Search USGS⌕ Search
Home/Archive/USGS

USGS · 70252698

Comparing modern identification methods for wild bees: Metabarcoding and image-based morphological taxonomic assignment

Cassandra Smith·Robert S. Cornman·Jennifer A. Fike·Johanna M. Kraus·Sara J. Oyler-McCance·Carrie E. Givens·Michelle L. Hladik·Mark W. Vandever·Dana W. Kolpin·Kelly L. Smalling
View original source ↗Export metadata ↓

Abstract

With the decline of bee populations worldwide, studies determining current wild bee distributions and diversity are increasingly important. Wild bee identification is often completed by experienced taxonomists or by genetic analysis. The current study was designed to compare two methods of identification including: (1) morphological identification by experienced taxonomists using images of field-collected wild bees and (2) genetic analysis of composite bee legs (multiple taxa) using metabarcoding. Bees were collected from conservation grasslands in eastern Iowa in summer 2019 and identified to the lowest taxonomic unit using both methods. Sanger sequencing of individual wild bee legs was used as a positive control for metabarcoding. Morphological identification of bees using images resulted in 36 unique taxa among 22 genera, and >80% of Bombus specimens were identified to species. Metabarcoding was limited to genus-level assignments among 18 genera but resolved some morphologically similar genera. Metabarcoding did not consistently detect all genera in the composite samples, including kleptoparasitic bees. Sanger sequencing showed similar presence or absence detection results as metabarcoding but provided species-level identifications for cryptic species (i.e., Lasioglossum ). Genus-specific detections were more frequent with morphological identification than metabarcoding, but certain genera such as Ceratina and Halictus were identified equally well with metabarcoding and morphology. Genera with proportionately less tissue in a composite sample were less likely to be detected using metabarcoding. Image-based methods were limited by image quality and visible morphological features, while genetic methods were limited by databases, primers, and amplification at target loci. This study shows how an image-based identification method compares with genetic techniques, and how in combination, the methods provide valuable genus- and species-level information for wild bees while preserving tissue for other analyses. These methods could be improved and transferred to a field setting to advance our understanding of wild bee distributions and to expedite conservation research.

Explore related subjects

PLoS ONEArticle

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Cassandra Smith, Robert S. Cornman, Jennifer A. Fike, Johanna M. Kraus, Sara J. Oyler-McCance, Carrie E. Givens, Michelle L. Hladik, Mark W. Vandever, Dana W. Kolpin, Kelly L. Smalling. 2024-04-02. Comparing modern identification methods for wild bees: Metabarcoding and image-based morphological taxonomic assignment. https://doi.org/10.1371/journal.pone.0301474

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related USGS reports

USGS2026

Analyses of eye lens stable isotopes across ontogeny of trophically diverse freshwater salmonids

Ontogenetic niche shifts in fishes are nearly universal but remain poorly understood in many species despite being fundamentally important for the persistence, management, and conservation of fish populations, including those of vulnerable salmonids. Eye lens stable isotope analysis has proven useful in studying ontogeny in some marine species but has rarely been applied in freshwater fishes. We conducted among the first applications of eye lens stable isotope analysis in two salmonids, Arctic Charr ( Salvelinus alpinus ) and Brook Trout ( Salvelinus fontinalis ), in four North American lakes at the southern extent of the range of Arctic Charr (Maine, USA). Our goal was to determine if ontogenetic patterns varied between individuals and populations in ways that relate to differential vulnerability. Like studies in marine systems, we found patterns in lens isotopic values that agree with expected ontogenetic patterns to reach known adult trophic niches. Within lakes and individuals examined in this study, Arctic Charr appeared more dependent on pelagic resources than co-occurring Brook Trout through life. Using Bayesian hierarchical linear regressions, we found evidence that ontogenetic shifts in trophic position (measured by δ 15 N) of Arctic Charr may vary among lakes. Arctic Charr in some populations increased in trophic position through life (population lifetime δ 15 N posterior mean slope estimate = 1.01) while others showed no substantial changes (population lifetime δ 15 N posterior mean slope = 0.05), which may relate to differences in habitat and fish assemblage among our study lakes. Our study suggests that individual life stages and populations of salmonids are likely to respond to climate variability (e.g., basal resource shifts) differentially, which could warrant population and life-stage-specific management.

Maine↗
USGS2026

Modeling chronic wasting disease transmission risk in mule deer related to habitat characteristics

Chronic wasting disease (CWD) is a prion disease of cervids that spreads to uninfected individuals through direct transmission (contact with infected individuals), vertical transmission (from mother to offspring), or indirect transmission (exposure to contaminated environments). The risk of indirect transmission is unevenly distributed on the landscape, and risk levels are expected to be controlled by patterns of habitat use by infected and uninfected individuals as well as environmental properties that alter the length of time prions remain infectious and available for uptake. Despite evidence from controlled or laboratory studies identifying environmental properties likely to affect patterns of CWD prion locations on the landscape, it remains difficult to connect mechanisms to realized increased or decreased risk of disease transmission, and few studies have attempted to detect patterns of different CWD risk in different environments. Using data from GPS-collared mule deer in Wyoming that were CWD-tested annually, we constructed models predicting annual probability of disease transmission contingent on environmental properties extracted from GPS use points. We compared models that emphasized different pathways of disease transmission by including or excluding sets of covariates that described deer density, habitat selection, and covariates expected to affect prion persistence in the environment. Results indicated that key habitat characteristics often selected by mule deer, such as proximity to secondary roads, were also associated with higher risk of testing positive for CWD, which supports the hypothesis that disease risk was correlated to patterns of habitat use by deer. We also found increased risk associated with spatial properties that were not selected-for by deer, such as areas where topography collects moisture, suggesting that prion retention mechanisms also play a role in risk. Incorporating these spatially-varying risk factors into our understanding of CWD transmission and outbreak progression can support managers in designing data collection and disease management strategies.

Wyoming↗
USGS2026

Estimating GPS-based social aggregation metrics using collar data

Understanding social aggregation patterns in ungulate herds is essential for gaining behavioral insights, optimizing resource use, reducing human-wildlife conflict, and managing disease risk. As chronic wasting disease is the preeminent disease-related threat to cervid populations in North America, knowledge of contact between individuals and spatiotemporal patterns of aggregation provides opportunity to understand and potentially reduce disease risk while supporting sustainable population sizes. Herd density metrics, derived from global positioning system (GPS) data, can be used to inform management decisions. To effectively compare aggregation behavior within and between herds, aggregation metrics must be accurate. However, the consistency of metrics across different GPS collar sample sizes remains unclear and robust studies of big game require understanding how these factors may vary in different contexts. We examined the minimum sample size necessary for reliable calculations of three aggregation metrics: pairwise inter-animal distances, daily proximity rates, and kernel density estimate (KDE) areas. We used GPS collar data from the Jackson and West Green River elk herds ( Cervus canadensis ) in western Wyoming, USA, that differ in herd size and group structure (single versus multiple sub-groups), representing common practical contexts. Elk locations were acquired for the Jackson herd between 2016 and 2019 and from 2005 to 2010 for the West Green River herd. Herd-specific characteristics substantially influence the sample size necessary for accurate density measurements. As predicted, larger herds with many groups require more GPS collars than small herds with fewer groups. The sample size needed to accurately estimate aggregation varies by metric, with KDE areas, useful for indexing environmentally transmitted disease risk, generally requiring fewer samples, especially in high-density contexts. The required sample size also varies with seasonal changes in density. During periods of highest density, similar sample sizes are required to estimate inter-animal distances and proximity rates regardless of herd characteristics. Our results have implications for costs associated with studying big game herds, indicating fewer collars may be sufficient in some cases. These insights can aid researchers and managers in determining the appropriate number of GPS collars required for effective herd monitoring and informing relevant aggregation metrics for their management goals.

Wyoming↗