How computers counting insects is essential in solving the biodiversity crisis
By building computers, cameras and microphones that automatically monitor insects, DK-based researcher Quentin Geissmann believes we’ll be able to restore our biodiversity and make agriculture sustainable.
Quentin Geissmanns lab at Aarhus University is not filled with plants, soil or cells as most biology labs are. Instead circuits, wires and microchips are scattered across the small room where he and his colleagues tinker with new technology.
He’s a biologist by training, but he’s always been fascinated by electronics and how to use technology to learn more about nature.
For centuries the study of life on our planet has been done by biologists going into the field to collect samples and spot new species. They’ve used nets, buckets, pins and most importantly sketch books to document the enormous variety of life.
But that is about to change.
In the future biologists, in addition to wading around forests, meadows or jungles, will use AI to assist their work. That is at least what Quentin Geissmann, assistant professor at Center for Quantitative Genetics and Genomics, is working towards.
- New technology enables us to monitor insects in ways we’ve never been able to before. Machine learning og affordable camera technology has made this possible. In the future biologists will have more time to interpret data and spend less time collecting it, he says.
Quentin Geissmann
Born in Dijon in France.
Although the city is famous for its mustard, the mustard seeds aren’t grown there anymore, as he explains. Instead he grew up surrounded by vineyards, beet fields and wheat farms.
A biologist by training, he’s always been fascinated by computers and technology. He knows a lot about insect physiology, but these days he’s more focused on the technology part of his research and not as much on the insects.
Came to Aarhus University two years ago and is now an assistant professor at QGG.
Before it’s too late
Across the globe biodiversity is threatened. Intensive farming, climate change and deforestation is rapidly devouring the natural habitats for thousands of species.
With species going extinct at an alarming rate, biodiversity researchers are in a hurry. They need to map and describe thousands of species before they disappear.
Without the technology that Quentin Geissmann and colleagues from around the world are developing, this would be an insurmountable task.
But by making computers do the counting and mapping it’s possible to see changes in biodiversity very accurately. And with this data in hand it will be possible to find the right solutions, he explains.
- Biodiversity and agriculture are very closely connected. When species disappear because of use of pesticides and deforestation, it affects the crops. Without insects we wouldn’t be able to grow many of our important crops.
- By mapping closely what happens to the biodiversity when we change crop rotation, our use of pesticides or irrigation we’ll be able to develop new and effective ways of making agriculture sustainable. Only by monitoring this very closely can we solve these problems.
Computers tracking bees
Right now Quentin Geissmann and his team are working on a camera-setup to monitor pollinators – that is, bees, butterflies, wasps and some types of flies.
The plan is to put up cameras next to agricultural areas in Denmark and across Europe. With this technology he hopes to gain new insights into what affects the pollinators. Is it small changes in the weather, the spraying of pesticides, harvesting or something different?
- Today it is very expensive and time consuming to map how insects are affected by different things. It is mostly done by hand by an entomologist walking the fields and woods observing and collecting specimens.
- The issue with this approach is that it's only a snapshot in time; it doesn't capture the ongoing changes in the population of the insects. With new technology we can monitor and document changes as they are happening. Fast or slow.
Before Quentin Geissmann came to Aarhus University he developed and deployed about 50 cameras in Canada. Devices that mapped some of the insects living in the Vancouver area. The technology he’s working on now is based on these earlier experiments.
- Two and half years ago I built this tool to detect insects in images and track them through time. That was what we used in Canada and with all this data we have developed an image recognition software that can classify 18 different groups of insects.
About the Center for Quantitative Genetics and Genomics
Together with 70 research colleagues at the Center for Quantitative Genetics and Genomics (QGG), Quentin Geissmann is working to deepen our understanding of biodiversity through the use of state of the art monitoring technology.
With the help of large statistical models, researchers at QGG identify the genetic traits in animals and plants valuable to agriculture. For example, a valuable trait would be cows that produce less methane during digestion or produce more milk. Or a trait enabling a plant to better resist a certain type of pest.
But as biodiversity and agriculture are interconnected the center also works with mapping biodiversity through genetic sampling, modeling and automated monitoring technology. At QGG they are part of the monitoring of the Danish wolf population for example.
More than 20 nationalities are employed at QGG working closely with industry, agriculture and governments throughout the world.
Like weather stations, but with insects
Quentin Geissmanns hope is that his technology will lead to some sort of permanent insect monitoring effort around the world. Kind of what we have with weather stations today.
- The technology is quite cheap. All it takes is a white surface, a camera and a Raspberry Pi, which is a small and very cheap computer chip. When the insects land on the white surface, the camera films them and through an app installed on the small computer sends the data to our server.
Because the technology is cheap, it is scalable. Setting up insect monitoring around all Denmark is possible in the near future, he thinks.
- The technology is almost there. I believe we’ll see much more of this kind of biodiversity monitoring in the years to come. And with all this data, we’ll be able to evaluate existing solutions to protect biodiversity, choose the most effective ones or try out and test new ideas.
Listening for worms
Bees, flies and butterflies are only a small part of the vast ecosystem that surrounds our cities and fields. In the earth beneath the plants there is a huge variety of worms, microorganisms and fungi.
A world that we know very little about.
Quentin Geissmann is currently working on technology to monitor part of this. He’s trying out a combination of cameras and microphones to map the comings and goings of the earthworms beneath our feet.
- If you walk on a lawn and really look hard, you’ll be able to see small casts from the burrows of the worms. My idea is to analyse these burrows with a camera and learn more about the worms, he says and continues:
- And then there’s the worms that stay underground. I’m trying to monitor them using sound. Ultrasound or ground penetrating radar combined with a software developed for detecting these worms, could revolutionize our understanding of these animals and how they affect the ecosystems.
If you ask Quentin Geissmann, he’s confident that this research area, which he’s dubbed computational entomology, will become important in the future. And his group at Aarhus University is right at the forefront.
- We have a unique research team in Aarhus. We are able to do stuff that few others around the world can. I really believe we can do a lot to improve biodiversity with our research, he concludes.
Contact
Quentin Geissmann
Tenure Track assistant professor
Center for Quantitative Genetics and Genomics at Aarhus University
Email: qgeissmann@qgg.au.dk
Jeppe Kyhne Knudsen
Journalist and science communicator
Faculty of Technical Sciences at Aarhus University
Phone: +45 93 50 81 48
Email: jkk@au.dk