How Camera-trap Data Can Help Us Predict Earthquakes

© Benjamin Drummond

Editor's note: Since 2007, the Tropical Ecology Assessment and Monitoring (TEAM) Network — a coalition of CI, the Smithsonian Institution and the Wildlife Conservation Society — has collected more than 2 million photos from camera traps in tropical forests across the globe. Although the primary goal of this collection is to assess ecosystem health and document change, it can be used for a remarkable number of unexpected purposes, as evidenced by this recent study.

Earthquakes can cause horrific damage to communities across the world. In the United States alone, scientists estimate that earthquakes are responsible for roughly US$ 2.5 billion in damage per year. For poorer developing nations, though, earthquakes can be even more devastating.

The inability of researchers to accurately predict earthquakes has been a major impediment to reducing the high number of fatalities caused by these catastrophic events. However, by using camera trap data from the TEAM Network, researchers at Anglia Ruskin University may have found a way to identify when and where earthquakes will occur.

The new study, led by Dr. Rachel Grant of Anglia Ruskin University and co-authored by Dr. Friedemann Freund of the SETI Institute and Dr. Jean-Pierre Raulin of the Centre of Radio Astronomy and Astrophysics, used TEAM camera-trap data from Peru’s Yanachaga-Chemillén National Park to analyze how animal behavior changes prior to a major earthquake to determine if those changes can help improve short-term seismic forecasting. Their findings were published by the journal Physics and Chemistry of the Earth last week.

The new study used data from motion-sensitive cameras placed in the park by the TEAM Network during the dry season every year and found significant changes in animal behavior 23 days before a magnitude 7.0 earthquake struck the nearby Contamana region in 2011. Within that 23 day period, fewer animals were captured on camera. Five days prior to the earthquake, animal activity essentially dropped to zero, indicating an aversion to the area being monitored.

Grant and her colleagues believe that the observed changes in animal behavior may be a response to the massive increase in the number of positive airborne ions generated by seismic activity leading up to the Contamana earthquake. Previous studies have shown that positive ions can cause disagreeable side effects in animals and humans, such as “serotonin syndrome,” which can lead to symptoms like restlessness, agitation, hyperactivity and confusion.

By comparing changes in animal behaviour to changes in very low frequency radio waves, which are impacted by the presence of positive ions, scientists found a particularly large fluctuation above the area surrounding the earthquake’s epicenter that coincided with a decrease in animal activity prior to the earthquake.

TEAM is currently discussing potential opportunities to further collaborate with Grant and expand on her research, including the possibility of adding magnetometers and ionization detectors to several national parks in which TEAM works. This research could help create a nature-based early warning system for earthquakes, with the potential to save thousands of lives and improve the livelihood of local communities.

Despite this big leap forward in our understanding of how natural disasters like earthquakes influence animal behavior, there is still so much we don’t know yet — not just about earthquakes, but the natural world and our connections to it. Fortunately, programs like TEAM are filling in the gaps.

Jimmy MacCarthy is the technical and communications manager for the TEAM Network.