A Golden Field Experience: Adventures of an ELP Research Analyst
ELP researchers have to be ready for anything. Exploring the soundscapes of African elephants often requires long hours of data-crunching and analysis in the Lab, but can also include exciting forays into the field. In 2018, for example, ELP research analyst Bobbi Estabrook traveled to eastern Senegal to investigate the potential negative effects of gold mining on elephants. ELP had been contacted by the Petowal Mining Company, who were interested in understanding how their mining operation, Toro Gold, could potentially affect the behavior of forest elephants in the nearby Niokolo-Koba National Park.
Once daily, the company inserts dynamite into holes in the ground to reveal gold in the rock¹. Although these blasts are relatively infrequent compared to other mining operations, the company was concerned that the sounds from these explosions could disrupt elephant communication. Transect surveys conducted in 2016 and 2017 suggested that elephants were active near the park boundary, about 10 kilometers from the mine². The low-frequency sounds created by the blast could possibly cover up the low rumbles that elephants use to communicate, a phenomenon known as “communication masking”². For these highly social animals, being unable to hear each other’s calls could seriously impact their survival and well-being².
In order to determine if the mine’s blasts were loud enough to disturb elephants, ELP designed a short-term study using passive acoustic monitoring technology. Recording units had to be placed at varying distances from the mine to assess how far the blast sounds traveled into the park. One recording unit was set up right next to the mine, to receive near-range noise levels. Another unit, 31 kilometers away from the mine, acted as a control to record ambient noise². The other four recorders were located at 5 and 10 kilometers from the mine, along the two elephant paths identified by the earlier transect study². These recorders, called Swifts, were developed in-house by Cornell’s Lab of Ornithology and could record sound autonomously with little maintenance required³.
In July 2018, Bobbi flew out on a plane paid for by the Petowal Mining Company to spend ten days at the mining site¹. Every day, Bobbi, two men from the mining company’s environmental team, and a local guide trekked for hours in the punishing sun to set up the Swift recorders at the selected locations. It was the beginning of the rainy season in Senegal, and the weather posed a serious challenge. As she explained, the frequent torrential downpours could flood the paths they had taken in the morning, so that they were uncrossable rivers when they tried to return in the evening.
The Swifts needed to be placed high up in the trees to avoid being damaged by curious animals. Although they saw no elephants on their treks through Niokolo-Koba National Park, Bobbi and the team often encountered signs of other wildlife, from nests left behind by chimpanzees to huge termite mounds. On one excursion, they found fresh prints from a lion — and almost ran into the maker of the prints later on. As Bobbi relates the story, they were walking through a patch of bamboo forest and saw a “large, tan-colored shape” through the stalks¹. The lion, a rare sight in the area, was startled by their presence and ran off.
Working with the mining company was also a unique experience. Bobbi stayed at the mine’s employee housing, which consisted of a row of shipping containers transformed into apartments¹. The apartments had the luxury of air conditioning and electricity — unusual, Bobbi said, for work in the field. After showing the environmental team how to replace the batteries and memory chips on the recorders, Bobbi was able to return to her previous research in Ghana. As Bobbi says, “that’s the beauty of passive acoustic monitoring”¹.
Although one of the six Swifts malfunctioned due to water damage, ELP was able to collect audio data for a period of 134 days. The amplitude (decibels) of blast signals at each recording site were only measured for the most common frequency range for elephant communication, 30 to 50 Hertz². Hertz and decibels are not synonymous measurements of sound; Hertz is a measure of a sound’s pitch, or frequency, while decibels measure the loudness or intensity level of sound. So, ELP only considered the intensity of sounds recorded within the low-frequency range of elephant calls. The mining company provided metadata for the times of each blast, which, when cross-referenced with the sound recordings, were used to locate and “tag” blast signals for analysis. The amplitude of each blast also needed to be compared to the normal “background noise” that an elephant would experience in their environment. The data was used to create long-term spectrograms, which displayed the average ambient sound levels of the area over time².
Ultimately, Bobbi and ELP director Peter Wrege concluded that the mining blasts were not likely to pose a threat at the distance where elephants had been identified. With each kilometer, the blast sound decreased by about 5 decibels². As expected, the recording site closest to the mine recorded extremely loud blasts. However, the second-closest site was not louder than even the farthest sites, suggesting that the sound did not travel far². This could be explained by acoustic shadowing; the sound may have been prevented from advancing into the park by the area’s relatively hilly landscape¹. Most likely, the blasts didn’t even register for the elephants in Niokolo-Koba National Park; in some cases, the thunderstorms were louder than the dynamite blasts. At 10 kilometers from the mine, the blasts registered at around 68 to 82 decibels, while the thunderclaps opportunistically recorded by the Swifts had an average amplitude of 79 decibels².
The Senegal project was a relatively short-term, narrowly focused study, and there remains more to investigate. For example, the study was limited to the low frequencies that elephants use to communicate; ELP’s sound data could be used to explore the possible effects of the blasts at higher frequencies, and on other species¹. Bobbi’s field experience is not only an example of the diversity of ELP research, but also evidence that monitoring the potential impact of human activity is not necessarily cost or time-intensive.
¹Estabrook, Bobbi. “Senegal Project.” Interview by Annalisse Eclipse. 21 May 2020.
²Wrege, Peter, and Bobbi Estabrook. Toro Gold: Blast Monitoring in Elephant Habitat of Niokolo-Koba Park. Technical Report. Cornell Lab of Ornithology Bioacoustics Research Program, Elephant Listening Project.
³SWIFT — Terrestrial Passive Acoustic Recording Unit. Cornell Lab of Ornithology, www.birds.cornell.edu/ccb/swift/.