Tree Crown Research Articles

YOLOTree-Individual Tree Spatial Positioning and Crown Volume Calculation Using UAV-RGB Imagery and LiDAR Data

Individual tree canopy extraction plays an important role in downstream studies such as plant phenotyping, panoptic segmentation and growth monitoring. Canopy volume calculation is an essential part of these studies. However, existing volume calculation methods based on LiDAR or based on UAV-RGB imagery cannot balance accuracy and real-time performance. Thus, we propose a two-step individual tree volumetric modeling method: first, we use RGB remote sensing images to obtain the crown volume information, and then we use spatially aligned point cloud data to obtain the height information to automate the calculation of the crown volume. After introducing the point cloud information, our method outperforms the RGB image-only based method in 62.5% of the volumetric accuracy. The AbsoluteError of tree crown volume is decreased by 8.304. Compared with the traditional 2.5D volume calculation method using cloud point data only, the proposed method is decreased by 93.306. Our method also achieves fast extraction of vegetation over a large area. Moreover, the proposed YOLOTree model is more comprehensive than the existing YOLO series in tree detection, with 0.81% improvement in precision, and ranks second in the whole series for mAP50-95 metrics. We sample and open-source the TreeLD dataset to contribute to research migration.

Synergizing a Deep Learning and Enhanced Graph-Partitioning Algorithm for Accurate Individual Rubber Tree-Crown Segmentation from Unmanned Aerial Vehicle Light-Detection and Ranging Data

The precise acquisition of phenotypic parameters for individual trees in plantation forests is important for forest management and resource exploration. The use of Light-Detection and Ranging (LiDAR) technology mounted on Unmanned Aerial Vehicles (UAVs) has become a critical method for forest resource monitoring. Achieving the accurate segmentation of individual tree crowns (ITCs) from UAV LiDAR data remains a significant technical challenge, especially in broad-leaved plantations such as rubber plantations. In this study, we designed an individual tree segmentation framework applicable to dense rubber plantations with complex canopy structures. First, the feature extraction module of PointNet++ was enhanced to precisely extract understory branches. Then, a graph-based segmentation algorithm focusing on the extracted branch and trunk points was designed to segment the point cloud of the rubber plantation. During the segmentation process, a directed acyclic graph is constructed using components generated through grey image clustering in the forest. The edge weights in this graph are determined according to scores calculated using the topologies and heights of the components. Subsequently, ITC segmentation is performed by trimming the edges of the graph to obtain multiple subgraphs representing individual trees. Four different plots were selected to validate the effectiveness of our method, and the widths obtained from our segmented ITCs were compared with the field measurement. As results, the improved PointNet++ achieved an average recall of 94.6% for tree trunk detection, along with an average precision of 96.2%. The accuracy of tree-crown segmentation in the four plots achieved maximal and minimal R2 values of 98.2% and 92.5%, respectively. Further comparative analysis revealed that our method outperforms traditional methods in terms of segmentation accuracy, even in rubber plantations characterized by dense canopies with indistinct boundaries. Thus, our algorithm exhibits great potential for the accurate segmentation of rubber trees, facilitating the acquisition of structural information critical to rubber plantation management.

Study on the daily thermal radiation iso-disturbance on a building by trees in summer

Many studies have demonstrated that the cooling and energy-saving effects of planting trees around buildings vary depending on their location. However, most research has focused on a limited number of representative planting sites. This study evaluates the reduction in thermal radiation absorbed by walls over a single day due to trees, defined as daily thermal radiation disturbance (TRDD). Using a combination of experiments and ENVI-met software simulations, we established a coordinate system to map the TRDD distribution of trees at various locations around a building. The findings reveal that trees at different locations exhibit isoTRDD lines, which are symmetrical about the building’s centerline. Trees planted along these isoTRDD lines exert equal effects on TRDD. The isoTRDD lines form a series of ellipses that decay in two stages with increasing distance between the tree crowns and the walls (DW-T): rapid decay when DW-T is less than 3 m and slower decay beyond 3 m. Additionally, we found a linear positive correlation between the change in the sky view factor (ΔSVF) on the wall surface caused by trees and the TRDD. Trees near the north and south walls, as well as those along the building's diagonal, primarily block sky-scattered radiation (R²>0.97). Trees around the east and west walls block both sky-scattered and direct solar short-wave radiation (R²>0.85). This indicates a significant relationship between isoTRDD and SVF of the building. The methodology proposed in this study allows for a comprehensive assessment of the spatial evolution of TRDD for trees around the building, providing a scientific basis for the spatial design of trees to optimize cooling and energy savings. Moreover, the relationship between ΔSVF and TRDD enables rapid calculations and comparisons of the cooling effects of trees at different locations.

Point cloud-based crown volume improves tree biomass estimation: Evaluating different crown volume extraction algorithms

Accurately estimating aboveground biomass (AGB) is increasingly crucial for a better understanding of terrestrial ecosystem carbon balance and its relationship with global climate change. Terrestrial laser scanning (TLS) enables detailed reconstruction of tree crown three-dimensional structures, such as tree crown volume (CV), which has been proven to have a strong correlation with AGB. In this study, a coupled algorithm framework for extracting “optimal biomass” CV to improve AGB estimation using allometric models is proposed. The coupled algorithm framework was compared with commonly used tree CV extraction algorithms (model prediction, convex hull, alpha-shape, and voxel) based on point cloud data from 19 artificial Korean pine plots. The results reveal that the alpha-shape coupled with the voxel algorithm provides the closest approximation to the “optimal biomass” CV, exhibiting the highest correlation with AGB (r = 0.7905). α values and voxel size significantly influence the accuracy of CV extraction, while small variations in point cloud density have minimal impact on the extracted CV. Furthermore, the explanatory power of the “optimal biomass” CV for tree AGB was tested using 30 destructively harvested sampled trees. It was observed that the allometric model incorporating both “optimal biomass” volume and DBH (R2 = 0.966, RMSE = 15.321 kg) outperformed the allometric model incorporating both H and DBH (R2 = 0.953, RMSE = 16.131 kg) and the DBH-only model (R2 = 0.948, RMSE = 16.615 kg), particularly in reducing underestimation of AGB for larger trees. These findings demonstrate that the proposed alpha-shape coupled with the voxel algorithm provides an effective solution for extracting “optimal biomass” CV, significantly improving the accuracy of tree-level AGB estimation.

Is Your Training Data Really Ground Truth? A Quality Assessment of Manual Annotation for Individual Tree Crown Delineation

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.

Assessing receptivity to malaria using case surveillance and forest data in a near-elimination setting in northeast Thailand

BackgroundThailand aimed to eliminate malaria by 2024, and as such is planning for future prevention of re-establishment in malaria free provinces. Understanding the receptivity of local areas to malaria allows the appropriate targeting of interventions. Current approaches to assessing receptivity involve collecting entomological data. Forest coverage is known to be associated with malaria risk, as an environment conducive to both vector breeding and high-risk human behaviours.MethodsGeolocated, anonymized, individual-level surveillance data from 2011 to 2021 from the Thai Division of Vector-Borne Disease (DVBD) was used to calculate incidence and estimated Rc at village level. Forest cover was calculated using raster maps of tree crown cover density and year of forest loss from the publicly available Hansen dataset. Incidence and forest cover were compared graphically and using Spearman’s rho. The current foci classification system was applied to data from the last 5 years (2017–2021) and forest cover for 2021 compared between the classifications. A simple risk score was developed to identify villages with high receptivity.ResultsThere was a non-linear decrease in annual cases by 96.6% (1061 to 36) across the two provinces from 2011 to 2021. Indigenous Annual Parasite Index (API) and approximated Rc were higher in villages in highly forested subdistricts, and with higher forest cover within 5 km. Forest cover was also higher in malaria foci which consistently reported malaria cases each year than those which did not. An Rc > 1 was only reported in villages in subdistricts with > 25% forest cover. When applying a simple risk score using forest cover and recent case history, the classifications were comparable to those of the risk stratification system currently used by the DVBD.ConclusionsThere was a positive association between forest coverage around a village and indigenous malaria cases. Most local transmission was observed in the heavily forested subdistricts on the international borders with Laos and Cambodia, which are where the most receptive villages are located. These areas are at greater risk of importation of malaria due to population mobility and forest-going activities. Combining forest cover and recent case surveillance data with measures of vulnerability may be useful for prediction of malaria recurrence risk.

Spatial digital twin framework for overheight vehicle warning and re-routing system

Overhead road obstacles present a significant logistical hazard to the heavy vehicle industry. Traditional overheight vehicle warning systems such as passive warning systems (PWS) and active warning systems (AWS) have not adequately reduced the frequency and impact of overheight incidents, encouraging transportation agencies to employ intelligent transport system (ITS) strategies using state-of-the-art advanced technologies. This research takes an innovative approach in developing an immersive user-focused experience, harnessing multi-disciplinary methods and tools to engineer a spatial digital twin prototype for a novel Internet-of-Things (IoT)-based active warning alert and re-routing system (AWARS). LiDAR and 3D GIS were used to model the complex road environment, tailored to the strict fiscal objectives sought by economically mindful organisations. Tree crowns were extracted from near-Infrared aerial imagery and digital elevation models, supplying the dimensions necessary for 3D tree modelling. IoT connectivity was configured using a real-time analytics approach to deliver alerts and re-routing options. The World Traffic Service with live and predictive traffic data was used for the routing application programming interface (API). A standard-configuration common rigid truck (CRT) was inserted into the 3D road environment model to simulate overheight collisions and to ascertain the effect of re-routing on estimated time of arrival (ETA). Longer ETA durations were observed for routes computed by the digital twin. Theoretically, enhanced situational awareness and subsequent reduction of risk likelihood suggests an optimised response to industry demands, despite extended travel times, cultivating a favorable impact on the supply chain through enhanced safety management.

Topgrafting as a tool in operational Scots pine breeding

Effective breeding requires multiplying desired genotypes, keeping them at a convenient location to perform crosses more efficiently, and building orchards to generate material for reforestation. While some of these aims can be achieved by conventional grafting involving only rootstock and scion, topgrafting is known to deliver all in a shorter time span. In this study, Scots pine scions were grafted onto the upper and lower tree crowns in two clonal archives with the aim of inducing early female and male strobili production, respectively. Their survival rates and strobili production were analyzed with generalized linear mixed models. Survival was low (14%) to moderate (41%), and mainly affected by the topgraft genotype, interstock genotype, crown position and weather conditions in connection with the grafting procedure. Survival was not affected by the cardinal position in the crown (south or north). Male flowering was ample three years after grafting and reached 56% in the first year among live scions, increasing to 62 and 59% in consecutive years. Female flowering was scarce and was 9% at first, later increasing to 26 and 20% of living scions but was strongly affected by the topgraft genotype. In one subset of scions, female flowering was observed 1 year after grafting. Overall, flowering success was mainly affected by the topgraft and interstock genotypes, and secondary growth of scions. This is one of few reports on topgrafting in functional Scots pine clonal archives.

Satellite monitoring of a pine forest affected by mountain pine beetle

The spread of bark beetles poses a significant threat to forest ecosystems, causing substantial damage to coniferous forests. The mountain pine beetle (Dendroctonus spp.) is considered to be one of the most harmful bark beetles, infesting pine trees and leading to severe consequences for forest ecosystems. This article aims to detect and analyze the dynamics of mountain pine beetle infestation in a pine forest stand, determine its affected area, and evaluate the effectiveness of sanitary and restoration measures. The study utilized satellite data from the Sentinel-2 mission and employed the Normalized Difference Vegetation Index (NDVI) and Normalized Difference Moisture Index (NDMI) to monitor the spread of mountain pine beetle infestation. The analysis was conducted for a pine forest stand in the Ostriv Forestry of the Vyshchedubechanske Forestry Enterprise in Kyiv Oblast, Ukraine. The results demonstrated the effectiveness of using satellite data and vegetation indices for monitoring mountain pine beetle infestation. The combined application of NDVI and NDMI enabled the detection of infested areas in the early stages by capturing changes in the spectral characteristics of tree crowns and vegetation moisture content. The satellite data analysis confirmed the complete infestation of a 0.01 km2 pine forest stand by the mountain pine beetle in 2017. A link between the activation of beetle spread and climate changes, particularly rising temperatures and decreasing precipitation, was observed, leading to the weakening of pine stands. The study revealed that in 2018, sanitary felling of the infested pine stand was carried out, as evidenced by the decrease in NDVI values. In 2019, a project to restore the forest ecosystem through the establishment of a new pine plantation was initiated. Satellite observations in April 2024 demonstrated successful growth of the young pine forest, with NDVI values indicating an increase from 0.25 in April 2020 to 0.45 in April 2024, reflecting active vegetation growth and successful ecosystem restoration. The obtained results highlight the importance of using satellite data for timely detection of pest outbreaks in forest ecosystems, determining the extent of infestation, monitoring the effectiveness of sanitary measures, and assessing the recovery of forest stands after restoration efforts. The study underscores the potential of satellite remote sensing techniques in forest ecosystem monitoring and supports decision-making processes for sustainable forest management.

Individual fertilization response with social status and crown characteristics for western hemlock (Tsuga heterophylla)

Due to their high variability, the growth responses of western hemlock ( Tsuga heterophylla (Raf.) Sarg.) stands to fertilization have been regarded as unresponsive or inconsistent. Tree-level fertilization response models for western hemlock were constructed to clarify the inconsistent stand-level responses, using extensive datasets from the United States and Canada. Tree growth in diameter and height, mortality, were assessed by fertilization, stand- and tree-level variables using generalized linear and nonlinear mixed-effects models. Western hemlock fertilization response was influenced not only by stand-level variables and application rates but also by tree social status and crown characteristics. Greater dominance and crown ratio in trees increased height growth responses but decreased diameter growth. The peak of fertilization effect on diameter growth was observed 3 years after application, while effect on height growth peaked at 6 years. Fertilization increased the overall mortality rate, but also ameliorated it for suppressed and steady-growing trees. These complicated fertilization responses of western hemlock may imply its tolerance and allocation strategy of resources at a given size and social status. The results highlight the importance of adopting a tree-level modeling approach to better understanding how western hemlock trees respond to fertilization, improving the accuracy of growth and yield prediction after fertilization.

Evaluating the transferability of airborne laser scanning derived stem size prediction models for Pinus taeda L. stem size estimation to two different locations and acquisition specifications

ABSTRACT Airborne laser scanning (ALS) datasets are used widely for estimating forest biometrics. The transferability of predictive models among ALS acquisitions is a topic of research due to differences in timing, flight parameters, equipment specifications, environmental conditions, and processing methods. The transferability of predictive models therefore is subject to uncertainty. This paper presents an evaluation of the transferability of models for the estimation of stem volume and diameter at breast height (DBH) based on individual tree crown size and competitive neighbourhood metrics derived for managed loblolly pine (Pinus taeda) and slash pine (Pinus elliottii) forest in the Southern USA. Two predictive models types were tested: multiple linear regression (MLR) and Rand Forest (RF). We also evaluated the inclusion of additional training data to model development. Models were able to be transferred to other locations with similar structural and management conditions as the original training dataset with little decrease in accuracy, specifically unthinned stands, despite different ALS acquisitions (Plot stem volume: R2 0.7–0.8; NRMSE 10–12%; mean DBH: R2 0.4–0.7; NRMSE 10–17%; plot basal area: R2 0.7–0.8; NRMSE 12%). Increases in structural differences between the training and test data, driven by age or thinning status, introduced unacceptable levels of uncertainty (Stem volume: R2 0.4–0.7; NRMSE 12–16%; mean DBH: R2 0.4–0.5; NRMSE 18–20%; plot basal area: R2 0.5–0.6; NRMSE 22–40%). Generally, RF models most accuracy estimated DBH, and MLR for stem volume. Improvements to estimate accuracy can be achieved through the addition of relatively small datasets, representing features which were not present in the original data. ALS’s ability to provide accurate and near-complete inventories of forests hold a great deal of potential for forest management. The existence of a transferable model that can be used across different acquisitions represents a saving in terms of cost and time, we would argue that future research is therefore warranted.

A Semi-Automatic Approach for Tree Crown Competition Indices Assessment From UAV LiDAR

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.

History of apple breeding and breeding achievements at FSBSO ARHCBAN

New apple varieties should possess a set of adaptive traits that provide yield, stable fruiting, and highly marketable fruits with a long period of postharvest storage, thus ensuring the profitability of their use in industrial plantings. According to the State Register, the central region of the Russian Federation is one of the leading regions in terms of the number of apple varieties approved for use. Long-term research of scientists and specialists from FSBSO ARHCBAN has greatly contributed to the development and improvement of the assortment both in the region and beyond. The key directions for apple breeding at the institution include improving winter hardiness, resistance to scab, and the columnar shape of the crown. The review article presents the history of shaping the directions for breeding work and modern achievements of FSBSO ARHCBAN in apple breeding, as well as an analysis of the long-term focused scientific activity of breeders. The most significant breeding results were achieved by Prof. V. Kichina, who, together with his students and followers, was the first in Russia to create several columnar apple varieties that offer high winter hardiness, resistance to scab, and marketable qualities of fruits (‘Dialog’, ‘Lukomor’, ‘Malyukha’, ‘Ostankino’, ‘President’, and ‘Chervonets’), as well as fundamentally new high-yielding varieties with a conventional crown shape that possess high adaptability and resistance to pathogens (‘Arkadik’, ‘Legendа’, ‘Marat Busurin’, and ‘Mayak Zagorye’). The article presents the main characteristics of breeding achievements of the last decades: tree crown shape; ripening time of fruits and their qualitative and quantitative indicators (tasting evaluation, average weight, shape, etc.); yield; resistance to abiotic and biotic factors. Data are presented on the use of apple varieties bred by FSBSO ARHCBAN both in agricultural production (on the basis of license agreements) and in scientific and technological cooperation with Russian National Research Universities (inclusion in breeding programs). The apple breeding activities conducted at FSBSO ARHCBAN are described, and the main directions for breeding work are outlined.

Effects of Ripening Phase and Cultivar under Sustainable Management on Fruit Quality and Antioxidants of Sweet Cherry

Sweet cherry grown under sustainable management produces highly valuable fruits, whose quality shows important biochemical and morphological changes during ripening. Research was carried out in Iasi (Romania), with the aim to assess the quality characteristics of the sweet cherry fruits of three cultivars (Van, Andreiaș, Margonia), grown in an inner or outer position inside the tree crown, at the pre-ripening or full ripeness phase. In 2022, the colour component a* showed higher values in cv. Van and Andreias red fruits and in an inner position, whereas the components L* and b* at the full ripeness phase were highest in cv. Margonia. The dry matter and total soluble solids contents increased from the pre-ripening to the full ripeness phase and were highest in cv. Van sweet cherry fruits; the DM of fruit from the outer part of crown was higher than that of fruit from the inner part at the pre-ripening phase. The content of phenolics was the highest in cv. Margonia fruits at the pre-ripening stage and in cv. Van at the full ripeness phase and higher in the inner tree crown zones. The cultivar Margonia generally showed the highest vitamin C content in both years and development phases. The yellow fruit cv. Margonia mostly showed the highest values of chlorophyll a and b. The fruit’s content of carotene, lycopene, and anthocyanins was generally the highest in the red fruits of cv. Andreias. The examined sweet cherry cultivars showed a high variability in fruit nutritional quality and proved to be a rich source of bioactive compounds with antioxidant potential.

Individual tree detection and crown delineation in the Harz National Park from 2009 to 2022 using mask R–CNN and aerial imagery

Forest diebacks pose a major threat to global ecosystems. Identifying and mapping both living and dead trees is crucial for understanding the causes and implementing effective management strategies. This study explores the efficacy of Mask R–CNN for automated forest dieback monitoring. The method detects individual trees, delineates their crowns, and classifies them as alive or dead. We evaluated the approach using aerial imagery and canopy height models in the Harz Mountains, Germany, a region severely affected by forest dieback. To assess the model's ability to track changes over time, we applied it to images from three separate flight campaigns (2009, 2016, and 2022). This evaluation considered variations in acquisition dates, cameras, post-processing techniques, and image tilting. Forest changes were analyzed based on the detected trees' number, spatial distribution, and height. A comprehensive accuracy assessment demonstrated the Mask R–CNN's robust performance, with precision scores ranging from 0.80 to 0.88 and F1-scores from 0.88 to 0.91. These results confirm the model's ability to generalize across diverse image acquisition conditions. While minor changes were observed between 2009 and 2016, the period between 2016 and 2022 witnessed substantial dieback, with a 64.57% loss of living trees. Notably, taller trees appeared to be particularly affected. This study highlights Mask R–CNN's potential as a valuable tool for automated forest dieback monitoring. It enables efficient detection, delineation, and classification of both living and dead trees, providing crucial data for informed forest management practices.

Trunk distortion weakens the tree productivity revealed by half-sib progeny determination of Pinus yunnanensis

Twisted trunks are not uncommon in trees, but their effects on tree growth are still unclear. Among coniferous tree species, the phenomenon of trunk distortion is more prominent in Pinus yunnanensis. To expand the germplasm of genetic resources, we selected families with excellent phenotypic traits to provide material for advanced generation breeding. The progeny test containing 93 superior families (3240 trees) was used as the research material. Phenotypic measurements and estimated genetic parameters (family heritability, realistic gain and genetic gain) were performed at 9, 15, and 18 years of age, respectively. The genetic evaluation yielded the following results (1) The intra-family variance component of plant height (PH) was greater than that of the inter-family, while the inter-family variance components of other traits (diameter at breast height (DBH), crown diameter (CD), height under branches (HUB), degree of stem-straightness (DS)) were greater than that of the intra-family, indicating that there was abundant variation among families and potential for selection. (2) At half rotation period (18 years old), there was a significant correlation among the traits. The proportion of trees with twisted trunks (level 1–3 straightness) reached 48%. The DS significantly affected growth traits, among which PH and DBH were the most affected. The volume loss rate caused by twisted trunk was 18.06-56.75%, implying that trunk distortion could not be completely eliminated after an artificial selection. (3) The influence of tree shape, crown width, and trunk on volume increased, and the early-late correlation between PH, DBH and volume was extremely significant. The range of phenotypic coefficient of variation, genetic variation coefficient and family heritability of growth traits (PH, DBH, and volume) were 44.29-127.13%, 22.88-60.87%, and 0.79–0.83, respectively. (4) A total of 21 superior families were selected by the method of membership function combined with independent selection. Compared with the mid-term selection (18 years old), the accuracy of early selection (9 years old) reached 77.5%. The selected families’ genetic gain and realistic gain range were 5.79-19.82% and 7.12-24.27%, respectively. This study can provide some useful reference for the breeding of coniferous species.

Observation of sunlight diffraction through tree twigs and leaves

This work reports the observation of sunlight diffraction through empty spaces in between twigs and leaves in tree crowns. The diffraction patterns are dominated by red wavelength on front followed by the trail of rest of spectral bands of white light rainbow bursts. Tree crowns act as obstacles and apertures causing diffraction/ interference of solar rays. Separation between leaves and twigs (d) is a few inches and distance between crown and camera (D) is in range of a few meters. Empty spaces in tree crown are much longer than red to violet wavelengths in sunlight therefore scattering, dispersion, reflection, diffraction and interference effects cannot be ruled out. Tree crown deflects solar rays into narrow angles all around to enable interference inside camera lens to create grating structure diffracting light into regular patterns. Diffraction pattern look like red petunia flower having 3 × 3 rose petals. Motorola mobile phone camera was used to record the sunlight diffraction patterns. Mobile camera consists of small focal length (≈5 mm) lens, a fixed opening aperture and an image sensor. Lens consists of few plastic or glass elements and sensor consists of charge coupled device (CCD) or CMOS technology. These diffraction/ interference patterns have potential to increase efficiency of Next Gen solar cells. This paper describes possible theoretical background of light diffraction and multiple ray interference through tree twigs and leaves.

Att-Mask R-CNN: an individual tree crown instance segmentation method based on fused attention mechanism

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.

Multitemporal airborne imaging spectrometry and fluorometry reveal contrasting photoprotective responses of trees

The Photochemical Reflectance Index (PRI) and solar induced fluorescence (SIF) provide information on plant photosynthetic activity. PRI and SIF are both strongly influenced by irradiance, but uncertainties related to the interpretation of these light responses at large spatial scales remain, partly due to a shortage of suitable data from aircraft or satellite platforms. The goal of this study was to explore interpretations of the PRI- and SIF-light responses of trees owing to species, functional types (evergreen and deciduous) and season. Using airborne hyperspectral and ultraspectral imagery in a North American urban forest, we derived PRI, SIF, and albedo (an indicator of illumination) at the 1-m pixel level. We then quantified crown-level PRI and SIF light responses of ten different tree species at three time points from late-summer to autumnal senescence using hierarchical models. Our results confirmed that both PRI and SIF were strongly influenced by illumination with PRI decreasing and SIF increasing with illumination. Both slope and intercept of the PRI-albedo relationship changed with season, but the pattern varied among species and functional types. SIF values decreased during autumnal senescence for all species, but evergreen species exhibited less seasonal decline in the slope of SIF-albedo relationship compared to deciduous species. The PRI and SIF light responses derived from the airborne imagery offer complementary information on dynamic photosynthesis responses presumably due to varying canopy structure, pigmentation and photoprotection among species and functional types. From airborne platforms, PRI- and SIF-light responses can be used to explore the contrasting physiological responses of individual tree crowns, providing a spatially and temporally explicit view of dynamic plant traits related to photoregulation and a novel view of functional diversity for entire landscapes.

Impact of urban tree arrangement on pedestrian exposure to the size-fractional particulate matter in a city boulevard

Trees act as natural filters that mitigate roadside air pollution. However, the filtration impact of different tree arrangements on traffic pollutants with different particle diameters has rarely been analysed in real street canyon environments. To quantify how roadside tree arrangements impact pedestrian exposure to particle number concentrations (PNCs) of different diameters (0.25–32 μm), in situ field measurements were carried out in a boulevard-type street canyon in the city of Xi'an, China. This study analysed the experimental data of PNCs collected along segments of a pedestrian lane under four typical tree arrangements: open space without trees, a sparse-spaced tree arrangement, a medium-spaced tree arrangement, and a dense-spaced tree arrangement in a street canyon. Our results reveal that the effect of tree arrangement on PNCs depended on the particle diameter. In general, trees can significantly reduce coarse PNC (particles with diameters >2.5 μm) but not the fine PNC. Quantitative analysis showed that a medium-spaced tree arrangement, in which tree crowns are adjacent to each other but do not overlap, is the most capable of reducing PNC, followed by a sparse-spaced tree arrangement, while a the dense-spaced tree arrangement has the least impact. The attenuation effect of trees on the PNCs increased with increasing particle diameter. Moreover, the presence of trees altered the local microclimate, which also affected how exposure to PNCs changed. Our empirical findings further highlight the complexity of how trees affect particulate pollutants in street canyons and provide timely insights for enhancing tree-planning management in cities from the perspective of air quality improvement.