Echo Friendly

Two UH researchers unravel the secrets of dolphin sonar and spawn two companies

January, 2004

In 2000, University of Hawaii researchers Paul Nachtigall and Whitlow Au published a scientific paper titled, “Dolphin Biosonar: A model for Biomimetic Sonars.” The paper laid out the most detailed description to date of the biological mechanisms that make the echolocation systems of dolphins extraordinarily efficient and powerful. And it provided a blueprint for the U.S. Navy to build better sonar systems, with improved capability, to spot dangerous mines. Ultimately, the dolphin-based sonar could have wide ranging uses, including everything from underwater geological surveys to noise-cancellation software in cell phones.

The secrets of sonar are just a few of the many things cetacean illuminated in the past 10 years by Nachtigall and Au during their tenures as top dogs of the University of Hawaii’s Marine Mammal Research Center at the Hawaii Institute of Marine Biology on Coconut Island. (Nachtigall is the director and Au is the chief scientist). The duo landed at UH in 1993, after the U.S. military decided to streamline operations at Kaneohe Marine Base in Windward Oahu and shutter the Naval Oceans Systems Center research laboratory, where Nachtigall and Au held research positions. Declining a Navy offer to transfer to a similar San Diego lab, Nachtigall and Au brought their coterie of cetaceans to pens on Coconut Island. They relocated their offices to a building they constructed out of eight shipping crates, with the help of the Marine Corps.

Over the subsequent decade, from this humble setting, Nachtigall and Au brought millions of dollars in research grants to Hawaii and produced prodigious amounts of research. Nachtigall estimates he currently brings in $1.2 million per year from a host of federal entities, including the National Marine Fisheries Service, the Office of Naval Research and the National Science Foundation. That budget supports a full- and part-time staff of 14, including graduate students and research assistants. It also buys plenty of fish to feed the MMRC’s in-house research subjects – two bottlenose dolphins and one false killer whale. The cetaceans get kid-gloved care, because it’s increasingly expensive to procure marine mammals for research purposes. “There is nobody that treats animals better than we do,” says Nachtigall.

Until the 1970s, scientists had only the barest understanding of how marine mammals used echo-location. Over time, Nachtigall, Au and other researchers around the country slowly learned how cetaceans have evolved unique biological structures, such as powerful vocalization muscles and curved surfaces, to magnify sound inside their heads. These structures help them create loud clicks, sometimes over 200 decibels (that’s louder than standing next to a lawnmower). The cetaceans then carefully listen to the echoes of their clicks bouncing off objects around them, to create a mental picture of their physical environment. This is no simple trick. The ocean is an inherently noisy place, and the brains of cetaceans have developed over millions of years as highly efficient, signal- processing systems. “The animals are excellent signal processors. They know exactly what they are looking at with their clicks, and they can look precisely at one particular target. Even if it moves back and forth, they can stay locked onto it,” explains Nachtigall.

The most obvious use for this type of sonar capability is to spot mines, something that the U.S. military and other militaries around the globe already do with trained dolphins. Ultimately, software mimicking cetacean methods of noise cancellation and signal processing could prove broadly useful. For example, cell-phone makers would love to build handsets that better cancel out street and wind noise. This would allow customers to use the phones in situations where they now cannot. Geologists would appreciate sonar systems that can more accurately lock onto promising targets during surveys of the ocean floor.

In fact, the research has already borne some commercial fruit. Two companies – one in Hawaii, one in California – have either spun off or directly benefited from Nachtigall and Au’s expertise. The Hawaii firm, Neptune Technologies, is currently working on a secret project to design a bionic dolphin for the U.S. Special Forces. The submersible dolphinbot would contain sophisticated sonar equipment and would be remotely controlled using a fiber-optic tether. “We are using the echo-location models that he developed over his 20 years of research. We mimic that to determine what’s in, on or around the seafloor, just as a dolphin would,” explains Jack Harmon, vice president of engineering for the firm, which received a Small Business Innovative Research grant from the Department of Defense to pursue dolphin sonar modeling. Harmon says that if the DOD likes his prototype, it could result in a production order of hundreds of these robotic dolphins. They would not be built in Hawaii, but the additional research and development and further production systems research could create dozens of high-paying, high-tech jobs here.

The California firm, DolphinSearch, develops software used to search through reams of text for relevant content. Founded in 1999 by then-UH researcher Herbert Roitblat, a colleague of Nachtigall and Au in the dolphin sonar-research field, DolphinSearch uses the same type of signal processing sonar content to separate the noise from the signal – in this case, useful news clippings or legal documents. (Roitblat still collaborates with Nachtigall and Au).

These days, the Coconut Island duo (along with their new colleague Alexander Supin from the Russian Academy of Sciences in Moscow) are busy figuring out how dolphins so effectively filter out louder noises, such as the echoes from their own clicks, to focus on more important, smaller ones, such as echoes from nearby fish. It’s all in the swim of things for Nachtigall and Au.

Related Stories

Magazine Promo

On Newsstands Now

HB-10-14Cover

HB October 2014 Wahine Issue

Author:

Alex Salkever