An approach to rapid processing of camera trap images with minimal human input

1. Camera traps have become an extensively utilized tool in ecological research, but the manual processing of images created by a network of camera traps rapidly becomes an overwhelming task, even for small camera trap studies.
2. We used transfer learning to create convolutional neural network (CNN) models for identification and classification. By utilizing a small dataset with an average of 275 labeled images per species class, the model was able to distinguish between species and remove false triggers.
3. We trained the model to detect 17 object classes with individual species identification, reaching an accuracy up to 92% and an average F1 score of 85%. Previous studies have suggested the need for thousands of images of each object class to reach results comparable to those achieved by human observers; however, we show that such accuracy can be achieved with fewer images.
4. With transfer learning and an ongoing camera trap study, a deep learning model can be successfully created by a small camera trap study. A generalizable model produced from an unbalanced class set can be utilized to extract trap events that can later be confirmed by human processors.

     

Spatial structure of reproductive success infers mechanisms of ungulate invasion in Nearctic boreal landscapes

1. Landscape change is a key driver of biodiversity declines due to habitat loss and fragmentation, but spatially shifting resources can also facilitate range expansion and invasion. Invasive populations are reproductively successful, and landscape change may buoy this success.
2. We show how modeling the spatial structure of reproductive success can elucidate the mechanisms of range shifts and sustained invasions for mammalian species with attendant young. We use an example of white‐tailed deer (deer; Odocoileus virginianus) expansion in the Nearctic boreal forest, a North American phenomenon implicated in severe declines of threatened woodland caribou (Rangifer tarandus).
3. We hypothesized that deer reproductive success is linked to forage subsidies provided by extensive landscape change via resource extraction. We measured deer occurrence using data from 62 camera traps in northern Alberta, Canada, over three years. We weighed support for multiple competing hypotheses about deer reproductive success using multistate occupancy models and generalized linear models in an AIC‐based model selection framework.
4. Spatial patterns of reproductive success were best explained by features associated with petroleum exploration and extraction, which offer early‐seral vegetation resource subsidies. Effect sizes of anthropogenic features eclipsed natural heterogeneity by two orders of magnitude. We conclude that anthropogenic early‐seral forage subsidies support high springtime reproductive success, mitigating or exceeding winter losses, maintaining populations.
5. Synthesis and Applications. Modeling spatial structuring in reproductive success can become a key goal of remote camera‐based global networks, yielding ecological insights into mechanisms of invasion and range shifts to inform effective decision‐making for global biodiversity conservation.

     

How citizen science could improve species distribution models and their independent assessment

1. Species distribution models (SDM) have been increasingly developed in recent years, but their validity is questioned. Their assessment can be improved by the use of independent data, but this can be difficult to obtain and prohibitive to collect. Standardized data from citizen science may be used to establish external evaluation datasets and to improve SDM validation and applicability.
2. We used opportunistic presence‐only data along with presence–absence data from a standardized citizen science program to establish and assess habitat suitability maps for 9 species of amphibian in western France. We assessed Generalized Additive and Random Forest Models’ performance by (1) cross‐validation using 30% of the opportunistic dataset used to calibrate the model or (2) external validation using different independent datasets derived from citizen science monitoring. We tested the effects of applying different combinations of filters to the citizen data and of complementing it with additional standardized fieldwork.
3. Cross‐validation with an internal evaluation dataset resulted in higher AUC (Area Under the receiver operating Curve) than external evaluation causing overestimation of model accuracy and did not select the same models; models integrating sampling effort performed better with external validation. AUC, specificity, and sensitivity of models calculated with different filtered external datasets differed for some species. However, for most species, complementary fieldwork was not necessary to obtain coherent results, as long as the citizen science data were strongly filtered.
4. Since external validation methods using independent data are considered more robust, filtering data from citizen sciences may make a valuable contribution to the assessment of SDM. Limited complementary fieldwork with volunteer''s participation to complete ecological gradients may also possibly enhance citizen involvement and lead to better use of SDM in decision processes for nature conservation.

     

Robust and simplified machine learning identification of pitfall trap‐collected ground beetles at the continental scale

1. Insect populations are changing rapidly, and monitoring these changes is essential for understanding the causes and consequences of such shifts. However, large‐scale insect identification projects are time‐consuming and expensive when done solely by human identifiers. Machine learning offers a possible solution to help collect insect data quickly and efficiently.
2. Here, we outline a methodology for training classification models to identify pitfall trap‐collected insects from image data and then apply the method to identify ground beetles (Carabidae). All beetles were collected by the National Ecological Observatory Network (NEON), a continental scale ecological monitoring project with sites across the United States. We describe the procedures for image collection, image data extraction, data preparation, and model training, and compare the performance of five machine learning algorithms and two classification methods (hierarchical vs. single‐level) identifying ground beetles from the species to subfamily level. All models were trained using pre‐extracted feature vectors, not raw image data. Our methodology allows for data to be extracted from multiple individuals within the same image thus enhancing time efficiency, utilizes relatively simple models that allow for direct assessment of model performance, and can be performed on relatively small datasets.
3. The best performing algorithm, linear discriminant analysis (LDA), reached an accuracy of 84.6% at the species level when naively identifying species, which was further increased to >95% when classifications were limited by known local species pools. Model performance was negatively correlated with taxonomic specificity, with the LDA model reaching an accuracy of ~99% at the subfamily level. When classifying carabid species not included in the training dataset at higher taxonomic levels species, the models performed significantly better than if classifications were made randomly. We also observed greater performance when classifications were made using the hierarchical classification method compared to the single‐level classification method at higher taxonomic levels.
4. The general methodology outlined here serves as a proof‐of‐concept for classifying pitfall trap‐collected organisms using machine learning algorithms, and the image data extraction methodology may be used for nonmachine learning uses. We propose that integration of machine learning in large‐scale identification pipelines will increase efficiency and lead to a greater flow of insect macroecological data, with the potential to be expanded for use with other noninsect taxa.

     

Automatic detection of fish and tracking of movement for ecology

1. Animal movement studies are conducted to monitor ecosystem health, understand ecological dynamics, and address management and conservation questions. In marine environments, traditional sampling and monitoring methods to measure animal movement are invasive, labor intensive, costly, and limited in the number of individuals that can be feasibly tracked. Automated detection and tracking of small‐scale movements of many animals through cameras are possible but are largely untested in field conditions, hampering applications to ecological questions.
2. Here, we aimed to test the ability of an automated object detection and object tracking pipeline to track small‐scale movement of many individuals in videos. We applied the pipeline to track fish movement in the field and characterize movement behavior. We automated the detection of a common fisheries species (yellowfin bream, Acanthopagrus australis) along a known movement passageway from underwater videos. We then tracked fish movement with three types of tracking algorithms (MOSSE, Seq‐NMS, and SiamMask) and evaluated their accuracy at characterizing movement.
3. We successfully detected yellowfin bream in a multispecies assemblage (F1 score =91%). At least 120 of the 169 individual bream present in videos were correctly identified and tracked. The accuracies among the three tracking architectures varied, with MOSSE and SiamMask achieving an accuracy of 78% and Seq‐NMS 84%.
4. By employing this integrated object detection and tracking pipeline, we demonstrated a noninvasive and reliable approach to studying fish behavior by tracking their movement under field conditions. These cost‐effective technologies provide a means for future studies to scale‐up the analysis of movement across many visual monitoring systems.

     

Linking camera‐trap data to taxonomy: Identifying photographs of morphologically similar chipmunks

Remote cameras are a common method for surveying wildlife and recently have been promoted for implementing large‐scale regional biodiversity monitoring programs. The use of camera‐trap data depends on the correct identification of animals captured in the photographs, yet misidentification rates can be high, especially when morphologically similar species co‐occur, and this can lead to faulty inferences and hinder conservation efforts. Correct identification is dependent on diagnosable taxonomic characters, photograph quality, and the experience and training of the observer. However, keys rooted in taxonomy are rarely used for the identification of camera‐trap images and error rates are rarely assessed, even when morphologically similar species are present in the study area. We tested a method for ensuring high identification accuracy using two sympatric and morphologically similar chipmunk (Neotamias) species as a case study. We hypothesized that the identification accuracy would improve with use of the identification key and with observer training, resulting in higher levels of observer confidence and higher levels of agreement among observers. We developed an identification key and tested identification accuracy based on photographs of verified museum specimens. Our results supported predictions for each of these hypotheses. In addition, we validated the method in the field by comparing remote‐camera data with live‐trapping data. We recommend use of these methods to evaluate error rates and to exclude ambiguous records in camera‐trap datasets. We urge that ensuring correct and scientifically defensible species identifications is incumbent on researchers and should be incorporated into the camera‐trap workflow.     

Tackling unresolved questions in forest ecology: The past and future role of simulation models

1. Understanding the processes that shape forest functioning, structure, and diversity remains challenging, although data on forest systems are being collected at a rapid pace and across scales. Forest models have a long history in bridging data with ecological knowledge and can simulate forest dynamics over spatio‐temporal scales unreachable by most empirical investigations.
2. We describe the development that different forest modelling communities have followed to underpin the leverage that simulation models offer for advancing our understanding of forest ecosystems.
3. Using three widely applied but contrasting approaches – species distribution models, individual‐based forest models, and dynamic global vegetation models – as examples, we show how scientific and technical advances have led models to transgress their initial objectives and limitations. We provide an overview of recent model applications on current important ecological topics and pinpoint ten key questions that could, and should, be tackled with forest models in the next decade.
4. Synthesis. This overview shows that forest models, due to their complementarity and mutual enrichment, represent an invaluable toolkit to address a wide range of fundamental and applied ecological questions, hence fostering a deeper understanding of forest dynamics in the context of global change.

Forest models can help understanding the processes that shape forest functioning, structure and diversity, since they can can simulate forest dynamics over spatio‐temporal scales unreachable by most empirical investigations. Here we describe the development of three widely applied but contrasting forest mo−delling approaches — species distribution models, individual‐based models and dynamic global vegetation models. We provide an overview of recent model applications and pinpoint ten key questions that could, and should, be tackled with forest models in the next decade.     

FishSizer: Software solution for efficiently measuring larval fish size

1. Length and depth of fish larvae are part of the fundamental measurements in many marine ecology studies involving early fish life history. Until now, obtaining these measurements has required intensive manual labor and the risk of inter‐ and intra‐observer variability.
2. We developed an open‐source software solution to semi‐automate the measurement process and thereby reduce both time consumption and technical variability. Using contrast‐based edge detection, the software segments images of a fish larva into “larva” and “background.” Length and depth are extracted from the “larva” segmentation while taking curvature of the larva into consideration. The graphical user interface optimizes workflow and ease of usage, thereby reducing time consumption for both training and analysis. The software allows for visual verification of all measurements.
3. A comparison of measurement methods on a set of larva images showed that this software reduces measurement time by 66%–78% relative to commonly used software.
4. Using this software instead of the commonly used manual approach has the potential to save researchers from many hours of monotonous work. No adjustment was necessary for 89% of the images regarding length (70% for depth). Hence, the only workload on most images was the visual inspection. As the visual inspection and manual dimension extraction works in the same way as currently used software, we expect no loss in accuracy.

     

Looking for the needle in a downsized haystack: Whole‐exome sequencing unravels genomic signals of climatic adaptation in Douglas‐fir (Pseudotsuga menziesii)

1. Conifers often occur along steep gradients of diverse climates throughout their natural ranges, which is expected to result in spatially varying selection to local climate conditions. However, signals of climatic adaptation can often be confounded, because unraveled clines covary with signals caused by neutral evolutionary processes such as gene flow and genetic drift. Consequently, our understanding of how selection and gene flow have shaped phenotypic and genotypic differentiation in trees is still limited.
2. A 40‐year‐old common garden experiment comprising 16 Douglas‐fir (Pseudotsuga menziesii) provenances from a north‐to‐south gradient of approx. 1,000 km was analyzed, and genomic information was obtained from exome capture, which resulted in an initial genomic dataset of >90,000 single nucleotide polymorphisms. We used a restrictive and conservative filtering approach, which permitted us to include only SNPs and individuals in environmental association analysis (EAA) that were free of potentially confounding effects (LD, relatedness among trees, heterozygosity deficiency, and deviations from Hardy–Weinberg proportions). We used four conceptually different genome scan methods based on F_ST outlier detection and gene–environment association in order to disentangle truly adaptive SNPs from neutral SNPs.
3. We found that a relatively small proportion of the exome showed a truly adaptive signal (0.01%–0.17%) when population substructuring and multiple testing was accounted for. Nevertheless, the unraveled SNP candidates showed significant relationships with climate at provenance origins, which strongly suggests that they have featured adaptation in Douglas‐fir along a climatic gradient. Two SNPs were independently found by three of the employed algorithms, and one of them is in close proximity to an annotated gene involved in circadian clock control and photoperiodism as was similarly found in Populus balsamifera.

Synthesis. We conclude that despite neutral evolutionary processes, phenotypic and genomic signals of adaptation to climate are responsible for differentiation, which in particular explain disparity between the well‐known coastal and interior varieties of Douglas‐fir.     

An automatic method for removing empty camera trap images using ensemble learning

Camera traps often produce massive images, and empty images that do not contain animals are usually overwhelming. Deep learning is a machine‐learning algorithm and widely used to identify empty camera trap images automatically. Existing methods with high accuracy are based on millions of training samples (images) and require a lot of time and personnel costs to label the training samples manually. Reducing the number of training samples can save the cost of manually labeling images. However, the deep learning models based on a small dataset produce a large omission error of animal images that many animal images tend to be identified as empty images, which may lead to loss of the opportunities of discovering and observing species. Therefore, it is still a challenge to build the DCNN model with small errors on a small dataset. Using deep convolutional neural networks and a small‐size dataset, we proposed an ensemble learning approach based on conservative strategies to identify and remove empty images automatically. Furthermore, we proposed three automatic identifying schemes of empty images for users who accept different omission errors of animal images. Our experimental results showed that these three schemes automatically identified and removed 50.78%, 58.48%, and 77.51% of the empty images in the dataset when the omission errors were 0.70%, 1.13%, and 2.54%, respectively. The analysis showed that using our scheme to automatically identify empty images did not omit species information. It only slightly changed the frequency of species occurrence. When only a small dataset was available, our approach provided an alternative to users to automatically identify and remove empty images, which can significantly reduce the time and personnel costs required to manually remove empty images. The cost savings were comparable to the percentage of empty images removed by models.     

Understanding the impact of correlation within pair‐bonds on Cormack–Jolly–Seber models

1. The Cormack–Jolly–Seber (CJS) model and its extensions have been widely applied to the study of animal survival rates in open populations. The model assumes that individuals within the population of interest have independent fates. It is, however, highly unlikely that a pair of animals which have formed a long‐term pairing have dissociated fates.
2. We examine a model extension which allows animals who have formed a pair‐bond to have correlated survival and recapture fates. Using the proposed extension to generate data, we conduct a simulation study exploring the impact that correlated fate data has on inference from the CJS model. We compute Monte Carlo estimates for the bias, range, and standard errors of the parameters of the CJS model for data with varying degrees of survival correlation between mates. Furthermore, we study the likelihood ratio test of sex effects within the CJS model by simulating densities of the deviance. Finally, we estimate the variance inflation factor c^ for CJS models that incorporate sex‐specific heterogeneity.
3. Our study shows that correlated fates between mated animals may result in underestimated standard errors for parsimonious models, significantly deflated likelihood ratio test statistics, and underestimated values of c^ for models taking sex‐specific effects into account.
4. Underestimated standard errors can result in lowered coverage of confidence intervals. Moreover, deflated test statistics will provide overly conservative test results. Finally, underestimated variance inflation factors can lead researchers to make incorrect conclusions about the level of extra‐binomial variation present in their data.

     

Underwater photogrammetry for close‐range 3D imaging of dry‐sensitive objects: The case study of cephalopod beaks

• Technical advances in 3D imaging have contributed to quantifying and understanding biological variability and complexity. However, small, dry‐sensitive objects are not easy to reconstruct using common and easily available techniques such as photogrammetry, surface scanning, or micro‐CT scanning. Here, we use cephalopod beaks as an example as their size, thickness, transparency, and dry‐sensitive nature make them particularly challenging. We developed a new, underwater, photogrammetry protocol in order to add these types of biological structures to the panel of photogrammetric possibilities.
• We used a camera with a macrophotography mode in a waterproof housing fixed in a tank with clear water. The beak was painted and fixed on a colored rotating support. Three angles of view, two acquisitions, and around 300 pictures per specimen were taken in order to reconstruct a full 3D model. These models were compared with others obtained with micro‐CT scanning to verify their accuracy.
• The models can be obtained quickly and cheaply compared with micro‐CT scanning and have sufficient precision for quantitative interspecific morphological analyses. Our work shows that underwater photogrammetry is a fast, noninvasive, efficient, and accurate way to reconstruct 3D models of dry‐sensitive objects while conserving their shape. While the reconstruction of the shape is accurate, some internal parts cannot be reconstructed with photogrammetry as they are not visible. In contrast, these structures are visible using reconstructions based on micro‐CT scanning. The mean difference between both methods is very small (10⁻⁵ to 10⁻⁴ mm) and is significantly lower than differences between meshes of different individuals.
• This photogrammetry protocol is portable, easy‐to‐use, fast, and reproducible. Micro‐CT scanning, in contrast, is time‐consuming, expensive, and nonportable. This protocol can be applied to reconstruct the 3D shape of many other dry‐sensitive objects such as shells of shellfish, cartilage, plants, and other chitinous materials.

     

Agent‐based modeling of the effects of forest dynamics,selective logging,and fragment size on epiphyte communities

1. Forest canopies play a crucial role in structuring communities of vascular epiphytes by providing substrate for colonization, by locally varying microclimate, and by causing epiphyte mortality due to branch or tree fall. However, as field studies in the three‐dimensional habitat of epiphytes are generally challenging, our understanding of how forest structure and dynamics influence the structure and dynamics of epiphyte communities is scarce.
2. Mechanistic models can improve our understanding of epiphyte community dynamics. We present such a model that couples dispersal, growth, and mortality of individual epiphytes with substrate dynamics, obtained from a three‐dimensional functional–structural forest model, allowing the study of forest–epiphyte interactions. After validating the epiphyte model with independent field data, we performed several theoretical simulation experiments to assess how (a) differences in natural forest dynamics, (b) selective logging, and (c) forest fragmentation could influence the long‐term dynamics of epiphyte communities.
3. The proportion of arboreal substrate occupied by epiphytes (i.e., saturation level) was tightly linked with forest dynamics and increased with decreasing forest turnover rates. While species richness was, in general, negatively correlated with forest turnover rates, low species numbers in forests with very‐low‐turnover rates were due to competitive exclusion when epiphyte communities became saturated. Logging had a negative impact on epiphyte communities, potentially leading to a near‐complete extirpation of epiphytes when the simulated target diameters fell below a threshold. Fragment size had no effect on epiphyte abundance and saturation level but correlated positively with species numbers.
4. Synthesis: The presented model is a first step toward studying the dynamic forest–epiphyte interactions in an agent‐based modeling framework. Our study suggests forest dynamics as key factor in controlling epiphyte communities. Thus, both natural and human‐induced changes in forest dynamics, for example, increased mortality rates or the loss of large trees, pose challenges for epiphyte conservation.

     

Intraspecific trait variability and community assembly in hawkmoths (Lepidoptera: Sphingidae) across an elevational gradient in the eastern Himalayas,India

1. We investigated some aspects of hawkmoth community assembly at 13 elevations along a 200‐ to 2770‐m transect in the eastern Himalayas, a little studied biodiversity hot spot of global importance. We measured the morphological traits of body mass, wing loading, and wing aspect ratio of 3,301 free‐ranging individuals of 76 species without having to collect or even constrain them. We used these trait measurements and T‐statistic metrics to assess the strength of intracommunity (“internal") and extra‐community (“external”) filters which determine the composition of communities vis‐a‐vis the regional pool of species.
2. The trait distribution of constituent species turned out to be nonrandom subsets of the community‐trait distribution, providing strong evidence for internal filtering in all elevational communities. The external filter metric was more ambiguous. However, the elevational dependence of many metrics including that of the internal filter provided evidence for external (i.e., environmental) filtering. On average, a species occupied as much as 50%–75% of the total community‐trait space, yet the T‐statistic metric for internal filter was sufficiently sensitive to detect a strong nonrandom structure in the trait distribution.
3. We suggest that the change in T‐statistic metrics along the environmental gradient may provide more clues to the process of community assembly than previously envisaged. A large, smoothly varying and well‐sampled environmental span would make it easier to discern them. Developing T‐statistics for combined analysis of multiple traits will perhaps provide a more accurate picture of internal/filtering and niche complementarity. Moths are a hyperdiverse taxon and a very important component of many ecosystems. Our technique for accurately measuring body and wing dimensions of free‐ranging moths can generate trait database for a large number of individuals in a time‐ and resource‐efficient manner for a variety of community assembly studies using this important taxon.

     

Animal Scanner: Software for classifying humans,animals, and empty frames in camera trap images

Camera traps are a popular tool to sample animal populations because they are noninvasive, detect a variety of species, and can record many thousands of animal detections per deployment. Cameras are typically set to take bursts of multiple photographs for each detection and are deployed in arrays of dozens or hundreds of sites, often resulting in millions of photographs per study. The task of converting photographs to animal detection records from such large image collections is daunting, and made worse by situations that generate copious empty pictures from false triggers (e.g., camera malfunction or moving vegetation) or pictures of humans. We developed computer vision algorithms to detect and classify moving objects to aid the first step of camera trap image filtering—separating the animal detections from the empty frames and pictures of humans. Our new work couples foreground object segmentation through background subtraction with deep learning classification to provide a fast and accurate scheme for human–animal detection. We provide these programs as both Matlab GUI and command prompt developed with C++. The software reads folders of camera trap images and outputs images annotated with bounding boxes around moving objects and a text file summary of results. This software maintains high accuracy while reducing the execution time by 14 times. It takes about 6 seconds to process a sequence of ten frames (on a 2.6 GHZ CPU computer). For those cameras with excessive empty frames due to camera malfunction or blowing vegetation automatically removes 54% of the false‐triggers sequences without influencing the human/animal sequences. We achieve 99.58% on image‐level empty versus object classification of Serengeti dataset. We offer the first computer vision tool for processing camera trap images providing substantial time savings for processing large image datasets, thus improving our ability to monitor wildlife across large scales with camera traps.     

A modified niche model for generating food webs with stage‐structured consumers: The stabilizing effects of life‐history stages on complex food webs

1. Almost all organisms grow in size during their lifetime and switch diets, trophic positions, and interacting partners as they grow. Such ontogenetic development introduces life‐history stages and flows of biomass between the stages through growth and reproduction. However, current research on complex food webs rarely considers life‐history stages. The few previously proposed methods do not take full advantage of the existing food web structural models that can produce realistic food web topologies.
2. We extended the niche model developed by Williams and Martinez (Nature, 2000, 404, 180–183) to generate food webs that included trophic species with a life‐history stage structure. Our method aggregated trophic species based on niche overlap to form a life‐history structured population; therefore, it largely preserved the topological structure of food webs generated by the niche model. We applied the theory of allometric predator–prey body mass ratio and parameterized an allometric bioenergetic model augmented with biomass flow between stages via growth and reproduction to study the effects of a stage structure on the stability of food webs.
3. When life‐history stages were linked via growth and reproduction, more food webs persisted, and persisting food webs tended to retain more trophic species. Topological differences between persisting linked and unlinked food webs were small to modest. The slopes of biomass spectra were lower, and weak interaction links were more prevalent in the linked food webs than the unlinked ones, suggesting that a life‐history stage structure promotes characteristics that can enhance stability of complex food webs.
4. Our results suggest a positive relationship between the complexity and stability of complex food webs. A life‐history stage structure in food webs may play important roles in dynamics of and diversity in food webs.

     

EDock‐ML: A web server for using ensemble docking with machine learning to aid drug discovery

EDock‐ML is a web server that facilitates the use of ensemble docking with machine learning to help decide whether a compound is worthwhile to be considered further in a drug discovery process. Ensemble docking provides an economical way to account for receptor flexibility in molecular docking. Machine learning improves the use of the resulting docking scores to evaluate whether a compound is likely to be useful. EDock‐ML takes a bottom‐up approach in which machine‐learning models are developed one protein at a time to improve predictions for the proteins included in its database. Because the machine‐learning models are intended to be used without changing the docking and model parameters with which the models were trained, novice users can use it directly without worrying about what parameters to choose. A user simply submits a compound specified by an ID from the ZINC database (Sterling, T.; Irwin, J. J., J Chem Inf Model 2015, 55[11], 2,324–2,337.) or upload a file prepared by a chemical drawing program and receives an output helping the user decide the likelihood of the compound to be active or inactive for a drug target. EDock‐ML can be accessed freely at edock‐ml.umsl.edu     

Biosurveillance for invasive insect pest species using an environmental DNA metabarcoding approach and a high salt trap collection fluid

1. With the increase in global trade and warming patterns, the movement, introduction, and establishment of non‐native insect species has increased. A rapid and effective early detection biosurveillance program to identify species of concern is needed to reduce future impacts and costs associated with introduced non‐native species. One of the challenges facing insect surveillance trapping methods is the sheer volume of individual specimens in the collections. Although molecular identification methods are improving, they currently have limitations (e.g., destructive processing of specimens) and a protocol addressing these limitations can support regulatory applications that need morphological evidence to corroborate molecular data.
2. The novel protocol presented here uses a metabarcoding approach to amplify environmental DNA from a saturated salt solution trap fluid, which retains trap specimens for downstream morphological identifications. The use of a saturated salt solution to preserve specimens in traps addresses issues with the high evaporation rate of ethanol in traps, and public safety concerns with other fluid preservation options with unattended traps in public settings.
3. Using a metabarcoding approach, a 407‐nucleotide segment of the cytochrome c oxidase subunit 1 (COI) animal barcode region was successfully amplified from Lindgren funnel trap collection fluids. These traps were placed in forested areas to survey for wood‐boring beetles of regulatory concern. Our results displayed successful amplification of target taxa, including the molecular identification of the Japanese Beetle Popillia japonica, a species regulated in Canada. A second species, Anisandrus maiche, recently introduced to North America, was identified in every trap. The genus Lymantria, which contains numerous species of concern to North American woodlands, was also detected. Also, there were six other species identified of interest due to their potential impacts on native and crop flora and fauna.
4. Our results show how this protocol can be used as an efficient method for the surveillance of insects using a trap with a saturated salt solution and eDNA metabarcoding to detect species of regulatory concern.