An official website of the United States government
This protocol describes the different methods to collect and preserve bark and ambrosia beetles, detailing collecting tools, recording relevant data, and optimizing step-by-step methods to extract beetles from twigs, branches, bark, and trunks. It elaborates on trapping techniques, tools, lures, baits, and beetle preservation. The main rule of manual collecting is to not attempt to pry the insect out of the wood or bark, but instead, remove the wood/bark away from the beetle: gently and systematically. The main rule of trapping is that there is no general attractant; instead, attractants and traps should reflect the ecology of the targeted beetle taxa.
more »
« less
This content will become publicly available on June 5, 2026
Progress in developing a bark beetle identification tool
This study presents an initial model for bark beetle identification, serving as a foundational step toward developing a fully functional and practical identification tool. Bark beetles are known for extensive damage to forests globally, as well as for uniform and homoplastic morphology which poses identification challenges. Utilizing a MaxViT-based deep learning backbone which utilizes local and global attention to classify bark beetles down to the genus level from images containing multiple beetles. The methodology involves a process of image collection, preparation, and model training, leveraging pre-classified beetle species to ensure accuracy and reliability. The model's F1 score estimates of 0.99 and 1.0 indicates a strong ability to accurately classify genera in the collected data, including those previously unknown to the model. This makes it a valuable first step towards building a tool for applications in forest management and ecological research. While the current model distinguishes among 12 genera, further refinement and additional data will be necessary to achieve reliable species-level identification, which is particularly important for detecting new invasive species. Despite the controlled conditions of image collection and potential challenges in real-world application, this study provides the first model capable of identifying the bark beetle genera, and by far the largest training set of images for any comparable insect group. We also designed a function that reports if a species appears to be unknown. Further research is suggested to enhance the model's generalization capabilities and scalability, emphasizing the integration of advanced machine learning techniques for improved species classification and the detection of invasive or undescribed species.
more »
« less
- Award ID(s):
- 2342481
- PAR ID:
- 10618362
- Editor(s):
- Kajtoch, Łukasz
- Publisher / Repository:
- Public Library of Science
- Date Published:
- Journal Name:
- PLOS One
- Volume:
- 20
- Issue:
- 6
- ISSN:
- 1932-6203
- Page Range / eLocation ID:
- e0310716
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)Bark beetles naturally inhabit forests and can cause large-scale tree mortality when they reach epidemic population numbers. A recent epidemic (1990s–2010s), primarily driven by mountain pine beetles ( Dendroctonus ponderosae ), was a leading mortality agent in western United States forests. Predictive models of beetle populations and their impact on forests largely depend on host related parameters, such as stand age, basal area, and density. We hypothesized that bark beetle attack patterns are also dependent on inferred beetle population densities: large epidemic populations of beetles will preferentially attack large-diameter trees, and successfully kill them with overwhelming numbers. Conversely, small endemic beetle populations will opportunistically attack stressed and small trees. We tested this hypothesis using 12 years of repeated field observations of three dominant forest species (lodgepole pine Pinus contorta , Engelmann spruce Picea engelmannii , and subalpine fir Abies lasiocarpa ) in subalpine forests of southeastern Wyoming paired with a Bayesian modeling approach. The models provide probabilistic predictions of beetle attack patterns that are free of assumptions required by frequentist models that are often violated in these data sets. Furthermore, we assessed seedling/sapling regeneration in response to overstory mortality and hypothesized that higher seedling/sapling establishment occurs in areas with highest overstory mortality because resources are freed from competing trees. Our results indicate that large-diameter trees were more likely to be attacked and killed by bark beetles than small-diameter trees during epidemic years for all species, but there was no shift toward preferentially attacking small-diameter trees in post-epidemic years. However, probabilities of bark beetle attack and mortality increased for small diameter lodgepole pine and Engelmann spruce trees in post-epidemic years compared to epidemic years. We also show an increase in overall understory growth (graminoids, forbs, and shrubs) and seedling/sapling establishment in response to beetle-caused overstory mortality, especially in lodgepole pine dominated stands. Our observations provide evidence of the trajectories of attack and mortality as well as early forest regrowth of three common tree species during the transition from epidemic to post-epidemic stages of bark beetle populations in the field.more » « less
-
ABSTRACT As the range of bark beetles expands into new forests and woodlands, the need to understand their effects on multiple trophic levels becomes increasingly important. To date, much attention has been paid to the aboveground processes affected by bark beetle infestation, with a focus on photoautotrophs and ecosystem level processes. However, indirect effects of bark beetle on belowground processes, especially the structure and function of soil microbiota remains largely a black box. Our study examined the impacts of bark beetle-induced tree mortality on soil microbial community structure and function using high-throughput sequencing of the soil bacterial and fungal communities and measurements of extracellular enzyme activities. The results suggest bark beetle infestation affected edaphic conditions through increased soil water content, pH, electrical conductivity, and carbon/nitrogen ratio and altered bulk and rhizosphere soil microbial community structure and function. Finally, increased enzymatic activity suggests heightened microbial decomposition following bark beetle infestation. With this increase in enzymatic activity, nutrients trapped in organic substrates may become accessible to seedlings and potentially alter the trajectory of forest regeneration. Our results indicate the need for incorporation of microbial processes into ecosystem level models. IMPORTANCE Belowground impacts of bark beetle infestation have not been explored as thoroughly as their aboveground counterparts. In order to accurately model impacts of bark beetle-induced tree mortality on carbon and nutrient cycling and forest regeneration, the intricacies of soil microbial communities must be examined. In this study, we investigated the structure and function of soil bacterial and fungal communities following bark beetle infestation. Our results show bark beetle infestation to impact soil conditions, as well as soil microbial community structure and function.more » « less
-
Abstract 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.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.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.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.more » « less
-
Abstract Biotic disturbances that overlap in space and time may result in important shifts in forest structure and composition, with potential effects on many ecosystem services. Starting in the late 1990s, outbreaks of multiple bark beetle species caused widespread mortality of three co‐occurring conifer species in the ca. 40,000‐km2subalpine zone of the southern Rocky Mountains (SRM), USA. To better understand the implications of such outbreaks, our goal was to determine if overlapping outbreaks of multiple bark beetle species caused greater tree mortality than single‐species outbreaks in stands with multiple susceptible host tree species. We mapped stand susceptibility to outbreaks of spruce beetle (SB,Dendroctonus rufipennis), mountain pine beetle (MPB,Dendroctonus ponderosae), and western balsam bark beetle (WBBB,Dryocoetes confusus) by combining aerial survey data and forest composition variables in a random forest modeling framework. Then, we used existing maps of cumulative forest mortality from bark beetles to investigate the extent and severity of overlapping outbreaks from 1999 to 2019. We found that 46% of stands with two or more of the three studied hosts species—Engelmann spruce (Picea engelmannii), lodgepole pine (Pinus contortavar.latifolia), or subalpine fir (Abies lasiocarpa)—were susceptible to overlapping outbreaks (25% of all sampled stands). Of those stands, 31% experienced outbreaks of two or more beetle species. Stands affected by outbreaks of both MPB and SB had higher tree mortality than stands affected by one species alone, though stands susceptible to both MPB and SB were uncommon (<4% of all sampled stands). No other combinations of beetle outbreaks increased tree mortality above levels caused by single‐species outbreaks. Thus, contrary to expectations, overlapping outbreaks were rarely more severe than single‐species outbreaks in the SRM. This suggests that diverse forest communities may buffer against the most severe effects of bark beetle outbreaks, even during warm, dry conditions.more » « less
