For the first time in the world, a tool was developed to detect the movement of bats at high altitude

Thanks to the new algorithm, which is now open to researchers from all over the world, it will be possible for the first time to monitor the movement of bats near wind turbines and develop new methods to protect them

A new algorithm developed by researchers from the university and Tel Aviv University succeeded for the first time in the world in detecting the movement of bats with the help of radar. from the findings published in the new study in the journal Methods in Ecology and Evolution It turned out that bats also migrate during the months of the year and that their migration patterns are fundamentally different from those of birds. Thanks to the new algorithm, which is now open to researchers from all over the world, it will be possible for the first time to monitor the movement of bats near wind turbines and develop new methods to protect them. "The use of wind turbines in the field of renewable energy around the world and in Israel is gaining momentum and the ability to operate them in an environmentally friendly manner is based on threshold values ​​for maximum damage to the wing owners by the turbines. But until now there was a significant gap in the understanding of the intensity of bat movement near the turbines, which caused a lack of effective removal methods and a high mortality of the bats. Thanks to our development, they will now be able to track and monitor the movement of bats all over the world - and thus save many of them," said doctoral student Yuval Warber from the university, who developed the algorithm under the guidance of Prof. Nir Sapir from the university and Prof. Yossi Yuval from Tel Aviv University.

In recent years, the use of radar to study aerial animals is gaining momentum. With the help of the radar you can get data on the size of the flying animal, flight speed, wing flap pattern and body structure. These data allow researchers to examine basic ecological issues in unprecedented detail, and they are used as mandatory data in infrastructure and development projects that may harm air routes with wings - from wind turbine farms to the construction of residential towers.

But so far, the radar data has not been able to separate birds and bats. In the end, in order to develop an algorithm that will know how to identify the animal, human learning is necessary. That is, that a human will enter data regarding each observation and even confirm that the radar has indeed identified the animal. Because bats fly mainly at night and at high altitude, it has not been possible until now to collect verified bat data to train a bat detection algorithm. The result: many infrastructure projects were built without considering bats, which also led to a lot of bat mortality.

In the current study, Prof. Nir Sapir and doctoral student Yuval Werber from the Department of Evolutionary and Environmental Biology at the university along with Prof. Yossi Yuval and research student Hadar Saxtin from the School of Zoology at Tel Aviv University asked to build an algorithm based on machine learning that can identify bats in radar systems for aerial animal research. To face the challenge of creating a database of verified bats for the machine learning stage, the researchers used a database collected by a network of bird radars deployed throughout the State of Israel from Hermon to Eilat over seven years. The database includes over 18,000,000 animal sightings that have been identified and recorded from 50 meters to 1,500 meters above the ground and is a unique research resource at a world level. In the first phase of the study, the researchers took advantage of the many years of familiarity with migration patterns in the Hula Valley area and isolated a period of the year - the first two weeks of June - when there is no nocturnal activity of migratory birds over the Hula Reed - that is, any nocturnal sighting that is not an insect at this time is most likely of a bat . In the second step, the estimated bat sightings were isolated based on activity times, wing movement biomechanics, and animal size. Thanks to the first two steps, the researchers were able to develop an algorithm that knows how to distinguish between birds and bats. After the development of the algorithm, various validation series were conducted for the algorithm and it was found that it reaches accuracy levels of over 90%. "With the help of the algorithm we developed, we created a unique database in the world of about 60,000 observations of bats in information collected throughout the year," said the researchers.

With the help of the unusual database, it was found that bat migration constitutes about 10-15% of the scope of songbird migration in Israel and that bat migration differs in its characteristics from bird migration - bat migration begins later compared to bird migration in both migration seasons and while songbirds migrate mainly in an altitude range of between 100 meters to 1000 meters above the ground, bats mainly fly in a much smaller height range, between 200 meters and 600 meters.

"Millions of bats die every year around the world as a result of the operation of wind turbines that aim to generate electricity in a non-polluting way. But even non-polluting sources of energy production have environmental consequences that can be serious. In this case, it is known from various studies in the world that in some farms more than ten times more bats die than birds. "Thanks to the algorithm we developed, it will be possible to monitor the movement of bats around the wind turbines and protect them as birds are protected," said research student Warber.

"The application of the algorithm we developed on radar data from other sites in the world is expected to produce a global database of bat movement, thus enabling the study of the environmental factors that affect bats and this for the purpose of an in-depth understanding of their behavior and the possible effects of climate change on many populations of bats around the world." concluded the researchers.

More of the topic in Hayadan:

• A quarter of the mammal species and about 10 percent of the bird species are in danger of extinction
• Meteorologists with wings
• The "super-sensory" watch of bats
• What is the connection between bats and Jews and Christians in antiquity?