Weakly Supervised Forest Canopy Extraction and Multi-dimensional Joint Canopy Entropy for Quantifying Canopy Structural Complexity Using Large-Scale Forest UAV LiDAR Data

The development and management of green open spaces are essential in overcoming environmental problems such as air pollution and urban warming. 3D modeling and biomass calculation are the example efforts in managing green open spaces. In this study, 3D modeling was carried out on point clouds data acquired by the UAV photogrammetry and UAV LiDAR methods. 3D modeling is done explicitly using the point clouds fitting method. This study uses three fitting methods: the spherical fitting method, the ellipsoid fitting method, and the spherical harmonics fitting method. The spherical harmonics fitting method provides the best results and produces an R2 value between 0.324 to 0.945. In this study, Above-Ground Biomass (AGB) calculations were also carried out from the modeling results using three methods with UAV LiDAR and Photogrammetry data. AGB calculation using UAV LiDAR data gives better results than using photogrammetric data. AGB calculation using UAV LiDAR data gives an accuracy of 78% of the field validation results. However, for visualization purposes with a not-too-wide area, a 3D model of photogrammetric data using the spherical harmonics method can be used.

Traditional forest inventories are based on field surveys of established sample plots, which involve field measurements of individual trees within a sample plot and the selection of proper allometric equations for tree volume calculation. Thus, accurate field measurements and properly selected allometric equations are two crucial factors for providing high-quality tree volumes. One key problem is the difficulty in accurately acquiring tree height data, resulting in high uncertainty in tree volume calculation when the diameter at breast height (DBH) alone is used. This study examined the uncertainty of tree height measurements using different means and the impact of allometric models on tree volume estimation accuracy. Masson pine and eucalyptus plantations in Fujian Province, China, were selected as examples; their tree heights were measured three ways: using an 18-m telescopic pole, UAV Lidar (unmanned aerial vehicle, light detection and ranging) data, and direct measurement of felled trees, with the latest one as a reference. The DBH-based and DBH–height-based allometric equations corresponding to specific tree species were used for the calculations of tree volumes. The results show that (1) tree volumes calculated from the DBH-based models were lower than those from the DBH–height-based models. On average, tree volumes were underestimated by 0.018 m3 and 0.117 m3 for Masson pine and eucalyptus, respectively, while the relative root-mean-squared errors (RMSEr) were 24.04% and 33.90%, respectively, when using the DBH-based model; (2) the tree height extracted from UAV Lidar data was more accurate than that measured using a telescopic pole, because the pole measurement method generally underestimated the tree height, especially when the trees were taller than the length of the pole (18 m in our study); (3) the tree heights measured using different methods greatly impacted the accuracies of tree volumes calculated using the DBH–height model. The telescopic-pole-measured tree heights resulted in a relative error of 9.1–11.8% in tree volume calculations. This research implies that incorporation of UAV Lidar data with DBH field measurements can effectively improve tree volume estimation and could be a new direction for sample plot data collection in the future.

Forest canopy structural complexity (CSC), the intricate arrangement and occupation of canopy elements in three-dimensional space, plays a critical role in shaping forest ecosystem productivity and stability by regulating light and water distribution within the canopy. However, the relationship between forest CSC and forest ecosystem productivity and stability remains controversial in current regional-scale studies, necessitating further investigation at broader spatial scales. Here, we introduce a novel entropy-based metric, canopy entropy, to quantify forest CSC from light detection and ranging (lidar) data. This metric effectively captures forest CSC variations arising from both horizontal and vertical arrangements and occupations of canopy elements. Notably, canopy entropy estimates from multiplatform lidar data demonstrate strong agreement, establishing its suitability for large-scale applications. Leveraging these advantages, as well as airborne lidar data from 4,000 forest plots worldwide and spaceborne lidar data from the Global Ecosystem Dynamics Investigation, we map the global distribution of forest CSC and investigate its relationships with forest ecosystem productivity and stability. We find climatic factors, especially water availability, play a critical role in driving the global distribution of forest CSC, while biotic factors exhibit a strong coupling impact with climatic and edaphic factors. From a global perspective, forest CSC predominantly enhances productivity and stability, although substantial variations are observed among forest ecoregions. The effects of forest CSC on productivity and stability are the balanced results of biodiversity and resource availability. These results offer valuable insights into understanding controversies in regional-scale studies. Furthermore, we found that managed forests generally exhibit lower CSC compared to intact forests but demonstrate stronger positive effects of CSC on ecosystem productivity and stability, underscoring the urgent need to incorporate CSC into forest management strategies to enhance climate change mitigation efforts.

Rivers and riparian areas are vital components of ecosystems, but accurately modeling their terrain presents challenges, especially in detecting the river surface. This paper proposes an integrated approach that combines UAV LiDAR and Single Beam Echo Sounder (SBES) data to construct a Digital Terrain Model (DTM) of river and riparian areas. The objective is to overcome the limitations posed by water, which absorbs near-infrared laser energy, resulting in weak or absent LiDAR returns. Different UAV LiDAR densities were examined to determine the optimal configuration for capturing riparian areas. Evaluation of the results utilized various metrics, including root mean square error (RMSE), mean square error (MSE), mean absolute error (MAE), mean bias error (MBE), and correlation coefficient (CC). Three ground filtering methods were implemented and assessed: morphological filters, adaptive triangulated irregular network (ATIN) filtering, and above-ground level (AGL) filtering. Among the evaluated methods, the DTM constructed using ATIN with an 80-meter flight configuration yielded the most accurate results. It achieved an RMSE of 0.18m, an MSE of 0.03m, an MAE of 0.17m, an MBE of 9.08m, and a CC of 1.00. Comparatively, other methods exhibited higher error values and lower correlation coefficients. The findings highlight the efficacy of ATIN filtering in conjunction with an 80-meter UAV LiDAR flight for obtaining reliable DTMs of river and riparian areas. This approach demonstrates significant improvement in accuracy, particularly in terms of RMSE and MSE. The derived DTM can be a valuable tool for safeguarding and managing these critical ecosystems. In summary, this paper successfully addresses the challenge of modeling river and riparian terrains by integrating UAV LiDAR and SBES data. By employing ATIN filtering with an 80-meter flight configuration, the study achieves a highly accurate DTM. By employing ATIN filtering with an 80-meter flight configuration, the study achieves a precise, high DTM with minimal error. The developed model contributes to protecting and preserving river and riparian ecosystems.

Effects of thinning on canopy structure, forest productivity, and productivity stability in mixed conifer-broadleaf forest: Insights from a LiDAR survey.

The even-aged northern hardwood forests of the Upper Great Lakes Region are undergoing an ecological transition during which structural and biotic complexity is increasing. Early-successional aspen (Populus spp.) and birch (Betula papyrifera) are senescing at an accelerating rate and are being replaced by middle-successional species including northern red oak (Quercus rubra), red maple (Acer rubrum), and white pine (Pinus strobus). Canopy structural complexity may increase due to forest age, canopy disturbances, and changing species diversity. More structurally complex canopies may enhance carbon (C) sequestration in old forests. We hypothesize that these biotic and structural alterations will result in increased structural complexity of the maturing canopy with implications for forest C uptake. At the University of Michigan Biological Station (UMBS), we combined a decade of observations of net primary productivity (NPP), leaf area index (LAI), site index, canopy tree-species diversity, and stand age with canopy structure measurements made with portable canopy lidar (PCL) in 30 forested plots. We then evaluated the relative impact of stand characteristics on productivity through succession using data collected over a nine-year period. We found that effects of canopy structural complexity on wood NPP (NPPw) were similar in magnitude to the effects of total leaf area and site quality. Furthermore, our results suggest that the effect of stand age on NPPw is mediated primarily through its effect on canopy structural complexity. Stand-level diversity of canopy-tree species was not significantly related to either canopy structure or NPPw. We conclude that increasing canopy structural complexity provides a mechanism for the potential maintenance of productivity in aging forests.

Forest-canopy closure (FCC) reflects the coverage of the forest tree canopy, which is one of the most important indicators of forest structure and a core parameter in forest resources investigation. In recent years, the rapid development of UAV LiDAR and photogrammetry technology has provided effective support for FCC estimation. However, affected by factors such as different tree species and different stand densities, it is difficult to estimate FCC accurately based on the single-tree canopy-contour method in complex forest regions. Thus, this study proposes a method for estimating FCC accurately using algorithm integration with an optimal window size for treetop detection and an optimal algorithm for crown-boundary extraction using UAV LiDAR data in various scenes. The research results show that: (1) The FCC estimation accuracy was improved using the method proposed in this study. The accuracy of FCC in a camphor pine forest (Pinus sylvestris var. mongolica Litv.) was 89.11%, with an improvement of 6.77–11.25% compared to the results obtained from other combined conditions. The FCC accuracy for white birch (White birch platyphylla Suk) was about 87.53%, with an increase of 3.25–8.42%. (2) The size of the window used for treetop detection is closely related to tree species and stand density. With the same forest-stand density, the treetop-detection window size of camphor pine was larger than that of white birch. The optimal window size of camphor pine was between 5 × 5~11 × 11 (corresponding 2.5~5.5 m), while that of white birch was between 3 × 3~7 × 7 (corresponding 1.5~3.5 m). (3) There are significant differences in the optimal-canopy-outline extraction algorithms for different scenarios. With a medium forest-stand density, the marker-controlled watershed (MCW) algorithm has the best tree-crown extraction effect. The region-growing (RG) method has better extraction results in the sparse areas of camphor pine and the dense areas of white birch. The Voronoi tessellation (VT) algorithm is more suitable for the dense areas of camphor pine and the sparse regions of white birch. The method proposed in this study provides a reference for FCC estimation using high-resolution remote-sensing images in complex forest areas containing various scenes.

Accurately understanding the stand composition of shelter forests is essential for the construction and benefit evaluation of shelter forest projects. This study explores classification methods for dominant tree species in shelter forests using UAV-derived RGB, hyperspectral, and LiDAR data. It also investigates the impact of individual tree crown (ITC) delineation accuracy, crown morphological parameters, and various data sources and classifiers. First, as a result of the overlap and complex structure of tree crowns in shelterbelt forests, existing ITC delineation methods often lead to over-segmentation or segmentation errors. To address this challenge, we propose a watershed and multi-feature-controlled spectral clustering (WMF-SCS) algorithm for ITC delineation based on UAV RGB and LiDAR data, which offers clearer and more reliable classification objects, features, and training data for tree species classification. Second, spectral, texture, structural, and crown morphological parameters were extracted using UAV hyperspectral and LiDAR data combined with ITC delineation results. Twenty-one classification images were constructed using RF, SVM, MLP, and SAMME for tree species classification. The results show that (1) the proposed WMF-SCS algorithm demonstrates significant performance in ITC delineation in complex mixed forest scenarios (Precision = 0.88, Recall = 0.87, F1-Score = 0.87), resulting in a 1.85% increase in overall classification accuracy; (2) the inclusion of crown morphological parameters derived from LiDAR data improves the overall accuracy of the random forest classifier by 5.82%; (3) compared to using LiDAR or hyperspectral data alone, the classification accuracy using multi-source data improves by an average of 7.94% and 7.52%, respectively; (4) the random forest classifier combined with multi-source data achieves the highest classification accuracy and consistency (OA = 90.70%, Kappa = 0.8747).

The fine classification of mangroves plays a crucial role in enhancing our understanding of their structural and functional aspects which has significant implications for biodiversity conservation, carbon sequestration, water quality enhancement, and sustainable development. Accurate classification aids in effective mangrove management, protection, and preservation of coastal ecosystems. Previous studies predominantly relied on passive optical remote sensing images as data sources for mangrove classification, often overlooking the intricate vertical structural complexities of mangrove species. In this study, we address this limitation by incorporating unmanned aerial vehicle-LiDAR (UAV-LiDAR) point cloud 3D data with UAV hyperspectral imagery to perform multivariate classification of mangrove species. Five distinct variable scenarios were employed: band characteristics (S1), vegetation index (S2), texture measures (S3), fused hyperspectral characteristics (S4), and a canopy height model (CHM) combined with UAV hyperspectral characteristics and LiDAR point cloud data (S5). To execute this classification task, an extreme gradient boosting (XGBoost) machine learning algorithm was employed. Our investigation focused on the estuary of the Pinglu Canal, situated within the Maowei Sea of the Beibu Gulf in China. By comparing the classification outcomes of the five variable scenarios, we assessed the unique contributions of each variable to the accurate classification of mangrove species. The findings underscore several key points: (1) The fusion of multiple features in the image scenario led to a higher overall accuracy (OA) compared to models that employed individual features. Specifically, scenario S4 achieved an OA of 88.48% and scenario S5 exhibited an even more impressive OA of 96.78%. These figures surpassed those of the individual feature models where the results were S1 (83.35%), S2 (83.55%), and S3 (71.28%). (2) Combining UAV hyperspectral and LiDAR-derived CHM data yielded improved accuracy in mangrove species classification. This fusion ultimately resulted in an OA of 96.78% and kappa coefficient of 95.96%. (3) Notably, the incorporation of data from individual bands and vegetation indices into texture measures can enhance the accuracy of mangrove species classification. The approach employed in this study—a combination of the XGBoost algorithm and the integration of UAV hyperspectral and CHM features from LiDAR point cloud data—proved to be highly effective and exhibited strong performance in classifying mangrove species. These findings lay a robust foundation for future research efforts focused on mangrove ecosystem services and ecological restoration of mangrove forests.

The accurate classification of single tree species in forests is important for assessing species diversity and estimating forest productivity. However, few studies have explored the influence of canopy morphological characteristics on the classification of tree species. Therefore, based on UAV LiDAR and hyperspectral data, in this study, we designed various classification schemes for the main tree species in the study area, i.e., birch, Manchurian ash, larch, Ulmus, and mongolica, in order to explore the effects of different data sources, classifiers, and canopy morphological features on the classification of a single tree species. The results showed that the classification accuracy of a single tree species using multisource remote sensing data was greater than that based on a single data source. The classification results of three different classifiers were compared, and the random forest and support vector machine classifiers exhibited similar classification accuracies, with overall accuracies above 78%. The BP neural network classifier had the lowest classification accuracy of 75.8%. The classification accuracy of all three classifiers for tree species was slightly improved when UAV LiDAR-extracted canopy morphological features were added to the classifier, indicating that the addition of canopy morphological features has a certain relevance for the classification of single tree species.

Vegetation structural complexity and biodiversity tend to be positively correlated, but understanding of this relationship is limited in part by structural metrics tending to quantify only horizontal or vertical variation, and that do not reflect internal structure. We developed new metrics for quantifying internal vegetation structural complexity using terrestrial LiDAR scanning and applied them to 12 NEON forest plots across an elevational gradient in Great Smoky Mountains National Park, USA. We asked (1) How do our newly developed structure metrics compare to traditional metrics? (2) How does forest structure vary with elevation in a high‐biodiversity, high topographic complexity region? (3) How do forest structural metrics vary in the strength of their relationships with vascular plant biodiversity? Our new measures of canopy density (Depth) and structural complexity (σDepth), and their canopy height‐normalized counterparts, were sensitive to structural variations and effectively summarized horizontal and vertical dimensions of structural complexity. Forest structure varied widely across plots spanning the elevational range of GRSM, with taller, more structurally complex forests at lower elevation. Vascular plant biodiversity was negatively correlated with elevation and more strongly positively correlated with vegetation structure variables. The strong correlations we observed between canopy structural complexity and biodiversity suggest that structural complexity metrics could be used to assay plant biodiversity over large areas in concert with airborne and spaceborne platforms.

Accurately estimating aboveground biomass (AGB) is crucial for assessing carbon storage in forest ecosystems. However, traditional field survey methods are time-consuming, and vegetation indices based on optical remote sensing are prone to saturation effects, potentially underestimating AGB in subtropical forests. To overcome these limitations, we propose an improved approach that combines three-dimensional (3D) forest structure data collected using unmanned aerial vehicle light detection and ranging (UAV LiDAR) technology with ground measurements to apply a binary allometric growth equation for estimating and mapping the spatial distribution of AGB in subtropical forests of China. Additionally, we analyze the influence of terrain factors such as elevation and slope on the distribution of forest biomass. Our results demonstrate a high accuracy in estimating tree height and diameter at breast height (DBH) using LiDAR data, with an R2 of 0.89 for tree height and 0.92 for DBH. In the study area, AGB ranges from 0.22 to 755.19 t/ha, with an average of 121.28 t/ha. High AGB values are mainly distributed in the western and central-southern parts of the study area, while low AGB values are concentrated in the northern and northeastern regions. Furthermore, we observe that AGB in the study area exhibits an increasing trend with altitude, reaching its peak at approximately 1650 m, followed by a gradual decline with further increase in altitude. Forest AGB gradually increases with slope, reaching its peak near 30°. However, AGB decreases within the 30–80° range as the slope increases. This study confirms the effectiveness of using UAV LiDAR for estimating and mapping the spatial distribution of AGB in complex terrains. This method can be widely applied in productivity, carbon sequestration, and biodiversity studies of subtropical forests.

Forests are crucial for biodiversity, climate regulation, and hydrological cycles, requiring sustainable management due to threats like deforestation and climate change. Traditional forest monitoring methods are labor-intensive and limited, whereas UAV LiDAR offers detailed three-dimensional data on forest structure and extensive coverage. This study primarily assesses individual tree segmentation algorithms in two forest ecosystems with different levels of complexity using high-density LiDAR data captured by the Zenmuse L1 sensor on a DJI Matrice 300RTK platform. The processing methodology for LiDAR data includes preliminary preprocessing steps to create Digital Elevation Models, Digital Surface Models, and Canopy Height Models. A comprehensive evaluation of the most effective techniques for classifying ground points in the LiDAR point cloud and deriving accurate models was performed, concluding that the Triangular Irregular Network method is a suitable choice. Subsequently, the segmentation step is applied to enable the analysis of forests at the individual tree level. Segmentation is crucial for monitoring forest health, estimating biomass, and understanding species composition and diversity. However, the selection of the most appropriate segmentation technique remains a hot research topic with a lack of consensus on the optimal approach and metrics to be employed. Therefore, after the review of the state of the art, a comparative assessment of four common segmentation algorithms (Dalponte2016, Silva2016, Watershed, and Li2012) was conducted. Results demonstrated that the Li2012 algorithm, applied to the normalized 3D point cloud, achieved the best performance with an F1-score of 91% and an IoU of 83%.

The monitoring of beach topographical changes and recovery processes under typhoon storm influence has primarily relied on traditional techniques that lack high spatial resolution. Therefore, we used an unmanned aerial vehicle light detection and ranging (UAV LiDAR) system to obtain the four time periods of topographic data from Tantou Beach, a sandy beach in Xiangshan County, Zhejiang Province, China, to explore beach topography and geomorphology in response to typhoon events. The UAV LiDAR data in four survey periods showed an overall vertical accuracy of approximately 5 cm. Based on the evaluated four time periods of the UAV LiDAR data, we created four corresponding DEMs for the beach. We calculated the DEM of difference (Dod), which showed that the erosion and siltation on Tantou Beach over different temporal scales had a significant alongshore zonal feature with a broad change range. The tidal level significantly impacted beach erosion and siltation changes. However, the storm surge did not affect the beach area above the spring high-tide level. After storms, siltation occurred above the spring high-tide zone. This study reveals the advantage of UAV LiDAR in monitoring beach changes and provides novel insights into the impacts of typhoon storms on coastal topographic and geomorphological change and recovery processes.

This study utilizes UAV-based LiDAR to analyze doline microtopography within a karst mountainous terrain. The study area, ‘Gulneomjae’ in Mungyeong City, South Korea, features steep slopes, limited accessibility, and abundant vegetation—conditions that traditionally hinder accurate topographic surveying. UAV LiDAR data were acquired using the DJI Matrice 300 RTK equipped with a Zenmuse L2 sensor, enabling high-density point cloud generation (98 points/m2). The point clouds were processed to remove non-ground points and generate a 0.25 m resolution DEM using TIN interpolation. A total of seven dolines were detected and delineated, and their morphometric characteristics—including area, perimeter, major and minor axes, and elevation—were analyzed. These results were compared with a 1:5000-scale DEM derived from the 2013 National Basic Map. Visual and numerical comparisons highlighted significant improvements in spatial resolution and feature delineation using UAV LiDAR. Although the 1:5000-scale DEM enables general doline detection, UAV LiDAR facilitates more precise boundary extraction and morphometric analysis. The study demonstrates the effectiveness of UAV LiDAR for detailed topographic mapping in complex karst terrains and offers a foundation for future automated classification and temporal change analysis.