Search

Recruitment limitation—the failure of a species to establish recruits at an available site—is a potential determinant of plant communities’ structure, causing local communities to be a limited subset of the regional species pool. Recruitment limitation results from three mechanisms: (i) lack of seed sources (i.e., source limitation), (ii) failure of available seeds to reach recruitment sites (i.e., dispersal limitation), and (iii) failure of arrived seeds to establish at a location (i.e., establishment limitation). Here, we evaluated the relative importance of these mechanisms in three co-occurring tree species (Dipteryx oleifera, Attalea butyracea, and Astrocaryum standleyanum) that share seed dispersers/predators. The study was set up on Barro Colorado Island (Panama) at 62 one-ha sites with varying tree densities. Source limitation was estimated as the proportion of sites that would be reached by seeds if seeds were distributed uniformly. Dispersal limitation was estimated from the number of sites with seeds in the soil bank. Establishment limitation was evaluated by measuring germination and 1-year survival in seed addition experiments. The effect of conspecific and heterospecific densities on the mechanisms was evaluated at three spatial scales (1, 5, and 9 ha). For all species, seed predation was the most important recruitment component (~80% decrease in seed survival). Establishment varied among species and was affected by conspecific and heterospecific species densities across spatial scales. Given that species identity, distribution, and seed dispersal/predation affect recruitment at multiple scales, multiscale studies are required to understand how recruitment limitation determines community structure in tropical forests. Abstract in Spanish is available with online material.

Native species can coexist with invasive congeners by partitioning niche space; however, impacts from invasive species often occur alongside other disturbances. Native species' responses to the interactions of multiple disturbances remain poorly understood. Here we study the impacts of urbanization and an invasive congener on a native species. Using abundance (catch-per-unit effort) and vertical distribution of native green anoles (Anolis carolinensis) and invasive brown anoles (Anolis sagrei) across a gradient of natural-to-urban forests, we ask if niche shifting (lability) is occurring, and if it can mitigate impacts from one or both disturbances. We use generalized linear models to relate species abundances across the landscape to urbanization, forest structural complexity, and congener abundances (i.e., A. sagrei); and test for an interaction between urbanization and congener abundance. Our data show that A. sagrei presence results in a 17-fold upward shift in vertical niche of A. carolinensis-an 8.3 m shift in median perch height, and models reveal urbanization also drives an increase in A. carolinensis perch height. A. carolinensis and A. sagrei abundances negatively and positively correlate with urbanization, respectively, and neither species' abundance correlate with congener abundance. Despite a positive correlation between A. sagrei abundance and urbanization, our results do not show evidence of this interaction affecting A. carolinensis. Instead, niche lability appears to enable the native species to mitigate the impact of one driver of decline (invasive competition) while our data suggest it declines with the second (urbanization).

Measuring forest biodiversity using terrestrial surveys is expensive and can only capture common species abundance in large heterogeneous landscapes. In contrast, combining airborne imagery with computer vision can generate individual tree data at the scales of hundreds of thousands of trees. To train computer vision models, ground-based species labels are combined with airborne reflectance data. Due to the difficulty of finding rare species in a large landscape, the majority of classification models only include the most abundant species, leading to biased predictions at broad scales. Extending classification models to include rare species requires targeted data collection and algorithmic improvements to overcome large data imbalances between dominant and rare taxa. In addition, large landscapes often require multiple acquisition events, leading to significant within-species variation in reflectance spectra. Using a multi-temporal hierarchical model, we demonstrate the ability to include species predicted at less than 1% frequency in landscape without losing performance on the dominant species. The final model has over 75% accuracy for 14 species with improved rare species classification compared to a baseline deep learning model. After filtering out dead trees, we generate landscape species maps of individual crowns for over 670,000 individual trees at the Ordway Swisher Biological Station within the National Ecological Observatory Network. We estimate the relative abundance of the species within the landscape and provide three measures of uncertainty to generate a range of counts for each species. These maps provide the first estimates of canopy tree diversity within NEON sites to include rare species and provide a blueprint for capturing tree diversity using airborne computer vision at broad scales.

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.

Scene understanding methodologies (e.g., classification, segmentation, and anomaly detection) can be costly when operating on a per-pixel basis. As remote sensing applications rely on imagery with ...

Advances in remote sensing imagery and machine learning applications unlock the potential for developing algorithms for species classification at the level of individual tree crowns at unprecedented scales. However, most approaches to date focus on site-specific applications and a small number of taxonomic groups. Little is known about how well these approaches generalize across broader geographic areas and ecosystems. Leveraging field surveys and hyperspectral remote sensing data from the National Ecological Observatory Network (NEON), we developed a continental-extent model for tree species classification that can be applied to the network, including a wide range of US terrestrial ecosystems. We compared the performance of a model trained with data from 27 NEON sites to models trained with data from each individual site, evaluating advantages and challenges posed by training species classifiers at the US scale. We evaluated the effect of geographic location, topography, and ecological conditions on the accuracy and precision of species predictions (72 out of 77 species). On average, the general model resulted in good overall classification accuracy (micro-F1 score), with better accuracy than site-specific classifiers (average individual tree level accuracy of 0.77 for the general model and 0.70 for site-specific models). Aggregating species to the genus-level increased accuracy to 0.83. Regions with more species exhibited lower classification accuracy. Predicted species were more likely to be confused with congeneric and co-occurring species and confusion was highest for trees with structural damage and in complex closed-canopy forests. The model produced accurate estimates of uncertainty, correctly identifying trees where confusion was likely. Using only data from NEON, this single integrated classifier can make predictions for 20% of all tree species found in forest ecosystems across the entire US, which make up to roughly 90% of the upper canopy of the studied ecosystems. This suggests the potential for integrating information from multiple datasets and locations to develop broad scale general models for species classification from hyperspectral imaging.

Delineating and classifying individual trees in remote sensing data is challenging. Many tree crown delineation methods have difficulty in closed-canopy forests and do not leverage multiple datasets. Methods to classify individual species are often accurate for common species, but perform poorly for less common species and when applied to new sites. We ran a data science competition to help identify effective methods for delineation of individual crowns and classification to determine species identity. This competition included data from multiple sites to assess the methods’ ability to generalize learning across multiple sites simultaneously, and transfer learning to novel sites where the methods were not trained. Six teams, representing 4 countries and 9 individual participants, submitted predictions. Methods from a previous competition were also applied and used as the baseline to understand whether the methods are changing and improving over time. The best delineation method was based on an instance segmentation pipeline, closely followed by a Faster R-CNN pipeline, both of which outperformed the baseline method. However, the baseline (based on a growing region algorithm) still performed well as did the Faster R-CNN. All delineation methods generalized well and transferred to novel forests effectively. The best species classification method was based on a two-stage fully connected neural network, which significantly outperformed the baseline (a random forest and Gradient boosting ensemble). The classification methods generalized well, with all teams training their models using multiple sites simultaneously, but the predictions from these trained models generally failed to transfer effectively to a novel site. Classification performance was strongly influenced by the number of field-based species IDs available for training the models, with most methods predicting common species well at the training sites. Classification errors (i.e., species misidentification) were most common between similar species in the same genus and different species that occur in the same habitat. The best methods handled class imbalance well and learned unique spectral features even with limited data. Most methods performed better than baseline in detecting new (untrained) species, especially in the site with no training data. Our experience further shows that data science competitions are useful for comparing different methods through the use of a standardized dataset and set of evaluation criteria, which highlights promising approaches and common challenges, and therefore advances the ecological and remote sensing field as a whole.

Broad scale remote sensing promises to build forest inventories at unprecedented scales. A crucial step in this process is to associate sensor data into individual crowns. While dozens of crown detection algorithms have been proposed, their performance is typically not compared based on standard data or evaluation metrics. There is a need for a benchmark dataset to minimize differences in reported results as well as support evaluation of algorithms across a broad range of forest types. Combining RGB, LiDAR and hyperspectral sensor data from the USA National Ecological Observatory Network’s Airborne Observation Platform with multiple types of evaluation data, we created a benchmark dataset to assess crown detection and delineation methods for canopy trees covering dominant forest types in the United States. This benchmark dataset includes an R package to standardize evaluation metrics and simplify comparisons between methods. The benchmark dataset contains over 6,000 image-annotated crowns, 400 field-annotated crowns, and 3,000 canopy stem points from a wide range of forest types. In addition, we include over 10,000 training crowns for optional use. We discuss the different evaluation data sources and assess the accuracy of the image-annotated crowns by comparing annotations among multiple annotators as well as overlapping field-annotated crowns. We provide an example submission and score for an open-source algorithm that can serve as a baseline for future methods., Author summary Combining RGB, LiDAR and hyperspectral sensor data from the USA National Ecological Observatory Network’s Airborne Observation Platform with multiple types of evaluation data, we created a benchmark dataset to assess crown detection and delineation methods for canopy trees covering dominant forest types in the United States. This benchmark dataset includes an R package to standardize evaluation metrics and simplify comparisons between methods. The benchmark dataset contains over 6,000 image-annotated crowns, 400 field-annotated crowns, and 3,000 canopy stem points from a wide range of forest types.

Lianas (woody climbing plants) often abound in tropical forests after selective logging and other disturbances. Liana cutting is an often-prescribed but seldom applied silvicultural treatment designed to foster the growth of future crop trees (hereafter: FCTs). Small-scale studies indicate that this treatment is effective, but little is known about its efficiency (i.e., proportions of lianas missed) or financial cost effectiveness at operational scales. To fill these gaps, we worked in a commercial forest concession in Belize where FCTs and trees to be felled were liberated from lianas in 500–1000 ha annual timber harvest blocks. We found that field crews assigned this and inventory-related tasks spent 11.8% of their time cutting lianas from FCTs at a cost of $0.11 per tree. Workers failed to cut 31.9% of the lianas that infested the 701 FCTs they were supposed to liberate; most of the missed lianas grew into the FCT crowns from neighbors or hung down far from the bole [an additional 48 FCTs (6%) were completely missed]. In a logging block treated 9-years prior to this study, 39% of the liberated Swietenia macrophylla FCTs 29–56 cm DBH were still liana-free whereas in an untreated stand, 94% of similar-sized conspecific control FCTs were liana-infested. Based on tree ring data for the same 9-year period, the liberated FCTs grew 38–63% faster than control FCTs. If the mean growth benefit is sustained over the entire 40-year cutting cycle, each liberated FCT will yield an average of 1.51 m3 (639 board feet) more roundwood than comparable FCTs in unliberated forest. Over this 40-year period with an annual discount rate of 4.5%, this added volume gives the $0.11 investment per tree a net present value of US $161.38 and a profitability index of 1467 for export quality timber. These results argue for the application of this inexpensive and effective treatment in managed forests where lianas abound.

Tree allometric relationships are widely employed for estimating forest biomass and production and are basic building blocks of dynamic vegetation models. In tropical forests, allometric relationships are often modeled by fitting scale-invariant power functions to pooled data from multiple species, an approach that fails to capture changes in scaling during ontogeny and physical limits to maximum tree size and that ignores interspecific differences in allometry. Here, we analyzed allometric relationships of tree height (9884 individuals) and crown area (2425) with trunk diameter for 162 species from the Barro Colorado Nature Monument, Panama. We fit nonlinear, hierarchical models informed by species traits – wood density, mean sapling growth, or sapling mortality – and assessed the performance of three alternative functional forms: the scale-invariant power function and the saturating Weibull and generalized Michaelis–Menten (gMM) functions. The relationship of tree height with trunk diameter was best fit by a saturating gMM model in which variation in allometric parameters was related to interspecific differences in sapling growth rates, a measure of regeneration light demand. Light-demanding species attained taller heights at comparatively smaller diameters as juveniles and had shorter asymptotic heights at larger diameters as adults. The relationship of crown area with trunk diameter was best fit by a power function model incorporating a weak positive relationship between crown area and species-specific wood density. The use of saturating functional forms and the incorporation of functional traits in tree allometric models is a promising approach for improving estimates of forest biomass and productivity. Our results provide an improved basis for parameterizing tropical plant functional types in vegetation models.

Supervised methods for object delineation in remote sensing require labeled ground-truth data. Gathering sufficient high quality ground-truth data is difficult, especially when targets are of irregular shape or difficult to distinguish from background or neighboring objects. Tree crown delineation provides key information from remote sensing images for forestry, ecology, and management. However, tree crowns in remote sensing imagery are often difficult to label and annotate due to irregular shape, overlapping canopies, shadowing, and indistinct edges. There are also multiple approaches to annotation in this field (e.g., rectangular boxes vs. convex polygons) that further contribute to annotation imprecision. However, current evaluation methods do not account for this uncertainty in annotations, and quantitative metrics for evaluation can vary across multiple annotators. In this paper, we address these limitations by developing an adaptation of the Rand index for weakly-labeled crown delineation that we call RandCrowns. Our new RandCrowns evaluation metric provides a method to appropriately evaluate delineated tree crowns while taking into account imprecision in the ground-truth delineations. The RandCrowns metric reformulates the Rand index by adjusting the areas over which each term of the index is computed to account for uncertain and imprecise object delineation labels. Quantitative comparisons to the commonly used intersection over union method shows a decrease in the variance generated by differences among multiple annotators. Combined with qualitative examples, our results suggest that the RandCrowns metric is more robust for scoring target delineations in the presence of uncertainty and imprecision in annotations that are inherent to tree crown delineation.

Riparian vegetation performs important ecosystems services, improving water quality, mitigating erosion, and maintaining regional plant and animal biodiversity. Regular annual flooding maintains riparian forests through an intermediate disturbance regime. In response, seasonally flooded vegetation has developed adaptations for seed dispersal and gas transfer to survive and reproduce while undergoing periods of flooding. In the Amazon, a dam building boom threatens the integrity of riparian vegetation by moving riparian corridors into dry-adapted ecosystems and reducing downstream flooding of riparian areas. Additionally, the region is undergoing intense development pressure resulting in the conversion of native riparian vegetation into agriculture. In this study, we measure how the installation of six large dams on the Tocantins River, coupled with land cover change from native forest and savanna to cattle pasture, has changed the land–water interface of this region. Using land cover data provided by MapBiomas, we quantified land cover change from 1985 to 2018 and measured changes in the riparian areas of the still free-flowing areas of the Tocantins River, riparian areas surrounding reservoirs, and in-stream vegetation dynamics. We found that deforestation in the riparian areas of the Tocantins River downstream of the dams is occurring at a higher rate than deforestation in the watershed. Additionally, reservoir filling resulted in creating hundreds of square kilometers of new riparian areas, pushing the riparian zone away from forest-dominated ecosystems into savanna-dominated areas. The quantity of in-stream vegetation throughout the study was dynamic and initially increased after damming before declining for the last decade of the study. Changes to native land cover in riparian areas of the Tocantins River threaten the integrity of ecosystem services provided by riparian vegetation and are likely to lead to further degradation of these areas.

Recently, remotely sensed measurements of the near-infrared reflectance (NIRv) of vegetation, the fluorescence correction vegetation index (FCVI), and radiance (NIRvrad) of vegetation have emerged as indicators of vegetation structure and function with potential to enhance or improve upon commonly used indicators, such as the normalized difference vegetation index (NDVI) and the enhanced vegetation index (EVI). The applicability of these remotely sensed indices to tropical forests, key ecosystems for global carbon cycling and biodiversity, has been limited. In particular, fine-scale spatial and temporal heterogeneity of structure and physiology may contribute to variation in these indices and the properties that are presumed to be tracked by them, such as gross primary productivity (GPP) and absorbed photosynthetically active radiation (APAR). In this study, fine-scale (approx. 15 cm) tropical forest heterogeneity represented by NIRv, FCVI, and NIRvrad and by lidar-derived height is investigated and compared to NIRv and EVI using unoccupied aerial system (UAS)-based hyperspectral and lidar sensors. By exploiting near-infrared signals, NIRv, FCVI, and NIRvrad captured the greatest spatiotemporal variability, followed by the enhanced vegetation index (EVI) and then the normalized difference vegetation index (NDVI). Wavelet analyses showed the dominant spatial scale of variability of all indicators was driven by tree clusters and larger-than-tree-crown size gaps rather than individual tree crowns. NIRv, FCVI, NIRvrad, and EVI captured variability at smaller spatial scales (∼ 50 m) than NDVI (∼ 90 m) and the lidar-based surface model (∼ 70 m). We show that spatial and temporal patterns of NIRv and FCVI were virtually identical for a dense green canopy, confirming predictions in earlier studies. Furthermore, we show that NIRvrad, which does not require separate irradiance measurements, correlated more strongly with GPP and PAR than did other indicators. NIRv, FCVI, and NIRvrad, which are related to canopy structure and the radiation regime of vegetation canopies, are promising tools to improve understanding of tropical forest canopy structure and function.

Functional traits are central to how organisms perform and influence ecosystem function. Although phylogenetic constraints and environmental conditions are both known to affect trait distributions, data limitations have resulted in large scale studies modeling traits either as species weighted averages (ignoring intraspecific variation) or as a function of the environment (ignoring phylogenetic constraints). As a result, large scale predictions for trait distributions do not include key drivers, likely resulting in biased predictions, and cannot be used to assess the relative contributions of inter- and intraspecific variation. To address these limitations, we developed a joint model integrating phylogenetic and environmental information to understand and predict the distribution of eight leaf traits across the eastern United States. This joint model explained 68% of trait variation, outperforming both species-only and environment-only models, with variance attributable to phylogeny alone (23%), the environment alone (18%), and their overlapping effects (26%). The importance of phylogenetic constraints and the environment varied by trait, with some traits associated predominantly with environmental variation and others with phylogeny. To make predictions more continuously across the eastern USA we combined this model with data from the large-scale Forest Inventory and Analysis survey to estimate traits for ~1.2 million trees. The combined model exhibited significant deviations in predictions from both species-only and environment-only models with variation in the direction and magnitude of these differences among ecoregions. These predictions demonstrate the importance of modeling both intra- and interspecific variation to understand and predict large scale gradients in species and ecosystem traits. Significance Statement Large scale trait studies typically rely on species level average trait values, but recent research suggests that intra-species variation plays a key role in the ecology of species across their range of distribution. We developed a new continental scale approach that integrates inter- and intra-specific variation on leaf traits by leveraging the interaction between the environment and the phylogenetic relationships across species. Using this approach to make predictions for over a million trees covering the Eastern United States shows that combining intra- and inter-species variation is crucial for capturing patterns at the continental scale. This approach represents a novel state of the art method with the potential to become the new standard for understanding traits at large scales.

Applications of unmanned aerial vehicles (UAVs) have proliferated in the last decade due to the technological advancements on various fronts such as structure-from-motion (SfM), machine learning, and robotics. An important preliminary step with regard to forest inventory and management is individual tree detection (ITD), which is required to calculate forest attributes such as stem volume, forest uniformity, and biomass estimation. However, users may find adopting the UAVs and algorithms for their specific projects challenging due to the plethora of information available. Herein, we provide a step-by-step tutorial for performing ITD using (i) low-cost UAV-derived imagery and (ii) UAV-based high-density lidar (light detection and ranging). Functions from open-source R packages were implemented to develop a canopy height model (CHM) and perform ITD utilizing the local maxima (LM) algorithm. ITD accuracy assessment statistics and validation were derived through manual visual interpretation from high-resolution imagery and field-data-based accuracy assessment. As the intended audience are beginners in remote sensing, we have adopted a very simple methodology and chosen study plots that have relatively open canopies to demonstrate our proposed approach; the respective R codes and sample plot data are available as supplementary materials.

ForestGEO is a network of scientists and long-term forest dynamics plots (FDPs) spanning the Earth's major forest types. ForestGEO's mission is to advance understanding of the diversity and dynamics of forests and to strengthen global capacity for forest science research. ForestGEO is unique among forest plot networks in its large-scale plot dimensions, censusing of all stems ≥1 cm in diameter, inclusion of tropical, temperate and boreal forests, and investigation of additional biotic (e.g., arthropods) and abiotic (e.g., soils) drivers, which together provide a holistic view of forest functioning. The 71 FDPs in 27 countries include approximately 7.33 million living trees and about 12,000 species, representing 20% of the world's known tree diversity. With >1300 published papers, ForestGEO researchers have made significant contributions in two fundamental areas: species coexistence and diversity, and ecosystem functioning. Specifically, defining the major biotic and abiotic controls on the distribution and coexistence of species and functional types and on variation in species' demography has led to improved understanding of how the multiple dimensions of forest diversity are structured across space and time and how this diversity relates to the processes controlling the role of forests in the Earth system. Nevertheless, knowledge gaps remain that impede our ability to predict how forest diversity and function will respond to climate change and other stressors. Meeting these global research challenges requires major advances in standardizing taxonomy of tropical species, resolving the main drivers of forest dynamics, and integrating plot-based ground and remote sensing observations to scale up estimates of forest diversity and function, coupled with improved predictive models. However, they cannot be met without greater financial commitment to sustain the long-term research of ForestGEO and other forest plot networks, greatly expanded scientific capacity across the world's forested nations, and increased collaboration and integration among research networks and disciplines addressing forest science.

Aim:\ud \ud Tree crowns determine light interception, carbon and water exchange. Thus, understanding the factors causing tree crown allometry to vary at the tree and stand level matters greatly for the development of future vegetation modelling and for the calibration of remote sensing products. Nevertheless, we know little about large‐scale variation and determinants in tropical tree crown allometry. In this study, we explored the continental variation in scaling exponents of site‐specific crown allometry and assessed their relationships with environmental and stand‐level variables in the tropics.\ud \ud Location:\ud \ud Global tropics.\ud \ud Time period:\ud \ud Early 21st century.\ud \ud Major taxa :\ud \ud Woody plants.\ud \ud Methods:\ud \ud Using a dataset of 87,737 trees distributed among 245 forest and savanna sites across the tropics, we fitted site‐specific allometric relationships between crown dimensions (crown depth, diameter and volume) and stem diameter using power‐law models. Stand‐level and environmental drivers of crown allometric relationships were assessed at pantropical and continental scales.\ud \ud Results:\ud \ud The scaling exponents of allometric relationships between stem diameter and crown dimensions were higher in savannas than in forests. We identified that continental crown models were better than pantropical crown models and that continental differences in crown allometric relationships were driven by both stand‐level (wood density) and environmental (precipitation, cation exchange capacity and soil texture) variables for both tropical biomes. For a given diameter, forest trees from Asia and savanna trees from Australia had smaller crown dimensions than trees in Africa and America, with crown volumes for some Asian forest trees being smaller than those of trees in African forests.\ud \ud Main conclusions:\ud \ud Our results provide new insight into geographical variability, with large continental differences in tropical tree crown allometry that were driven by stand‐level and environmental variables. They have implications for the assessment of ecosystem function and for the monitoring of woody biomass by remote sensing techniques in the global tropics.

Broad scale remote sensing promises to build forest inventories at unprecedented scales. A crucial step in this process is designing individual tree segmentation algorithms to associate pixels into delineated tree crowns. While dozens of tree delineation algorithms have been proposed, their performance is typically not compared based on standard data or evaluation metrics, making it difficult to understand which algorithms perform best under what circumstances. There is a need for an open evaluation benchmark to minimize differences in reported results due to data quality, forest type and evaluation metrics, and to support evaluation of algorithms across a broad range of forest types. Combining RGB, LiDAR and hyperspectral sensor data from the National Ecological Observatory Network’s Airborne Observation Platform with multiple types of evaluation data, we created a novel benchmark dataset to assess individual tree delineation methods. This benchmark dataset includes an R package to standardize evaluation metrics and simplify comparisons between methods. The benchmark dataset contains over 6,000 image-annotated crowns, 424 field-annotated crowns, and 3,777 overstory stem points from a wide range of forest types. In addition, we include over 10,000 training crowns for optional use. We discuss the different evaluation sources and assess the accuracy of the image-annotated crowns by comparing annotations among multiple annotators as well as to overlapping field-annotated crowns. We provide an example submission and score for an open-source baseline for future methods.

Forests provide essential biodiversity, ecosystem and economic services. Information on individual trees is important for understanding the state of forest ecosystems but obtaining individual-level data at broad scales is challenging due to the costs and logistics of data collection. While advances in remote sensing techniques allow surveys of individual trees at unprecedented extents, there remain significant technical and computational challenges in turning sensor data into tangible information. Using deep learning methods, we produced an open-source dataset of individual-level crown estimates for 100 million trees at 37 sites across the United States surveyed by the National Ecological Observatory Network’s Airborne Observation Platform. Each canopy tree crown is represented by a rectangular bounding box and includes information on the height, crown area, and spatial location of the tree. Tree crowns identified using this technique correspond well with hand-labeled crowns, exhibiting both high levels of overlap and good correspondence in height estimates. These data have the potential to drive significant expansion of individual-level research on trees by facilitating both regional analyses at scales of ~10,000 ha and cross-region comparisons encompassing forest types from most of the United States.

Functional ecology has increasingly focused on describing ecological communities based on their traits (measurable features affecting individuals' fitness and performance). Analyzing trait distributions within and among forests could significantly improve understanding of community composition and ecosystem function. Historically, data on trait distributions are generated by (1) collecting a small number of leaves from a small number of trees, which suffers from limited sampling but produces information at the fundamental ecological unit (the individual), or (2) using remote-sensing images to infer traits, producing information continuously across large regions, but as plots (containing multiple trees of different species) or pixels, not individuals. Remote-sensing methods that identify individual trees and estimate their traits would provide the benefits of both approaches, producing continuous large-scale data linked to biological individuals. We used data from the National Ecological Observatory Network (NEON) to develop a method to scale up functional traits from 160 trees to the millions of trees within the spatial extent of two NEON sites. The pipeline consists of three stages: (1) image segmentation, to identify individual trees and estimate structural traits; (2) an ensemble of models to infer leaf mass area (LMA), nitrogen, carbon, and phosphorus content using hyperspectral signatures, and DBH from allometry; and (3) predictions for segmented crowns for the full remote-sensing footprint at the NEON sites. The R2 values on held-out test data ranged from 0.41 to 0.75 on held-out test data. The ensemble approach performed better than single partial least-squares models. Carbon performed poorly compared to other traits (R2 of 0.41). The crown segmentation step contributed the most uncertainty in the pipeline, due to over-segmentation. The pipeline produced good estimates of DBH (R2 of 0.62 on held-out data). Trait predictions for crowns performed significantly better than comparable predictions on pixels, resulting in improvement of R2 on test data of between 0.07 and 0.26. We used the pipeline to produce individual-level trait data for ~5 million individual crowns, covering a total extent of ~360 km2 . This large data set allows testing ecological questions on landscape scales, revealing that foliar traits are correlated with structural traits and environmental conditions.

Forecasting rates of forest succession at landscape scales will aid global efforts to restore tree cover to millions of hectares of degraded land. While optical satellite remote sensing can detect regional land cover change, quantifying forest structural change is challenging. We developed a state-space modeling framework that applies Landsat satellite data to estimate variability in rates of natural regeneration between sites in a tropical landscape. Our models work by disentangling measurement error in Landsat-derived spectral reflectance from process error related to successional variability. We applied our modeling framework to rank rates of forest succession between 10 naturally regenerating sites in Southwestern Panama from about 2001 to 2015 and tested how different models for measurement error impacted forecast accuracy, ecological inference, and rankings of successional rates between sites. We achieved the greatest increase in forecasting accuracy by adding intra-annual phenological variation to a model based on Landsat-derived normalized difference vegetation index (NDVI). The best-performing model accounted for inter- and intra-annual noise in spectral reflectance and translated NDVI to canopy height via Landsat-lidar fusion. Modeling forest succession as a function of canopy height rather than NDVI also resulted in more realistic estimates of forest state during early succession, including greater confidence in rank order of successional rates between sites. These results establish the viability of state-space models to quantify ecological dynamics from time series of space-borne imagery. State-space models also provide a statistical approach well-suited to fusing high-resolution data, such as airborne lidar, with lower-resolution data that provides better temporal and spatial coverage, such as the Landsat satellite record. Monitoring forest succession using satellite imagery could play a key role in achieving global restoration targets, including identifying sites that will regain tree cover with minimal intervention.

Agricultural land now exceeds forests as the dominant global biome. Because of their global dominance, and potential expansion or loss, methods to estimate biomass and carbon in agricultural areas are necessary for monitoring global terrestrial carbon stocks and predicting carbon dynamics. Agricultural areas in the tropics have substantial tree cover and associated above ground biomass (AGB) and carbon. Active remote sensing data, such as airborne LiDAR (light detection and ranging), can provide accurate estimates of biomass stocks, but common plot-based methods may not be suitable for agricultural areas with dispersed and heterogeneous tree cover. The objectives of this research are to quantify AGB of a tropical agricultural landscape using a tree-based method that directly incorporates the size of individual trees, and to understand how landscape estimates of AGB from a tree-based method compare to estimates from a plot-based method. We use high-resolution (1.12 m) airborne LiDAR data collected on a 9280-ha region of the Azuero Peninsula of Panama. We model individual tree AGB with canopy dimensions from the LiDAR data. We apply the model to individual tree crown polygons and aggregate AGB estimates to compare with previously developed plot-based estimates. We find that agricultural trees are a distinct and dominant part of our study site. The tree-based approach estimates greater AGB in pixels with low forest cover than the plot-based approach, resulting a 2-fold difference in landscape AGB estimates between the methods for non-forested areas. Additionally, one third of the total landscape AGB exists in areas having

The environmental impacts of energy generation plants, especially those with large dams, have been widely discussed in the Amazon region, but little attention has been paid to the impacts of the associated transmission lines. These impacts are likely to be substantial given the wide geographic extent of the lines and the relatively high forest cover in the traversed areas. Publicly available information about the location and extent of the transmission line network in the Amazon is neither accurate nor current, and its environmental impacts on terrestrial ecosystems have not been assessed on a large scale. This study estimates the scale of the impact of the current and planned transmission and distribution line network using a hand-digitized dataset and the predicted impact area determined from Environmental Impact Assessments. The Legal Amazon region contains 39,625 km of verified transmission and distribution lines, estimated to directly impact 23,467 km2 of land. We find that the transmission line network directly impacts double the area flooded by hydroelectric reservoirs in the Legal Amazon. Of the direct impact area, 5.1% is within protected areas and 10.3% overlaps with intact forest. By 2026, the transmission line network is estimated to grow by 37% in the Legal Amazon, increasing the direct impact to forests by 70% and to protected lands by 29%. Transmission lines are impacting enough land to be considered a serious conservation threat and should be treated as such in research and environmental planning in the Amazon region.

Tree allometric relationships are widely employed to estimate forest biomass and production, and are basic building blocks of dynamic vegetation models. In tropical forests, allometric relationships are often modeled by fitting scale-invariant power functions to pooled data from multiple species, an approach that fails to reflect finite size effects at the smallest and largest sizes, and that ignores interspecific differences in allometry. Here, we analyzed allometric relationships of tree height (9884 individuals) and crown area (2425) with trunk diameter using species-specific morphological and life history data of 162 species from Barro Colorado Island, Panamá. We fit nonlinear, hierarchical models informed by species traits and assessed the performance of three alternative functional forms: the scale-invariant power function, and the saturating Weibull and generalized Michaelis-Menten (gMM) functions. The relationship of tree height with trunk diameter was best fit by a saturating gMM model in which variation in allometric parameters was related to interspecific differences in sapling growth rates, a measure of regeneration light demand. Light-demanding species attained taller heights at comparatively smaller diameters as juveniles and had shorter asymptotic heights at larger diameters as adults. The relationship of crown area with trunk diameter was best fit by a power function model incorporating a weak positive relationship between crown area and species-specific wood density. The use of saturating functional forms and the incorporation of functional traits in tree allometric models is a promising approach to improve estimates of forest biomass and productivity. Our results provide an improved basis for parameterizing tropical tree functional types in vegetation models.

Riparian forests are ecotones that link aquatic and terrestrial habitats, providing ecosystem services including sediment control and nutrient regulation. Riparian forest function is intimately linked to river hydrology and floodplain dynamics, which can be severely altered by dams. The Tocantins River in the eastern Amazon has six mega-dams along its course. To understand the large-scale and cumulative impacts of multiple dams on the Tocantins floodplain, we quantified landscape-scale changes in floodplain extent, hydroperiod, and flood timing on a 145-km stretch of the river downstream of five dams. We used water level data from 1985 to 2019 to compare daily floodplain inundation dynamics before and after damming. We also developed models to examine the impacts of climate and land use change on hydrology of the Tocantins River. After installation of the first dam in 1998, an average of 82.3 km2 (63%) of the floodplain no longer flooded, overall average hydroperiod decreased by 15 days (11%), and flooding started an average of five days earlier. After all five dams were installed, 72% of the average pre-dam flooded area no longer flooded, average hydroperiod had decreased by 35%, and average inundation onset occurred 12 days later. These changes in floodplain hydrology appeared to be driven primarily by dam operations as we found no significant changes in precipitation over the study period. Increasing loss of natural vegetation in the watershed may play a role in changed hydrology but cannot explain the abrupt loss of floodplain extent after the first dam was installed. This is one of few studies to quantify dam-induced floodplain alteration at a landscape scale and to investigate impacts of multiple dams on a landscape. Our results indicate that the Tocantins River floodplain is undergoing drastic hydrologic alteration. The impacts of multiple dams are cumulative and non-linear, especially for hydroperiod and flood timing.

Tree detection is a fundamental task in remote sensing for forestry and ecosystem ecology applications. While many individual tree segmentation algorithms have been proposed, the development and testing of these algorithms is typically site specific, with few methods evaluated against data from multiple forest types simultaneously. This makes it difficult to determine the generalization of proposed approaches, and limits tree detection at broad scales. Using data from the National Ecological Observatory Network we extend a recently developed semi-supervised deep learning algorithm to include data from a range of forest types, determine whether information from one forest can be used for tree detection in other forests, and explore the potential for building a universal tree detection algorithm. We find that the deep learning approach works well for overstory tree detection across forest conditions, outperforming conventional LIDAR-only methods in all forest types. Performance was best in open oak woodlands and worst in alpine forests. When models were fit to one forest type and used to predict another, performance generally decreased, with better performance when forests were more similar in structure. However, when models were pretrained on data from other sites and then fine-tuned using a small amount of hand-labeled data from the evaluation site, they performed similarly to local site models. Most importantly, a universal model fit to data from all sites simultaneously performed as well or better than individual models trained for each local site. This result suggests that RGB tree detection models that can be applied to a wide array of forest types at broad scales should be possible.

Tropical forests exhibit complex but poorly understood patterns of leaf phenology. Understanding species- and individual-level phenological patterns in tropical forests requires datasets covering large numbers of trees, which can be provided by Unmanned Aerial Vehicles (UAVs). In this paper, we test a workflow combining high-resolution RGB images (7 cm/pixel) acquired from UAVs with a machine learning algorithm to monitor tree and species leaf phenology in a tropical forest in Panama. We acquired images for 34 flight dates over a 12-month period. Crown boundaries were digitized in images and linked with forest inventory data to identify species. We evaluated predictions of leaf cover from different models that included up to 14 image features extracted for each crown on each date. The models were trained and tested with visual estimates of leaf cover from 2422 images from 85 crowns belonging to eight species spanning a range of phenological patterns. The best-performing model included both standard color metrics, as well as texture metrics that quantify within-crown variation, with r2 of 0.84 and mean absolute error (MAE) of 7.8% in 10-fold cross-validation. In contrast, the model based only on the widely-used Green Chromatic Coordinate (GCC) index performed relatively poorly (r2 = 0.52, MAE = 13.6%). These results highlight the utility of texture features for image analysis of tropical forest canopies, where illumination changes may diminish the utility of color indices, such as GCC. The algorithm successfully predicted both individual-tree and species patterns, with mean r2 of 0.82 and 0.89 and mean MAE of 8.1% and 6.0% for individual- and species-level analyses, respectively. Our study is the first to develop and test methods for landscape-scale UAV monitoring of individual trees and species in diverse tropical forests. Our analyses revealed undescribed patterns of high intraspecific variation and complex leaf cover changes for some species.

Functional ecology has increasingly focused on describing ecological communities based on their traits (measurable features affecting individuals fitness and performance). Analyzing trait distributions within and among forests could significantly improve understanding of community composition and ecosystem function. Historically, data on trait distributions are generated by (1) collecting a small number of leaves from a small number of trees, which suffers from limited sampling but produces information at the fundamental ecological unit (the individual); or (2) using remote sensing images to infer traits, producing information continuously across large regions, but as plots (containing multiple trees of different species) or pixels, not individuals. Remote sensing methods that identify individual trees and estimate their traits would provide the benefits of both approaches, producing continuous large-scale data linked to biological individuals. We used data from the National Ecological Observatory Network (NEON) to develop a method to scale up functional traits from 160 trees to the millions of trees within the spatial extent of two NEON sites. The pipeline consists of three stages: 1) image segmentation, to identify individual trees and estimate structural traits; 2) ensemble of models to infer leaf mass area (LMA), nitrogen, carbon, and phosphorus content using hyperspectral signatures, and DBH from allometry; and 3) predictions for segmented crowns for the full remote sensing footprint at the NEON sites.The R2 values on held out test data ranged from 0.41 to 0.75 on held out test data. The ensemble approach performed better than single partial least squares models. Carbon performed poorly compared to other traits (R2 of 0.41). The crown segmentation step contributed the most uncertainty in the pipeline, due to over-segmentation. The pipeline produced good estimates of DBH (R2 of 0.62 on held out data). Trait predictions for crowns performed significantly better than comparable predictions on pixels, resulting in improvement of R2 on test data of between to 0.26. We used the pipeline to produce individual level trait data for ∼5 million individual crowns, covering a total extent of ∼360 km2. This large dataset allows testing ecological questions on landscape scales, revealing that foliar traits are correlated with structural traits and environmental conditions.

Remote sensing can transform the speed, scale, and cost of biodiversity and forestry surveys. Data acquisition currently outpaces the ability to identify individual organisms in high resolution imagery. We outline an approach for identifying tree-crowns in RGB imagery while using a semi-supervised deep learning detection network. Individual crown delineation has been a long-standing challenge in remote sensing and available algorithms produce mixed results. We show that deep learning models can leverage existing Light Detection and Ranging (LIDAR)-based unsupervised delineation to generate trees that are used for training an initial RGB crown detection model. Despite limitations in the original unsupervised detection approach, this noisy training data may contain information from which the neural network can learn initial tree features. We then refine the initial model using a small number of higher-quality hand-annotated RGB images. We validate our proposed approach while using an open-canopy site in the National Ecological Observation Network. Our results show that a model using 434,551 self-generated trees with the addition of 2848 hand-annotated trees yields accurate predictions in natural landscapes. Using an intersection-over-union threshold of 0.5, the full model had an average tree crown recall of 0.69, with a precision of 0.61 for the visually-annotated data. The model had an average tree detection rate of 0.82 for the field collected stems. The addition of a small number of hand-annotated trees improved the performance over the initial self-supervised model. This semi-supervised deep learning approach demonstrates that remote sensing can overcome a lack of labeled training data by generating noisy data for initial training using unsupervised methods and retraining the resulting models with high quality labeled data.

In the last twenty years, multiple large and small hydroelectric dams have begun to transform the Amazonian region, spawning a growing volume of academic research across diverse disciplinary and interdisciplinary fields. In this article, we offer a critical review of recent research related to hydropower and sustainability with a focus on the Brazilian Amazon. We revisit the sustainability concept to include the contribution of various knowledge fields and perspectives for understanding, managing and making decisions about social-ecological systems transformed by dams. We conducted a literature review in Web of Science of academic publications centered in the past 5 years (2014–2019), on diverse aspects of hydropower planning, construction, operation and monitoring in the Brazilian Amazon. We present results of a co-occurrence network analysis of publications, highlighting bridging fields, network disconnections, and opportunities for interdisciplinary research. Finally, we report recent advances in the understanding and management of social-ecological systems in Amazonian watersheds, including biophysical, socio-economic, governance and development processes linked to hydropower planning and implementation. This review identifies knowledge gaps and future research directions, highlighting opportunities for improved communication among scientists, practitioners, decision-makers, indigenous peoples and local communities.