NRL Oceanographers Develop Optical Water Mass Classification System for Coastal Waters


3/30/2003 - 30-03r
Contact: Public Affairs Office, (202) 767-2541


Related Visuals

Oceanographers at the Naval Research Laboratory field site at the Stennis Space Center in Mississippi have developed new algorithms and tools to characterize the optical constituents of coastal waters from satellites. Furthermore, using these new satellite products, they have developed an optical water mass classification system to identify and track coastal features. The classification system can be automatically applied to satellite imagery received at NRL, and will facilitate estimates of the transport of dissolved and particulate material in the water.

Sunlight strikes the ocean surface, penetrates into the water column, interacts with the dissolved and particulate material in the water, and is then reflected back out. Coastal areas typically have higher concentrations of these materials than open-ocean regions, causing coastal waters to appear greenish and open-ocean regions to appear blue. The quantity and quality of the light reflected from the ocean varies over time and space and represents the "ocean color signature" of the water, which can be measured by satellite sensors. Ocean optical properties provide the link between the "color of water" and the "composition" of the water. By unraveling the color signature, a wealth of information can be obtained about the individual components that contribute to the signal, including the concentrations of the dissolved and particulate matter in the water.

According to Dr. Rick Gould, Acting Head of the Ocean Optics Section (Code 7333), ocean color can be measured from space using satellites so that the researchers can synoptically map spatial distributions of bio-optically active components at ecologically-relevant time scales. "For the first time, using new satellite bio-optical algorithms that we have developed to divide the dissolved and particulate matter into organic and inorganic components, we are unraveling the complicated spectral signature and gaining valuable insight into a variety of physical and biogeochemical processes occurring in coastal and shelf regions," said Dr. Gould.

Starting with the CZCS (Coastal Zone Color Scanner), and more recently with SeaWiFS (Sea-viewing Wide Field-of-view Sensor), MODIS (Moderate Resolution Imaging Spectroradiometer), and MERIS (MEdium Resolution Imaging Spectrometer), ocean color satellites have provided synoptic estimates of bio-optical parameters such as chlorophyll for nearly 25 years. It is possible to estimate water composition from space because the water-leaving radiances at visible wavelengths measured by the satellite sensors are related to the particulate and dissolved substances in the water through the radiative transfer equations. In other words, light in the water is affected by both absorption and scattering processes, and these processes are tightly coupled to the type and amount of the materials in the water (materials such as phytoplankton pigments, suspended sediments, and dissolved organic matter).

These relationships enable researchers to estimate the inherent optical properties (IOPs) of the water, such as the absorption and backscattering coefficients, which affect phytoplankton primary production, biomass, heat flux, convective mixing, and naval applications (laser mine detection systems, diver visibility). Furthermore, the optical properties vary over short spatial and temporal scales in the coastal environments. So, if the satellite remote sensing estimates of the IOPs are available at hourly and daily time intervals, and they can be followed for periods of weeks, then the satellite estimates can be used to optically track and classify water masses at similar time scales to the processes controlling the distribution patterns (phytoplankton growth, advection, particle settling and resuspension). Current ocean color satellites provide two or three "looks" per day at a given area, so the temporal resolution is adequate to resolve some of these processes, but not processes that operate at hourly time scales, such as tides. Also, current ocean color satellites generally provide data in just a few spectral bands and at 1 km spatial resolution, so processes and patterns at finer spatial scales will not be resolved. New sensors are planned with both higher spectral and spatial resolution. New algorithms will also be required to exploit these improved capabilities and provide new products in coastal areas.

"The concentration of total suspended solids (TSS), and its partitioning into particulate organic and inorganic matter (POM, PIM) is of interest from both remote sensing and modeling aspects," stated Dr. Gould. "The concentration and space/time distribution of the inorganic component (including silts, clays, sand, and phytoplankton debris) can be used to trace river plumes and fronts, and can indicate regions of particle resuspension from wave and storm-induced turbulence."

The distribution of the organic component (including living phytoplankton, zooplankton, and their decay products) impacts the development of anoxic "dead zones," and is required for carbon flux estimates. (Carbon is the currency used in the exchange processes between the biosphere, hydrosphere, atmosphere, and lithosphere, so its distribution has far-reaching effects on global warming and ocean circulation.)

New algorithms to estimate the concentrations of PIM, POM, and TSS have been developed by NRL and applied to SeaWiFS satellite ocean color imagery. These new satellite products can be used to optically characterize and trace water masses. For example, if two sequential images of the TSS product are taken and a "difference" image is taken (i.e., at each pixel in the image, subtract the value on day one from the value on day two), the river plume advection can be followed and the fate of the associated effluent can be traced. (Figure 1A). Furthermore, if images of the ratio of the PIM and POM products are created, and another difference image is created for the same two days, the relative changes in the particle composition can be assessed (Figure 1B). An increase in TSS between two days could be due to an increase in the organic component of the particulate load, the inorganic component, or both.

"The PIM/POM difference image helps us determine which case occurred, and even helps us distinguish between competing physical and biological processes," Dr. Gould said (for example, an increase in the inorganic component indicates wave resuspension of bottom sediments or river discharge, whereas an increase in the organic component indicates phytoplankton growth).

Historically, oceanographers have used physical properties of the water, such as temperature and salinity, to track and classify water masses. Optical classification schemes have also been proposed, but recent advances in optical instrumentation in the past decade permit new measurements to be made on a routine basis with unprecedented accuracy. Coupling these new in situ optical measurements with advances in bio-optical algorithms and remote sensing capabilities has allowed us to explore new optical water mass classification systems. Additional work remains to determine how closely the optical water mass delineations correspond to the traditional physical ones, and if there are differences, why.

The total absorption coefficient can be partitioned into individual absorption components due to phytoplankton (aj), detritus (ad), and colored dissolved organic matter (aCDOM). New algorithms were developed to estimate each of these components from the satellite ocean color imagery to facilitate optical water mass classification. A combined image of these three parameters was created to help visualize the spatial distribution of the components. First, the three coefficients are summed together at each pixel in an image, and the percentage of the total due to each component is calculated. In Figure 2A, the red pixels indicate areas of relatively high detrital and CDOM absorption, the blue pixels correspond to relatively high CDOM absorption and lower phytoplankton and detrital absorption, and the green pixels represent areas of relatively high phytoplankton and CDOM absorption and low detrital absorption. For a more quantitative characterization of the optical characteristics of the water masses, ternary diagrams were formed (Figure 2B) to classify each image pixel into one of 16 classes based on the percentages of each of the absorption components. The temporal and spatial variability of the water masses in a region can then be classified and traced by performing these analyses on multiple scenes over time.

This research has led to the development of new algorithms to assess water optical properties from space, and to classify water masses based on these optical properties. For the first time, researchers have the capability to monitor geochemical and optical processes and the impact of human activity on the coastal zone. The development of this capability to remotely estimate both the concentrations and the optical characteristics of the organic and inorganic constituents of the water, coupled with the new ocean color sensors coming online, helps trace and classify water masses, monitor river discharge, circulation patterns, sediment resuspension, phytoplankton growth, and the carbon cycle.



Get NRL News: RSS


About the U.S. Naval Research Laboratory

The U.S. Naval Research Laboratory is the Navy's full-spectrum corporate laboratory, conducting a broadly based multidisciplinary program of scientific research and advanced technological development. The Laboratory, with a total complement of nearly 2,800 personnel, is located in southwest Washington, D.C., with other major sites at the Stennis Space Center, Miss., and Monterey, Calif. NRL has served the Navy and the nation for over 90 years and continues to meet the complex technological challenges of today's world. For more information, visit the NRL homepage or join the conversation on Twitter, Facebook, and YouTube.

Comment policy: We hope to receive submissions from all viewpoints, but we ask that all participants agree to the Department of Defense Social Media User Agreement. All comments are reviewed before being posted.