An innovative light-based tool deciphers hover fly wing spectra in real time, addressing insect surveillance gaps and monitoring the unfolding biodiversity crisis.
Insects are the most diverse group of animals on Earth, maintaining complex relationships with the environment and playing important ecological roles such as decomposing organic matter and pollination. The deepening biodiversity crisis is evident in insect populations, whose decline has a dramatic domino effect on both plants and animals.
Monitoring insects and analyzing their distribution over time is important to understand the factors that influence their abundance and diversity, but suitable monitoring tools are very limited.
“As a physicist entering the world of entomology, I was quite shocked to learn that there was actually very little data on insect populations,” said Benjamin Thomas, an associate professor at the New Jersey Institute of Technology. As an expert in optical sensors, he has spent the past decade developing photon tools to study insects.
lack of means
The lack of insect monitoring tools is also a problem that Mikkel Breidgaard and his team at Lund University’s Department of Physics are working to solve.in recent papers was announced on cutting edge sciencethey report an easy-to-use approach that analyzes how an insect’s wings reflect light to determine its species and sex, while also providing hints about the ecological niche the insect occupies.
Traditional monitoring techniques rely on captured insects, which are labor-intensive and error-prone, especially among species that are very similar to each other, although the number of observations is small; The approach used can make a difference, and researchers can do this. Specialized cameras will be brought to the site to monitor free-flying populations in real time.
“This is a step toward improving the identification accuracy of these sensors and could make a big difference in our ability to monitor insect populations,” said Thomas, who was not involved in the study. he said. “When combined with new AI tools and machine learning classifiers, this technology holds great promise.”
Unique spectral fingerprints of hoverfly wings
Similar to humans’ unique fingerprints, the delicate feathers of insects can also be used for identification. When imaged with a hyperspectral camera, which can capture a wide spectrum of light far beyond the capabilities of our eyes, we can reconstruct the wing structure based on how strongly and evenly it reflects light.
“We can see how each subtle color is affected by wing thickness, even for colors outside the visible spectrum,” said Dr. Meng Li. Student in Bridegoal’s group and first author of the paper. “This detailed color information allows us to more accurately study and identify insects based on their unique wing patterns.”
To test their idea, Meng Li analyzed 600 hoverflies from 30 species provided by the Lund University Museum of Biology. Hover flies, the second largest pollinating insect after honeybees in the wild, constitute a large family of insects consisting of approximately 6,000 species that vary widely in morphology, larval diet, habitat, and behavior. One of their special skills is imitating other insects such as wasps and bees, and they are often seen “hovering” over their favorite flowers.
When asked whether their findings could be directly applied to free-flying insects (such as those that exist in the field), Lee said that the thickness of an insect’s wings does not change throughout its life, and that “the thickness of an insect’s wing does not change throughout its life, and that it is Even the dried specimens on display do not change.”Without moisture, interference patterns appear [similar to those] of living insects of the same species. ”
To demonstrate that their camera could accurately identify different species of hoverflies, the researchers hand-picked datasets from each species containing parameters that define wing structure. They also estimated the flapping frequency of the species, which is currently commonly used to identify insects, based on their weight and wing size. Combining all these parameters, the team achieved his identification accuracy of 91%. significantly higher than using just the wing beat frequency.
They also found that the spectral characteristics of hoverfly wings vary widely not only between species but also between sexes. This also applies to species where female and male flies are very similar to each other and are therefore difficult to distinguish by eye. Additional sex differentiation is therefore invaluable when monitoring insect populations and reproduction.
The enormous size of the hoverfly family is reflected in an impressive variety of different behaviors between species, so Breidgaard and his team also investigated the correlation between these different behaviors and wing spectral characteristics. Now you can. To do so, they first compared closely related species and then correlated wing structure with factors such as behavioral mimicry and habitat type. The researchers found, for example, that hoverflies with thick wings tend to mimic honey bees, are active in the summer, and are primarily found in meadows and forests.
Interestingly, although closely related species generally had similar plumage, behavioral traits correlated even more strongly with changes in wing thickness than genetics. This means that the spectral characteristics of insect wings are not only unique identifiers of species and sex, but also help to further characterize the insect’s ecological niche. This is particularly useful when monitoring the impact of external factors such as agriculture and deforestation on insect diversity.
The future of insect monitoring
Of course, this is not the end of the story. Their proof-of-concept study is now paving the way for further research to improve the method and confirm that it can be applied to diverse habitats. The technique could also be tested and improved on other types of insects, which have very different wing structures.
According to Thomas, “[The analysis] To achieve high species identification accuracy, it must be combined with other information. This means that large databases of optical properties for each species need to be created so that they can be used to identify diverse populations of insects. ”
Bridegaard and his team say they are optimistic that they will continue to develop insect monitoring tools and further improve the sensors, for example to increase the distance at which insects can be captured.
According to Brydegaard, their method can be implemented using sensors already in use in the field and, importantly, on free-flying insects. This could be a promising start for managing large insect databases and ultimately monitoring insects in rapidly changing habitats.
References: Meng Li et al. Identification of hover fly species and sex using wing interference signals, advanced science (2023). Doi: 10.1002/advs.202304657
