), K-nearest neighbor (K-NN), support vector machine (SVM), and multiple linear regression (MLR)), using three single types of variables and combinations of variables with different strategies. The results imply that the spatial heterogeneity of spectral reflectance is significantly positively correlated with tree CD values and is more sensitive to tree CD values than traditional spectral features and textural features. Furthermore, the ability of spatial heterogeneity to map tree CD values is significantly higher than traditional variable sets after obtaining stable features with appropriate filter window sizes. The results also demonstrate that the accuracy of mapped tree CD values is significantly improved using combined variable sets with different feature extraction methods. For example, in our experiments, the R2 and rRMSE values of the optimal results ranged from 0.60 to 0.66, and from 15.76% to 16.68%, respectively. It is confirmed that spatial heterogeneity with high sensitivity can effectively map tree CD values, and the accuracy of mapping tree CD values can be greatly improved using a combination of spectral features extracted by an object-based method and spatial heterogeneity extracted by a filter-based method. [ABSTRACT FROM AUTHOR]

The automatic detection of tree crowns and estimation of crown areas from remotely sensed information offer a quick approach for grasping the dynamics of forest ecosystems and are of great significance for both biodiversity and ecosystem conservation. Among various types of remote sensing data, unmanned aerial vehicle (UAV)-acquired RGB imagery has been increasingly used for tree crown detection and crown area estimation; the method has efficient advantages and relies heavily on deep learning models. However, the approach has not been thoroughly investigated in deciduous forests with complex crown structures. In this study, we evaluated two widely used, deep-learning-based tree crown detection and delineation approaches (DeepForest and Detectree2) to assess their potential for detecting tree crowns from UAV-acquired RGB imagery in an alpine, temperate deciduous forest with a complicated species composition. A total of 499 digitized crowns, including four dominant species, with corresponding, accurate inventory data in a 1.5 ha study plot were treated as training and validation datasets. We attempted to identify an effective model to delineate tree crowns and to explore the effects of the spatial resolution on the detection performance, as well as the extracted tree crown areas, with a detailed field inventory. The results show that the two deep-learning-based models, of which Detectree2 (F1 score: 0.57) outperformed DeepForest (F1 score: 0.52), could both be transferred to predict tree crowns successfully. However, the spatial resolution had an obvious effect on the estimation accuracy of tree crown detection, especially when the resolution was greater than 0.1 m. Furthermore, Dectree2 could estimate tree crown areas accurately, highlighting its potential and robustness for tree detection and delineation. In addition, the performance of tree crown detection varied among different species. These results indicate that the evaluated approaches could efficiently delineate individual tree crowns in high-resolution optical images, while demonstrating the applicability of Detectree2, and, thus, have the potential to offer transferable strategies that can be applied to other forest ecosystems. [ABSTRACT FROM AUTHOR]

Mapping the aboveground biomass (AGB) of forests in a spatially continuous manner is essential to comprehend the carbon sequestration capacity of forest ecosystems. Automatic extraction of tree crown and height is an effective approach to estimating biomass at the tree level. In this study, we employed high-resolution unmanned aerial vehicle (UAV) images and a mask region-based convolutional neural network (Mask R-CNN) to automatically identify the tree crown and height of Pinus sylvestris, we added tree height information to the model training by setting special tree height label samples and determined the optimal model by comparing the tree crown and height detection accuracy of four different combinations (RGB-CHM-DSM, RGB-CHM, RGB-DSM, RGB) of UAV images. The findings demonstrate that the model with RGB-CHM-DSM combination has the highest crown extraction accuracy (Precision = 0.896, Recall = 0.916, F1 Score = 0.906, IoU = 0.822), and the tree height extraction results with RGB-CHM combination has the highest correlation with the tree height of UAV images (R2 = 0.93, RMSE = 0.25, rRMSE = 3.10). Higher extraction accuracy is achieved due to the inclusion of CHM features in the model, and the results can be used to estimate forest AGB at different scales, improve the accuracy and efficiency of forest carbon sink research, minimize the workload of field investigations, and reduce the cost of manual methods. • A Mask-RCNN model was proposed to detect tree crown and height simultaneously. • Spatially continuous individual tree level aboveground biomass was mapped. • The detection accuracy of different feature combination models was compared. • Adding CHM features can improve the tree crown and height detection accuracy. [ABSTRACT FROM AUTHOR]

In light of the current dearth of strip intermittent fertilization devices in standardized orchards, this study presents the design of an intermittent fertilization control system based on fruit tree crown detection, developed with the objective of meeting the agronomic requirements of strip furrow fertilization in standardized orchards. The initial stage of the process entails the design of the essential components of the fertilization apparatus, followed by the construction of the intermittent fertilization control system. The ultrasonic sensor was employed as the fruit tree crown detection module, and a mathematical model of fertilization speed was constructed to achieve uniform intermittent fertilization. Furthermore, in order to enhance the responsiveness and reliability of the fertilization servo motor, MATLAB Simulink was employed to assess the dynamic performance of the system under disparate control strategies. Ultimately, to validate the simulation outcomes, a field trial was conducted to assess the precision and uniformity of intermittent fertilization. The results demonstrate that the dynamic performance of the system under the fuzzy PID control strategy is optimal, and the coefficient of variation of the fertilizer uniformity of the intermittent fertilization device is less than 7%. The mean effective fertilization rate exceeded 85%, with the primary indices satisfying the agronomic criteria. [ABSTRACT FROM AUTHOR]

Coastal forest acts as a natural solution against floods and tsunami events; however, the difference in the management techniques can significantly affect the forest effectiveness. This study clarified the role of tree crown height and forest thickness by elucidating the flow hydrodynamics and tsunami energy reduction around coastal forests. Through physical laboratory tests in an experimental flume with quasi-steady subcritical flow conditions, this paper elucidated the flow hydrodynamics while investigating the mitigation function of coastal forest and the reduction potential of approaching flow velocity and fluid force around it. The outcomes demonstrated that a forest model with a low tree crown height i.e. crown height to tree height ratio of 0.3 (lowest considered among 0.7, 0.5, and 0.3), caused a strong reflection (back movement of water wave due to backwater rise) of flow, followed by large flow resistance capability in case of large forest thickness (W) than that in half forest thickness (W/2) configuration. The highest reduction potential of approaching flow velocity and fluid force i.e., around 50%, was observed in the case of the lowest tree crown height and largest forest thickness, and such forest configuration had shown to dissipate the maximum energy i.e., around 40%, of the inundating tsunami current. The study found a strong relation of flow energy loss to that of low tree crown height and large forest thickness, highlighting the improved reduction potential efficiency of such coastal forest designs against tsunami disasters. [ABSTRACT FROM AUTHOR]

Automatic collection of tree-level crown information is essential for sustainable forest management and fine carbon stock estimation. UAV-based light detection and ranging (LiDAR) and UAV-based multi-angle photogrammetry (UMP) data depict the 3D structure of forests at a fine-grained level by generating detailed point clouds, making them potential alternatives to labor-intensive forest inventories. However, the accuracy of the individual tree crown segmentation algorithms that have been developed is unstable in forest stands with high terrain undulation and high canopy density, mainly due to the various crown sizes and interlocking crowns resulting in varying degrees of over- or under-segmentation. Here, we propose self-similarity cluster grouping (SCG) algorithm for individual tree crown segmentation that integrates multivariable calculus of crown surfaces and spectral-texture-color spatial information of crown. Firstly, according to the property that DSM and its multi-order gradient information can characterize the crown surface variation and concavity-convexity features, first- and second-order edge detection operators were used to preliminarily determine the crown patch edges in order to reduce under-segmentation. Then, we developed a self-similarity weight function controlled by the spectral, texture and color spatial information of the tree crown patches to increase the similarity difference between adjacent crown patches of the same tree and those of neighboring trees, and designed the strategy for cluster grouping crown patches to complete individual tree crown segmentation. The performance of the proposed SCG algorithm was verified in Mytilaria, Red oatchestnu, Chinese fir and Eucalyptus plots in subtropical forests of China using LiDAR and UMP data. The overall accuracy of F-score (f) was above 0.85 for crown segmentation, and the rRMSE for crown width, crown area and crown circumference extractions reached 0.13, 0.22 and 0.14, respectively. On this basis, we evaluated the effect of spatial resolution of DSM on the segmentation accuracy of SCG algorithm, and found that the crown segmentation accuracy was proportional to the spatial resolution. Compared to the normalized cut algorithm, marker-controlled watershed algorithm and threshold-based cloud point segmentation algorithm, the SCG algorithm improved the overall accuracy f of individual tree crown segmentation by 0.06, 0.13 and 0.05 for LiDAR and 0.06, 0.21 and 0.10 for UMP, respectively. Furthermore, the effectiveness and generalizability of the SCG algorithm was verified in other Mytilaria, Red oatchestnut, Chinese fir and Eucalyptus plots in subtropical forests and Larch and Chinese pine plots in temperate forests using UMP data. The crown segmentation accuracy was better than 0.82, and the crown width extraction accuracy was up to 89 %. Overall, our proposed SCG algorithm reduces the over- and under-segmentation in complex forest structures and provides technical support for accurate crown information extraction at both plot and forest stand levels. • Multivariable calculus theory was used for individual tree crown segmentation. • Super-segmentation controlled by multi-gradient data reduced under-segmentation. • Merging strategy based on similarity weighting functions helped in segmentation. • Verifying the generality of our algorithm applied to UMP and LiDAR data. [ABSTRACT FROM AUTHOR]

In the case of tree species delineation with very high spatial resolution (VHR) images, is desirable that each segment corresponds to one individual tree crown (ITC). However, in order to have a segmentation algorithm that generates segments matching to ITCs, its parameters ought to be properly tuned. Aiming to avoid time-consuming trial-and-error procedures associated with this task, some initiatives for the automatic search of segmentation parameters have been developed, such as metaheuristic methods. The objective of this work was to test the automatic tuning of segmentation parameters of three segmentation algorithms for the delineation of ITCs belonging to a native endangered species in a subtropical forest area, comparing this method with the traditional trial-and-error approach. Two datasets (WorldView-2 and an orthoimage) and three segmentation algorithms (multiresolution, mean-shift and graph-based) were tested. For the automatic approach, a hybrid metaheuristic method was applied to accomplish the automatic search of parameters for the segmentation algorithms, while for the trial-and-error, a visual assessment was conducted for each set of parameters tested. Four supervised metrics were used to assess the quality of the segmentation results for the optimization approach and for the final set of parameters chosen in the trial-and-error approach. Results showed that none of the algorithms, datasets or approaches differ too much. The evaluation metrics values were lower, indicating that the reference ITCs polygons matched with the segmentation results. Despite the similar results, the automatic tuning of segmentation parameters proved to be a feasible alternative to reduce the subjectivity and the human effort in the choice of segmentation parameters as compared to the trial-and error approach. [ABSTRACT FROM AUTHOR]

Copyright of Forestry Chronicle is the property of Canadian Institute of Forestry and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

We present a deep learning-based framework for individual tree crown delineation in aerial and satellite images. This is an important task, e.g., for forest yield or carbon stock estimation. In contrast to earlier work, the presented method creates irregular polygons instead of bounding boxes and also provides a tree cover mask for areas that are not separable. Furthermore, it is trainable with low amounts of training data and does not need 3D height information from, e.g., laser sensors. We tested the approach in two scenarios: (1) with 30 cm WorldView-3 satellite imagery from an urban region in Bengaluru, India, and (2) with 5 cm aerial imagery of a densely forested area near Gartow, Germany. The intersection over union between the reference and predicted tree cover mask is 71.2% for the satellite imagery and 81.9% for the aerial images. On the polygon level, the method reaches an accuracy of 46.3% and a recall of 63.7% in the satellite images and an accuracy of 52% and recall of 66.2% in the aerial images, which is comparable to previous works that only predicted bounding boxes. Depending on the image resolution, limitations to separate individual tree crowns occur in situations where trees are hardly separable even for human image interpreters (e.g., homogeneous canopies, very small trees). The results indicate that the presented approach can efficiently delineate individual tree crowns in high-resolution optical images. Given the high availability of such imagery, the framework provides a powerful tool for tree monitoring. The source code and pretrained weights are publicly available at https://github.com/AWF-GAUG/TreeCrownDelineation. [ABSTRACT FROM AUTHOR]

Protecting the future of forests in the United States and other countries depends in part on our ability to monitor and map forest health conditions in a timely fashion to facilitate management of emerging threats and disturbances over a multitude of spatial scales. Remote sensing data and technologies have contributed to our ability to meet these needs, but existing methods relying on supervised classification are often limited to specific areas by the availability of imagery or training data, as well as model transferability. Scaling up and operationalizing these methods for general broadscale monitoring and mapping may be promoted by using simple models that are easily trained and projected across space and time with widely available imagery. Here, we describe a new model that classifies high resolution (~1 m2) 3-band red, green, blue (RGB) imagery from a single point in time into one of four color classes corresponding to tree crown condition or health: green healthy crowns, red damaged or dying crowns, gray damaged or dead crowns, and shadowed crowns where the condition status is unknown. These Tree Crown Health (TCH) models trained on data from the United States (US) Department of Agriculture, National Agriculture Imagery Program (NAIP), for all 48 States in the contiguous US and spanning years 2012 to 2019, exhibited high measures of model performance and transferability when evaluated using randomly withheld testing data (n = 122 NAIP state x year combinations; median overall accuracy 0.89–0.90; median Kappa 0.85–0.86). We present examples of how TCH models can detect and map individual tree mortality resulting from a variety of nationally significant native and invasive forest insects and diseases in the US. We conclude with discussion of opportunities and challenges for extending and implementing TCH models in support of broadscale monitoring and mapping of forest health. [ABSTRACT FROM AUTHOR]

The tree crown, with its functionality of assimilation, respiration, and transpiration, is a key forest ecosystem structure, resulting in high demand for characterizing tree crown structure and growth on a spatiotemporal scale. Airborne laser scanning (ALS) was found to be useful in measuring the structural properties associated with individual tree crowns. However, established ALS-assisted monitoring frameworks are still limited. The main objective of this study was to investigate the feasibility of detecting species-specific individual tree crown growth by means of airborne laser scanning (ALS) measurements in 2009 (T1) and 2014 (T2). Our study was conducted in southern Finland over 91 sample plots with a size of 32 × 32 m. The ALS crown metrics of width (WD), projection area (A2D), volume (V), and surface area (A3D) were derived for species-specific individually matched trees in T1 and T2. The Scots pine (Pinus sylvestris), Norway spruce (Picea abies (L.) H. Karst), and birch (Betula sp.) were the three species groups that studied. We found a high capability of bi-temporal ALS measurements in the detection of species-specific crown growth (Δ), especially for the 3D crown metrics of V and A3D, with Cohen's D values of 1.09–1.46 (p-value < 0.0001). Scots pine was observed to have the highest relative crown growth (rΔ) and showed statistically significant differences with Norway spruce and birch in terms of rΔWD, rΔA2D, rΔV, and rΔA3D at a 95% confidence interval. Meanwhile, birch and Norway spruce had no statistically significant differences in rΔWD, rΔV, and rΔA3D (p-value < 0.0001). However, the amount of rΔ variability that could be explained by the species was only 2–5%. This revealed the complex nature of growth controlled by many biotic and abiotic factors other than species. Our results address the great potential of ALS data in crown growth detection that can be used for growth studies at large scales. [ABSTRACT FROM AUTHOR]

Urban trees play an important role in cooling urban microclimates and regulating outdoor thermal comfort. To better understand their contribution to these processes, it is crucial to elucidate urban trees' radiative thermal performance, especially in the infrared (IR) region (approximately 50% of solar radiation). Yet, owing to significant conceptual and methodological challenges, studies on the radiative performance of trees have mainly focused on individual leaves rather than crown-level characteristics. Here we applied a novel conceptual and methodological framework to characterise the crown-level IR radiative performance of 10 lime trees (Tilia cordata), a common urban tree in the UK and Europe. Our results show that reflected and transmitted solar energy from leaves is dominated (>70%) by IR radiation. At the leaf level, transmission and reflection spectra are similar between trees (differences typically < 10% in IR region), including those under significantly different urban stress conditions. However, at the crown-level, substantial variations in IR transflectance spectra (maximum difference > 40% in IR region) were found between trees. These variations were largely due to crown structural differences (leaf number, density, angles), rather than leaf solar interaction character (leaf-level transmittance or reflectance, leaf colour). Crown transflectance measured from the four cardinal directions was significantly different in the IR region (maximum differences circa 30%), and changed substantially with solar time. Hence, a tree's surroundings received very different, and time dependent, levels of solar IR radiation. These findings have significant implications for species selection and control of environmental stress factors in urban microclimates. • Established the conceptual and methodological framework of tree crown spectroscopy. • Infrared radiation dominates transmitted and reflected solar radiation of trees. • Identified the important factors affecting tree-solar radiation interactions. • Trees' radiative performance is mainly affected by crown structural differences. [ABSTRACT FROM AUTHOR]

Chinese olive trees (Canarium album L.) are broad-leaved species that are widely planted in China. Accurately obtaining tree crown information provides important data for evaluating Chinese olive tree growth status, water and fertilizer management, and yield estimation. To this end, this study first used unmanned aerial vehicle (UAV) images in the visible band as the source of remote sensing (RS) data. Second, based on spectral features of the image object, the vegetation index, shape, texture, and terrain features were introduced. Finally, the extraction effect of different feature dimensions was analyzed based on the random forest (RF) algorithm, and the performance of different classifiers was compared based on the features after dimensionality reduction. The results showed that the difference in feature dimensionality and importance was the main factor that led to a change in extraction accuracy. RF has the best extraction effect among the current mainstream machine learning (ML) algorithms. In comparison with the pixel-based (PB) classification method, the object-based image analysis (OBIA) method can extract features of each element of RS images, which has certain advantages. Therefore, the combination of OBIA and RF algorithms is a good solution for Chinese olive tree crown (COTC) extraction based on UAV visible band images. [ABSTRACT FROM AUTHOR]

A tree crown is a mass of foliage and significantly correlated to biomass and growth of other parts of a tree. Crown width (CW) is defined as an average of two crown diameters perpendicular to each other, obtained from measurements of four crown radii (crown components). The CW is often used as a covariate predictor in various forest models and therefore, its accurate estimate can be useful for forest management. Although various CW models have been developed so far, all of them lack a detailed investigation of the additivity properties of crown components and inherent correlations among the components and total CW. This study compared three alternative procedures of developing the systems of CW models: nonlinear simultaneous equations (NSE), disaggregated model structure with both one-step proportional weighting system (DMS&OSPWS), and two-step proportional weighting system, and aggregated model structure with one constraint on the total CW. Other two commonly used additivity methods: adjustment in proportion and ordinary least square with separating regression were also evaluated. The comparisons were based on the analyses of data obtained from 2207 trees of Mongolian oak ( Quercus mongolica Fisch.) on 116 permanent sample plots located in northern China. Results showed that each method ensured a summation of crown components equal to twice as much as the total CW. The NSE better accounted for the inherent correlations among the crown components and total CW. Out of the model structures evaluated, DMS&OSPWS provided the best performance. This article emphasized more on the methodology and it was expected that the method could be applied by other researchers to develop similar systems of the CW models for other species as well. [ABSTRACT FROM AUTHOR]

Airborne light detection and ranging (LiDAR) data are increasingly used to inform sustainable forest management practices. Information about species composition is needed for a range of applications; however, commonly used area-based summaries of LiDAR data are limited to accurately differentiate tree species. The objective of this study was to map dominant species groups across a large (>580,000 ha) boreal forest by combining area-based and individual tree metrics derived from single photon LiDAR data with multispectral information derived from Sentinel-2 imagery. Classification of the forest into jack pine (Pinus banksiana), black spruce (Picea mariana), mixed conifer, mixedwood, and hardwood species groups resulted in an overall accuracy of 67.8% compared with field data. A more generalized classification into softwood, hardwood, and mixedwood had an overall accuracy of 83.2%. The reflectance in the red edge region of the electromagnetic spectrum (λ = 740 nm), the area and volume of tree crowns, and the cumulative distribution of LiDAR returns in the canopy were particularly important variables to discriminate between species groups. Wall-to-wall predictions of species groups based on remotely sensed data—as derived herein—could provide a spatially-detailed alternative to stand-level species composition information traditionally derived from photo-interpreted aerial imagery. [ABSTRACT FROM AUTHOR]

Crown fire damage is a mixture of three principal fire-related components: charred material, scorched foliage, and unaltered green canopy. This study estimated the abundance of these physical alterations in two immediate post-fire Mediterranean forest contexts (Portugal and Italy) by applying linear spectral mixture analysis (LSMA) on Sentinel-2 imagery. The tree crowns fire damage was subsequently mapped, integrating fractional abundance information in a random forest (RF) algorithm, comparing the accuracy resulting from the adoption of generic or image spectral libraries as the primary investigative goal. Although image-derived endmembers resulted in more effectiveness in terms of fire-related components abundance quantification (LMSA-derived RMSE < 0.1), the F-scores always were ≥ 90% whether generic endmembers or image endmembers derived information was employed. The environmental heterogeneity of the two study areas affected the fire severity gradients, with a prevalence of the charred (PT) (45–46%) and green class (IT) (44–53%). Post-fire temporal monitoring was initialized by applying the proposed strategies, and the preliminary results showed a positive recovery trend in forest vegetation from the first year following the fire event, with a reduced charcoal predominance and an increasing proportion of green components. [ABSTRACT FROM AUTHOR]

Crown damage can account for over 23% of canopy biomass turnover in tropical forests and is a strong predictor of tree mortality; yet, it is not typically represented in vegetation models. We incorporate crown damage into the Functionally Assembled Terrestrial Ecosystem Simulator (FATES), to evaluate how lags between damage and tree recovery or death alter demographic rates and patterns of carbon turnover. We represent crown damage as a reduction in a tree's crown area and leaf and branch biomass, and allow associated variation in the ratio of aboveground to belowground plant tissue. We compare simulations with crown damage to simulations with equivalent instant increases in mortality and benchmark results against data from Barro Colorado Island (BCI), Panama. In FATES, crown damage causes decreases in growth rates that match observations from BCI. Crown damage leads to increases in carbon starvation mortality in FATES, but only in configurations with high root respiration and decreases in carbon storage following damage. Crown damage also alters competitive dynamics, as plant functional types that can recover from crown damage outcompete those that cannot. This is a first exploration of the trade‐off between the additional complexity of the novel crown damage module and improved predictive capabilities. At BCI, a tropical forest that does not experience high levels of disturbance, both the crown damage simulations and simulations with equivalent increases in mortality does a reasonable job of capturing observations. The crown damage module provides functionality for exploring dynamics in forests with more extreme disturbances such as cyclones and for capturing the synergistic effects of disturbances that overlap in space and time. [ABSTRACT FROM AUTHOR]

Mangrove forests are vulnerable ecosystems that require broad-scale monitoring. Various solutions based on satellite imagery have emerged for this purpose but still suffer from the lack of methods to accurately delineate individual tree crowns (ITCs). Within-stand variability in crown size and shape, crown clumping and fragmentation, and understory vegetation hamper the delineation in these ecosystems. To cope with these factors, the proposed method combines a deep learning-based enhancement of ITCs with a marker-controlled watershed segmentation algorithm. The MT-EDv3 neural network is employed to compute the normalized Euclidean distance of crown pixels to treetops and a Laplacian of Gaussian filter is applied to the resulting image to enhance crown borders before segmentation. The method was applied to WorldView imagery over four mangrove sites worldwide and compared to previously published methods using standardized metrics. Accurate detection (Overall Accuracy ≥ 0.93 and Kappa ≥ 0.87) and area estimation (R 2 ≥ 0.66, NRMSE ≤ 12%) of crowns was achieved for all sites using either the panchromatic band or a combination of the pan-sharpened visible-near-infrared bands. Based on Precision, Recall, and F1-score, the proposed method outperformed previous watershed segmentation and software-based algorithms of crown delineation, as well as the Mask R-CNN segmentation framework. The viewing geometry of images and the forest heterogeneity were identified as important contributors to the delineation accuracy. This study is the first to achieve accurate delineation of ITCs in mangrove forests across sites, opening perspectives of applications to satellite-based monitoring. The method shows promising transferability to other very-high-resolution satellite sensors as well as to aerial and unmanned aerial vehicle imagery and could be improved by including more spectral information and LiDAR-derived canopy height models. [ABSTRACT FROM AUTHOR]

Tree-crown and height are primary tree measurements in forest inventory. Convolutional neural networks (CNNs) are a class of neural networks, which can be used in forest inventory; however, no prior studies have developed a CNN model to detect tree crown and height simultaneously. This study is the first-of-its-kind that explored training a mask region-based convolutional neural network (Mask R-CNN) for automatically and concurrently detecting discontinuous tree crown and height of Chinese fir (Cunninghamia lanceolata (Lamb) Hook) in a plantation. A DJI Phantom4-Multispectral Unmanned Aerial Vehicle (UAV) was used to obtain high-resolution images of the study site, Shunchang County, China. Tree crown and height of Chinese fir was manually delineated and derived from this UAV imagery. A portion of the ground-truthed tree height values were used as a test set, and the remaining measurements were used as the model training data. Six different band combinations and derivations of the UAV imagery were used to detect tree crown and height, respectively (Multi band-DSM, RGB-DSM, NDVI-DSM, Multi band-CHM, RGB-CHM, and NDVI-CHM combination). The Mask R-CNN model with the NDVI-CHM combination achieved superior performance. The accuracy of Chinese fir's individual tree-crown detection was considerable (F1 score = 84.68%), the Intersection over Union (IoU) of tree crown delineation was 91.27%, and tree height estimates were highly correlated with the height from UAV imagery (R2 = 0.97, RMSE = 0.11 m, rRMSE = 4.35%) and field measurement (R2 = 0.87, RMSE = 0.24 m, rRMSE = 9.67%). Results demonstrate that the input image with an CHM achieves higher accuracy of tree crown delineation and tree height assessment compared to an image with a DSM. The accuracy and efficiency of Mask R-CNN has a great potential to assist the application of remote sensing in forests. [ABSTRACT FROM AUTHOR]

To sustainably manage forest biodiversity and monitor changes in species patterning, mapping the spatial distribution of tree species is indispensable. Remote sensing can provide powerful tools for mapping species, but this task is complex in areas with high plant diversity and multi-layered canopies. This paper addresses the issue of classifying wet eucalypt forest plants by examining tree crown segmentation and species classification using different combinations of remote sensing datasets against mapped tree locations. This study explores optimal segmentation parameters for tree crown delineation compared to manually digitized tree crowns. The best segmentation accuracy of 88.71%, resulted from segmenting a combined Minimum Noise Fraction (MNF) dataset derived from hyperspectral imagery (HSI) and a LiDAR-derived Canopy Height Model (CHM). Object-based classification of tree species was performed using a random forest classifier. The fused dataset of MNF and CHM produced the highest overall accuracy of 78.26% for four vegetation classes, while the fused HSI, indices, and CHM performed best (66.67%) with five vegetation classes. However, both approaches had a high overall performance. The CHM contributed to tree crown segmentation and species classification accuracy, and fused datasets were more robust to spatially discriminate wet eucalypt forest species compared to a single dataset. Eucalyptus obliqua was classified with the highest accuracy of 90.86% for four classes using the fused MNF and CHM dataset, and 86.11% for five classes using the fused HSI, indices, and CHM dataset. An important understorey species – the tree fern (Dicksonia antarctica) – was classified with the highest accuracy of 83.54% for four classes using HSI. Therefore, fusing hyperspectral and LiDAR data could classify both the overstorey and dominant understorey species, and thus play a crucial role in identifying forest biological diversity. This approach will be useful for forest managers and ecologists to plan sustainable management of eucalypt forest biodiversity and produce maps for monitoring species of interest. [ABSTRACT FROM AUTHOR]

Recurrent climate-driven disturbances impact on the health of European forests that reacted with increased tree dieback and mortality over the course of the last four decades. There is therefore large interest in predicting and understanding the fate and survival of forests under climate change. Forest conditions are monitored within the pan-European ICP Forests programme (UN-ECE International Co-operative Programme on Assessment and Monitoring of Air Pollution Effects on Forests) since the 1980s, with tree crown defoliation being the most widely used parameter. Defoliation is not a cause-specific indicator of tree health and vitality, and there is a need to connect defoliation levels with the physiological functioning of trees. The physiological responses connected to tree crown defoliation are species-specific and concern, among others, water relations, photosynthesis and carbon metabolism, growth, and mineral nutrients of leaves. The indicators to measure physiological variables in forest monitoring programs must be easy to apply in the field with current state-of-the-art technologies, be replicable, inexpensive, time efficient and regulated by ad hoc protocols. The ultimate purpose is to provide data to feed process-based models to predict mortality and threats in forests due to climate change. This study reviews the problems and perspectives connected to the realization of a systematic assessment of physiological variables and proposes a set of indicators suitable for future application in forest monitoring programs. [ABSTRACT FROM AUTHOR]

This article describes algorithms to extract tree crowns using two-dimensional (2D) and three-dimensional (3D) segmentation. As a first step, a 2D-search detected the tallest trees but was unable to detect trees located below other trees. However, a 3D-search for local maxima of model fits could be used in a second step to detect trees also in lower canopy layers. We compared tree detection results from ALS carried out at 1450 m above ground level (high altitude) and tree detection results from ALS carried out at 150 m above ground level (low altitude). For validation, we used manual measurements of trees in ten large field plots, each with an 80 m diameter, in a hemiboreal forest in Sweden (lat. 58°28' N, long. 13°38' E). In order to measure the effect of using algorithms with different computational costs, we validated the tree detection from the 2D segmentation step and compared the results with the 2D segmentation followed by 3D segmentation of the ALS point cloud. When applying 2D segmentation only, the algorithm detected 87% of the trees measured in the field using high-altitude ALS data; the detection rate increased to 91% using low-altitude ALS data. However, when applying 3D segmentation as well, the algorithm detected 92% of the trees measured in the field using high-altitude ALS data; the detection rate increased to 99% using low-altitude ALS data. For all combinations of algorithms and data resolutions, undetected trees accounted for, on average, 0–5% of the total stem volume in the field plots. The 3D tree crown segmentation, which was using crown density models, made it possible to detect a large percentage of trees in multi-layered forests, compared with using only a 2D segmentation method. [ABSTRACT FROM AUTHOR]

In old-growth Sequoia sempervirens forests, reiterated trunks and limbs provide required habitat elements for specialized arboreal species, including an endangered seabird, Brachyramphus marmoratus . The oldest second-growth redwood forests—established after 19th century logging—lack species dependent on complex structure, presumably because redwoods maintain simple, model-conforming crowns for centuries unless damaged by wind or fire. We imposed a factorial experiment on 24 redwoods 59–75 m tall in six second-growth forests to determine if trunk reiteration and limb formation can be induced by removing treetops (topping) and branch tips (tipping) to disrupt apical control. We also increased light availability in the upper crown by pruning branches. After intensively mapping trunks and branches and imposing treatments, we re-mapped trees three years later to quantify growth increments. Topping stimulated trunk reiteration from the cut, and reiterated tops gained height more rapidly than controls. Tipping also stimulated trunk reiteration from branches, especially when combined with topping, resulting in formation of limbs (i.e., branches giving rise to reiterated trunks). Pruning had consistently negative effects on trunk and branch growth increments even after accounting for post-treatment variation in leaf area and light availability, suggesting that photosynthate was diverted to compartmentalization of wounds. Strategic injury of trees may have long-term conservation value in second-growth redwood forests if limbs can be initiated high enough in the crown to persist as trees approach maximum height. Topping and tipping treatments should be combined with silvicultural thinning of neighboring trees to increase light availability far more than can be achieved by pruning. [ABSTRACT FROM AUTHOR]

Functional traits offer insights into plant performance. However, linking traits to individual tree performance requires considering the overall phenotypic context and utilizing traits measured at the individual level. Tree growth rates are informative metrics about performance, however how it is measured provides different information on tree or community growth dynamics and structure and can alter the relationships or their strength with functional traits and individual phenotypes. Moreover, at the community level, trees can adjust their crown shapes and sizes to optimize canopy space utilization, and a high level of canopy packing can enhance the individual to community-level growth rate. In this study, we assessed the crown area of 1144 individual trees, a simple trait measured at the individual level, and the leaf mass per area (LMA), a common functional trait, to test trait-growth relationships, considering absolute and relative growth rate. The observed total photosynthetic mass (integrating crown area and LMA) positively explained absolute tree growth rates, interacting with tree height. As higher the trees, more intense was the effect. Importantly, this effect was not solely based on LMA or crown size. This implies that LMA's predictive power could improve by integrating leaf-level traits with whole-plant allocation to leaf area. These variables, however, failed to predict the tree's relative growth rates. Additionally, our study found that increased canopy packing levels raised the community-level growth rate within our plots. This overall community growth seems be facilitated by denser tree arrangements and efficient light interception in the forest due to the higher canopy packing level. We highlight the importance of considering crown area as a critical variable to be measured during floristic inventories and in studies focused on tree performance. [ABSTRACT FROM AUTHOR]

Geostatistics was applied to estimate mean tree crown diameter using high spatial resolution aerial photo of three different mangrove forest structure in the Mahakam Delta, East Kalimantan, Indonesia. The variogram analysis of the aerial photo was succeeded to estimate mean tree crown diameter of mangrove forest of Mahakam Delta. The Range of variogram was used to predict mean tree crown diameter and came to result with the error about 6.19% or less than 0.5 m. Based on this study, green band looks promising but still need more exploration and further test especially for complex canopy structure such as lowland Dipterocarp forest which completely distinct to Mangrove forest. Despite of its successness, variogram was not succeed to estimate mean crown diameter of high density mangrove forest plantation sites. Close-spacing plantation suppressed tree crown growth and development. In the aerial photo, it produced smooth texture feature of tree crown with less edge of shadow. Consequently, tree crown became indistinguishable. [ABSTRACT FROM AUTHOR]

This article describes a new algorithm for the detection and delineation of tree crowns using optical sub-meter resolution satellite images. The algorithm focuses on detecting individual semi-isolated trees in a variety of environments defined as trees outside forests (TOF). The concept of Marked Point Processes (MPPs), which alternates phases of “birth” and “death” iterations to satisfy a density factor was used as a theoretical basis. The “mark” in the MPP represents the object being sought. Unlike most applications of MPP to object recognition, the mark used in our algorithm is computed from a 3D geometrical optical model artificially lit using the same illumination parameters as the image itself. Because trees differ in size, the process also incorporates a tree crown radius variable. The algorithm is tested on four sub-meter satellite images, each in a different environment. Validation was performed on both detection and delineation. The detection was based on tree crown counting and yielded an accuracy ranging from 0.81 to 0.95 for the four images. The delineation accuracy was estimated based on the crown pixel count and yielded an accuracy of ≈ 0.63 (0.57–0.72). [ABSTRACT FROM AUTHOR]

The invasive phytopathogen Phytophthora ramorum has caused extensive infection of larch forest across areas of the UK, particularly in Southwest England, South Wales and Southwest Scotland. At present, landscape level assessment of the disease in these areas is conducted manually by tree health surveyors during helicopter surveys. Airborne laser scanning (ALS), also known as LiDAR, has previously been applied to the segmentation of larch tree crowns infected by P. ramorum infection and the detection of insect pests in coniferous tree species. This study evaluates metrics from high-density discrete ALS point clouds (24 points/m 2 ) and canopy height models (CHMs) to identify individual trees infected with P. ramorum and to discriminate between four disease severity categories (NI: not infected, 1: light, 2: moderate, 3: heavy). The metrics derived from ALS point clouds include canopy cover, skewness, and bicentiles (B60, B70, B80 and B90) calculated using both a static (1 m) and a variable (50% of tree height) cut-off height. Significant differences are found between all disease severity categories, except in the case of healthy individuals (NI) and those in the early stages of infection (category 1). In addition, fragmentation metrics are shown to identify the increased patchiness and intra-crown height irregularities of CHMs associated with individual trees subject to heavy infection (category 3) of P. ramorum . Classifications using a k-nearest neighbour (k-NN) classifier and ALS point cloud metrics to classify disease presence/absence and severity yielded overall accuracies of 72% and 65% respectively. The results indicate that ALS can be used to identify individual tree crowns subject to moderate and heavy P. ramorum infection in larch forests. This information demonstrates the potential applications of ALS for the development of a targeted phytosanitary approach for the management of P. ramorum . [ABSTRACT FROM AUTHOR]

The width of individual tree crowns is a critical parameter of canopy structure that can indirectly determine the stand diameter at breast height, tree height, competition, and understory light. Additionally, accurate estimation of crown width is gainful in improving forest carbon storage estimations. However, detecting automatic and accurate crown widths based on visible images is still challenging owing to the influence of crown overlap, shadows, and background. Therefore, we proposed a revised local transect method to detect crown widths from three high-spatial-resolution unmanned aerial vehicle images. Two parameters, the boundary point adjustment parameter (α) and percentile parameter (β), were added to remove the outliers for crown width estimate, reducing the influence of crown overlap, background, and shadow on boundary point detection and delineating the crown width more accurately. The results showed that α had a more significant effect on the crown-width detection accuracy than β. A larger smoothing window size was necessary to improve the accuracy of crown width detection for images with very high spatial resolution. With the optimal α (10%) and smoothing window size, the coefficients of determination between the detected crown width estimates and referenced crown width were between 0.75 and 0.78, and the relative root mean square errors were between 7.50% and 14.65%. Overall, the revised local transect method for detecting crown width is feasible and provides technical support for the accurate detection of crown width based on visible images from unmanned aerial vehicles. • A revised local transect method (RLTM) was proposed to detect crown widths. • Two parameters for improving crown width delineation were used in the RLTM. • The RLTM algorithm achieved high accuracies between 85.35% and 92.50%. • The RLTM was a feasible method for accurately detecting crown width. [ABSTRACT FROM AUTHOR]