Voices for Conservation

Using the sounds of forest elephants to aid in their protection

By Katy Payne and Karyn Rode

Video footage shows Eli charging at Greyboy; a sonogram shows Greyboy's scream as he retreats. The sonogram was generated by Raven software from the Lab's Bioacoustics Research Program.

Why in the world is the Cornell Lab of Ornithology the home of a thriving project on elephants? Elephants may not be birds, but they are part of a larger whole with significance for birds and all of nature—so they are included in the Lab's broad mission of conservation.

In the Lab's Bioacoustics Research Program, we study bioacoustics—the sounds of life. Sounds—whether made by birds, elephants, or whales—can provide key information to be used in conservation. This is particularly valuable when wildlife can be heard but not seen, such as night-time songbird migrants, whales beneath the ocean, or nocturnal bats. Forest elephants are another example—they live under the thick canopy of the central and west African rainforest, where they are difficult to see, but they are highly vocal. Along with other studies in the Lab's Bioacoustics Research Program, these efforts are helping to push forward the whole remarkable process of acoustic monitoring, something we view as a quantum leap in the conservation of nature.

Already, the Elephant Listening Project is gathering information about a keystone species in one of the greatest forests on earth (more than two million square kilometers in central Africa). To decimate forest elephants is to degrade that forest, and vice versa. Such degradation is occurring at an alarming rate as a consequence of a trio of human activities: logging, immigration, and poaching—illegally killing elephants for bushmeat and ivory. The help we are poised to offer is coming in the nick of time. New bioacoustics technologies are making it possible to locate, count, and describe the threatened elephants as well as detect gunshots, chainsaws, and vehicle noise—evidence of human activities that are related to the elephants' apparent decline.

Aida IV takes shelter from danger under her mother?s belly.

Photo by Andrea Turkalo

Of the three living elephant species, the African forest elephant is the only one that still has extensive habitat and little direct conflict with human beings in much of that habitat. For these reasons, we suspect it has the greatest chance of surviving as sustainable wild populations. However, the inability to see through the forest means there are large gaps in information.

Estimates of the number of African forest elephants vary fivefold (between 19,000 and 109,000 individuals), according to the International Union for Conservation of Nature (IUCN). No information exists on the structure—and therefore health—of most populations, and surveillance is difficult, making it challenging for law enforcement to suppress poaching. For conservation organizations and local governments, lack of information about elephant numbers, health, patterns of activity, and poaching, results in a dangerous ignorance about how to protect these animals effectively. This is the area in which we can offer improvement.

Over the past five years, the Elephant Listening Project has explored ways of using elephant vocalizations as indicators of population abundance, behavior, and demography (age and sex structure). In 2000 and 2002, the Elephant Listening Project worked in the Dzanga forest clearing in the Central African Republic, an unusual site where large numbers of forest elephants can be seen on any day of the year. There we were guests of Andrea Turkalo of the Wildlife Conservation Society, who has monitored the clearing visually for 15 years and recognizes some 3,200 elephants by eye.

Our purpose was to learn as much as possible about patterns and meanings of elephant vocalizations where we could simultaneously watch and record the animals without disturbing their natural behavior. After hauling a set of autonomous recording units into trees (out of reach of curious elephants), we recorded continuously during each three-month period, capturing several hundred thousand elephant calls. Simultaneous observations and video recordings placed thousands of these calls in context—illuminating the age, sex, and identity of the calling elephant and the circumstances surrounding each call.

This enormous amount of data has still not been fully mined, but our work has shown that elephant numbers can be estimated over large areas using acoustic information alone—without the presence of humans. We can use the rate of the elephants' low-frequency vocalizations to assess their density in the sampled area. The elephants' calls are also traveling far enough to make them useful in large-scale population censuses. (An average forest elephant call is discernable in roughly a two-square-kilometer forested area.)

Much more information is still being processed, including the extent to which particular elephant call types can be automatically recognized by a computer program as it browses through long recordings. This may enable us to document the age, sex, and behavior of unseen callers, indicating whether there is acoustic evidence of reproduction and a good balance of males, females, and infants. No other method of surveying elusive animals in forests has such potential to deliver information on the structure and health of endangered species.

Researcher Karyn Rode in the field.

Photo courtesy of Karyn Rode

Over the next year we are focusing our efforts in Gabon, which has recently established 13 new national parks, some of which are brimming with forest elephants. Our monitoring efforts will first focus on mineral-rich forest clearings, which—although few are as elephant-rich as Dzanga—are common throughout central Africa and are also important areas for gorillas, chimpanzees, bongo, forest buffalo, Gray Parrots, turacos, and other noisy animals.

Sad but obvious to say, these areas are also hotspots for poaching. Remote recording units will eavesdrop on forest clearings to identify areas of poaching activity and gather quantitative baseline data on seasonal and daily patterns of elephant use. Recordings made in later seasons will be compared to these baseline data. It is only with such information that the effectiveness of conservation programs can be determined and specific strategies for promoting elephant conservation can be identified.

We have set our sights on these immediate goals, but we will close by mentioning a pair of loftier hopes that reflect other developments now underway in the Lab's Bioacoustics Research Program. The first is a hope based on a monitoring system for whales in the waters of Cape Cod. Computerized buoys will recognize calls of right whales and transmit the information to satellites overhead which in turn beam them to managers of commercial shipping lanes. These managers can decide to detour ships when the whales are in harm's way.

A comparable effort would be real-time monitoring of gunshots and elephants to allow timely responses to poaching activity in remote national parks. This could make a major difference in the incidence of poaching. We also envision a time when the recordings from the Elephant Listening Project will be used to determine the presence or absence of other vocal species, in addition to elephants. All this potential keeps us hard at work.

As with many programs at the Lab, research for the Elephant Listening Project is completely supported by donations and grants. Time is very much in short supply, and we deeply appreciate all the help we can get.

To make a special gift for the Elephant Listening Project, see the article How You Can Help.

Katy Payne is founder and past principal investigator of the Elephant Listening Project. Conservation biologist Karyn Rode is the project's new principal investigator. The authors express their gratitude for all the help they are receiving as they attempt to confront this great challenge in the company of so many superb friends and collaborators.

For permission to reprint all or part of this article, please contact Laura Erickson, editor, Cornell Lab of Ornithology, 159 Sapsucker Woods Rd., Ithaca, NY, 14850. Phone: (607) 254-1114. email: lle24@cornell.edu