Scientists use aerial imaging to find hidden invaders in Hawaiian rain forest

Courtesy of Gregory Asner Invade

The middle and top images reveal canopy water content and leaf nitrogen concentration from high-altitude airborne imaging spectroscopy.

Peter Vitousek Vitousek

Scientists from Stanford and the Carnegie Institution have developed a remote-sensing technique that identifies invasive plant species before they dominate a landscape. Writing in the March online edition of the Proceedings of the National Academy of Sciences (PNAS), the researchers describe how they used novel measurement techniques from a high-altitude aircraft to detect two species of invading plants that are changing the ecology of the rainforest near Kilauea Volcano in Hawaii Volcanoes National Park.

"This is the first time where remote sensing showed me something new concerning how an ecosystem works," said Peter Vitousek, the Clifford G. Morrison Professor in Population and Resource Studies at Stanford and co-author of the PNAS study. "Up to this point, remote sensing has been invaluable for understanding how features or processes that have been observed in one or a few places are distributed in space and time. These new methods discovered a consequence of biological invasion that had not been detected before, and showed how it varies across the landscape."

The new technology is exciting because it can detect the impact of biological invasions on entire ecosystems, say the authors. It may prove especially useful surveying islands such as Hawaii, where introduced species can wreak havoc very quickly.

Invasive chemistry

The PNAS study focused on two tree species found on Kilauea Volcano—the native 'ohi'a (Metrosideros polymorpha), which has a low concentration of nitrogen in its leaves (0.6-0.8 percent), and the exotic Myrica faya from the Canary Islands, which has relatively high nitrogen concentration (1.5-1.8 percent).

The researchers used the recently upgraded NASA Airborne Visible and Infrared Imaging Spectrometer (AVIRIS) to measure leaf nitrogen and water content from the aircraft, and corroborated the data on the ground.

"We found chemical fingerprints from the plant leaves and used them to tell which species dominated specific areas," said lead author Gregory Asner, a member of the Carnegie Institution's Department of Global Ecology faculty and an assistant professor, by courtesy, of geological and environmental sciences at Stanford.

"The fingerprints showed where the native dominant 'ohi'a tree has been taken over by the invading Myrica, and more importantly identified areas where Myrica invasion is in its early stages," he said. "The aircraft imagery also showed us how the forest canopy chemistry is changing as a result of the invader."

The high leaf nitrogen associated with the invading tree means that it is basically fertilizing the forest with more nitrogen, Asner noted: "The leaves turn over faster, and there is more nitrogen in the soil. However, the invader shades out nearly all other species, so this excess nitrogen is not available to other species. Although we don't know exactly what the domino effects of this invasion will be, we are in a good position to predict them as we learn more about the chemical changes the forest is undergoing."

Invasive ginger

Aerial surveys revealed another invader, the Kahili ginger plant (Hedychium gardnerianum), growing under the forest canopy. Ginger cannot be detected from above the canopy using traditional aircraft or satellite approaches, but AVIRIS is sensitive to the high water content of the ginger plant. Aircraft-based analysis also found that ginger reduces the amount of nitrogen in the Metrosideros forest canopy—a finding later corroborated by ground-based sampling.

"This is the first time in my experience that remote sensing has detected an understory species, Kahili ginger, one of the most disruptive weeds in Hawaiian rain forests," said Tim Tunison, chief of resource management at Hawaii Volcanoes National Park. "We need to understand the ecological effects of invasions over the landscape to develop effective control strategies, and the Asner/Vitousek work gives us valuable insights about this problem. On a more practical level, we need to know the distribution of invasives. Weeds are often difficult to find in dense, wet forest in Hawaii. This study has helped us with a particularly difficult-to-map species with confusing signatures, Myrica faya."

Funding for the PNAS study was provided by the National Science Foundation, the NASA New Investigator Program, the Mellon Foundation and the Carnegie Institution. Asner, Vitousek and their colleagues from Stanford and Carnegie have begun working with the National Park Service, NASA and the Nature Conservancy on a new project to map the chemical and structural composition of Hawaiian ecosystems and to find exotic species and track their ecological impacts.