Coral Paparazzi—Looking for a Quick Way to Spot Evidence of Coral Disease


		Search all stories:

	Home \|\| Sections: Fieldwork \| Research \| Outreach \| Meetings \| Awards \| Staff & Center News \| Publications \|\| Archives

Coral Paparazzi—Looking for a Quick Way to Spot Evidence of Coral Disease

By Christina A. Kellogg and David G. Zawada
Oct. / Nov. 2010

in this issue:
previous story \| next story

Lights, camera, action! That's surely what it felt like to be a diseased coral in the Florida Keys during the first two weeks of August 2010. Using both an underwater, diver-operated spectrometer and an imaging spectrometer, U.S. Geological Survey (USGS) scientists collected data to determine whether certain light wavelengths or spectral signatures can be used as diagnostic indicators of coral disease. The spectrometer measures the distribution of light energy over a range of wavelengths—in this case, just visible light reflected off the coral. The imaging spectrometer collects a series of images that correspond to specific wavelengths across the electromagnetic spectrum—spanning visible to near-infrared light in this study—and combines them in an "image data cube," which can be thought of as a deck of cards in which each card has a unique color.

Above: Great star coral (Montastraea cavernosa) affected by black-band disease. Coral tissue is light brown, black line is disease lesion, and white area is dead coral skeleton. Area circled in red is the part of the coral collected for imaging spectroscopy, histology, and microbiology sampling. Inset shows the detail of the disease interface captured by the imaging spectrometer. [larger version]

Mitigation of impacts from coral diseases and syndromes requires better diagnostic techniques to detect the onset of disease; and, ideally, these techniques would be rapid and non-destructive. The ultimate goal is to develop an imaging system that could remotely sense disease-induced stress on the coral reef while being towed behind a boat, allowing much greater coverage in a day than current diver surveys. To this end, Chris Kellogg and Dave Zawada of the USGS St. Petersburg Coastal and Marine Science Center in St. Petersburg, Florida, collected diseased corals and small healthy corals (as controls) to scan with these various instruments. To ground-truth any optical signatures that might be discovered, they also sampled disease lesions for laboratory studies of cell and tissue anatomy (histology) and examination of microbes living on the coral. Analyses of this unique dataset will be carried out over the next year. This research was conducted under National Marine Sanctuary Permit FKNMS-2010-084.

Above left: Small, healthy colony of great star coral (Montastraea cavernosa). Inset shows detail of several polyps captured by the imaging spectrometer. Black squares on the ruler are 1 cm wide; black bars are 1 mm wide. [larger version]

Above right: The imaging spectrometer captures a series of 640 images, each one corresponding to a different wavelength. The stacked images form an "image data cube," much like a deck of cards in which each card has a unique color. As a result, every pixel in the image has its own spectrum. The example spectra on the right correspond to the points marked by colored dots in the image. In each spectrum, the horizontal axis is light wavelength, and the vertical axis is reflectance. Note that the spectrum for a point in the diseased, "black band" region (red dot) differs from the spectrum for a point in healthy coral tissue (blue dot). Observing spectral changes while moving from healthy to diseased to dead areas of the coral is a key aspect of this study. [larger version]