Data science competition for cross-site individual tree species identification from airborne remote sensing data.

Data on individual tree crowns from remote sensing have the potential to advance forest ecology by providing information about forest composition and structure with a continuous spatial coverage over large spatial extents. Classifying individual trees to their taxonomic species over large regions from remote sensing data is challenging. 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 the task of classification of individual crowns to species identity. The competition included data from three sites to assess each methods' ability to generalize patterns across two sites simultaneously and apply methods to an untrained site. Three different metrics were used to assess and compare model performance. Six teams participated, representing four countries and nine individuals. The highest performing method from a previous competition in 2017 was applied and used as a baseline to understand advancements and changes in successful methods. The best species classification method was based on a two-stage fully connected neural network that significantly outperformed the baseline random forest and gradient boosting ensemble methods. All methods generalized well by showing relatively strong performance on the trained sites (accuracy = 0.46-0.55, macro F1 = 0.09-0.32, cross entropy loss = 2.4-9.2), but generally failed to transfer effectively to the untrained site (accuracy = 0.07-0.32, macro F1 = 0.02-0.18, cross entropy loss = 2.8-16.3). Classification performance was influenced by the number of samples with species labels available for training, with most methods predicting common species at the training sites well (maximum F1 score of 0.86) relative to the uncommon species where none were predicted. Classification errors were most common between species in the same genus and different species that occur in the same habitat. Most methods performed better than the baseline in detecting if a species was not in the training data by predicting an untrained mixed-species class, especially in the untrained site. This work has highlighted that data science competitions can encourage advancement of methods, particularly by bringing in new people from outside the focal discipline, and by providing an open dataset and evaluation criteria from which participants can learn.

Species-level tree crown maps improve predictions of tree recruit abundance in a tropical landscape.

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.

A benchmark dataset for canopy crown detection and delineation in co-registered airborne RGB, LiDAR and hyperspectral imagery from the National Ecological Observation Network.

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.

A remote sensing derived data set of 100 million individual tree crowns for the National Ecological Observatory Network.

Forests provide biodiversity, ecosystem, and economic services. Information on individual trees is important for understanding 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 technical challenges in turning sensor data into tangible information. Using deep learning methods, we produced an open-source data set 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. These data have the potential to drive significant expansion of individual-level research on trees by facilitating both regional analyses and cross-region comparisons encompassing forest types from most of the United States.

Estimating individual-level plant traits at scale.

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.

A data science challenge for converting airborne remote sensing data into ecological information.

Ecology has reached the point where data science competitions, in which multiple groups solve the same problem using the same data by different methods, will be productive for advancing quantitative methods for tasks such as species identification from remote sensing images. We ran a competition to help improve three tasks that are central to converting images into information on individual trees: (1) crown segmentation, for identifying the location and size of individual trees; (2) alignment, to match ground truthed trees with remote sensing; and (3) species classification of individual trees. Six teams (composed of 16 individual participants) submitted predictions for one or more tasks. The crown segmentation task proved to be the most challenging, with the highest-performing algorithm yielding only 34% overlap between remotely sensed crowns and the ground truthed trees. However, most algorithms performed better on large trees. For the alignment task, an algorithm based on minimizing the difference, in terms of both position and tree size, between ground truthed and remotely sensed crowns yielded a perfect alignment. In hindsight, this task was over simplified by only including targeted trees instead of all possible remotely sensed crowns. Several algorithms performed well for species classification, with the highest-performing algorithm correctly classifying 92% of individuals and performing well on both common and rare species. Comparisons of results across algorithms provided a number of insights for improving the overall accuracy in extracting ecological information from remote sensing. Our experience suggests that this kind of competition can benefit methods development in ecology and biology more broadly.

A hyperspectral image can predict tropical tree growth rates in single-species stands.

Remote sensing is increasingly needed to meet the critical demand for estimates of forest structure and composition at landscape to continental scales. Hyperspectral images can detect tree canopy properties, including species identity, leaf chemistry and disease. Tree growth rates are related to these measurable canopy properties but whether growth can be directly predicted from hyperspectral data remains unknown. We used a single hyperspectral image and light detection and ranging-derived elevation to predict growth rates for 20 tropical tree species planted in experimental plots. We asked whether a consistent relationship between spectral data and growth rates exists across all species and which spectral regions, associated with different canopy chemical and structural properties, are important for predicting growth rates. We found that a linear combination of narrowband indices and elevation is correlated with standardized growth rates across all 20 tree species (R2  = 53.70%). Although wavelengths from the entire visible-to-shortwave infrared spectrum were involved in our analysis, results point to relatively greater importance of visible and near-infrared regions for relating canopy reflectance to tree growth data. Overall, we demonstrate the potential for hyperspectral data to quantify tree demography over a much larger area than possible with field-based methods in forest inventory plots.

Outer bark thickness decreases more with height on stems of fire-resistant than fire-sensitive Floridian oaks (Quercus spp.; Fagaceae).

UNLABELLED: • PREMISE OF THE STUDY: In ecosystems maintained by low-intensity surface fires, tree bark thickness is a determinant of fire-survival because it protects underlying tissues from heat damage. However, it has been unclear whether relatively thick bark i S: maintained at all heights or only near the ground where damage is most likely.• METHODS: We studied six Quercus species from the red and white clades, with three species characteristic of fire-maintained savannas and three species characteristic of forests with infrequent fire. Inner and outer bark (secondary phloem and rhytidome, respectively) thicknesses were measured at intervals from 10 to 300 cm above the ground. We used linear mixed-effects models to test for relationships among height, habitat, and clade on relative thickness (stem proportion) of total, inner, and outer bark. Bark moisture and tissue density were measured for each species at 10 cm.• KEY RESULTS: Absolute and relative total bark thickness declined with height, with no difference in height-related changes between habitat groups. Relative outer bark thickness showed a height-by-habitat interaction. There was a clade effect on relative thickness, but no interaction with height. Moisture contents were higher in inner than outer bark, and red oaks had denser bark than white oaks, but neither trait differed by habitat.• CONCLUSIONS: Quercus species characteristic of fire-prone habitats invest more in outer bark near the ground where heat damage to outer tissues is most likely. Future investigations of bark should consider the height at which measurements are made and distinguish between inner and outer bark.

Women are underrepresented on the editorial boards of journals in environmental biology and natural resource management.

Despite women earning similar numbers of graduate degrees as men in STEM disciplines, they are underrepresented in upper level positions in both academia and industry. Editorial board memberships are an important example of such positions; membership is both a professional honor in recognition of achievement and an opportunity for professional advancement. We surveyed 10 highly regarded journals in environmental biology, natural resource management, and plant sciences to quantify the number of women on their editorial boards and in positions of editorial leadership (i.e., Associate Editors and Editors-in-Chief) from 1985 to 2013. We found that during this time period only 16% of subject editors were women, with more pronounced disparities in positions of editorial leadership. Although the trend was towards improvement over time, there was surprising variation between journals, including those with similar disciplinary foci. While demographic changes in academia may reduce these disparities over time, we argue journals should proactively strive for gender parity on their editorial boards. This will both increase the number of women afforded the opportunities and benefits that accompany board membership and increase the number of role models and potential mentors for early-career scientists and students.