Tropical forests under surveillance

How can new technologies help to reveal the hidden lives of tropical animals and plants?

Difficult environmental conditions and complex biological interactions make it tough for tropical biologists to understand animal behavior, climate change effects and highly biodiverse plants and forest organisms. On Monday, July 29, 2002, the Smithsonian Tropical Research Institute (STRI) convened a diverse group of rainforest biologists with engineers at the forefront of sensor and communications technologies to brainstorm new ways to apply technology in tropical research and conservation efforts.

According to the symposium organizer, Julio Escobar, President of Centauri Technologies Corporation, field biologists may be unaware of the latest technological innovations as they plan research strategies, or may be unable to afford novel equipment. Furthermore, new technologies designed for commercial purposes may not easily lend themselves to biological applications.

Advanced sensor technologies in concert with modern communications platforms will be vastly more useful to researchers if they are intentionally designed to monitor an incredibly complex tropical environment where a huge variety of organisms interact.

Elusive ocelots, bats flying through underbrush and male beetles who secretly manipulate their mates were the study subjects mentioned as biologists came up with a “wish list” of questions to be answered if the appropriate technology were available. Roland Kays, Curator of Mammals at the New York State Museum commented that the first challenge to an understanding of animal behavior in the tropics is simply the ability to find the same animal several times in order to make multiple observations. With Martin Wikelski, Assistant Professor of Zoology at Princeton University, Kays will implement an automated telemetry system on Barro Colorado Island to locate animals via an array of radio towers designed to replace chasing individually radio-collared animals through the forest with a hand-held antenna with the ability to locate an animal on a palm-sized computer, or to be paged when an animal leaves the reserve(http://www.princeton.edu/~wikelski/research).

High-quality sound recordings, automated analyses of sound sequences and radiotransmitters with ultrasound sensors were on the wish list of Elisabeth Kalko, Staff Scientist at STRI and the University of Ulm who studies bats: nocturnal, highly mobile inhabitants of inaccessible, obstacle-rich habitats. Bats play an extremely important role in seed dispersal, pollination and insect predation in tropical forests and an understanding of their community structure and behavior is essential to conservation efforts.

In order to survive in nature animals must reproduce, but their secret sex lives may be nearly impossible for biologists to observe. William Eberhard, from STRI and the University of Costa Rica, currently relies on a microscope and a notebook for most of his research, but dreams of tiny surveillance cameras, chemical sensors and x-ray like technology to enable him to watch the intimate interactions of spiders, beetles, birds and bees.

Conservation of the Amazon basin, where 40% of the world’s remaining rainforest is under immediate threat of destruction, has benefitted from initial efforts to monitor land use via remote sensing devices and satellite technology.

However, the current challenge, according to William Laurance, STRI staff and Biological Dynamics of Forest Fragments Project, is to link information available at very large scales with local land-use information. Current technologies still cannot accurately discriminate small clearings caused by logging in Brazil where the government estimates that over 80% of timber harvest operations are illegal.

What technologies could be immediately put into place to address urgent biological questions and what will be coming down the line? Deborah Estrin, Professor of Computer Science at UCLA and Director of the new NSF-funded Center for Embedded Networked Sensing designs systems to observe phenomena that were previously unobservable(http://www.cs.ucla.edu/csd/CENS.html). She discussed a system at the James Reserve, of tiny environmental sensors in place which continuously monitor micro-climatic factors. Communication of data consumes the majority of the energy needed to run these systems, so systems designed to eliminate unnecessary information don’t waste energy communicating these to users. Minimizing the weight of the power source to run a sensor is critical. Monitoring small-bodied animals like birds, bats and insects is limited by the animal’s ability to carry even the lightest batteries.

Vertically oriented radar automatically monitors the density and distribution of insects flying at altitude and has been used by Joseph Riley, who headed the Natural Resources Institute’s Radar Unit, to monitor insect migration and behavior in the British Isles.

While cluttered tropical environments may be less amenable to the use of traditional radar, new transponders allow radar to detect signals reflected off of small targets flying in front of a cluttered background like the tropical forest understory.

Peter Mansfield, of Mansfield Technical Consulting explained the ins and outs of directional sensors and range sensors (GPS, RADAR, LIDAR) as well as proximity detectors, accelerometers and micro electronic mechanical systems (MEMS), their limitations and their capabilities for use in biological applications.

Charles Perkins, Nokia Corporation, commented on the use of Ad Hoc Networks as platforms to provide biologists with immediate access to visual and audio information about their research subjects. “Consider how much information we could gain from a set of cell phones mounted on trees, which automatically call in relevant data to researchers as it becomes available, ” commented Frank Levinson, Finisar Corporation founder.

Micromachining technologies have tremendously reduced the size and cost of sensors combining small mechanical components with electronic components, and making it much more feasible and cost-effective to monitor wilderness habitats.

Jack Judy, Assistant Professor of Electrical Engineering at the University of California, Los Angeles discussed the development of a variety of micro-electromechanical systems (MEMS), including electrical noses and other chemical sensors to monitor evironmental contamination.

Finally, Dennis Shaw, Chief Information Officer at the Smithsonian Institution, discussed the current capability of telecommunications systems at the Smithsonian Tropical Research Institute to support scientific research. The Smithsonian Institution has placed new emphasis on updating computing and telephone systems to connect remote field sites and facilitate access to the most recent data generated by staff scientists and guest researchers. STRI us committed to pushing the development of new technologies for improving understanding and conservation of tropical organisms and communities.

Workshop participants continued their informal discussions the following day on Barro Colorado Island, STRI’s original field station in Panama and “the best studied piece of tropical real estate in the New World,” where they experienced tropical forest conditions first hand.