Biodiversity by DNA — Life in a Southern Forest

Our patch of forest is now part of a global program sampling insect biodiversity. Over the next year we will be collecting thousands of flying insects for DNA analysis. Some will be species we know well, others will be new to us – and no doubt many will be new to science.

So how did this come about?

Late last year we received an email from Paul Hebert, CEO of the Centre for Biodiversity Genomics at the University of Guelph, Ontario, instigator of the Global Malaise Trap Program (GMP), and founder/Scientific Director of the International Barcode of Life initiative – inviting us to take part. We were both surprised and delighted.

Of course we jumped at the chance!

The GMP began in 2012. It now involves nearly 50 countries and hundreds of sampling sites.

"The program represents a major step toward understanding how arthropod communities vary across space and time and how they respond to environmental change." (https://biodiversitygenomics.net/projects/gmp/).

So on 3rd January we set up two Malaise traps, and they are already yielding some impressive results.

Featured

Trap 1, one of two Malaise traps we erected on 3rd January 2026. Both will be in place for the next 12 months. The elevation at this site is the highest across our home forest, about 26m above sea level.

Trap 2, on a track that runs through a low-lying part of the forest (~16m above sea level). The barrier surrounding it is to discourage wombats and wallabies from barging on through. After all, the traps do straddle the forest tracks regularly used by the local wildlife … and by us.

Within a couple of hours of initial set-up, it was clear Trap 1 was working! Insects collect in the upper chamber and are preserved in the ethanol filled collection bottle.

Here’s a glimpse of the first week’s harvest from Trap 1. Even we were amazed by the extraordinary diversity! Granted, conditions were particularly ‘good’ for insects that week, with above average temperatures. A heatwave, in fact.

What the project is all about

At the heart of the project is a taxonomic tool called DNA barcoding. This method compares short, standardised DNA sequences to distinguish between species.

extract from BOLD https://portal.boldsystems.org/record/AUSCL324-12 … DNA barcode for a specimen from South Australia, a crabronid wasp in the genus *Cerceris*.

Individual specimens of a single species share a barcode. That is, the sequence of a particular stretch of DNA is virtually identical between individuals ... males, females, juveniles or adults. But that particular sequence is unique to that species. For a more detailed explanation (link. https://ibol.org/about/dna-barcoding/).

BOLD – a barcode reference library

The Barcode of Life Data Systems (BOLD) portal provides access to a suite of web-based resources for publishing, sharing, curating and analysing DNA barcode records. And importantly from our perspective, it provides ready access to this information for the general public. Paul and I regularly check BOLD records for specific taxa, with a particular interest in those with associated images. Indeed, many users of iNaturalist will be familiar with BOLD.

Featured

The BOLD library currently holds barcode data for more than 800 specimens of Cerceris, spanning 45 across Europe, Africa, North America, Asia … and a few from Australia. These represent 133 species, based on their DNA barcodes.
https://portal.boldsystems.org/result?query=Cerceris[tax]

Currently, filtering the search for Cerceris in Australia yields 19 specimens, representing 8 species (as determined by their DNA barcodes … BINS in this summary table).
https://portal.boldsystems.org/result?query=Cerceris[tax],Australia[geo]

For one species of Cerceris, Cerceris minuscula, there are currently 7 records. The specifics of where, when and how each specimen was collected are all recorded on BOLD, along with other relevant details. This information is freely available to the public (with the exception of any data under embargo while awaiting publication by the researchers involved).
https://portal.boldsystems.org/result?query=%22Cerceris%20minuscula%22[tax]

Another example of search results by genus, this one a page Paul has made good use of in his ongoing study of Australian sawflies.
https://portal.boldsystems.org/result?query=Perga[tax]

When a specimen record is added to BOLD, the system seeks to match its barcode with those already in the reference library. Sometimes there will be a match, nearly always identified to family, and sometimes to genus or even species (Steinke et al. 2024). Often there will be no match ... and that in itself is informative. It indicates a species not previously 'known' to BOLD.

A species barcode is a powerful addition to the taxonomist's toolkit. It supplements the morphological information that has long been relied upon to describe and distinguish species (Stevens et al. 2011). For example, differences in barcode can uncover a 'cryptic species' - one that appears identical to another, yet is genetically distinct. Conversely, some species are so morphologically varied that they might be considered separate species – yet their barcodes suggest they are one and the same. In addition, similarities in barcode sequence can help determine which species are closely related, and which more distant. Of course, the barcode is only a tiny fragment of the genome. Taxonomists will utilise a variety of genetic markers to further test theories of relatedness.

Global Malaise Trap Program

Specimens and their barcode sequences are input to BOLD from a variety of sources. And one of the major contributors is the Global Malaise Trap Project (GMP).

The GMP is one of a range of projects coordinated by the Centre for Biodiversity Genomics in Canada. The Program’s scope is truly global, seeking to “unite researchers worldwide in monitoring terrestrial arthropod diversity through standardized sampling and DNA barcoding” (GMP website).

Featured

GMP coverage, as of October 2025

Ecoregions currently sampled (with at least 10,000 records and 1,000 BINS) as part of the GMP.
https://public.tableau.com/app/profile/evgeny.zakharov/viz/EcoRegionCoveragePublic/EcoRegionCoverage

A standard survey method is essential if data is to be compared over time, and between places. Importantly, GMP employs the same design insect trap at each site (Uhler et al. 2022) and samples are collected and processed following a specified protocol. Once in place, a trap should not be moved during the course of the study ... which in most cases is 12 months, as it is for us.

Featured

The GMP employs the same Malaise trap design, worldwide. For example:
top left: Mt. Kitanglad Range, PHILIPPINES
middle left: SINT EUSTATIUS (Dutch Caribbean island)
middle right: Ziarat, PAKISTAN
bottom left: Beirut, LEBANON
bottom right: Isabela Island, GALAPAGOS / ECUADOR
[images from GMP website]

The aim is to set the trap across an insect flight path. The GMP sampling protocol includes the following explanation:

Site Selection - Deploy the trap at a site which is subject to minimal disturbance and ideally in a habitat’s climax vegetation (i.e. placement in a national park or other protected area is preferred). When possible, position the trap at a right angle to an insect flight line, in areas with low undergrowth; forest edges or clearings and elevated sites are recommended.

Samples are collected weekly and stored cold (ideally in a freezer). Batches are then shipped to the Centre for Biodiversity Genomics for imaging and sequencing.
[image from GMP website]

At CBG, the samples are analysed in detail. This includes a system for imaging individual specimens before sequencing.
[extract Steinke et al. 2024, Figure 1, p. 122]

The Malaise trap is a passive sampling method, designed primarily to intercept insects in flight. It involves no particular insect attractants, and functions 24 hours a day with only limited need for human involvement. This contrasts with sampling techiques such as light traps, coloured water traps, or sweep netting.

Featured

The black barrier net of the trap is set across the insect flight path. Many flying insects move towards light when they encounter an obstacle. Hence the white roof of the tent.

The Malaise trap was initially developed by René Malaise, back in 1934. The version in use across the GMP is a later model, known as Malaise Trap II, Townes Style.
photo: Trap 1, 30th Jan. 2026

Once inside the tent, all upward movement leads to the collection opening at the apex.
photo: Trap 1, 30th Jan. 2026

Once an insect ‘escapes’ the tent through the apex opening, it is trapped in the collecting chamber. Most soon drop into the ethanol filled bottle below.
photo: Trap 1, 30th Jan. 2026

Not all insects are created equal ... at least when it comes to Malaise trapping.

extract from GMTP website https://biodiversitygenomics.net/projects/gmp/ … Project Outcomes (as at October 2025), showing the 20 taxonomic orders with the highest BIN (ie species) counts. Note that this is a log scale.

The most likely to be caught in the Townes-style Malaise traps are flies (Diptera) and wasps/bees (Hymenoptera) (Seymour et al. 2024). Both tend to fly towards light when they strike an obstacle. Upon hitting the barrier net they move upwards to the white roof and from there towards the tent apex and the collection bottle. In contrast, beetles often drop to the ground when disturbed (Uhler et al. 2022). This probably contributes to the marked disparity in catch across these three megadiverse insect orders. For every beetle there are three wasp species, and for every wasp there are two flies!

Despite the ‘flight intercept / positive phototaxis’ model, the traps still harvest a wide range of insect types. Including many wingless species. It seems that some simply climb their way to the top, and then stumble into the bottle.

Featured

Trap 1, Week 4

Flies and wasps dominate each of our weekly collections. Many are large and showy, and there is a huge variety. Others are extremely small – but they all play a role in the forest ecosystem. And all will be barcoded.

With special permission, we view each sample collection before bagging and storage. Any specimens removed for further study are carefully labelled, and we take great care in handling. It is critical to the validity of the GMP data that every insect, no matter how small, is recorded and processed as part of the appropriate sample.

Trap 1, Week 4

It’s not ALL flies, wasps, ants and bees. Many other orders are represented in the mix. In this image alone there are various beetles (Coleoptera), bugs (Hemiptera), moths (Lepidoptera), a cockroach (Blattodea) and grasshoppers (Orthoptera).

Trap 1, Week 4

Trap 1, Week 3

The raspy cricket in the centre of this image was a surprising find. Not only do we see this species only rarely … but it is flightless. It must have climbed the trap mesh and ultimately fallen into the collection bottle. And it wasn’t alone! This single sample contained an adult female (centre, arrow), an adult male, and at least five nymphs (eg arrow top left).

Note too the many tiny insects, some circled in red. Probably flies and wasps … we’ll know soon enough!

Trap 1, Week 3

Trap 2, Week 3

Another postcard for biodiversity. By my reckoning there are six insect orders in shot here, and apart from a few beetles, no two are alike.

With a pair of small spiders for extra variety.

I didn’t arrange them this way. It’s just a random closeup of a part of the (much!) larger weekly sample.

Trap 2, Week 3

Our contributions

There are at least three ways in which we can contribute to GMP and IBOL.

1. Bulk samples from an under surveyed ecosystem

We are located on the very far south coast of NSW, largely surrounded by native forests. Our home patch is also relatively undisturbed forest, comprising a diverse mix of almost entirely indigenous plant species. It is a natural place. And as we live on site, we can readily monitor the traps and make weekly collections.

And importantly, our 'ecoregion' has not previously been sampled as part of GMP.

Featured

We are located in a GMP ecoregion (as per Dinerstein et al. 2017) that includes the far south coast of NSW, the east coast of Victoria, a large part of south western Victoria, and south western slopes of NSW.

Our samples will be the first for our ecoregion!

2. Individual reference specimens

In addition to the unidentified insects in the bulk samples, we will be submitting other select insects for barcoding. These will be specimens where we can confidently provide an identification based on their morphology.

Featured

BEMBICIDAE, Bembicinae

Austrogorytes spryi
There are currently no reference examples on BOLD for this endemic Australian genus.

Collected here by hand, 30th December 2025. Currently stored dry, pinned. (#2512E)

BEMBICIDAE, Bembicinae

CRABRONIDAE, Crabroninae, Miscophini

Sphodrotes punctuosa
Another Australian endemic genus not well represented in the BOLD reference collection is Sphodrotes. There are currently 5 specimens, and all a single species – S. nemoralis.

Having examined our local species in detail, I’m confident it is a different species – S. punctuosa.

Collected here by hand, 24th December 2025 (#2512C). Currently stored in ethanol, but soon to be pinned.

CRABRONIDAE, Crabroninae, Miscophini

MEGALYRIDAE

Now this is one of several rather special specimens – genus Megalyra.

The family Megalyridae is currently represented by just 22 specimens on BOLD, and the genus Megalyra just 7 – from anywhere in the world. With just one from Australia!

Collected here by hand, 19th August 2024. Currently stored dry. (#2408B)

MEGALYRIDAE

3. Knowledge of Australian species

Through our ongoing studies, Paul and I are building our expertise in the identification of particular insect groups. For Paul these are the sawflies, for me crabronid wasps. And for some of these taxa ... the crabronids, in particular ... many of the Australian specimens on BOLD are in need of curation. For those species we know well, and for which there are good images on the BOLD database, we are able to offer suggested identifications. Often this will just be to subfamily or genus, but it all helps to refine the reference database.

Multiple benefits

Our knowledge of local insect diversity will certainly grow through our involvement with GMP and BOLD. For example, we know there are hundreds of species of tiny flies and wasps in the forest .... but just how many? And what families are represented? And how does this change across the seasons?

We will be able to compare our ecosystem to adjacent bioregions. In what ways do the insect biota differ, and what ways are they the same?

In addition, by sequencing select specimens that we have studied in detail, we will be able to compare their barcodes to the BOLD reference library. Is there a match, or to what are they most similar?

And just having the Malaise traps in place has already yielded several specimens we are keen to investigate further! If Paul is very lucky, the traps may yet turn up an interesting sawfly or two ... the adults of which can be frustratingly elusive.

Featured

Proctotrupidae I think, based on the venation and ovipositor. BOLD currently has just 2 specimens of a single Australian species, recognised at the family level. This may be the same species … but it may be different. The barcode will be the test.

Either way, it’s a new family for our home list! No wonder really, given its small size.

(Temporarily isolated from the Trap 1, Week 2 collection, with special permission from Paul Hebert.)

This looks like Pseudoturneria, a crabronid wasp I’ve studied nearby but until now had not found here in the forest. Once I set and pin it, I’ll be better placed to confirm the ID.

Pseudoturneria is yet another Australian genus not currently represented in the BOLD database. We’ll soon change that!

From Trap 1, Week 1.

Now who is this intriguing crabronid?!? At first glance I thought another Pseudoturneria … but perhaps not. The legs are a different colour, so it’s almost certainly a different species. Definitely worth a detailed study before it goes off to BOLD for sequencing.

From Trap 1, Week 3.

The missing female!

Arrolla lawrencei was described from a single male, the female unknown at that time (1990). We have sighted pairs of this species here in the past (details here), but not collected them. It seems we have now! A description of the female is worthy of a short publication … and barcoding, of course!

From Trap 1, Week 3

Irrespective of any insights we gain regarding our local insects, simply being part of such a vast, collaborative effort to better understand the biodiversity of life on earth is its own reward.

References and links

Dinerstein, E. et al. 2017. An ecoregion-based approach to protecting half the terrestrial realm. BioScience, 67(6): 534–545 open access https://doi.org/10.1093/biosci/bix014

Ratnasingham, S. et al. 2024. BOLD v4: A Centralized Bioinformatics Platform for DNA-Based Biodiversity Data. In: DeSalle, R. (eds) DNA Barcoding. Methods in Molecular Biology, vol 2744. Humana, New York, NY. https://doi.org/10.1007/978-1-0716-3581-0_26

Seymour, M. et al. 2024. Global arthropod beta-diversity is spatially and temporally structured by latitude. Communications Biology 7:552 open access https://www.nature.com/articles/s42003-024-06199-1

Steinke, D.; Ratnasingham,S.; Agda, J.; Ait Boutou, H.; Box, I.C.H.; Boyle, M.; Chan, D.; Feng, C.; Lowe, S.C.; McKeown, J.T.A.; et al. 2024. Towards a Taxonomy Machine: A Training Set of 5.6 Million Arthropod Images. Data 9, 122 open access https://doi.org/10.3390/data9110122

Stevens, M.I., Porco, D., D’Haese, C.A. & Deharveng, L. 2011. Comment on ‘Taxonomy and the DNA Barcoding Enterprise’ by Ebach (2011). Zootaxa 2838: 85-88 open access https://www.mapress.com/zt/article/view/zootaxa.2838.1.6

Uhler, J. et al. 2022. A comparison of different Malaise trap types. Insect Conservation and Diversity 15, 666-672 DOI: 10.1111/icad.12604 open access https://resjournals.onlinelibrary.wiley.com/doi/10.1111/icad.12604

Global Malaise Trap Program: learn more

International Barcode of Life: learn more

Search BOLD