Next Article in Journal
Molecular Updates on the ‘Warty Dorid’ Doris verrucosa Linnaeus, 1758 (Mollusca, Nudibranchia) from the Mediterranean Sea
Previous Article in Journal
Population Genetic Structure: Where, What, and Why?
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Diversity
Volume 17
Issue 8
10.3390/d17080585
diversity-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Farmland Biodiversity Monitoring Using DNA Metabarcoding

by
Dirk Steinke
1,2,*,
Muhammad Ashfaq
1,
Chris Y. Ho
1,
Kate H. J. Perez
1,
Jayme E. Sones
1,
Stephanie L. DeWaard
1,
Jeremy R. DeWaard
1,
Sujeevan Ratnasingham
1,2,
Evgeny V. Zakharov
1,2 and
Paul D. N. Hebert
1,2
1
Centre for Biodiversity Genomics, University of Guelph, 50 Stone Road East, Guelph, ON N1G 2W1, Canada
2
Department of Integrative Biology, University of Guelph, 50 Stone Road East, Guelph, ON N1G 2W1, Canada
*
Author to whom correspondence should be addressed.
Diversity 2025, 17(8), 585; https://doi.org/10.3390/d17080585
Submission received: 20 June 2025 / Revised: 14 August 2025 / Accepted: 16 August 2025 / Published: 20 August 2025
(This article belongs to the Section Biodiversity Loss & Dynamics)
Browse Figures
Versions Notes

Abstract

Although 5–20% of global crop production is lost to arthropod damage, current biomonitoring programs are extremely limited. This study evaluates the feasibility of using metabarcoding to assess overall insect diversity and detect pest species in agricultural settings. It introduces a curated DNA barcode reference library for Canadian insects that are agricultural pests and applies it to metabarcoding data from the analysis of Malaise trap samples from two experimental farms in Southern Ontario. A total of 7707 arthropod species were collected across the two farms, and projections indicate that another 4000 await detection. These taxa included 231 registered pest species. The composition of the overall arthropod community composition was more heavily influenced by site location than crop type, but pest species composition was influenced by the crop. This study confirms that metabarcoding enables the evaluation of the species composition of arthropod communities in agroecosystems, allowing pest species to be tracked.

1. Introduction

Arthropods are important contributors to agroecosystems, as they provide critical services, including nutrient recycling, pollination, biological control, and food for other organisms. Although they only considered four services, ref. [1] estimated an annual value of USD 57 billion for the services provided by arthropods in the United States. Alarmingly, recent reports suggest large declines in their biomass, with knock-on effects across ecosystems [2,3,4,5,6]. The four primary drivers of these declines are thought to be the following: (i) habitat loss to agriculture and urbanization; (ii) chemicals—especially pesticides and fertilizers; (iii) pathogens and introduced species; and (iv) climate change [6]. If current trends are sustained, 40% of all insect species may become extinct within the next few decades [6].
In addition to declines in abundance, the species composition of arthropod communities is shifting, with specialists being replaced by pollution-tolerant dietary generalists [6]. Although generalists aid stability in local food webs and, by extension, entire ecosystems [7], this is offset by the declines in diversity. Many specialists that play important functional roles in ecosystems are extremely vulnerable to land-use modification and pollution. Their loss increases interaction strengths within more simplified food webs, which are known to be less resilient [8]. As a result, species are declining in abundance or becoming extirpated, particularly in heavily modified areas such as intensively farmed lands, raising the risk of pest outbreaks [9].
While agroecosystems need arthropods to deliver essential services, farm management practices reduce their abundance. The decline of beneficial arthropod species also leads to an increase in pest species. Currently, about 5–20% of annual global crop production is lost to arthropods, and this is likely to increase with climate change [10]. Furthermore, the control of insect pests and crop diseases is increasingly threatened by rising pesticide resistance [11,12,13].
Past efforts to monitor farmland biodiversity have been limited, as they typically relied on indirect broad-scale multispecies assessments of terrestrial vertebrates (e.g., Agriculture and Agri-Food Canada’s “Wildlife Capacity on Farmland Indicator”) or observations of farmland birds and butterfly populations (e.g., European Union). A recent assessment found no coordinated biomonitoring of agricultural lands in North America and Europe [14]. Conventional biomonitoring utilizing morphological identification is labor-intensive, and agricultural lands are vast, so it is no surprise that governments and the agricultural sector have hesitated to invest. While some biomonitoring data exists, there is a lack of consistent information, even for common species; however, they often provide important ecosystem services or provoke crop damage. Large-scale surveillance programs for agricultural pests are also limited to a few species that are responsible for the most economic damage. As a consequence, intensive agriculture usually promotes the blanket use of pesticides at high concentrations without assessing the presence of target pests. Some modern pesticides seek to reduce non-target impacts [15], but this specificity also means that resistance often evolves quickly in the target pests.
DNA metabarcoding provides an alternative to conventional approaches for biodiversity monitoring. As it can rapidly generate georeferenced occurrence data for many species at low cost, it has been increasingly adopted to monitor populations of aquatic and terrestrial arthropods [16,17,18,19,20], vertebrates [21], pollen [22], diatoms [23], and fungi [24,25,26]. However, the accuracy and reliability of these results hinge on the quality and completeness of the DNA barcode reference libraries [27,28,29]. This is also true for pest monitoring using DNA barcodes (e.g., [30,31,32,33]). Although major programs are underway to build these comprehensive reference libraries [34], the number of unregistered species is still far larger than those with coverage (1.28 M of estimated 10 M [35]). The DNA barcode reference libraries for the terrestrial arthropods of Canada [36] are particularly comprehensive [20,28,37]. However, this is the exception, so studies on other regions often focus on specific taxonomic groups with a limited geographic scope (e.g., [38,39,40]). Smaller, targeted libraries, e.g., for arthropods of biosecurity concern, such as agricultural pests, are more attainable and contribute to the overarching goal of building a fully parameterized library of all species [30,41]. But are the currently available libraries capable of delivering reliable pest identification?
This study examines the feasibility of using metabarcoding to assess insect diversity and detect pest species in agricultural settings. It introduces a curated reference library for most registered pests of Canadian agriculture and applies it to metabarcoding datasets obtained from Malaise trap samples collected at two experimental farms in southern Ontario to determine the composition and dynamics of the pest species community throughout a full growing season.

2. Materials and Methods

2.1. Sample Collection

Two research farms were sampled in 2017: Arkell Research Station in Guelph, Ontario, and Elora Research Station in Elora, Ontario (Figure 1A). At each farm, three fields with different crops were monitored. In Arkell, a 20.2-hectare (ha) soy field (AS), a 35.6 ha corn field (AC), and a 42.1 ha wheat field (AW) were sampled (Figure 1B). In Elora, a 6.1 ha soy field (ES), an 8.1 ha corn field (EC), and a 4.0 ha alfalfa field (EA) were sampled (Figure 1C).
Five 4-headed Sea Land Air Malaise (SLAM) traps were deployed at each field; one at the midpoint of each edge and one mid-field. The SLAM traps were positioned so each collecting wedge faced a cardinal direction (Figure 1D). The exact location of each trap is provided in Table S1. In total, 30 traps were deployed across the six fields from May 3 (pre-planting) until November 1 (post-harvest). Weekly samples (24 weeks) were collected in 250 mL plastic bottles filled with 95% ethanol for a total of 2880 samples (24 weeks × 30 traps × 4 bottles).
Because many traps (21/30) interfered with farm activities (tilling, planting, spraying, harvesting), they had to be removed on several occasions for a 7-day duration. Only weekly samples with a complete 7-day sampling duration for all five SLAM traps on a field with no disturbances or additional issues (including trap damage or evaporation of ethanol) were processed to ensure equal sampling times. This resulted in a total of 1540 samples that were selected for analysis from each location and crop type to maximize sampling coverage over the season.

2.2. Reference Library Assembly

The list of Canadian arthropod pests of agricultural crops was compiled from multiple sources, starting with the list of pests regulated by the Canadian Food Inspection Agency (CFIA—https://inspection.canada.ca/plant-health/invasive-species/eng/1299168913252/1299168989280, accessed 1 November 2024) and the Digital Library of the Centre for Agriculture and Biosciences International (CABI—https://www.cabidigitallibrary.org/journal/cabicompendium, accessed 1 November 2024), as well as other sources (Table S2). This approach generated a checklist (https://doi.org/10.5281/zenodo.16894353) with 928 species. BOLD was queried for all BINs (Barcode Index Number; [42]) associated with each species name and its synonyms. The search returned 111,993 records representing 1403 BINs mapping to 841 of the 928 species in the checklist (https://doi.org/10.5883/DS-CAPP).
The dataset was filtered to exclude COI sequences shorter than 450 bp, records lacking latitude and longitude, records containing stop codons, contaminated sequences, or records flagged as problematic on BOLD. Only public records were used. The data was subsequently subsampled to retain five records per distinct BIN/taxon combination, with a preference for the longest sequence length between 600 bp and 800 bp and an associated image. The final reference library contained 5103 records representing 1185 BINs, mapping to 783 Linnean species, and is available at https://doi.org/10.5883/DS-PSCA.

2.3. DNA Extraction and PCR

DNA extraction employed a membrane-based protocol [43] modified for bulk samples [20]. Specimens were removed from ethanol by filtration through a sterile Microfunnel 0.45 µM Supor Membrane Filter (Pall Corporation, Port Washington, NY, USA) using a 6-Funnel Manifold (Pall Corporation, Port Washington, NY, USA). The wet weight of each sample was then measured in grams to allow standardization of the ratio of lysis buffer to biomass. After the addition of buffer, each sample was incubated overnight at 56 °C while gently mixed on a shaker. Two 50 μL aliquots (technical replicates) from each of the 1540 lysates were then transferred into separate wells in 96-well microplates, and DNA extracts were generated using Acroprep 3.0 µm glass fiber/0.2 µm Bio-Inert membrane plates (Pall Corporation, Port Washington, NY, USA). Each plate contained 88 lysate samples (2 technical replicates of 44 samples), 2 technical replicates of a positive control (lysate from a bulk sample whose component specimens were individually Sanger sequenced—public BOLD dataset—https://doi.org/10.5883/DS-RRNGS), and 6 negative controls. Each lysate was mixed with 100 μL of binding mix, transferred to a column plate, and centrifuged at 5000× g for 5 min. DNA was then purified with three washes; the first employed 180 μL of protein wash buffer, centrifuged at 5000× g for 5 min. Each column was then washed twice with 600 μL of wash buffer and centrifuged at 5000× g for 5 min. Columns were transferred to clean tubes and spun dry at 5000× g for 5 min to remove residual buffer before their transfer to clean collection tubes, followed by incubation for 30 min at 56 °C to dry the membrane. DNA was eluted by adding 60 μL of 10 mM Tris-HCl, pH 8.0, followed by centrifugation at 5000× g for 5 min.
PCR reactions employed a standard protocol [19]. Briefly, each reaction included 5% trehalose (Honeywell Fluka, Charlottem NC, USA), 1× Platinum Taq reaction buffer (Thermo Fisher Scientific, Waltham, MA, USA), 2.5 mM MgCl2 (Thermo Fisher Scientific, Waltham, MA, USA), 0.1 μM of each primer (Integrated DNA Technologies, Coralville, IA, USA), 50 μM of each dNTP (KAPA Biosystems, Wilmington, MA, USA), 0.3 units of Platinum Taq (Thermo Fisher Scientific, Waltham, MA, USA), 2 μL of DNA extract, and Hyclone ultra-pure water (Thermo Fisher Scientific, Waltham, MA, USA) for a final volume of 12.5 μL. Two-stage PCR was used to generate amplicon libraries for sequencing on an Ion Torrent S5 platform. The first round of PCR used the primer combination AncientLepF3 [44] and LepR1 [45] to amplify a 463 bp fragment of COI. Prior to the second PCR, first-round products were diluted 2× with ddH2O. Fusion primers were then used to attach platform-specific unique molecular identifiers (UMIs) along with the sequencing adaptors required for Ion Torrent S5 libraries. Both rounds of PCR employed the same thermocycling conditions: initial denaturation at 94 °C for 2 min, followed by 20 cycles of denaturation at 94 °C for 40 s, annealing at 51 °C for 1 min, and extension at 72 °C for 1 min, with a final extension at 72 °C of 5 min.

2.4. HTS Library Construction

For each plate, labelled products were pooled prior to sequencing. In total, 35 libraries were assembled. Each included two technical replicates of 44 samples plus six technical replicates of an extraction negative and two positive controls, respectively (i.e., 96 samples). Samples, together with positive and negative controls, were pooled after UMI tagging to create a library that was analyzed on a 530 chip (35 chips in total). Amplicon libraries were prepared on an Ion Chef (Thermo Fisher Scientific, Waltham, MA, USA) and sequenced on an Ion Torrent S5 platform at the Centre for Biodiversity Genomics, following manufacturer’s instructions (Thermo Fisher Scientific, Waltham, MA, USA).

2.5. Data Analysis

Reads were uploaded to mBRAVE (http://mbrave.net/, accessed 1 November 2024) for quality filtering and subsequent queries using several reference libraries in an open reference approach. Reads were queried against the Canadian Agricultural Pest library (DS-PSCA) and against five additional system libraries: bacteria (SYS-CRLBACTERIA) to screen for endosymbionts such as Wolbachia, chordates (SYS-CRLCHORDATA), insects (SYS-CRLINSECTA), non-insect arthropods (SYS-CRLNONINSECTARTH), and non-arthropod invertebrates (SYS-CRLNONARTHINVERT). All non-arthropod reads were discarded from further analysis. Sequences were only retained if they were >350 bp and met three quality criteria: mean QV > 20; <25% positions with a QV <20; <5% positions with QV <10. Reads were trimmed 30 bp from their 5′ terminus with a set trim length filter of 450 bp. Reads were matched to sequences in each reference library with an ID distance threshold of 3% but were only retained for further analysis if at least five reads matched an OTU in the reference database. This number is based on earlier benchmarking of the assignment algorithm on mBRAVE, where IonTorrent-generated sequences provided the best compromise between removing errors and retaining real matches [20]. All reads failing to match any sequence in the five reference libraries were clustered at an OTU threshold of 1% with a minimum of five reads per cluster, again a value based on initial benchmarking. All raw data are available in the NCBI Short Read Archive PRJNA892122.
mBRAVE was used to generate BIN (and OTU) tables, including all library queries for each individual plate/run (88 samples, plus 6 negative and 2 positive controls, so 96 for each run). Read counts for any BINs recovered from the negative control on a plate were subtracted from the counts for the same BIN in the 88 sample wells in the run. When this subtraction reduced the read count for a BIN to zero, its occurrence was removed. This step reduced the effects of rare tag switching [18] and background contamination.
Datasets downloaded from mBRAVE were converted into OTU tables and presence/absence matrices for further analysis using an R script 4.4.2 (suppl data) [46]. To determine the completeness of sampling, accumulation curves and the Chao-1 estimator for total diversity [47] were calculated using the vegan 2.7 package [48]. Differences in BIN composition between the two farms and among the four crop types were examined using non-metric multidimensional scaling (NMDS).
To estimate the percentage of species that occur infrequently in the sampled fields and could be considered transients, we counted BINs that were not present in all five traps of a field. BINs present in all five traps were considered core species. In addition, we determined surrounding landscape features by mapping site locations on the 2020 Land Cover of Canada dataset [49] using QGIS 3.4 [50]. A 2 km buffer was created for each farm before QGIS’s zonal histogram tool determined the number of pixels per buffer of each of eight landcover types (three types of forest, wetland, cropland, barren land, urban, water). These were converted into percent cover. Forest cover types were amalgamated into a single forest designation.

3. Results

Sequence analysis of the 1540 samples produced 317,849,360 reads across 35 S5 runs (mean reads per run = 9.08 million; see Table S3). After filtering, 125,384,173 reads were assigned to a BIN. Only 0.3% of reads did not find a BIN match on the Barcode of Life Datasystems (BOLD, [51]). These unmatched reads were de novo clustered using mBRAVE with a 99% similarity threshold. This analysis recognized an average of six additional OTUs per sample, but >98% were chimeras, sequences with multiple indels, or NUMTs, so they were excluded from further analysis.
A total of 7707 BINs was detected at the two farms (Figure 2A), with 5911 BINs at Arkell and 5227 BINs at Elora. The Chao 1 estimate for the total number of BINs at both sites was 11,631 (Figure 2A), with 9428 and 8039 BINs estimated for Arkell and Elora, respectively. Both sites showed many transient species (Arkell 88%, Elora 85%). Table 1 shows the percentage of each landcover type in the 2 km buffer zones around each farm. While both farms are embedded in large cropland/urban areas, a third of Arkell’s surroundings is forest versus 7% at Elora.
Mapping reads against the Canadian Agricultural Pest library (DS-PSCA) revealed matches to 231 BINs when both farms were considered (Figure 2B), with 225 BINs at Arkell and 165 at Elora. The Chao 1 estimate for the total number of pest BINs at both sites was 261 (Figure 2B), with 288 and 188 BINs for Arkell and Elora, respectively. We also detected one species (Dichrorampha aeratana) that was released as a biocontrol agent to counteract an invasive weed species (Leucanthemum vulgare). The two sites shared 3341 BINs overall and 165 of the pest BINs. On average, 0.39 million sequences were recovered per trap per week with an average of 78 BINs (range 4 to 354 BINs, Table S3) per sample. Considering crop types, both soy and alfalfa fields had about 50% higher average richness for the total community (144 and 146, respectively) and the pest assemblage (46 and 47, respectively) than corn and wheat (116 and 107 for all arthropods, 32 and 29 for pests). Average richness of pest species ranged from 32 to 47 per sample (mean = 38.5).
The BIN richness for samples collected from the four cardinal directions averaged 132 (range = 27–346). Richness was very similar for all directions, ranging from 128 to 135 BINs. BIN overlap among the four bottles on a trap averaged 23%, meaning that 77% of the BINs recovered from each bottle were unique to it.
A NMDS Ordination plot revealed that BIN assemblages for the two farms formed distinct cohesive groupings (Figure 3A), while no divergence was apparent among crop types (Figure 3C). NMDS plots for the observed pest assemblages showed the opposite—the two farms were not separable (Figure 3B), but there was separation for certain crop types (Figure 3D). A PERMANOVA analysis suggested that overall community structure varied between farms (R2  =  0.14, p  =  0.0001) and crop type (R2  =  0.17, p  =  0.0001). Pest communities varied weakly between sites (R2  =  0.09, p  =  0.0001) but more strongly between crop types (R2  =  0.30, p  =  0.0001) (Table 2). Multilevel pairwise comparisons did not show any significant differences between pairs.
Taxonomic composition at an ordinal level was similar among the samples, with over half of the BINs being flies (Diptera), followed by Hymenoptera, Lepidoptera, Coleoptera, and Hemiptera (Figure 4). Composition shifted within the pest assemblages (Figure 4), as most pest BINs were Lepidoptera, followed by Coleoptera, Hemiptera, and Diptera. Only one third of the detected pest species feed on the crops in which they were found (Wheat—29.7%, Corn—36.1%, Soy—28.4%, Alfalfa—38.2%). About 22% of all detections are pests of trees; most of those (96%) were identified as transient species.

4. Discussion

This study used metabarcoding to examine the species composition of 1540 samples (4 samples per trap and week = 385 trap weeks) derived from 30 SLAM traps deployed at two experimental farms in Southern Ontario. The results not only confirm the feasibility of a DNA-based biomonitoring to measure the species composition of arthropod communities but also demonstrate its capacity to monitor and identify pest species using a curated reference library for all known Canadian plant pests. With a cost of about CAD 150 per trap sample (four collecting bottles due to SLAM design, using two replicate samples of each) and a processing time of a week (extraction to sequencing) for sets of 88–132 samples at a time, this approach is fast, cost-effective, and scalable (compared to estimated costs for morphology-based assessments of CAD 800–1100 [52]). These advantages are critical for the large-scale, rapid detection of registered pest species [53] necessary to effectively monitor agroecosystems.
We observed a rather large difference between the number of arthropod species detected (N = 7707) and the Chao estimator of the likely true richness (N = 11,631). The low overlap among bottles from individual traps reinforces this conclusion. However, the trade-off is four times the effort and cost for DNA extraction and sequencing, as well as a smaller bottle size requiring shorter sampling intervals. Overall, our sampling was insufficient to collect all BINs at each farm. Earlier work suggests that decreasing the distances between individual traps can increase diversity coverage (e.g., [54]), but this would require deploying many more traps, which would not be feasible in active farming operations, as traps represent an obstacle in daily operations. Environmental DNA (eDNA) approaches might be a solution, as the collection of airborne DNA or DNA from leaf surfaces would interfere less with farming operations [55].
Flying insects are highly mobile, and our estimate of transient species suggested that most species collected from the farm fields were likely just passing through. This is supported by the fact that 37% of the species were only collected once. By comparison, 12–15% of species were collected by all traps at a farm, suggesting these represent core members of the local community. Transient species likely originated from neighboring forests, shrubland, and wetlands. Although both sites are mainly surrounded by cropland and urban development, the surrounding 2 km buffer zone also includes forests. Arkell had considerably more forest cover than Elora (31% vs. 7%), which likely explains its higher BIN richness. Ref. [2] hypothesized that such areas, which serve as insect sources, are negatively affected and drained by the neighboring agricultural fields, which serve as ecological traps that expose insects to pesticides.
The current approach to crop protection is Integrated Pest Management (IPM), which seeks to minimize crop damage by the most economical means and the least impact on non-target organisms [56,57]. Successful IPM begins with the reliable identification of pest species, and DNA barcoding has been slowly incorporated into some routine workflows of regulatory agencies [58,59]. However, to facilitate this, there is a need for comprehensive reference libraries on key pest species. Our reference library for registered arthropod pests in Canada covers federal and provincial regulatory requirements and includes 816 species or 1193 BINs. The higher BIN than species count likely reflects the observation that nearly a third of the registered agricultural pest species appear to be species complexes [30]. By mapping BINs to a species name, we increased the power of this already well-parameterized library (88% of all registered species) sourced from earlier large-scale efforts to register all Canadian species [28,37]. Further parameterization will require more targeted approaches, such as the sequencing of specimens held in Natural History Collections [60].
About 25% (231) of the registered pest species were detected at both sites. A Chao estimator puts this number closer to 261, which means we were only missing about 11% of the predicted pest diversity. The presence of so many BINs poses a challenge to alternative approaches of pest identification, such as morphological inspection or even machine learning supported by computer vision [61,62,63]. Morphological assignments are limited by the lack of taxonomists and by the lack of diagnostic morphological characters [64]. The challenge for machine learning systems is the need for training data (hundreds, if not thousands of images) for each potential pest species [63], which poses a challenge because the list of Canadian Agricultural pests contains >900 nominal species. On the other hand, metabarcoding lacks proper abundance information, which is crucial for determining the density of pests in a given area.
It is known that both the rotation and abundance of crops have local impacts on pest infestations, leading to the idea that the manipulation of crop structure could alleviate insect damage, promote biological control, and allow reduced pesticide use [65,66,67]. However, it has also been shown that pests vary in their response to crop abundance [68]. The high diversity of pests observed in this study makes crop consolidation for this assemblage a major challenge for impactful management strategies; however, only one third of the detected pest species are known to attack the crop growing where they were collected. Although many of the pest species do not feed on any of the crops, their presence can pose indirect risks through disruption of food webs [69], potential disease transmission [70], or community alteration [71]. In addition, 22% of observed harmful species were forest pests, which might not be the direct target of a focused agricultural monitoring system but nevertheless represent potential threats to other adjacent ecosystems. For example, the Emerald Ash Borer (Agrilus planipennis) is responsible for approximately CAD 1.2 billion annual damage [72], and collateral observations such as ours could aid in monitoring efforts. The species was frequently collected at Arkell, but just once at Elora, reflecting its lower forest cover. Again, this and many other observed pest species can be considered transient species originating from nearby forests, shrubland, and wetlands.
While the overall community composition differed significantly between the sites, which are about 30 km apart (Figure 3A), we found few compositional differences between pest species (Figure 3B). The opposite was true for associations with crop types. Pest assemblies differed between crop types (Figure 3D), while overall arthropod communities largely overlapped across crop types except for communities from wheat fields (Figure 3C). In general, arthropod diversity was about 50% higher in soy and alfalfa fields than in wheat and corn, as they represent structurally more complex habitats (branching stems, dense growth). As habitat complexity is considered a key driver of biodiversity [73,74,75], both soy and alfalfa provide a more suitable environment for many arthropods, with the unintended side-effect that pest species thrive under these conditions. Especially for alfalfa, we found that 38% of the observed pest species feed on the crop. Given that soybean and alfalfa production in Canada have increased in recent years (Statistics Canada, Tables 32-10-0359-01, 32-10-0043-01), this could be reason for concern.

5. Conclusions

This study confirms that DNA metabarcoding enables cost-effective biomonitoring, which can measure species diversity at multiple levels of arthropod community organization. A very large proportion of detected species likely represents insects from nearby forests. This ongoing transfer of species might add additional harmful species to the farming operation or indicate expansion patterns, but these movements also expose forest species to pesticides. We continuously monitored and identified pest species and secured results within a week using a curated reference library for all known and registered Canadian plant pests. Our results also showed a diverse community of pest species present in farm fields over the course of a growing season, with preferences for specific crop types. These findings pose some challenges because management practices must consider an array of species rather than just a few key pests.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/d17080585/s1.

Author Contributions

D.S., J.R.D., J.E.S., E.V.Z., and P.D.N.H. designed the study. D.S., J.E.S., and K.H.J.P. coordinated the study. S.L.D. conducted bench work and contributed to analyses. C.Y.H., M.A., and D.S. assembled and edited the Canadian Pest Reference Library. D.S. performed the analyses and wrote the original manuscript, while P.D.N.H., J.E.S., and S.R. revised it. All authors have read and agreed to the published version of the manuscript.

Funding

This study was enabled by awards to PDNH from the Canada First Research Excellence Fund to the University of Guelph’s “Food From Thought” research program (Project 000054), the New Frontiers in Research Fund (NFRFT-2020-00073), the Canada Foundation for Innovation (MSI 42450), as well as the Ontario Ministry of Economic Development, Job Creation and Trade, and the Ontario Ministry of Colleges and Universities.

Acknowledgments

We thank the collections, sequencing, and informatics staff at the Centre for Biodiversity Genomics for acquiring and processing the specimens analyzed in this study. We are very grateful to the staff at Elora Research Station and Arkell Research Station for facilitating collections.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Losey, J.E.; Vaughan, M. The economic value of ecological services provided by insects. BioScience 2006, 56, 311–323. [Google Scholar] [CrossRef]
  2. Hallmann, C.A.; Sorg, M.; Jongejans, E.; Siepel, H.; Hofland, N.; Schwan, H.; Stenmans, W.; Müller, A.; Sumser, H.; Hörren, T.; et al. More than 75 percent decline over 27 years in total flying insect biomass in protected areas. PLoS ONE 2017, 12, e0185809. [Google Scholar] [CrossRef]
  3. Vogel, G. Where have all the insects gone? Science 2017, 356, 576–579. [Google Scholar] [CrossRef]
  4. Lister, B.C.; Garcia, A. Climate-driven declines in arthropod abundance restructure a rainforest food web. Proc. Natl. Acad. Sci. USA 2018, 115, E10397–E10406. [Google Scholar] [CrossRef] [PubMed]
  5. Seibold, S.; Gossner, M.M.; Simons, N.K.; Blüthgen, N.; Müller, J.; Ambarli, D.; Ammer, C.; Bauhus, J.; Fischer, M.; Habel, J.C.; et al. Arthropod decline in grasslands and forests is associated with drivers at landscape level. Nature 2019, 574, 671–674. [Google Scholar] [CrossRef] [PubMed]
  6. Sánchez-Bayo, F.; Wyckhuys, K.A.G. Worldwide decline of the entomofauna: A review of its drivers. Biol. Conserv. 2019, 232, 8–27. [Google Scholar] [CrossRef]
  7. Benton, T.G.; Bieg, C.; Harwatt, H.; Pudasaini, R.; Wellesley, L. Food System Impacts on Biodiversity Loss; Energy, Environment and Resources Programme, Chatham House, The Royal Institute of International Affairs: London, UK, 2021.
  8. Benton, T.G.; Vickery, J.A.; Wilson, J.D. Farmland biodiversity: Is habitat heterogeneity the key? Trends Ecol. Evol. 2003, 18, 182–188. [Google Scholar] [CrossRef]
  9. Dainese, M.; Isaac, N.J.B.; Powney, G.D.; Bommarco, R.; Öckinger, E.; Kuussaari, M.; Pöyry, J.; Benton, T.G.; Gabriel, D.; Hodgson, J.A.; et al. Landscape simplification weakens the association between terrestrial producer and consumer diversity in Europe. Glob. Change Biol. 2016, 23, 3040–3051. [Google Scholar] [CrossRef]
  10. Deutsch, C.A.; Tewksbury, J.J.; Tigchelaar, M.; Battisti, D.S.; Merrill, S.C.; Huey, R.B.; Naylor, R.L. Increase in crop losses to insect pests in a warming climate. Science 2018, 361, 916–919. [Google Scholar] [CrossRef]
  11. Powles, S.B.; Yu, Q. Evolution in action: Plants resistant to herbicides. Annu. Rev. Plant Biol. 2010, 61, 317–347. [Google Scholar] [CrossRef]
  12. Bass, C.; Denholm, I.; Williamson, M.S.; Nauen, R. The global status of insect resistance to neonicotinoid insecticides. Pestic. Biochem. Physiol. 2015, 121, 78–87. [Google Scholar] [CrossRef] [PubMed]
  13. Lucas, J.A.; Hawkins, N.J.; Fraaije, B.A. The evolution of fungicide resistance. Adv. Appl. Microbiol. 2015, 90, 29–92. [Google Scholar]
  14. Herzog, F.; Franklin, J. State-of-the-art practices in farmland biodiversity monitoring for North America and Europe. Ambio 2016, 45, 857–871. [Google Scholar] [CrossRef] [PubMed]
  15. Vyas, S.C. Nontarget Effects of Agricultural Fungicides; CRC Press: Boca Raton, FL, USA, 1988. [Google Scholar]
  16. Ji, C.; Chng, K.R.; Hui Boey, E.J.; Ng, A.H.Q.; Wilm, A.; Nagarajan, N. INC-Seq: Accurate single molecule reads using nanopore sequencing. Gigascience 2016, 5, 34. [Google Scholar]
  17. Beng, K.C.; Tomlinson, K.W.; Shen, X.H.; Surget-Groba, Y.; Hughes, A.C.; Corlett, R.T.; Slik, J.W.F. The utility of DNA metabarcoding for studying the response of arthropod diversity and composition to land-use change in the tropics. Sci. Rep. 2016, 6, 24965. [Google Scholar] [CrossRef] [PubMed]
  18. Elbrecht, V.; Steinke, D. Scaling up DNA metabarcoding for freshwater macrozoobenthos monitoring. Freshw. Biol. 2018, 64, 380–387. [Google Scholar] [CrossRef]
  19. Braukmann, T.W.A.; Prosser, S.J.R.; Ivanova, N.V.; Elbrecht, V.; Steinke, D.; Ratnasingham, R.; deWaard, J.R.; Sones, J.E.; Zakharov, E.V.; Hebert, P.D.N. Metabarcoding a diverse arthropod mock community. Mol. Ecol. Resour. 2019, 19, 711–727. [Google Scholar] [CrossRef]
  20. Steinke, D.; deWaard, S.L.; Sones, J.E.; Ivanova, N.V.; Prosser, S.W.J.; Perez, K.; Braukmann, T.W.A.; Milton, M.; Zakharov, E.V.; deWaard, J.R.; et al. Message in a bottle—Metabarcoding enables biodiversity comparisons across ecoregions. GigaScience 2022, 11, giac040. [Google Scholar] [CrossRef]
  21. Sato, H.; Sogo, Y.; Doi, H.; Yamanaka, H. Usefulness and limitations of sample pooling for environmental DNA metabarcoding of freshwater fish communities. Sci. Rep. 2017, 7, 14860. [Google Scholar] [CrossRef]
  22. Bell, K.L. Applying pollen DNA metabarcoding to the study of plant-pollinator interactions. Appl. Plant Sci. 2017, 5, apps.1600124. [Google Scholar] [CrossRef]
  23. Vasselon, V.; Bouchez, A.; Rimet, F.; Jacquet, S.; Trobajo, R.; Corniquel, M.; Tapolczai, K.; Domaizon, I. Avoiding quantification bias in metabarcoding: Application of a cell biovolume correction factor in diatom molecular biomonitoring. Methods Ecol. Evol. 2018, 9, 1060–1069. [Google Scholar] [CrossRef]
  24. Bellemain, E.; Davey, M.L.; Kauserud, H.; Epp, L.S.; Boessenkool, S.; Coissac, E.; Geml, J.; Edwards, M.; Willerslev, E.; Gussarova, G.; et al. Fungal palaeodiversity revealed using high-throughput metabarcoding of ancient DNA from arctic permafrost. Environ. Microbiol. 2012, 15, 1176–1189. [Google Scholar] [CrossRef] [PubMed]
  25. Aas, A.B.; Davey, M.L.; Kauserud, H. ITS all right mama: Investigating the formation of chimeric sequences in the ITS2 region by DNA metabarcoding analyses of fungal mock communities of different complexities. Mol. Ecol. Resour. 2017, 17, 730–741. [Google Scholar]
  26. Tedersoo, L.; Tooming-Klunderud, A.; Anslan, S. PacBio metabarcoding of Fungi and other eukaryotes: Errors, biases, and perspectives. New Phytol. 2018, 217, 1370–1385. [Google Scholar] [CrossRef]
  27. Collins, R.A.; Trauzzi, G.; Maltby, K.M.; Gibson, T.I.; Ratcliffe, F.C.; Hallam, J.; Rainbird, S.; Maclaine, J.; Henderson, P.A.; Sims, D.W.; et al. Meta-Fish-Lib: A generalized, dynamic DNA reference library pipeline for metabarcoding of fishes. J. Fish Biol. 2021, 99, 1446–1454. [Google Scholar] [CrossRef]
  28. deWaard, J.R.; Ratnasingham, S.; Zakharov, E.V.; Borisenko, A.V.; Steinke, D.; Telfer, A.C.; Perez, K.H.J.; Sones, J.E.; Young, M.R.; Levesque-Beaudin, V.; et al. A reference library for Canadian invertebrates with 1.5 million barcodes, voucher specimens, and DNA samples. Sci. Data 2019, 6, 308. [Google Scholar] [CrossRef]
  29. Weigand, H.; Beermann, A.J.; Ciampor, F.; Costa, F.O.; Csabai, Z.; Duarte, S.; Geiger, M.F.; Grabowski, M.; Rimet, F.; Rulik, B.; et al. DNA barcode reference libraries for the monitoring of aquatic biota in Europe: Gap-analysis and recommendations for future work. Sci. Total Environ. 2019, 678, 499–524. [Google Scholar] [CrossRef]
  30. Ashfaq, M.; Hebert, P.D.N. DNA barcodes for bio-surveillance: Regulated and economically important arthropod plant pests. Genome 2016, 59, 11. [Google Scholar] [CrossRef]
  31. Lee, T.R.C.; Anderson, S.J.; Tran-Nguyen, L.T.T.; Sallam, N.; Le Ru, B.P.; Conlong, D.; Powell, K.; Ward, A.; Mitchell, A. Towards a global DNA barcode reference library for quarantine identifications of lepidoptera stemborers, with an emphasis on sugarcane pests. Sci. Rep. 2019, 9, 7039. [Google Scholar] [CrossRef]
  32. Madden, M.J.L.; Young, R.G.; Brown, J.W.; Miller, S.E.; Frewin, A.J.; Hanner, R.H. Using DNA barcoding to improve invasive pest identification at U.S. ports-of-entry. PLoS ONE 2019, 14, e0222291. [Google Scholar] [CrossRef] [PubMed]
  33. Batovska, J.; Piper, A.M.; Valenzula, I.; Cunningham, J.P.; Blacket, M.J. Developing a non-destructive metabarcoding protocol for detection of pest insects in bulk trap catches. Sci. Rep. 2021, 11, 7946. [Google Scholar] [CrossRef]
  34. Hobern, D. BIOSCAN: DNA barcoding to accelerate taxonomy and biogeography for conservation and sustainability. Genome 2021, 64, 161–164. [Google Scholar] [CrossRef]
  35. Larsen, A.E.; Noack, F. Identifying the landscape drivers of agricultural insecticide use leveraging evidence from 100,000 fields. Proc. Natl. Acad. Sci. USA 2017, 114, 5473–5478. [Google Scholar] [CrossRef] [PubMed]
  36. Pentinsaari, M.; Levesque-Beaudin, V.; Blagoev, G.; deWaard, J.R.; Ho, C.; Steinke, D.; Manjunath, R.; Ratnasingham, S.; Hebert, P.D.N. Constructing a Highly Validated DNA Barcode Reference Library for Canadian Animals: Implications for DNA-Based Identifications. Available online: https://id.boldsystems.org (accessed on 1 November 2024).
  37. Hebert, P.D.N.; Ratnasingham, S.; Zakharov, E.V.; Telfer, A.C.; Levesque-Beaudin, V.; Milton, M.M.; Pederson, S.; Janetta, P.; deWaard, J.R. Counting animal species with DNA barcodes: Canadian insects. Philos. Trans. R. Soc. B 2016, 371, 20150333. [Google Scholar] [CrossRef] [PubMed]
  38. Delrieu-Trottin, E.; Williams, J.T.; Pitassy, D.; Driskell, A.; Hubert, N.; Vivani, J.; Cribb, T.H.; Espiau, B.; Galzin, R.; Kulbicki, M.; et al. A DNA barcode reference library of French Polynesian shore fishes. Sci. Data 2019, 6, 114. [Google Scholar] [CrossRef] [PubMed]
  39. Moriniere, J.; Balke, M.; Doczkal, D.; Geiger, M.F.; Hardulak, L.A.; Haszprunar, G.; Hausmann, A.; Hendrich, L.; Regalado, L.; Rulik, B.; et al. A DNA barcode library for 5,200 German flies and midges (Insecta:Diptera) and its implications for metabarcoding-based biomonitoring. Mol. Ecol. Resour. 2019, 19, 900–928. [Google Scholar] [CrossRef]
  40. Lin, X.L.; Mo, L.; Bu, W.J.; Wang, X.H. The first comprehensive DNA barcode reference library if Chinese Tanytarsus (Diptera: Chironomidae) for environmental DNA metabarcoding. Divers. Distrib. 2020, 27, 1932–1941. [Google Scholar] [CrossRef]
  41. Piper, A.M.; Batovska, J.; Cogan, N.O.I.; Weiss, J.; Cunningham, J.P.; Rodoni, B.C.; Blacket, M.J. Prospects and challenges of implementing DNA metabarcoding for high-throughput insect surveillance. GigaScience 2019, 8, giz092. [Google Scholar] [CrossRef]
  42. Ratnasingham, S.; Hebert, P.D.N. A DNA-based registry for all animal species: The Barcode Index Number (BIN) System. PLoS ONE 2013, 8, e66213. [Google Scholar] [CrossRef]
  43. Ivanova, N.V.; deWaard, J.R.; Hebert, P.D.N. An inexpensive, automation-friendly protocol for recovering high-quality DNA. Mol. Ecol. Resour. 2006, 6, 998–1002. [Google Scholar] [CrossRef]
  44. Prosser, S.W.J.; deWaard, J.R.; Miller, S.E.; Hebert, P.D.N. DNA barcodes from century-old type specimens using next-generation sequencing. Mol. Ecol. Resour. 2016, 16, 487–497. [Google Scholar] [CrossRef] [PubMed]
  45. Hebert, P.D.N.; Penton, E.H.; Burns, J.M.; Janzen, D.H.; Hallwachs, W. Ten species in one: DNA barcoding reveals cryptic species in the neotropical skipper butterfly Astraptes fulgerator. Proc. Natl. Acad. Sci. USA 2004, 101, 14812–14817. [Google Scholar] [CrossRef]
  46. R Core Team. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria, 2020; Available online: https://www.R-project.org/ (accessed on 1 November 2024).
  47. Magurran, A.E. Measuring Biological Diversity; Wiley-Blackwell: Malden, MA, USA, 2003. [Google Scholar]
  48. Oksanen, J.; Blanchet, F.G.; Friendly, M.; Kindt, R.; Legendre, P.; McGlinn, D.; Minchin, P.R.; O’Hara, R.B.; Simpson, G.L.; Solymos, P.; et al. vegan: Community Ecology Package; R Package Version 2.5-1. 2018. Available online: https://vegandevs.github.io/vegan/ (accessed on 1 November 2024).
  49. Natural Resources Canada, Canada Centre for Remote Sensing. 2020 Land Cover of Canada. Natural Resources Canada, Federal Geospatial Platform. 2025. Available online: https://osdp-psdo.canada.ca/dp/en/search/metadata/NRCAN-FGP-1-ee1580ab-a23d-4f86-a09b-79763677eb47 (accessed on 1 November 2024).
  50. QGI.S.org. QGIS Geographic Information System. QGIS Association. 2025. Available online: http://www.qgis.org (accessed on 1 November 2024).
  51. Ratnasingham, S.; Hebert, P.D.N. BOLD: The Barcode of Life Data System (www.barcodinglife.org). Mol. Ecol. Notes 2007, 7, 355–364. [Google Scholar] [CrossRef]
  52. Stein, E.D.; Martinez, M.C.; Stiles, S.; Miller, P.E.; Zakharov, E.V. Is DNA Barcoding Actually Cheaper and Faster than Traditional Morphological Methods: Results from a Survey of Freshwater Bioassessment Efforts in the United States? PLoS ONE 2014, 9, e95525. [Google Scholar] [CrossRef]
  53. Ball, S.L.; Armstrong, K.F. Rapid, one-step DNA extraction for insect pest identification by using DNA barcodes. J. Econ. Entomol. 2008, 101, 523–532. [Google Scholar] [CrossRef]
  54. Steinke, D.; Braukmann, T.W.A.; Manerus, L.; Woodhouse, A.; Elbrecht, V. Effects of Malaise trap spacing on species richness and composition of terrestrial arthropod bulk samples. Metabarcoding Metagenomics 2021, 5, e59201. [Google Scholar] [CrossRef]
  55. Kestel, J.H.; Field, D.L.; Bateman, P.W.; White, N.E.; Allentoft, M.E.; Hopkins, A.J.M.; Gibberd, M.; Nevill, P. Applications of environmental DNA (eDNA) in agricultural systems: Current uses, limitations and future prospects. Sci. Total Environ. 2022, 847, 157556. [Google Scholar] [CrossRef]
  56. Bottrell, D.R.; Bottrell, D.G. Integrated Pest Management; Council on Environmental Quality: Washington, DC, USA, 1979. [Google Scholar]
  57. Deguine, J.P.; Aubertot, J.N.; Flor, R.J.; Lescourret, F.; Wyckhuys, K.A.G.; Ratnadass, A. Integrated pest management: Good intentions, hard realities. A review. Agron. Sustain. Dev. 2021, 41, 38. [Google Scholar] [CrossRef]
  58. Floyd, R.; Lima, J.; deWaard, J.; Humble, L.; Hanner, R. Common goals: Policy implications of DNA barcoding as a protocol for identification of arthropod pests. Biol. Invasions 2010, 12, 2947–2954. [Google Scholar] [CrossRef]
  59. Jones, Y.L.; Peters, S.M.; Weland, C.; Ivanova, N.V.; Yancy, H.F. Potential use of DNA barcodes in regulatory science: Identification of the U.S. Food and Drug Administration’s “Dirty 22,” contributors to the spread of foodborne pathogens. J. Food Prot. 2013, 76, 144–149. [Google Scholar] [CrossRef] [PubMed]
  60. Levesque-Beaudin, V.; Miller, M.E.; Dikow, T.; Miller, S.E.; Prosser, S.W.J.; Zakharov, E.V.; McKeown, J.T.A.; Sones, J.E.; Redmond, N.E.; Coddington, J.A.; et al. A workflow for expanding DNA barcode reference libraries through ‘museum harvesting’ of natural history collections. Biodivers. Data J. 2023, 11, e100677. [Google Scholar] [CrossRef]
  61. Li, R.; Wang, R.; Xie, C.; Liu, L.; Zhang, J.; Wang, F.; Liu, W. A coarse-to-fine network for aphid recognition and detection in the field. Biosyst. Eng. 2019, 187, 39–52. [Google Scholar] [CrossRef]
  62. Park, Y.-H.; Choi, S.H.; Kwon, Y.-J.; Kwon, S.-W.; Kang, Y.J.; Jun, T.-H. Detection of soybean insect pest and a forecasting platform using deep learning with unmanned ground vehicles. Agronomy 2023, 13, 477. [Google Scholar] [CrossRef]
  63. Schneider, S.; Taylor, G.W.; Kremer, S.C.; Fryxell, J.M. Getting the bugs out of AI: Advancing ecological research on arthropods through computer vision. Ecol. Lett. 2023, 26, 1247–1258. [Google Scholar] [CrossRef] [PubMed]
  64. Sweeney, B.W.; Battle, J.M.; Jackson, J.K.; Dapkey, T. Can DNA barcodes of stream macroinvertebrates improve descriptions of community structure and water quality? J. N. Am. Benthol. Soc. 2011, 30, 195–216. [Google Scholar] [CrossRef]
  65. Tscharntke, T.; Klein, A.M.; Kruess, A.; Steffan-Dewenter, I.; Thies, C. Landscape perspectives on agricultural intensification and biodiversity-ecosystem service management. Ecol. Lett. 2005, 8, 857–874. [Google Scholar] [CrossRef]
  66. Larsen, B.B.; Miller, E.C.; Rhodes, M.K.; Wiens, J.J. Inordinate Fondness Multiplied and Redistributed: The Number of Species on Earth and the New Pie of Life. Q. Rev. Biol. 2017, 92, 229–265. [Google Scholar] [CrossRef]
  67. Haan, N.L.; Zhang, Y.; Landis, D.A. Predicting landscape configuration effects on agricultural pest suppression. Trends Ecol. Evol. 2020, 35, 175–186. [Google Scholar] [CrossRef] [PubMed]
  68. Rosenheim, J.A.; Cluff, E.; Lippey, M.K.; Cass, B.N.; Paredes, D.; Parsa, S.; Karp, D.S.; Chaplin-Kramer, R. Increasing crop field size does not consistently exacerbate insect pest problems. Proc. Natl. Acad. Sci. USA 2022, 119, e2208813119. [Google Scholar] [CrossRef]
  69. Frank, S.D.; Tooker, J.F. Neonicotinoids pose undocumented threats to food webs. Proc. Natl. Acad. Sci. USA 2020, 117, 22609–22613. [Google Scholar] [CrossRef]
  70. Wielkopolan, B.; Jakubowska, M.; Obrępalska-Stęplowska, A. Beetles as Plant Pathogen Vectors. Front. Plant Sci. 2021, 12, 748093. [Google Scholar] [CrossRef]
  71. Chailleux, A.; Mohl, E.K.; Teixeira Alves, M.; Messelink, G.J.; Desneuz, N. Natural enemy-mediated indirect interactions among prey species: Potential for enhancing biocontrol services in agroecosystems. Pest Manag. Sci. 2014, 70, 1769–1779. [Google Scholar] [CrossRef]
  72. Aukema, J.E.; Leung, B.; Kovacs, K.; Chivers, C.; Britton, K.O.; Englin, J.; Frankel, S.J.; Haight, R.G.; Holmes, T.P.; Liebhold, A.M.; et al. Economic Impacts of Non-Native Forest Insects in the Continental United States. PLoS ONE 2011, 6, e24587. [Google Scholar] [CrossRef]
  73. MacArthur, R.H. Patterns of species diversity. Biol. Rev. 1965, 4, 510–533. [Google Scholar] [CrossRef]
  74. Badgley, C.; Smiley, T.M.; Terry, R.; Davis, E.B.; DeSantis, L.R.; Fox, D.L.; Hopkins, S.S.B.; Jezkova, T.; Matocq, M.D.; Matzke, N.; et al. Biodiversity and topographic complexity: Modern and geohistorical perspectives. Trends Ecol. Evol. 2017, 32, 211–226. [Google Scholar] [CrossRef] [PubMed]
  75. Loke, L.H.L.; Chisholm, R.A. Measuring habitat complexity and spatial heterogeneity in ecology. Ecol. Lett. 2022, 15, 2269–2288. [Google Scholar] [CrossRef] [PubMed]
Diversity 17 00585 g001
Figure 1. (A) Map of sampling sites within Canada and within southern Ontario (insert). (B) Map of six crop fields sampled at the Elora and (C) Arkell research stations. Trap locations are indicated by red dots and trap numbers. Diagram insert (D) shows trap placement and orientation at each field; each collecting wedge faces a cardinal direction.
Figure 1. (A) Map of sampling sites within Canada and within southern Ontario (insert). (B) Map of six crop fields sampled at the Elora and (C) Arkell research stations. Trap locations are indicated by red dots and trap numbers. Diagram insert (D) shows trap placement and orientation at each field; each collecting wedge faces a cardinal direction.
Diversity 17 00585 g001
Diversity 17 00585 g002
Figure 2. Accumulation curves per site for (A) the 1672 Malaise trap samples collected at two experimental farms in Southern Ontario and (B) a subset including only matches to a Canadian Agricultural Pest library.
Figure 2. Accumulation curves per site for (A) the 1672 Malaise trap samples collected at two experimental farms in Southern Ontario and (B) a subset including only matches to a Canadian Agricultural Pest library.
Diversity 17 00585 g002
Diversity 17 00585 g003
Figure 3. Non-metric multidimensional scaling (NMDS) plots for all samples (A,C) and the Canadian Agricultural Pest subset (B,D) using the Bray–Curtis index coefficient. Color coding is based on sites (A,B) or crop type (C,D).
Figure 3. Non-metric multidimensional scaling (NMDS) plots for all samples (A,C) and the Canadian Agricultural Pest subset (B,D) using the Bray–Curtis index coefficient. Color coding is based on sites (A,B) or crop type (C,D).
Diversity 17 00585 g003
Diversity 17 00585 g004
Figure 4. Taxonomic composition of the full dataset and the Canadian Agricultural Pest subset.
Figure 4. Taxonomic composition of the full dataset and the Canadian Agricultural Pest subset.
Diversity 17 00585 g004
Table 1. Percentage landcover type within 2 km radius around site, according to the 2020 Land Cover of Canada dataset.
Table 1. Percentage landcover type within 2 km radius around site, according to the 2020 Land Cover of Canada dataset.
ArkellElora
Cropland35.95%85.45%
Urban27.28%7.51%
Forest31.04%6.69%
Barren lands2.50%0.12%
Shrubland1.26%0.10%
Grassland0.05%0%
Wetland1.04%0.10%
Water0.89%0.03%
Table 2. PERMANOVA results for both the full dataset and a Canadian Pest subset.
Table 2. PERMANOVA results for both the full dataset and a Canadian Pest subset.
dfSum of SquaresR2FPr (>F)
Full dataset
Site10.580.145.141.00 × 10−4***
Crop type30.700.172.081.00 × 10−4***
Residual252.800.69
Total294.071
Pest
Site10.260.093.773.00 × 10−4***
Crop type30.850.304.151.00 × 10−4***
Residual251.710.61
Total292.831
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Steinke, D.; Ashfaq, M.; Ho, C.Y.; Perez, K.H.J.; Sones, J.E.; DeWaard, S.L.; DeWaard, J.R.; Ratnasingham, S.; Zakharov, E.V.; Hebert, P.D.N. Farmland Biodiversity Monitoring Using DNA Metabarcoding. Diversity 2025, 17, 585. https://doi.org/10.3390/d17080585

AMA Style

Steinke D, Ashfaq M, Ho CY, Perez KHJ, Sones JE, DeWaard SL, DeWaard JR, Ratnasingham S, Zakharov EV, Hebert PDN. Farmland Biodiversity Monitoring Using DNA Metabarcoding. Diversity. 2025; 17(8):585. https://doi.org/10.3390/d17080585

Chicago/Turabian Style

Steinke, Dirk, Muhammad Ashfaq, Chris Y. Ho, Kate H. J. Perez, Jayme E. Sones, Stephanie L. DeWaard, Jeremy R. DeWaard, Sujeevan Ratnasingham, Evgeny V. Zakharov, and Paul D. N. Hebert. 2025. "Farmland Biodiversity Monitoring Using DNA Metabarcoding" Diversity 17, no. 8: 585. https://doi.org/10.3390/d17080585

APA Style

Steinke, D., Ashfaq, M., Ho, C. Y., Perez, K. H. J., Sones, J. E., DeWaard, S. L., DeWaard, J. R., Ratnasingham, S., Zakharov, E. V., & Hebert, P. D. N. (2025). Farmland Biodiversity Monitoring Using DNA Metabarcoding. Diversity, 17(8), 585. https://doi.org/10.3390/d17080585

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop