Machine Learning-Empowered Students Monitor and Safeguard Whale Communities

In the summer of 2021, Chinmay Govind and Nihar Ballamudi, students at the University of Pennsylvania, have embarked on an exciting journey as part of the Penn Undergraduate Research Mentoring Program (PURM). Their project, focused on whale monitoring, is sponsored by Joseph Kroll, a professor in the Department of Physics and Astronomy.

Their research, led by Govind and Ballamudi, uses AI to locate and census whales across broader areas, rather than at a single fixed site. Govind's focus is on locating individual whales, while Ballamudi's focus is on censusing whale populations to more effectively monitor their movements.

One of the challenges they face is the time difference of arrival for a whale call. Ocean noise, multiple whales producing sounds within close proximity, and multipath (excessive sound that echoes off the ocean floor and surface) are some barriers to this. However, Govind has been able to record the origin points of whale calls with a median error of 20 milliseconds, thanks to the AI models he uses.

Govind's models estimate the time difference of arrival of whale calls from the recorded audio data to calculate the whale's coordinates. Meanwhile, Ballamudi uses machine learning models and NOAA sound data to simulate sea environments and census whale populations.

Their research project combines mathematics, signal processing, animal behaviour, and machine learning to map the locations of whales and determine the size of whale populations in target areas. Once they can record the exact number of whales in a target range, they could retroactively pinpoint the precise location of each whale.

John Spiesberger, a visiting scholar in the Department of Earth & Environmental Science, and his son, Ari Spiesberger, a recent Penn graduate with expertise in machine learning models, are mentoring Govind and Ballamudi. Spiesberger has trained them to use AI models, build realistic simulations, and prime the software to predict unknown variables. He also invites Govind and Ballamudi to present their results to U.S. Navy sponsors and share opportunities for expanding upon this research.

Spiesberger aims to secure grants for Govind and Ballamudi so they can continue this impactful work long after summer concludes. The findings of their research could help policymakers and environmental groups make informed decisions about policies involving whales. Moreover, the research could potentially extend to the monitoring and conservation of other sea animals.

In addition to this, Spiesberger has also coached Govind and Ballamudi on enhancing their science communication skills. The research uses data from the National Oceanic and Atmospheric Administration (NOAA) sound receivers north of Cape Cod Bay, but it can apply to any location. Govind expresses interest in developing a model that can recognize multiple sources at the same time.

This innovative project is a testament to the potential of AI in conservation efforts and the exciting research being carried out by the next generation of scientists at the University of Pennsylvania.