Measuring aquatic vegetation goes high-tech
Using people to manually characterize aquatic vegetation in the coastal environment may no longer be the best method of capturing features in a marine environment.
Scientists at PNNL's Marine Sciences Laboratory are combining satellite remote sensing technology and computer modeling to create better and faster analysis of vegetation coverage. "With one satellite image, you may be able to characterize an entire area by applying an algorithm to that image and converting it to a vegetation map," said Lyle Hibler, MSL computer modeling expert. "Two scenes, taken at different times, allow an analyst to observe changes over time, or trend analysis."
Characterizing submerged aquatic vegetation in the near shore environment is important for coastal resource managers who make decisions regarding ferry terminal dredging, water traffic and other issues. "Ferry terminal construction or maintenance can be detrimental to submerged vegetation, which is important habitat for fish and other marine life," Hibler said.
Although this combination of technologies has been used successfully to characterize floating vegetation offshore and to estimate the sea's surface temperature, using it to characterize vegetation in shallow water is a developing field. "We wanted to map submerged vegetation using the chlorophyll response signal. However, an abundance of free-floating material containing chlorophyll, such as dead plants and plankton, which are common in shallower water, can confound attempts to map the rooted vegetation," Hibler said.
Near shore characterization also has several coastal security applications. The Marine Sciences Laboratory has worked on research and development projects supported by the National Imagery and Mapping Agency (NIMA), a national intelligence service, to better characterize the ever-changing features in the coastal zone.
Another coastal security application involves using remote sensing to detect algal bloom die-off, which could signal a toxic material in the water. A vegetation map also could help the military infer the water depth and vegetation characteristics of a coastal environment.
MSL researchers used satellite imagery from the U.S. Department of Energy's Multispectral Thermal Imager. MTI is a sensor with 15 spectral bands, ranging from visible to long-range infrared. It is used to derive a broad range of information on facilities, activities and characteristics, including surface temperature, materials, water quality and vegetation health. By covering several discrete wavelengths of light, MTI can provide a more continuous representation of the light spectrum compared to most other satellite sensors, and ultimately, finer image detail. The MTI project, sponsored by DOE's Office of Nonproliferation and Research Engineering, demonstrates advanced multispectral and thermal imaging, image processing and related technologies.
"Further studies will need to be conducted to integrate other sensors and observational data sets, in addition to the information from MTI, into full model applications," Hibler said. "A Willapa Bay, Wash., study has shown that additional coupling of remote sensing and computer modeling can provide a clearer understanding of some of the processes and water quality issues that affect marine habitat, such as light attenuation, impact of tides, salinity and turbidity."