NSF Award
2423087
Sound plays an important role in all organisms’ lives from detecting predators, prey and ambient conditions to communicating to potential mates, offspring or competitors. Thus, vast insight can be gained by understanding patterns of sound in animal communities. Soundscape analysis has traditionally focused on airborne sound, but most ‘sound’ does not travel through air but instead through solid substrates like plant stems and leaves—e.g., 93% of insects and numerous other animals use vibrational communication. Thus, studying vibrations can provide unprecedented insight into animal lives. This study develops novel tools to analyze entirely vibrational acoustic scenes—vibroscapes—to test hypotheses about how animal communication evolves and how noise dictates the daily lives of animals. The team—a midcareer scientist and the Cornell Lab of Ornithology—will use existing recordings of vibroscapes across different grassland sites to develop artificial intelligence approaches to automating vibrational sound analysis. Then, they will apply the new tools to generate data to test ecological and evolutionary hypotheses about when and how insects vibrationally communicate and test the potential for anthropogenic change to impact animal communities. In addition to unprecedented insight into animal lives, the research will develop vibrational analysis tools that will be made freely available to the scientific community. These tools will expand access to the study of vibrations (Biotremology) and open the door to studies providing a vastly better understanding of how animals navigate their worlds and communicate with one another.<br/> <br/>Our understanding of the behavioral processes shaping vibrational signaling communities lags far behind that of airborne acoustic communities. In parallel, the toolsets needed to process large vibrational data sets is currently non-existent compared to the rapidly increasing number of Artificial Intelligence models used for airborne soundscape analysis. By generating tools for analyzing vibroscapes this work will reveal unquantifiable insight into animal acoustics and behavior, including the ecological and evolutionary processes shaping communication and how noise dictates animal behavior. Furthermore, this work tests fundamental principles of animal communication by analyzing communication networks and interactions among vibrational signalers. Furthermore, this work will train a mid-career PI in AI approaches, infuse the Cornell Lab of Ornithology with vibrational expertise, and develop tools that Biotremology researchers globally can use for their own data sets. The PI will also incorporate AI approaches into their advanced evolution course and will train students from a variety of stages and institutions in cutting edge techniques. Overall, this work is well-positioned to shed unprecedented insight into the most common form of acoustic communication and test basic principles of communication and human impacts on animals’ lives.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
